User talk:Dallan/Archive 2009

Topics


Firefox Glitch??? [14 April 2009]

Not sure if this is a problem with only my computer but...
There are times when the blue menu bar will display alternate characters along with the WeRelate, Search, Add... text. So, instead of WeRelate, I see &lt;werelate&gt; and instead of Add, I see &lt;add&gt; (NOTE: I don't see < or >, I edited the code in this note to reflect what I actually see on the screen.) Their respective pull-down menus will also feature these extra characters.

The line below also has issues, showing Place, Discussion, Edit, History, Move, Unwatch, &lt;treeadd&gt;, More instead of Place, Talk, Edit, History, Rename, Unwatch, Tree +, More. FYI, I'm using Firefox 3.0.5.--Jennifer (JBS66) 18:38, 1 January 2009 (EST)

I see this every once in awhile too. It's not a firefox glitch. To make it possible to convert the interface to different languages the menu words are read from the database to construct the page. What you're seeing is the "raw" form of the menu words, as if the database lookup failed. I'm not sure why this happens. When it happens to me, it goes away if I refresh the page. I'm hoping that the bug will go away when I update to the latest version of MediaWiki in a couple of months.--Dallan 19:51, 1 January 2009 (EST)

Thanks for your response. For me, refresh doesn't work. I need to shut Firefox down and then open it again. And I thought it was getting cranky because I had so many tabs open at once :~) At least it wasn't just me.--Jennifer (JBS66) 19:58, 1 January 2009 (EST)


Have you tried holding down the "shift" key while clicking on the refresh button? That should tell FF to re-fetch the page instead of using its cached version.--Dallan 20:44, 1 January 2009 (EST)


WEbadministrator. Just trying to review my GEDCOM and when I get to Family Matches I get a TIP popup that I cannot get rid of. It follows cursor everywhere until I go to help or get completely out of the overview. It needs to have an exit capability after it is used.--Allensharon 17:50, 14 April 2009 (EDT)


Sorry about that. There should be an "ok" button at the bottom-right of the dialog, and a "Don't show this again" checkbox on the bottom left. Maybe the entire dialog isn't showing? (You might want to lower the separator-bar to make the top pane larger.) I'll add a X close-button to the upper-right corner later this week.--Dallan 19:37, 14 April 2009 (EDT)


Category oops [2 January 2009]

Hi Dallan,

Ronni did all this great work in doing all the county census pages for 1850 Missouri here. But I goofed in writing the help, and said that the category should be with a capital letter instead of lower case --"1850 Missouri Census" instead of "1850 Missouri census". So every other state, including Missouri, has a page with a lower case "census" (see here), and then there's this one with a capital letter. I tried redirecting, but it doesn't work. Can this be fixed without editing every page again?--Amelia 20:51, 2 January 2009 (EST)

I'll have my son (User:Taylor) take care of it.--Dallan 11:39, 3 January 2009 (EST)
This has been fixed.--Dallan 23:14, 3 January 2009 (EST)

maps and books [3 January 2009]

IS there a place for maps and books?--Mawgli 12:23, 3 January 2009 (EST)


Are you asking if we have maps and books available here? The answer to that question is generally "very few". But wiki pages in the Source namespace tell you where you can find maps on other websites or in microfilm. Select "WeRelate" from the "Search" menu, then click on the "Source" link at the left, and then click on the "Map/Gazetteers" link and you'll see a bunch of them. If you have some maps that you'd like to share, that are in the public domain due to their age, please feel free to upload them as images, create Source pages for them, and include links to the images on the Source pages.--Dallan 12:41, 3 January 2009 (EST)


Using ref tags for footnotes. [3 January 2009]

I have a question about reference tags.
I'm going through the Friesland place pages (needed a break from Québec for a while). I'm sourcing some of my findings with a <ref> tag. I'm noticing that sometimes, a page will show footnotes that I did not tag on the page see Place:Surhuisterveen, Achtkarspelen, Friesland, Netherlands. The only footnote that should be there is #5, the others are all ones that I used in previous pages. If I go back in to edit the page, then save again, the footnotes are then correct. Any idea why this is happening?--Jennifer (JBS66) 16:32, 3 January 2009 (EST)


I'm going to have to think about this one. I'm not sure how the references tag is picking up ref's from previous pages. I'll look into it next week.--Dallan 17:22, 3 January 2009 (EST)


Merge solution - DFree [5 January 2009]

Hello Dallan,

Thank You so much for your detailed instructions. I will give it a try tomorrow.

I cleaned up (deleted) the obvious Mysources into Title, then I hope to change them into Sources instead.

Census as example.

I have one question about that though.

I have a habit of adding notes to the Mysources. Person born in IL, parents born in NY as example. Right now I am changing the Mysources to Title only. When I add them to Sources will I not add my notes to the Shared Source such as the 1850 census? Do you have a suggestion so I can keep the note, but not screw up the Shared Source?

Thanks DFree------DFree 01:09, 4 January 2009 (EST)


Right - it wouldn't make sense to add your notes to the shared source. One possibility would be to add your notes to the text of the source citation on the Person or Family page. That's what I've been thinking that people would do. Would that work?

Alternatively, I've been thinking about this problem (coincidentally), and another possibility could be to allow you to link a MySource page to a Source page for the bibliographic details, in which case you could link your MySource page to the Source page for the 1850 census, keep your notes on the MySource page, and continue to link your Person and Family pages to the MySource page.--Dallan 11:33, 5 January 2009 (EST)


Hello Dallan,

As you know I am a low tech user. I think I know what you mean by adding the text of the source citation on the person page. That will work and not add to the Werelate memory. It is something that I can do this year as Werelate grows, that will not be hard to adapt to the tech changes too. I will make a Source Note section on the Person Page text box. That should work. If other users want to change it, or as Werelate changes it is easily changed.

How does that sound?

DFree--DFree 12:34, 5 January 2009 (EST)

Example

Person Page: William Forth (#)

Source Notes for William Forth:

1850 census - this wife is probably Mary Warren.


If that works, it will keep things simple. Part of each source citation on a Person or Family page is a big text area labeled "Text / Transcription location" where you can enter whatever information you want. You could add your notes there, or you could add them to the big Person Page text box. I'd probably add them to the "Text / Transcription location" box that's part of the source citation.--Dallan 15:10, 5 January 2009 (EST)


Hello Dallan,

Thank You for your suggestion. I figured out what you meant. It works great. I can add my notes, and use the Werelate Source page.

Have a good day, Debbie Freeman --DFree 16:02, 5 January 2009 (EST)


US Population Census - Source Pages [6 January 2009]

Hello Dallan,

Sorry to bother you again. I have two questions you could probably pass on if needed.

I located the US Pop Census Source pages. The 1850 US Pop Census Source page is great. There does not seem to be one for the 1820 US Pop Census, and the 1930 Source one is under "Category: 1930 US Census" with very little information.

There are so many pages already existing under the subject of census I wonder if I should create a 1820 Source page similar to the 1850 one? If I could figure it out that is.

I could not figure out to add NARA as a Repository for the 1930 census. It looked like there was a link to the FHL catalog for some of the rolls. How do you do add a Repository ?

Thank You, Debbie Freeman --DFree 17:10, 5 January 2009 (EST)


Please go ahead and create a Source page for the 1820 census similar to the 1850 one. You can create a Source page or a Repository page by clicking on "Add" in the blue menu bar, then on Source or Repository. It looks like there is already a repository for the National Archives: Repository:The National Archives (United States). I just added a "redirect" so that Repository:NARA points to this repository.--Dallan 18:11, 5 January 2009 (EST)


Hope you don't mind if I jump in here...
WeRelate does have a page for the 1820 census, it's at Source:United States. 1820 U.S. Census Population Schedule. All of the U.S. censuses have similar titles, with only the dates changing. I saw that you created one with a slightly different title, I think that will need to be redirected to the page I indicated above.--Jennifer (JBS66) 18:34, 5 January 2009 (EST)


Hello Jennifer,

Thank You for chipping in. I do appreciate it. There are so many US census pages it is hard to see the one tree through the forest. I unfortunately found the 1820 page too late. Now I can't delete the 1820 source page I created.

I will go back to lurking, and clean up my People pages now that I have the correct census pages "watched". I do try not to create too much of a mess. Oh well. Debbie Freeman --DFree 18:48, 5 January 2009 (EST)


Oh - PLEASE don't go back to lurking!!! Mistakes around here really aren't a big deal - they can easily be fixed - no mess created. Your contributions and efforts are much more important.

As for the source you created, Source:United States. 1820 U.S. Population Census Schedule, you can edit the page, go to the big text box, type #REDIRECT [[Source:United States. 1820 U.S. Census Population Schedule]] into it, and then save. This will redirect your source page to the existing one. We have a lot of these redirect pages around here - no harm done.--Jennifer (JBS66) 19:01, 5 January 2009 (EST)


Hello Dallan & Jennifer,

I tried a redirect. I think it worked. Thank You for the help. I wish it was easier to locate the right source pages. Do you have a trick to make it easier? Debbie Freeman --DFree 19:18, 5 January 2009 (EST)


I agree with Jennifer -- mistakes are very easy to correct in a wiki, so don't worry about making them! Everyone has made them, especially starting out.

As for finding an existing census source, two things:

  • First, when you add a source citation on a Person/Family page, choose "Source" for the source namespace, then start typing "United States. 1820" in the title field and wait a few seconds. You should see a drop-down list of the sources that start with the words you've entered.
  • If that fails, you could select WeRelate from the Search menu, change the namespace to "Source", check "exact matches only", and enter words in the source title.

--Dallan 10:06, 6 January 2009 (EST)


BLM Werelate pages? [7 January 2009]

Hello Dallan,

I have a question to ask you or others if they want to chip in.

This is something I do not have the computer skills to figure out.

There is a Werelate Repository page for the BLM. There is also two Source pages for the BLM (one has a comma, one does not) Example - United States. Bureau of Land Management, General Land Office Records. There is a good Werelate page (?article) called "Getting more from online land records" about the BLM.

I also recall some times ago there was a highlighted page on this topic too.

Would this be something that should be linked, or merged?

It would seem to be a good candidate.

If so I could change my BLM Mysource pages and use that as my source.

Opinions?

Two small notes.

I did not see on any of these pages any mention that NARA in Washington DC has the (paper version) land files, except for the cash sale files. Those are still in the BLM.

If I remember right a University in Nebraska (can't remember the name, I have the link though) has a good web page where you can input in the legal land description for the western states into their form and it will tell you the longitude and latitude for that BLM land, and other interesting things too.

Thank You, Debbie Freeman --DFree 14:54, 7 January 2009 (EST)


Incorrect child [9 January 2009]

Dallan, Which child did you eliminate from Samuel and Clarissa Hancock Alger - and why? Please write to me at (address removed)

Thanks--Dixieboy32 21:06, 8 January 2009 (EST)


In a previous edit it looks like the husband Samuel Alger had been inadvertently added also as a child of the family. I noticed this today and removed Samuel as a child of this family. I kept him as a husband. And Samuel H. Alger born 1826 is still listed as a child.

By the way, it looks like you and User:Kristyspillane are related -- you both have a page for Samuel Alger in your tree: Person:Samuel Alger (1) and Person:Samuel Alger (7). You can combine both of these pages into a single page for Samuel by navigating to one of the pages for Samuel, clicking on "Find duplicates" in the "More" menu in the upper-right corner of the page, checking the box next to the other page for Samuel and clicking the "Compare" button, then following the instructions for merging the pages. We're doing a big push lately to combine multiple pages for the same individuals.--Dallan 22:56, 8 January 2009 (EST)


Interactive genie program [14 January 2009]

Hi Dallan,

Thanks so much for WeRelate; I am excited about the upcoming addition of gedcom export. Are you familiar with the genealogy program Family Historian? Here is a link to download a free trial: [1].

The US price is rather high because of the exchange rate. The most interesting features of this program to me are the charts and the fact that the file format is the gedcom standard. Rather than have another genie program with yet another file extension; I thought the concept of using the gedcom extension for the WeRelate interactive genie program has possibilities.

Roots Magic 4 is in the community preview stage; after that follows public beta testing and then release. One of the most recent disclosures about this new program is this recent posting on the Roots Magic blog;

And now for the final sneak preview

One of the most common comments weve seen (after When is this going to be released?) has to do with the size and speed of the new version. Weve all seen (and probably bought) programs that claimed to be the next great thing, only to find that it was bloated and slow because it was based on a framework like .NET, or based on some huge database engine that had to be installed on the computer before the program could run. (Note to .NET programmers we know there is a place for .NET and large database engines it just isnt in genealogy software.)

Well, we dont roll that way here at RootsMagic. RM4 is a native Windows program (no .NET bloat), and uses an embedded database engine (no huge database engine required to be installed on your computer).

So what does this programming jargon mean to you? It means we are able to add one extra feature that other bloated programs will never be able to copy RootsMagic-To-Go.

Thats right. Not only can you have your data on a flash drive, but you can run the full RootsMagic program and your data directly from the flash drive.

Are you going to a library that doesnt have RootsMagic on their computers? Dont worry. Just use RootsMagic-To-Go to install RM4 and your data on your flash drive, and take it with you. When you get to the library just plug your flash drive into the computer and get to work. When you get home just move the data right back onto your computer.


Have a great day.--Beth 09:48, 9 January 2009 (EST)


Thanks for the tip on family-historian. I'll have to check it out. I hadn't heard about RM-to-go; that's interesting. Although it may be just as easy for people to update WeRelate when they're at the library and then have their changes automatically copy to their local computer when they get home.--Dallan 18:12, 14 January 2009 (EST)


Source:Castle Garden - Double checking [14 January 2009]

Hello Dallan,

I wanted to double check before I did this in case I am wrong.

I have found a Werelate Source called Source:Castle Garden. It has the web site name (org), and is listed as the Repository. I searched for a person I knew, and he came up on the website with the same information I have from the offline NARA source. So I know the NARA off line source. So it looks like they are using this source for the search able index. Would you like to keep the source page as Castle Garden? Should I create a new Source page called "Passenger Lists of Vessels Arriving at New York, New York, 1820-1897"? Should I then Redirect it to the new page?

Suggestions?

Thank You, Debbie Freeman --DFree 15:37, 10 January 2009 (EST)


Hello Dallan,

I answered one of my questions. The NARA Werelate source page exits. I just found it.

Debbie Freeman --DFree 15:42, 10 January 2009 (EST)


Hi Debbie, I'm not sure I quite understand the question, but you could create a "Source:United States, New York, New York. Passenger Lists of Vessels Arriving at New York, New York, 1820-1897" if you wanted. If all of the records at Castle Garden are from this one NARA offline source, you could redirect the Castle Garden source to this source.--Dallan 18:12, 14 January 2009 (EST)


Country Abbreviations [14 January 2009]

I was wondering if there would be negative side-effects of adding the 2/3-letter abbreviations for countries see WorldAtlas. For the Netherlands, my GEDCOM contained NL instead of Netherlands, and the matcher didn't recognize that. I can add NL as an alternate name on Place:Netherlands if that's ok.--Jennifer (JBS66) 14:21, 11 January 2009 (EST)


There is a recently uploaded GEDCOM that contains many references to places in the Netherlands. They are using NL for Netherlands and a 2-letter code for the province (like ND for Noord-Holland). I also noticed where a couple of places were being matched to towns in Place:Newfoundland and Labrador, Canada due to that abbreviation being NL.
Would it cause a problem if I added NL to the Netherlands and added the various abbreviations to the provinces? Thank You!--Jennifer (JBS66) 07:14, 13 January 2009 (EST)


Hi Jennifer,

I guess that you're refering to my GEDCOM's here, right? I just removed them again, and uploaded a new one that should be more stable than the previous ones. I'm doing that while trying to find a way to upload a relevant excerpt of my database, which has slightly more than 20 000 persons, and has lots of duplicates in remote areas too.

I have no idea what Dallan thinks about these abbreviations, but here's a short list of provinces for your reference:

GR Groningen
FR Friesland
DR Drenthe
OV Overijssel
GE Gelderland
NH Noord Holland
ZH Zuid Holland
UT Utrecht
ZE Zeeland
NB Noord Brabant
LI Limburg
FL Flevoland

I listed them in the order that I learnt them in, many years ago. The last one, Flevoland, didn't exist by then. And I apologize for not using a proper table syntax here. :-) --Enno 16:53, 13 January 2009 (EST)


Thank you for the list of abbreviations. I would also add a few additions that I have seen used:

  • GL Gelderland
  • ZL Zeeland
  • LB Limburg

Interestingly Enno, I was not referring to your GEDCOM :>) There is another user that recently joined, User:Bergsmit that is also researching the Netherlands. Due to the common interest in Netherlands place page names, I am moving this conversation over to the Netherlands Talk Page. I have been working on Friesland pages, and would love advice from people who actually live in the Netherlands!--Jennifer (JBS66) 06:09, 14 January 2009 (EST)


Hi Jennifer,

For me the country codes are more important than those provinces. That's because I don't want to use an English country name in my repository. I publish most of my data on Dutch sites, so I will always use Nederland as my country name, not (The) Netherlands.

NL is sort of language neutral, and it's shorter, so I would happily accept that.

cheers,

--Enno 15:07, 14 January 2009 (EST)


I'm going to answer on Place talk:Netherlands.--Dallan 18:12, 14 January 2009 (EST)


Here how LDS wiki site is doing it [16 January 2009]

I thought you might be interested in how the LDS wiki site is doing it. They've got a large pool of folks to draw from. But it looks like it's well organized: https://wiki.familysearch.org/en/Maryland_Barn_Raising_Tasks#Simple_tasks --Janiejac 00:10, 16 January 2009 (EST)


Yes, the LDS have just assigned Family History Library employees to spend 30% of their time working on the wiki (they estimate they will put in 20,000 hours this year). It's already a terrific resource for research guidance, and this will make it much better in the future. Using wiki pages to track progress is what we use for some of the projects here as well (see WeRelate:Source review for example). It's a great way to track projects, and anyone is welcome to set up a project page this way.--Dallan 10:45, 16 January 2009 (EST)


More Netherlands Place Name Matching Questions [16 January 2009]

Dallan, this issue is related to the Netherlands discussion above. User:Bergsmit and User:Borgsteede have uploaded massive individuals/families in the last week. The problem is, when the abbreviations were added to Canada, it caused a problem to all of these GEDCOM's. One example, Person:Johannes Sturkenboom (1) death occurred in Houten, UT, NL but this red-links to Place:Houten, Ut, Newfoundland and Labrador, Canada instead of to the correct Place:Houten, Utrecht, Netherlands.
Now that I've added the abbreviations for Netherlands and its provinces, the newest GEDCOM upload appears to be ok. However, having thousands of red-links like Place:Amsterdam, Nh, Newfoundland and Labrador, Canada is not good. This certainly isn't either of the above user's fault. Is the only recourse to locate and redirect each of these through Special:Wantedpages? So, one would need to go into Place:Amsterdam, Nh, Newfoundland and Labrador, Canada and #REDIRECT to Place:Amsterdam, Noord-Holland, Netherlands. Or... is there some way to do a search&replace through these particular GEDCOM's and replace Newfoundland and Labrador, Canada with NL?
Thank you!--Jennifer (JBS66) 09:17, 16 January 2009 (EST)


Please don't set up redirects for all of these pages. That would be a lot of work, and the new abbreviations make the redirects unnecessary going forward. As place-matching has matured over time, there are quite a few gedcom's that have red links for their places that would now be blue links if they were uploaded again. In the next few months (early Spring) I'm going to write a program to re-match all of the places on existing person and family pages. Most of the red links should turn blue at that time. So if User:Bergsmit and User:Borgsteede don't mind waiting until then, that's an effortless solution to the problem.

If they didn't want to wait, another thing they could do if they haven't made a lot of on-line changes to their pages (by editing or merging), is to delete their trees and re-upload them.--Dallan 10:45, 16 January 2009 (EST)


Gedcom import [17 January 2009]

Hi Dallan,

When I first commenced using WeRelate, you kindly increase my ability to upload a GEDCOM file containing approx 9000 people.

Well, I have just deleted my tree on WeRelate in order to upload the updated GEDCOM and now I get an error message stating the GEDCOM file is too large.

Does the increase in GEDCOM size stay once you action or do I need to apply each time? Thanks.

Regards, Andrew--nastrond 18:04, 16 January 2009 (EST)


If I increase the limit and your gedcom ends up generating a lot of duplicates with existing pages in the wiki, I'm going to get in trouble with the very dedicated people who are trying to merge duplicate pages. (They're already feeling a little put out due to the recent flurry of uploads we had a couple of weeks ago.) How much of a hurry are you in? Could you wait 2-3 weeks? If so, we should have the new upload process in place and you can upload it then. Merge will be part of the upload process at that time.--Dallan 18:40, 16 January 2009 (EST)


Hi Dallan,

Major additional information over the Xmas period, so I deleted the tree in preparation for the upload. I would prefer to upload now if possible? I am also recommending this site to a one-name researcher for reference, so the tree available would help. Thanks.--nastrond 23:01, 16 January 2009 (EST)


Ok, I've approved your tree. Please check your Special:ShowDuplicates page (available on the MyRelate menu) the day after you upload your gedcom and merge any duplicates you find. You'll probably also need to check for duplicates again the next day to merge any duplicates found after you merge the first batch. Thank-you!--Dallan 00:25, 18 January 2009 (EST)


Sources & Surnames [17 January 2009]

Source pages have fields for surnames covered. However, these sources then do not link back to the surname page. Example: In Source:Wanzer, William David. History of the Wanzer Family in America, Wanzer is listed in the surnames covered field, but does not link to Surname:Wanzer (Source:Ancestry Message Boards ( wanzer ) does though).
Would this be a beneficial addition to the surname pages?--Jennifer (JBS66) 10:41, 17 January 2009 (EST)


That's a good idea. I'll add it on Monday. Thanks!--Dallan 00:25, 18 January 2009 (EST)


Source:England and Wales. Index to the civil registration of Births, Marriages and Deaths for England and Wales, 1837-1980 [17 January 2009]

Hello Dallan, and other Werelate users,

I thought I would bring this to your attention. I hope I did not make a big mess. I tried to cleanup my MySource and Redirect it to this Source FHL community source. I retitled it to the . also. I added websites, (?Repository), etc.

It might need to be undone.

Hope this is done the right way.

Thanks, Debbie Freeman --DFree 14:53, 17 January 2009 (EST)


Hi, it looks like you got some help from User:JBS66; let one of us know if you have any more questions.--Dallan 00:25, 18 January 2009 (EST)


Renaming Source to Repository [21 January 2009]

FYI - It appears that when a Source is renamed to a Repository, that those who were watching the Source are not automatically added as watching the Repository. (This is based on only one incident that I noticed though.)--Jennifer (JBS66) 17:16, 17 January 2009 (EST)


Thank-you for letting me know. I'll fix it on Monday.--Dallan 00:25, 18 January 2009 (EST)

I checked this out: the watchers are being added to the repository page, but they're just not listed on the page right away. This is because because the repository page is cached by the server before the watchers are added. Eventually the cached page expires and the next time it's viewed the watchers appear. This is part of an existing bug that I'll fix later this year. Thank-you for pointing it out! (There have been other cases where the watchers really weren't added, so it's good to check into these things.)--Dallan 09:47, 21 January 2009 (EST)

International ambitions [27 January 2009]

Hi Dallan,

I'm very curious about the international ambitions of this site, or its sponsors. Can you tell me something about that?

The reason I ask this is that for me there are a lot more people to relate to when we can attract more users from The Netherlands, and in my case also Germany, and I guess that the same holds true for many other countries with immigrant links to the US and Canada.

That said, putting more foreign GEDCOM's here will also put a bigger load on projects like the documentation of places and such. Moreover, users will also upload foreign notes and sources, and may even ask for localized help pages, etc.

The Wiki software speaks a lot of languages already, so that's not the problem. My question is more like whether you and your sponsors have the resources to accommodate more users from my part of the world, and others too, of course.--Enno 14:40, 18 January 2009 (EST)


First of all, I would love long-term for WeRelate.org to become an international resource. Interestingly enough, based upon a crawl of genealogical web pages that I did a couple of years ago (click on Web on the Search menu), The Netherlands and Germany have the largest number of non-English genealogy web pages, so they would both be at the top of the list.

What it will take to go international is two things:

  • You already mentioned that the software needs to be localized. This is relatively easy with the MediaWiki (wikipedia) software, but as I've been making changes I've gotten lazy and hard-coded messages into the software itself. I estimate it would take about a week for me to extract the messages into a table that others could translate.
  • The larger need is for people who would be willing to manage the international content: translating system messages, help pages, managing the fixing up of place pages, run a German-language Watercooler, etc.

I haven't pushed forward any internationalization efforts because we're in such a state of flux with the basic software, but once things settle down, which I'm hoping will be toward the end of the year, I'd love to work with people who would be interested in helping to set up different languages. The hardware could certainly accommodate more users; it's not being stressed right now. What's needed is a group of people who would be willing to do the translation work, starting say in the Fall.--Dallan 23:21, 20 January 2009 (EST)


I see. Well, I mainly asked this to figure out whether you would appreciate it when I recommend this site on a major Dutch forum. Attracting more Dutch and German users would give me more chances to connect to other trees, but in your answer I read that you still have to focus on the concepts and the software of the site itself, so I think I'd better wait till later this year.--Enno 16:36, 27 January 2009 (EST)

Realistically I think it's a good idea to wait for now unless they'd be ok with an English-language website. This Fall will be a good time to expand into other languages.--Dallan 17:52, 27 January 2009 (EST)


Genealogy Software & Printing Pedigree Charts [21 January 2009]

In researching my Netherlands lines, I saw a few sites that utilized this software (here is an example of it in action [2]). What I especially like about it is the ability to print pedigree charts in pdf format. I'm currently using WeRelate as the only genealogical program I update, and I could really use a clean printed copy to work from... I'm not sure that it would integrate with WeRelate, but some of the features are very user friendly.--Jennifer (JBS66) 08:11, 21 January 2009 (EST)


Watchers after Merge [21 jan 2009]

I did a merge this morning Special:ReviewMerge/630, where User:Jlhasty was watching Family:Robert Drouin and Marie Chapelier (6), Person:Robert Drouin (15), Person:Marie Chapelier (6), in addition to all of the children's pages. After the merge, this user is not shown watching the newly merged pages.
These 2 earlier merges worked though Special:ReviewMerge/629 and Special:ReviewMerge/628.--Jennifer (JBS66) 08:41, 21 January 2009 (EST)--Jennifer (JBS66) 08:43, 21 January 2009 (EST)

This is part of the same bug as you mentioned above -- the pages get generated and cached by the server before the new watchers are added, so when you first look at them you're seeing the cached versions. Jlhasty is listed now for example. I'm planning to upgrade to the latest version of the MediaWiki (wikipedia) software next month; I'll fix this bug after I've upgraded.--Dallan 09:47, 21 January 2009 (EST)

Bug in warning window when editing a page [26 January 2009]

Hi Dallan,

Don't remember where this discussion was originally; but the warning only allows one to make one edit at the time if I change the type of source. As an example, if I change a source page type to something different, I am not allowed to also add a repository at the same time because the warning prevents me from editing another category on the page. --Beth 21:40, 23 January 2009 (EST)


Hi Beth,

Sorry - I'm not sure I understand what you mean. I just tried changing the type of a Source page, adding a repository, and saving the page, and it worked just fine. For a lot of the Source pages, if you edit the page the source type and the repository type show up in red because they've never been set. You should be able to change any other field on the page so long as you also put something in both of these red fields before saving the page.--Dallan 16:00, 26 January 2009 (EST)

Hi Dallan, Well, first of all it is not a big deal; but the page I was editing already had a repository and source type. I changed the source type and the reminder window popped up; so I could not add an additional repository on the same edit; but I could of course add the repository on a second edit. --Beth 16:42, 26 January 2009 (EST)


Place names in GEDCOM files [28 January 2009]

I am a newly-registered user who would like to create and downlaod several GEDCOM files from my database. I have over 10,000 names and currently use Family Tree Maker v. 13.0.281. My question is what to do about place names? I have always followed the convention of attaching the word "County" in each place name and using the postal abbreviations for U.S. states and Canadian provinces, i.e. "Rutland, Rutland County, VT". Unfortunately FTM does not have a utility for easily editing place names on a macro basis.

From reading through the Talk topics it appears that I will be creating duplicate place names with each of my WeRelate pages. What problems does this create and do you have any suggestions for how to go about editing my place names either before or after downloading to WeRelate?--VTJohn 13:25, 24 January 2009 (EST)


We've come a long way in matching places over the past six months. Your places containing the word County and using US and Canadian postal abbreviations should be matched correctly now. One known exception: if you end a place with just ", NL" instead of ", NL, Canada", the system will likely match it to the Netherlands instead of Newfoundland. This is because matching happens right-to-left and NL is also an abbreviation for Netherlands, and countries generally take precedence over states/provinces in matching.

Having said this, we're currently working on a major update to the GEDCOM upload process, where you'll be able to see exactly which wiki places the places in your GEDCOM will be matched to, and you'll be able to correct the matches if necessary (or tell me about improvements to the place matcher that might need to be made). You'll be able to do this during a "review" step after your GEDCOM is uploaded but before the wiki pages are actually created. We expect to have this ready by Feb 15. So one option is to wait until then to upload your GEDCOM. Another note: we currently allow GEDCOMs with a maximum of 5,000 people to be uploaded. Once the new upload process is in place we'll likely increase that number.--Dallan 16:00, 26 January 2009 (EST)


Dallan - Thank you for the response. I expect I'll wait until after Feb. 15 to do my upload. Even then, it may be easier to break my gedcom into several pieces but the edit function seems worth waiting for.

John--VTJohn 19:54, 27 January 2009 (EST)


Suggestion - Union - Civil War Pension files (United States) [27 January 2009]

Hello Dallan,

I created a new Source using "Evidence Explained" it is called Source:United States. United States. Civil War and Later Pension Files". Sorry about the double United States. After I typed it in the Title, a double came up maybe a new feature. I then located an already created Source called Source:General Index to pension files, 1861-1934 (forgot the ##) which was probably created from the FHL card catalog, I checked and it was T288 from NARA.

I then linked the Source:General Index to the Source:Civil War pension Source.

I do not know if it would help in the computer work, but it might be a help. All (Union) Civil War Pensions, and later wars mentioned in this Source would be under this "Source:United States. United States. Civil War and Later Pension Files" if we used the NARA system.

Hope this helps, Debbie Freeman User: DFree--DFree 13:21, 25 January 2009 (EST)


This looks good. I just edited Source:United States. General Index to pension files, 1861-1934 (184124) to add a link to Source:United States. United States. Civil War and Later Pension Files as well.

What would you think about renaming Source:United States. United States. Civil War and Later Pension Files to just Source:United States. Civil War and Later Pension Files? The double United States came up because the place covered by the source is automatically pre-pended to the source title. But since the source title already starts with United States in this case, it seems redundant.--Dallan 16:00, 26 January 2009 (EST)


Hello Dallan,

Gratefully!! Thank You please. I does bug me to see the double United States.

Debbie Freeman --DFree 16:39, 27 January 2009 (EST)


Poland Place Names [29 January 2009]

I'm starting to think about outlining project goals for a future Eastern European Place Renaming/Merging project.
I have an initial question first though. I noticed for the Poland place pages, they appear to be missing a hierarchy level. This info. is contained in each FHLC page. Is this a project that needs to be planned as a manual process (for the approximately 10,500 pages), or is this something that can be mined with a bit of code?--Jennifer (JBS66) 09:22, 26 January 2009 (EST)


For places in Germany, the FHLC often included the district but sometimes included all sorts of other things in parentheses at the end of the place name, so I couldn't trust parenthesized names to always be district names. If you would review Polish places in the FHLC and tell me whether the place names in parentheses always (or mostly) correspond to the districts the places are located in, I'll be happy to write a program to rename polish places accordingly. That would be great news. I've got to get GEDCOM export working first, but I could write the program about this time next month. Let me know what you think.

Another problem with Eastern European places that you've probably noticed is that we have separate Place pages for the same place in different countries. Ideally we'd have just one place page, or at least links between the various pages. The problem is that in the initial creation of the place wiki, I checked for duplicates and merged only within the same country. I investigated the possibility of checking across countries, but many places were named differently enough that it wasn't something I felt comfortable doing automatically. Still, I'm open to ideas for some automated renaming/merging as you mention above, especially involving hundreds or thousands of places.--Dallan 16:00, 26 January 2009 (EST)


It appears that about 2% of the information contained in the parentheses is either unnecessary information (ie the word diecezja or powiat), are abbreviations of locations where the full location name also appears elsewhere, or contain misspellings or incorrect characters. This being said, not only are we dealing with changes in countries, many of these towns changed the districts they resided in as well. I'm not sure if adding the hierarchy level helps or muddies the mess.
I put some very basic ideas on this page. The idea of renaming Eastern European place pages is such a large project, we could be talking well over 100,000 pages, that a collaborative effort from people familiar with the area would be best. So, I'll just put ideas and resources in one place, and see what happens.--Jennifer (JBS66) 10:09, 29 January 2009 (EST)


Ok, let me know whether people decide that it's worth adding the district level. I don't really know either. But if it's just a 2% problem rate, and we can avoid some of the problems simply by creating a list of words to avoid (like diecezja or powiat), then it's worth adding it automatically. I couldn't do it right now, but I could in 2-3 months.--Dallan 23:05, 29 January 2009 (EST)


WeRelate.org and New FamilySearch [27 January 2009]

Hi Dallan,

How do you see the concepts of your site when you compare them to New FamilySearch? I learned about the latter through the FamilySearch Labs pages, and I see similarities, but also differences.

Since I'm not a member of the LDS, I can only see how it works in Ancestral Quest demo movies, but I will probably start using it when it is available to the general public, because it may better fit with the way I work with PAF right now.

thanks, --Enno 16:30, 27 January 2009 (EST)


I'm fairly familiar with New FamilySearch (nFS). I've been using it since last September and I'm friends with several of the developers and support people on the project. Here are some differences:

Similarities

  • both have the development of a common tree, with one "page" for each unique individual represented in the database, as a goal.
  • both have merge/unmerge capability.
  • both have web-based interfaces, with a flash-based add-on that allows you to view more generations of your tree at once.
  • both are developed by non-profit organizations and are free to use.

Differences

  • in nFS nobody else can edit your information. You can't edit or remove anybody else's information either. This is the basic fundamental difference between WeRelate and nFS. A lot of people in nFS have dozens of possible names, birthdates, etc. One of these names/dates/places can be selected as the information that appears on the "summary" screen, but the "details" screen shows all of them. It can get a little confusing to figure out which of possibly-dozens of alternatives is correct. It would be nice to remove incorrect ones, but all you can do is "dispute" them, which just flags that someone disagrees with them.
  • in nFS if someone else modifies a page you have contributed to, you don't get notified. The only way you can tell is if you periodically re-visit every one of your pages. (It may be that if you use the latest version of Ancestral Quest to interface with nFS you do get notified - I just downloaded the new version and haven't had a chance to try it yet.)
  • nFS doesn't have page histories - you can't see what the page looked like in the past or who changed what when and why. If someone else has changed a birthdate for example it can be difficult to tell what the birthdate was beforehand.
  • nFS doesn't have comment pages; you collaborate with other contributors by sending them email if they have chosen to make their email address available to others.
  • nFS doesn't have a "recent changes" screen. As far as I know, nobody watches the changes that are taking place to make sure that obvious problems aren't being introduced.
  • nFS doesn't allow uploading pictures or stories.
  • the pedigree map and timeline functionality at WeRelate is more advanced.
  • nFS doesn't have place, source, or repository wiki pages. They do however have a good set of research guides at http://wiki.familysearch.org.
  • if you upload a GEDCOM to nFS, you have to merge duplicates one person at a time; they don't have the concept of family merges, and they won't tell you which people in your GEDCOM have likely duplicates unless you visit every person that you uploaded and click on "Find duplicates".
  • I believe GEDCOM uploads are currently limited to around 1000 people.
  • nFS has been in development since 2002. It was supposed to be available to all LDS people in 2005 and to non-LDS people shortly thereafter. It's not available to all LDS people yet.
  • nFS has around 500-600M people in it; WeRelate has a little under 2M people in it. I'm guessing that roughly 100M of the people in nFS are duplicates.
  • nFS is "prettier" than WeRelate, and somewhat easier to use although we're working to change that.
  • nFS allows you to enter living people. Nobody else can see the living people that you enter except you, but WeRelate doesn't allow living people except for a nearly content-less "Living Surname" wiki page.
  • The latest version of Ancestral Quest and FamilyInsight both allow you to "sync" data from a desktop genealogy program to nFS. I haven't played around with either of these programs yet so I can't tell you how well the synchronization works. WeRelate won't add synchronization with a desktop genealogy program until later this year.

--Dallan 17:14, 27 January 2009 (EST)


Ancestry.com and FHLC sources that duplicate [29 January 2009]

Are there plans to automatically merge duplicate sources from Ancestry and FHLC, or is this a potential manual project? Example: source from FHLC and source from Ancestry. NOTE: There are just around 18,000 Source pages prefaced with Ancestry.com - that are potentially duplicated. --Jennifer (JBS66) 10:15, 29 January 2009 (EST)


I think it would be worth trying to merge them semi-automatically. We could at least have the system come up with a "potential matches" list and have people mark whether they are truly matches or not. A project for Summer perhaps?--Dallan 23:05, 29 January 2009 (EST)


Source Place Issued Field [29 January 2009]

Quick FYI - When entering a place issued in a source, if you use a | to display a shorter name of the location, the words "Place Issued" disappears from the Source page. See Source:Doubleday, John M. Our ancestors Samuel Terry, Benjamin Throope, John Swift, Isaac Williams, Clark Mills, Wright Nicholson, Elisha Doubleday.--Jennifer (JBS66) 18:30, 29 January 2009 (EST)


Thanks for letting me know. This is fixed now.--Dallan 23:05, 29 January 2009 (EST)


Have you been tweaking Search? [4 February 2009]

I notice that characters - like : or ( ), are no longer causing a parsing error in searching. But, now quotations are not working to limit a search by exact phrase. So, I get the same number of hits whether or not I enclose a phrase in quotations, and the words of the phrase are not all together.--Jennifer (JBS66) 17:45, 31 January 2009 (EST)


Hmm, I made a minor change. It shouldn't have affected search, but it might have. I just tried putting quotations around a phrase in the 'keywords' field and it seemed to work though. Could you give me a link to a search screen where it doesn't work and I'll look into it? Thanks.--Dallan 19:21, 31 January 2009 (EST)


Sorry for not including a link originally - that's not like me :>) Isn't this interesting... I tried a bunch of searches to recreate it and each search worked as it should! I guess I noticed the problem on a rare case, here it is:

When I mentioned a parsing error, this is what I meant (it doesn't happen in the Title field, only the keyword field) : [3]. When I remove the :, it's ok though. Thank you!!--Jennifer (JBS66) 19:40, 31 January 2009 (EST)


Here's another example:


Ok, I see what's wrong. In order to reduce the size of the index, short words (and the word genealogy) are not indexed. So a search for "Who's who in Steuben" becomes just "Steuben", and "Ziegler Genealogy" becomes just "Ziegler". I don't want to add short words into the index because it will make the index too large, but your example has shown me that adding "genealogy" to the index would be useful. I'll add that to my todo list.

The other problem is that a : in the keywords field isn't handled very well. I'll also add that to my todo list.--Dallan 11:06, 4 February 2009 (EST)


Extra period in title following initials [31 January 2009]

Hi Dallan. You're probably already aware of this, but there's an extra period following the author's name when it ends in an initial in source titles. Example: Source:Pixley, R.B.. Official War History of Ozaukee County, Wisconsin. --Ronni 20:08, 31 January 2009 (EST)


Thank-you. I wasn't aware of it. I'll fix it on Monday.--Dallan 21:46, 31 January 2009 (EST)


Agent Run of Sunday... [4 February 2009]

Thanks for the Agent run - I picked up over 400 pages today! The 70 or so wikipedia templates left empty last time remain so, but I'm still psyched about the 400 more I got.

I also added a draft of a proposed guideline for wikipedia use. I noticed everyone on that at the watercooler, you'll see it at Proposed Guidelines for use of Wikipedia. I really want to move it along so that a set of conventions gets adopted on this - we really ought to boot-strap our work on wikipedia, but also give back to wikipedia.

Thanks again...--Jrm03063 20:14, 1 February 2009 (EST)


Thanks! I'll read through the proposal tonight. Could you give me examples of 2-3 wikipedia templates that are empty? I want to make sure that they get updated in the next "full" refresh.--Dallan 11:06, 4 February 2009 (EST)


spiritual wife-ism not ready for prime time [4 February 2009]

Dallan, while your recent request inspired me to draft a wiki article version of my ms on the above-named topic, rereading it this a.m. makes me realize much work has to be done (links and refs) before it is ready for prime time.

On a related note I see that while my other user pages are automatically listed on my user page, this one is not. You can find, tho', on my contributions page.

And yet another related question, I see that renamed user pages appear on one's page under both the old and new name. Is there a way to fix that? (And is this why renaming is to be avoided?)

jillaine 08:34, 2 February 2009 (EST)


Interesting article! It shows up on your user page now. In certain situations pages get cached at the server in their old state and it takes awhile for them to show the updated version (this is a bug that I haven't fixed yet).

Yes, the issue with renaming is that it causes two pages to be created: one with the old name that points to the new name, and one with the new name. Both of these pages will show up in various places: in watchlists, in drop-down menus, when browsing pages from the Admin menu, etc. Renaming is ok in general, but if it happens too often it can clutter up these lists.

I see you put the old name on the speedy-delete category. I went ahead and deleted it. Just an FYI, you can also delete your own user pages (unless I have a bug) by selecting "Delete" from the "More" menu.--Dallan 11:06, 4 February 2009 (EST)


Search with Special Characters [4 February 2009]

Hello there, another small FYI
When I conducted this search, which contains the word Düsseldorf, there were no returns. When I changed it to Dusseldorf, or removed Exact Match, it worked.--Jennifer (JBS66) 12:35, 2 February 2009 (EST)


Argh, that shouldn't be happening. I'll add this to the list of things to fix for search. Thank-you for pointing it out.--Dallan 11:06, 4 February 2009 (EST)


Please import GEDCOM [4 February 2009]

Dallan,

Please import my new GEDCOM so I can start merging my Ancestry.com tree information with my TMG information tomorrow.

Thanks, --Lauren 21:15, 3 February 2009 (EST)


I released the hold on it. It should finish processing shortly.--Dallan 11:06, 4 February 2009 (EST)


Thank you! --Lauren 11:08, 4 February 2009 (EST)


Suggestion on African American - Slave People Pages [9 February 2009]

Hello,

I am hoping for suggestions. I am finding the names of slaves on the will of the Owner. I would like to create People Pages for these African Americans. So far I only have first names. Is there a tutorial, or agreed to system on how to create these People Pages? How are other Werelate Users handling this situation?

Debbie Freeman --DFree 11:57, 4 February 2009 (EST)


This is probably a good question for the Watercooler; my thinking is that perhaps you could put "of" followed by the name of the plantation they lived on (with a preference for the plantation they were born on if you know it) for the surname?--Dallan 19:50, 8 February 2009 (EST)


Thanks Dallan, I copied it over to the Watercooler

Debbie Freeman --DFree 21:00, 8 February 2009 (EST)


User DNA page [8 February 2009]

Hi, I'm new to WeRelate. I've mostly been reading the Help and FAQ pages over the last couple of days. However, I noted you have some DNA tables in your Sandbox, so I thought you might like to see my user DNA page: User:Parsa/DNA. ---Parsa 22:27, 6 February 2009 (EST)

Wow - very nice!

I'm waiting to see if DNA research catches on with the community here. If it does, I was thinking I could come up with a way to search for people with similar DNA markers.--Dallan 19:50, 8 February 2009 (EST)


Odd template behavior [8 February 2009]

The Bu template does not seem to be working the way it's supposed to. I see it used in other template descriptions, and they all seem messed up. ---Parsa 22:16, 8 February 2009 (EST)


This file is larger than the maximum allowable size for a GEDCOM. [9 February 2009]

I am trying to upload my gedcom Spencer_Taylor_Family.ged. It has 12,000+ names, and contains all the work that my great-great aunt started in the 1970's, and that I finished in the 1980's and 1990's. But it is too big. Can you help?

Drew Spencer drewspen@gmail.com--Drewspen 10:25, 9 February 2009 (EST)


We're currently working on a major update to the GEDCOM upload process. The new process should be in place soon - within the next 1-2 weeks. If all goes well with the new process, we'll increase the limit on uploaded GEDCOM's. My expectation is that the limit will be increased by the end of the month. In the meantime, you could upload a portion (less than 5000 people) of your GEDCOM now, and upload the remainder next month.--Dallan 10:42, 9 February 2009 (EST)


Source Title question [10 February 2009]

Dallan, is it possible for you to check out at conversation at User talk:Quolla6? It involves how we should title a book reprint that does not fit into WR naming conventions. Thank you!--Jennifer (JBS66) 10:52, 10 February 2009 (EST)


Will do. Thanks for bringing it to my attention.--Dallan 18:05, 10 February 2009 (EST)


Embeddable Google Books [15 February 2009]

I was wondering if it is possible/advisable to Embed Google Books here at WeRelate.--Jennifer (JBS66) 14:05, 14 February 2009 (EST)


It shouldn't be too hard to add this functionality. I'll put it on my todo list right after merge-during-upload and gedcom export.--Dallan 14:37, 14 February 2009 (EST)


Thank you, I think this would be a neat feature.
On another loosely related note: Is it possible to embed an external image on our pages? (instead of uploading).

Google Books Share this Clip suggests the following code:

<a href="http://books.google.com/books?id=gbsNAAAAQAAJ&dq=1732%20Gosselin&pg=PA328&ci=115,754,443,238&source=bookclip"><img src="http://books.google.com/books?id=gbsNAAAAQAAJ&pg=PA328&img=1&zoom=3&hl=en&sig=ACfU3U3XGmFJrxd0pQDASzINQUKEFJrU7w&ci=115%2C754%2C443%2C238&edge=1" border="0" alt="Text not available"/></a><br/><a href="http://books.google.com/books?id=gbsNAAAAQAAJ&dq=1732%20Gosselin&pg=PA328&ci=115,754,443,238&source=bookclip">Dictionnaire généalogique des familles canadiennes depuis la fondation de la colonie jusqu'à nos jours By Cyprien Tanguay</a>

Thank you!!--Jennifer (JBS66) 16:23, 14 February 2009 (EST)


Currently you can't enter "a" or "img" tags in wiki text; this has been a bit of an annoyance. I'll take a look at this after GEDCOM export to see what I can do.--Dallan 19:50, 14 February 2009 (EST)


I researched my second question a bit more (embedding images), and I think it's probably not such a good idea after all. I see that a wiki needs to allow external images. Wikipedia, however, does not allow inline linking for a variety of reasons. So, while my first question might still be plausible, it's probably best that I either upload the image or link to the outside source.--Jennifer (JBS66) 07:22, 15 February 2009 (EST)


It may still be possible to allow inline image linking for specific websites like google books.--Dallan 20:46, 15 February 2009 (EST)


I see you're prepping for the agent run... [14 February 2009]

In addition to the "source-wikipedia" people that didn't get quite completely refreshed, there are still some un-refreshed templates out there from the initial round. If you search templates for the string "Pending refresh", you'll find 80+/- that didn't get updated. Not sure what happened there...--Jrm03063 15:58, 14 February 2009 (EST)


I checked the first 5 of these pages. In all cases the page was renamed after October 13th 2008, which is the date of the most-recent Wikipedia download. So our refresh agent can't find the article in the download because it's still listed under the old name.

I'm kind of surprised that a newer download isn't available yet from Wikipedia. They usually produce a new download about every three months. Hopefully we'll get a new download next month, and we'll refresh all of the wikipedia templates from it.

I'll update the source-wikipedia templates tomorrow. Looks like there are nearly 150 to do.--Dallan 20:06, 14 February 2009 (EST)


Place Name in Marriage Field [19 February 2009]

Dallan, I have a question regarding this family page. I created a new Place page. However, when I go to enter this place into the family page, the location is available in the system-created list, but when I preview, it has entered lower case letters Saint-amant-de-boixe, and the town red-links. I couldn't quite figure it out - watch it be a caching problem again! :>) (though I did wait a few hours to see if it resolved itself).--Jennifer (JBS66) 18:14, 16 February 2009 (EST)

Please disregard - guess I didn't wait long enough!--Jennifer (JBS66) 09:37, 17 February 2009 (EST)


There is something you can do to clear the server cache -- I do it every once in awhile but I haven't advertised it because it is kind of "tech-y" :-) -- in the URL field, at the end of the page title add "?action=purge" and press enter. Clicking on this link: http://werelate.org/wiki/User_talk:Dallan?action=purge will remove my talk page from the server cache. It's worth trying in case you have a question about whether something's just badly cached or not.--Dallan 21:37, 19 February 2009 (EST)


MediaWiki Gallery Tag [19 February 2009]

I was playing around with images and galleries, and noted that some of the functions within the MediaWiki syntax for the <gallery> tag don't seem to function on WeRelate. The syntax for the initial container tag is given on this Wikipedia page: Wikipedia:Gallery_tag. The "caption" and "perrow" seem to work correctly, but the "widths" and "heights" don't do anything to the size of the images in the gallery.

Another thing I found is that if I try to float an image "right", it drops down low on the page so that it's below the Google ads rather than to the right on the top of the page. It would be nice if the image was in the correct position on the right side (and just to the left of the ads). -- Parsa 16:52, 18 February 2009 (EST)


I'm currently running an earlier version of MediaWiki (I need to upgrade). This might explain why some of the gallery options don't work correctly. You should be able to float images right without having them flow below the ads though; just make them smaller - e.g., [[Image:Home of Col. William Edmiston.jpg|300px|right]].--Dallan 21:37, 19 February 2009 (EST)


Renamed Tree [24 February 2009]

I renamed one of my trees a few days ago. Now, when I search by keyword +Tree:"JBS66/Brien" (the new tree name), it only has a fraction of the pages that should be under that new name. When I search using the old tree name, +Tree:"JBS66/Normandin&Smereczanski", the rest of the pages are there. I'm trying to use trees as a tool to keep track of the pages I've cleaned up. So, not that it's a big deal, just wondering why searches would need to be conducted using the old name. Thank you!--Jennifer (JBS66) 07:08, 20 February 2009 (EST)


Wow, I just realized that renaming a tree doesn't force the pages in the tree to be re-indexed. That's why most of them are still listed under the old name. I just forced all of the pages in your tree to be re-indexed, so they should be listed under the new name in about an hour. And I'll fix this tree renaming bug as soon as I get done with gedcom export.--Dallan 12:50, 20 February 2009 (EST)


Dallan, I really appreciate you re-indexing this tree for me! Sorry about the added to-do list item :>) --Jennifer (JBS66) 08:33, 24 February 2009 (EST)


No problem - I'm glad you pointed out the problem! :-) --Dallan 15:33, 24 February 2009 (EST)


Argh! please complete merge-upon-upload [1 March 2009]

Someone recently uploaded a GEDCOM with a bunch of dupes of existing colonial people. Waaaaaahhh.... please please please complete the merge=upon=upload feature. And and and... can't we puhleeze stop all gedcom uploads until the merge=upon=upload is complete? huh? can we? please???? (just whining... can you tell?) -- jillaine 22:06, 25 February 2009 (EST)


How about if I delete the GEDCOM and ask the person to re-upload after we get merge-during-upload finished? I don't think that would be asking too much from them.--Dallan 22:15, 26 February 2009 (EST)


yes, please!! Thanks, Dallan! -- jillaine--jillaine 06:52, 27 February 2009 (EST)

Hi Jillaine, I've leave the uploader a message and removed the tree.--Dallan 11:10, 1 March 2009 (EST)


Sources: Place Issued [1 March 2009]

I know that you recently tweaked the Place Issued field for Sources based upon a question of mine. I'm wondering if it's possible/advisable to change this field to a straight text instead of a fill-in. My reasoning is that when you go to What Links Here for a town, it is less than useful to have a list of books published in that town (rather, it should just contain books referencing that town). Maybe this would be a discussion for Watercooler? Thanks!--Jennifer (JBS66) 14:39, 27 February 2009 (EST)


Good point. I've taken out the link.--Dallan 11:10, 1 March 2009 (EST)


Pages that are Protected [1 March 2009]

Dallan, I protected this page against edits. Jillaine noticed that when the page is protected, users do not see an Edit option on their page. At Wikipedia, when a page is protected, users still have the option of viewing and copying the source code, which is what I wanted to have happen in this case. Thoughts??--Jennifer (JBS66) 09:27, 28 February 2009 (EST)


Jillaine has since noted that View Source is there - under a Views Pulldown.--Jennifer (JBS66) 16:43, 28 February 2009 (EST)


CORRECTION: This "View" drop-down menu is viewable on the Blackberry interface, but is NOT included on the "normal" web interface. I can't find it at all now. Dallan? -- jillaine 22:32, 28 February 2009 (EST)

Darn it. Things are different on protected pages. There is now a "View Source" along the 2nd-from-top menu. Sheesh. Never mind. It's there after all. I'm going batty. ("going?" my husband would ask.) -- jillaine 22:33, 28 February 2009 (EST)

I'm not sure what happened. The "View source" appears where the "Edit" option used to be in Firefox. I haven't tested it on Blackberry though.

I've just set up a "test" site where we can play with things (upload gedcom's, create test pages, create multiple logins, etc.): http://sandbox.werelate.org/ I'll announce it on the watercooler and in the news tonight or tomorrow. Feel free to try out things like this on that website if you want.

The new merge-during-upload software is running on that site, as well as a new GEDCOM review option under the Admin menu that lists all GEDCOMs uploaded.--Dallan 11:10, 1 March 2009 (EST)


Review of merge of Mehitable Cobb [2 March 2009]

Discussion moved to User_talk:Jrich#Could_you_review_a_merge.3F_.5B2_March_2009.5D. Hope this link (the # part) gets updated as the topic is edited? --Jrich 15:52, 2 March 2009 (EST)


Merging and sources [5 March 2009]

I wanted to pass along something that I noticed while merging Kenz's duplicates. Where there was data supported by a source, the merging program didn't recognize that the fields were identical. One example is Family:Finlay McKenzie and Jane Hook (1). You can see where it kept the alt. marriages, even though they were the same. I know there was a choice to unclick them at the time of merge (I didn't do this particular merge). My question is why it wasn't automatically blanked out (as happens when the fields match). Thanks.--Jennifer (JBS66) 06:34, 5 March 2009 (EST)


The issue is that if you drop the event with the source attached but keep the source, the source doesn't get attached to the existing event. Until I fix this issue I thought that it would be better to have the source-attached-event checked by default, since it would be better to keep both events than to lose the link between the source and the event. Eventually I need to fix this problem however.--Dallan 19:12, 6 March 2009 (EST)


Source - Repository request [13 April 2009]

Hello Dallan,

I would like to suggest if possible if you could add an extra choose on the Repository list.

You can now choose free, paid, FHC, other.

Would you be willing to add a new one, something like Salt Lake City only, or FHL.

There is a percentage of the catalog that cannot be sent out of the library, like some books.

Other does not seem to fit, and FHC is inaccurate.

Thanks for thinking about this suggestion.

Debbie Freeman


DFree 20:47, 16 March 2009 (EDT)

This seems like a good idea. I'll add it later this week. (I need to get the new gedcom upload ready first.)--Dallan 23:04, 16 March 2009 (EDT)

I thought about this further. I don't want to get into the situation where we have different availability types for different libraries. Could we say "Other" for the repository type and list the FHL as the repository?--Dallan 23:25, 13 April 2009 (EDT)

User's Network View [21 March 2009]

Hello Dallan,

User:Bergsmit brought this to my attention this morning. If you go to his Network page, you can see where he is hardly connected to anybody (these should number in the 1000's). I believe what may have caused the problem is that on 13 Mar, he deleted his Userpage, then on 17 Mar he restored it.

Is this, perhaps, an indexing problem with an easy fix?? Thank you!!--Jennifer (JBS66) 07:17, 18 March 2009 (EDT)

Additional Details: User:Bergsmit is watching the Person:Charlemagne (1) page along with 33 other users. I asked him to go into the page, make an edit, and then save it. I wanted to see if his Network page would then add these users - but the Network page didn't change. He also states that this was happening before he deleted his Userpage.--Jennifer (JBS66) 16:18, 19 March 2009 (EDT)


Argh, I just checked and it looks like Bergsmit should be connected to 82 other people. I doubt it has anything to do with deleting his user page. There must be a problem updating the network links. It's not a fix I can do right away; I'll fix it next week.--Dallan 02:20, 21 March 2009 (EDT)


Great - thank you very much for looking into this! I wanted to also direct your attention to this page, where Fred posted details about the bug.--Jennifer (JBS66) 06:46, 21 March 2009 (EDT)


marshall large gedcom [20 March 2009]

I would like to add my Marshall gedcom but it has more than 5000. How do I do it?

thanks Joe Marshall--JoeMarshel 15:19, 18 March 2009 (EDT)


I'm currently in the final stages of preparing a new gedcom import program that should be able to handle larger gedcoms. I hope to have it ready by the end of next week. I'd like to test it on smaller gedcom's for about a week after that, then open it up to larger gedcom's the first part of April 3. You could upload a partial GEDCOM now or wait a few weeks and upload your GEDCOM then.--Dallan 02:20, 21 March 2009 (EDT)


Losing credentials? [2 April 2009]

Every once in a while I seem to get logged out while in the middle of something. Sometimes I can recover using the back button on my browser, but most times I need to sign in again. Not real predictable so not sure I can possibly recreate. Most recent time I did a save on a page I was editing, the saved page was displayed showing I was logged in, but when I tried to go to my watchlist, it said I needed to sign in to see my watchlist. Not a serious problem, just occasional annoyance. The page in question, in case you got some logs that can help you analyze this, is Family:Daniel Billing and Lydia Wheeler (1). --Jrich 14:37, 19 March 2009 (EDT)


This is odd. The log shows you editing the family page using Firefox 2.0, and then accessing your watchlist without the session information (so you'd need to log in again) using Firefox 3.0. Are you using two different browsers by any chance?--Dallan 02:20, 21 March 2009 (EDT)


Yes, I do use two versions, but on different computers, and not at the same time. I don't think that is it, as it just happened again, and I am only on one computer right now. But it may have something to do with tabs in firefox. Perhaps a bug on their part? Usually when I am adding a comment, I end up opening another tab so I can refer to the original topic, or to search for a reference, etc. That is what I was doing just now when it happened, and this is a common modus operandi for me, to be accessing WeRelate in two different tabs at the same time (only one instantiation of the browser though). --Jrich 12:55, 21 March 2009 (EDT)


I'm not sure what's going on; I use multiple tabs in Firefox without being logged out. What about checking the "Remember me on this machine" checkbox when you sign in - do you check that? Also, if it happens again, could you let me know which page you were looking at right before and right after it happened? That was very helpful when looking through the log files. Thanks.--Dallan 00:52, 22 March 2009 (EDT)


I may have enough detail to help you find this session in your logs. I was just bouncing around in a single tab between Family:John Libby and Unknown (5) and the talk page for the same family. It was the only thing going on in my browser, no other tabs at all. I then opened a new tab to visit books.google.com, came back to my first tab, switched between Talk and Family page, and then went to edit, and was told to sign in again. I had been signed in because I had done a minor wordsmithing edit back when I first came to the page. All WeRelate work was done from the same tab, the second tab was only used to visit books.google.com. If I had to guess, I would say that something in opening a new tab is messing up one or the other side of the communication. If it starts sending a different version string as you indicated before, one would suspect this is a Firefox bug? It is version 3.0.1 of Firefox on Windows XP. --Jrich 09:32, 23 March 2009 (EDT)


I'm stumped on this. I can see in the log that you successfully edit the page, and the browser request contains a session cookie indicating that you're signed in. This automatically redirects you to the new version of page. But when your browser follows the redirect, the new request not only doesn't contain the session cookie, it appears to contain a "logged-out" cookie from when you logged out roughly 13 hours previous. It looks like the software only sets the logged-out cookie when you click on the sign-out button at the top of the screen; there's no other time that I can see in the code where that cookie is set, and you're clearly not logging out in between pressing the "save page" button and following the redirect to view the edited page. It's as if the browser has decided to go back in time and send up an old cookie. I can think of two things:

  • try clearing your browser cache and your cookies,
  • don't click sign-out when you leave. The logged-out cookie should expire in 24 hours, so if you don't log out I'm hoping that after 24 hours the browser won't send the logged-out cookie to the server. I can't be certain though.

Sorry I can't identify anything more specific. By the way, I use Firefox 3 every day, often with multiple tabs, and I've never had this problem. It's very odd.--Dallan 18:49, 24 March 2009 (EDT)


I have noticed this problem too, though it's happened pretty rarely. I too use Firefox, 3.0.7. I'll let you know the details next time it happens.--Jennifer (JBS66) 19:09, 24 March 2009 (EDT)


BTW, I sometimes experience this as well; it's rare, and I have not been able to identify any particular pattern that causes it. I'm on a Mac using Firefox. -- jillaine 08:01, 28 March 2009 (EDT)

Just ran into this in circumstances that don't appear to be related to tabs. My bookmark for WeRelate takes me to my watchlist. It always pops up with a login screen, and then I return to my watchlist. So I logged in and used the handy link provided to return to my Watchlist. Once there, I clicked on show me all pages changes since last visited, and it asked me to login again. It seems to me from previous incidents, now that I have logged in twice, I won't have any further problems.

I believe I have stopped logging out per your advice from before. But I do know my previous session earlier today was on a different computer running Windows 98. This double login was on my Windows XP computer, since for sessions anticipated to be short, I do not bother powering up the old computer. They are most likely running different versions of Firefox (the other is not on, so can't check right now).

--Jrich 19:17, 31 March 2009 (EDT)


I'm not sure what is going on. I plan to upgrade to the latest version of the MediaWiki (wikipedia) software that powers WeRelate later this Spring -- hopefully that will solve the issue. Sorry about the trouble.--Dallan 10:25, 2 April 2009 (EDT)


It's not really a problem, very minor extra work. But I just wanted to check that you saw both logins and one isn't being stolen or any of those kind of things. I assume so, since this problem really seems to be related to my common practice of switching between two computers... --Jrich 10:41, 2 April 2009 (EDT)


Table Template [14 April 2009]

Dallan, I brought over a table template from wikipedia. It uses an #if function that doesn't seem to work here. It looks it's a part of a MediaWiki extension. Is this something that is even possible to do? Thank you!--Jennifer (JBS66) 06:38, 20 March 2009 (EDT)


I'm sorry, the #if function was added in a later release of MediaWiki than what we have. I need to update to the latest version but it's going to take several weeks so I've been putting it off until some of the higher-priority things are implemented. Hopefully sometime this summer.--Dallan 02:20, 21 March 2009 (EDT)


Hi there! I just wanted to double check that this [4] parser function extension wouldn't work with our MediaWiki version. It says that it's compatible with 1.7+... Not meaning to bug you about it or anything :-) just eager to bring over a few interesting templates from Wikipedia that use the #if statement. Thank you!--Jennifer (JBS66) 10:53, 14 April 2009 (EDT)


I know. We're going to have to think about whether we want to upgrade to the latest version of MediaWiki or we want to rename and clean up the source database this Summer.--Dallan 17:23, 14 April 2009 (EDT)


Please archive Watercooler again [1 April 2009]

The Watercooler has become so large again that my Blackberry can't pull it in. (I like to catch up on werelate while I'm commuting on the bus.) Could you or Solveig please trim it again? Thanks! -- jillaine 07:59, 28 March 2009 (EDT)


I just archived topics before Feb 16th. Wow - it's been too long since I've visited the watercooler. I really hope to get this new GEDCOM upload program finished soon. It's taking way too long.--Dallan 19:15, 30 March 2009 (EDT)


Thanks Dallan. The watercooler is still too big for my bberry but sheesh, you've got better things to do. -- jillaine 18:48, 1 April 2009 (EDT)

I left on the conversations I haven't replied to yet :-(. The new gedcom uploader will be ready this weekend (FINALLY!); then I'll get back to the watercooler and see if I can pare it down further.--Dallan 10:25, 2 April 2009 (EDT)

Blocked user [28 March 2009]

Hi Dallan,

Our problem user has returned under a new user name. I blocked the ISP address and selected infinite. --Beth 09:33, 28 March 2009 (EDT)


sandbox new upload feature [6 April 2009]

Hi Dallan, I just uploaded another GEDCOM at the sandbox to check the new upload feature again. Whew! Getting complicated; but the helps are much more focused and available.

I clicked on persons to see if a person I had would be rejected as possibly living because there was no birth or death dates. It wasn't and the wiki page looked good. But I didn't see a way to get from the wiki page back to the people and help page. So I clicked again on Overview and that got me back to help; then I had to click again on people page to continue checking the people. Seems to be an unnecessary and not indicated step to get back to checking the people.

The warnings caught a couple of problems I hadn't seen, so that was good. Then I removed the GEDCOM. Keep up the good work!--Janiejac 18:24, 2 April 2009 (EDT)


Thanks! There's a little "Help" button in the upper-right corner that pops open a help window. BTW, I just finished "Edit" yesterday. There's a link below just below the title on Person and Family page that lets you edit them. The edits aren't currently taken into account when the wiki pages are created, but I'm hoping they'll be applied by the end of the day today. We're getting close I think.

I plan to add a 5-minute video tutorial before this goes onto the main website.--Dallan 09:21, 3 April 2009 (EDT)


Just an FYI, the new GEDCOM upload is finally ready for final testing in the sandbox. If no one finds any more bugs or usability problems I plan to move it over to this site in a few days. Wow - spent over two months doing something I thought would take one -- story of my life :-)...Dallan 11:05, 6 April 2009 (EDT)


lost 500 [10 April 2009]

Hello,

I have uploaded the same gedcom twice now and both times it shows that my tree has 2218 people when the gedcom file that I uploaded had 2718. I have not gone through and found where the missing people are but in having decided that I quite like werelate and that I would like to use it as the place to put my family tree I find this to be a difficulty in that 500 people have been trimmed from my tree.

I thought you should know that this is occurring and if possible i will help you identify what and where the problem is.

Regards Jeffrey--JeffreyRLehrer 04:35, 8 April 2009 (EDT)


It sounds like the system thinks they are living. People who have a birth date within a hundred years or so and no death date get marked as living. And when all members of the families that a person belongs to are marked living, then a wiki page for the person is not created.

One way to see what's going on is to go to the sandbox site, create an account for yourself there, and upload your GEDCOM. We're testing a new GEDCOM upload process on the sandbox that will be moved over to the main website later this week. The new process includes a review program that allows you to see which people are living, what the wiki pages will look like, and will also allow you to link to existing people rather than create duplicate pages.

If you upload your GEDCOM to the sandbox and you find that pages aren't being created for non-living people, would you let me know? I can use the review program there to figure out what's going on.--Dallan 08:40, 8 April 2009 (EDT)


Hi Dallan, I have gone to the sandbox and not even 1 name is being imported from my gedcom? I have tried twice and now I am trying with a different gedcom file format. Jeffrey--JeffreyRLehrer 21:47, 9 April 2009 (EDT)

Jeffery, I just noticed your comment and wondered if you've gone through the review process? The sandbox is testing the new process, which requires that you go to your talk page and click on the review gedcom link there to review the gedcom before it is uploaded. The latest link posted is here and it looks like there are people imported.--Amelia 00:05, 10 April 2009 (EDT)
I should have mentioned that the new import process is a two-step process. After the first step, you click on the link that was left on your talk page to review what the wiki pages will look like, review any warnings and possibly correct dates if necessary, match families to existing families, etc. In addition to the link that Amelia mentioned, here's another link to review the earlier GEDCOM you uploaded. After you've reviewed a GEDCOM and you're happy with the results, you press "Import" to finalize the import.--Dallan 09:11, 10 April 2009 (EDT)
I just looked at the earlier GEDCOM, and under the first tab in the review program it says that 500 people will be excluded, so that explains the missing 500 people. I looked at several of the people who are excluded, and they do appear to be living.--Dallan 09:23, 10 April 2009 (EDT)

User with 2 accounts [8 April 2009]

Dallan,

I was wondering if you could take a look at User:Mymadsenwikipages and User:MyMadsenwikipages. There are many dupes within each and between the two.

Thank you!--Jennifer (JBS66) 12:29, 8 April 2009 (EDT)


Yes, it looks like they stopped using the second account and started using the new account. I'll ask them if I can delete the tree in their earlier account and remove the account.--Dallan 12:38, 8 April 2009 (EDT)


Wikipedia linking bug? [9 April 2009]

Hi Dallan,

I thought it might be a good way to draw attention to categories and other profiles to put some Wikipedia templates on category pages, like Category:U.S. Senators. The Wikipedia template program picks up the fact that that page draws content from the U.S. Senate page, but it doesn't add the colon to properly create the link. See, for example, this version of the John Bell template before I fixed it [5]. Can that be fixed, or should I stop adding content to category pages?--Amelia 17:19, 8 April 2009 (EDT)

I see what you're saying. I'll fix the bug - go ahead and keep putting wp templates in categories. Regarding the pages that were updated incorrectly in this refresh, can you tell me which categories you've added wikipedia templates to? That will help me figure out if I should have Taylor edit the templates manually to correct them or if I should restart the refresh after I fix the bug. If it's more than a couple hundred pages I'll restart the refresh.--Dallan 17:59, 8 April 2009 (EDT)
I think it was just Senators and Category:U.S. Secretaries of State until I saw if it worked. If you look at those categories, though, something else weird is going on. They have all sorts of templates and places in them now.--Amelia 18:02, 8 April 2009 (EDT)
Since I didn't put the colon before the category, Templates for wikipedia pages that reference the United States Senate article have been placed into that category. And if those templates are included in Place places (many of them are), then the Place pages are also placed in the category. Once the WP refresh is finished (hopefully tomorrow night) I'll have Taylor edit the templates and add the colon in front of the category. That should remove the templates and the place pages from the category.--Dallan 18:35, 8 April 2009 (EDT)
Another issue I'm just going to put here because it's mildly related. I've done some section linking, and it doesn't seem to work at times. I got a notification that this was changed this morning, and there's nothing there: Template:Wp-Cornelius P. Comegys-Early life and family ??--Amelia 11:15, 9 April 2009 (EDT)
I found the problem. Matching section headings was case-sensitive, so "Early life and family" didn't match "Early Life and Family". I've changed it to be case-insensitive, so the template should be updated correctly when we do the next full refresh in 3-4 months.--Dallan 15:23, 9 April 2009 (EDT)

Source bug on upload (I hope) [13 April 2009]

I posted on the sandbox because I thought this was just an annoying usability issue. But it got worse. I have a general 1900 census in my gedcom. Two of them, actually (for some unknown reason). The first time I went looking for the 1900 U.S. Census Population Schedule source, it was ridiculously hard to find. I tried about six different searches before finally getting it to come up on the second page of the search. It should be first when you use exact title searching. Otherwise, why have we gone through all this pain to name these things this way. But then when I got to the second source, I can't get the 1900 census source to come up at all. It's not there, no matter what I search for or what choices I use. I can only assume that the system has withheld it because I've used it already. But assuming that someone doesn't have two of the same source assumes entirely too much orderliness I think. Plus I know from painful experience that some systems generate duplicative sources in gedcoms just because of the way they are designed.

On a related note that comes up when I try the source matching, someday it would be nice to have a way to edit the sources on upload to be able to specify that my specific source is the 'record' in a general source, such as an article in NEHGR.--Amelia 20:27, 12 April 2009 (EDT)


Dallan, I believe Amelia is describing the same problem I was attempting to describe on your sandbox talk page. jillaine 21:09, 12 April 2009 (EDT)

I'll work on an idea for getting US sources to come up first in searches for the entire US later this afternoon (i.e., when you're searching for sources in a higher-level place, sources for that place should appear above sources for lower-level places). The system doesn't hide already-used sources so it's got to be part of the source-ranking problem. Saying that your GEDCOM source is 'record' in a general source is a good idea; I'll think more about that for the future.--Dallan 12:55, 13 April 2009 (EDT)

I changed how sources are ranked in search results. If you set Namespace to Source and search for sources for a high-level place, sources covering the entire place are ranked above sources covering lower-level places. So if you enter "United States" in the Place field for example, sources for the entire US should be ranked above county-level sources. Please let me know if this solves the problem.--Dallan 18:44, 13 April 2009 (EDT)

Sandbox - no email after signup

I tried to signup in sandbox last night. Too new to be in the Feb refresh. I never did get the email that has to be verified to let me do stuff. I'd like to test the gedcom before I do my big upload. The first little one created a mess. Used same username as production. --Jlanoux 10:04, 13 April 2009 (EDT)


I confirmed your email on the sandbox and we'll look into the problem of emails not being sent out. Thank you for letting me know!--Dallan 12:55, 13 April 2009 (EDT)


New GEDCOM import process -- Wow! (but with questions) [15 April 2009]

Boy, you really spent some time on this import machine. When you get all the bugs out, you ought to consider marketing it to Ancestry. God know, they need something like this. (Assuming they care. . . .) But I have some observations and questions, too.

Thanks :-)

People tab

I saw no problems here. I didn't deliberately exclude living people before setting up my GEDCOM file because I wanted to see what would happen. They all were marked both "Living" and "Excluded," which is good. "Living" should be an automatic "exclude."

You're the second person to ask for this. I plan to bring it up at the watercooler for a vote later this week.

Families tab

Example: John E. Cockrell & ___________. John had 5 wives and he appears in the families tab list 6 times -- once with each wife and once by himself, with no spouse. Several other males in the list with multiple spouses have the same problem -- but other males with multiple spouses do not.

Your GEDCOM includes a sixth family for John, one where he has 7 children but no wife. That's where the extra family is coming from. (The MRIN of the sixth family is F187 if that helps.)

Similar problem with a few females. Example: ___________ & Nancy Ellison. She had 3 husbands -- none of whom appear on the persons tab list, so I guess they weren't imported. I suspect that problem is on the outgoing end, not at WeRelate.

According to your GEDCOM, Nancy was married three times (MRIN F24, F244, and F245), but the GEDCOM contains a husband for only one of the marriages: Phillip Frakes, who does appear in the person tab list. The other two marriages don't list a husband, only a wife.

Warnings tab

Example: Theodore Adams & Mary J. Cockrell -- Husband has the same surname as his wife. Obviously, they don't, nor do they on the displayed family page. My import was ~600 persons. More than 200 have this message, none of them accurate that I can see. Do I really need to click "Edit" on each of these, then click "Save" so the review will show "Edited: Yes"?

Ah. My wife, looking over my shoulder, tells me this is because TMG creates an "alternate" name for a woman's married name. I may have to go back and re-do this GEDCOM without those married names. . . . And going to the last tab, it tells me I do have to go through each of those 200+ false warnings before I can complete the import! Any ideas? (At the least, there needs to be a filter, or trap, or whatever, for this.)
Yes, this is a bug. It looks like we should be comparing just the primary husband and wife names. Thank-you for pointing it out! No, you don't need to click Edit on these. Normally, the uploader wants you to click on each of the warnings to read them, but given that these are all spurious, please let me know when you're ready to import and I'll mark it "ready to import". Admins can import gedcom's despite the warnings. We've found several bugs with the warnings this week; hopefully we'll get them straightened out by the weekend.

Places tab

Several place names with included street addresses in my database were pushed to the top as "not matched." No surprise -- so I did a search in the displayed page in the lower pane, removing the street address info -- assuming that allow them to match up. But on all of them I got "No match." I know perfectly well that "Toledo, Lucas County, Ohio" is in the list of place names. So is "Indianapolis, Marion County, Indiana," and so is "Dallas, Dallas County, Texas." The search engine finds them when I'm in another iteration of the website -- but not through the Search function on the GEDCOM review, apparently.

Another bug - thanks for pointing it out. I'll take care of this tomorrow.

Also, I have a habit of abbreviating township as "Twp" in my database and that's what was imported. But the process couldn't match, e.g., "Cass Twp, Dubois County, Illinois" with "Cass Township, Dubois County, Illinois." Since the "twp" abbreviation is pretty common, you might want to include a conversion for it in the import process.

Twp is indeed a problem. Hopefully I'll fix this tomorrow. But we have another problem as well: there are very few township pages in the wiki. I need to add them but it will be awhile.

Most of my other unmatched places were cemeteries, which is not unexpected.


Sources tab

I have 120+ sources in the file, none of them with matches. This doesn't surprise me, actually, for a number of reasons that have been under discussion. I'll fix those as I go through the new pages, tidying them up; some of the sources I'm altering anyway, to suit the purposes of WeRelate, which isn't always identical to the reasons for citing a source in my database.

Match Families tab

It says: "Match Related: Match the family and also related families up and down the tree. Related families (spouses of children, parents of husband and wife) where all family members are good matches are matched automatically." It's the "automatically" part of this that gives me pause. I don't think I want anything happening in a merge that I can't approve of individually, step by step. (I don't like surprises.)

It's meant to give you pause :-). Some people upload very large GEDCOM's that contain a lot of people they didn't research personally. I think the automatic matching will be a benefit to them, but for people you've researched personally I assume you'll want to be more careful. (The automatic matching is pretty strict though -- related families that aren't a pretty good match are added instead to the list of potential matches for your review.)

If I choose to deal with all my potential duplicates later, . . . can I just select "Not a Match"? Or will that disallow those newly created (and now duplicated) pages from ever showing up on my list of duplicates under My Relate?

You can, but you've got to promise that you will indeed merge them later :-). If they end up having the same title as an existing family (so first given name and surname of husband and wife match) then they'll show up on your list of duplicates. The GEDCOM family-matcher is more encompassing -- it can identify matching families where the husband or wife might have a different name if the birthdates match for example. It's generally faster to get them out of the way in the GEDCOM uploader. The duplicates list is refreshed only once a day, so if you match one family today, their relatives might not show up in the duplicates list until tomorrow.

In the detailed match-up process, I note that a certain person's "S1" source information is completely different in my GEDCOM from the "S1" source information in the family I'm matching with. That is, my S1 refers to a book, their S1 refers to a census listing. This isn't surprising, since my research wasn't identical to some other genealogist, with identical sources found and cited. On the one hand, I don't want to lose either of these sources -- but I don't want one version of the S1 text to overwrite the other, either. And I don't know about having two separate "Source 1" entities. The help note says it will renumber sources and notes -- but this isn't like an alternate birthdate. Any thoughts here?

Just mark the checkbox next to your S1 source in the update screen. When it gets copied to the existing page it will be renumbered as S2 (and any references to it will be renumbered as well). That should be ok; the source numbers don't have any special significance. They're just used to distinguish one source reference from another.

--Mike (mksmith) 18:07, 14 April 2009 (EDT)

--Dallan 19:27, 14 April 2009 (EDT)
Going back to the Warnings tab, I found that all that was required was to touch the items in the list. I didn't have to open and edit each one. Most of the remaining (non-married name) warnings concerned things like "events after death" (mostly notes on probate), children born more than 25 years after marriage (it happens, esp on the frontier), and marriages in which the bride was 15 yrs old (ditto). I checked the others against my TMG database and found no errors in the original, so I believe those were GEDCOM-construction problems -- like the missing spouses you noted. So the import has into the queue for approval, and I'll fix the noted errors by hand by hand in the created pages. And I'll merge any duplicates that show up, promise! I also have a lot of sources to work on. --Mike (mksmith) 22:19, 14 April 2009 (EDT)
I've cleared it for import. It should finish shortly. Also, I'll change it to ignore probate events after death, increase the "children born after marriage" warning to 30 years, and decrease the "wife marriage age" to 12 years. That will hopefully result in fewer false warnings in the future. It looks like you've merged all of the potential duplicates that the system found, so you should be all set! Thank you for working through all of this and giving feedback!--Dallan 22:45, 14 April 2009 (EDT)

Dallan---
      Sitting down to begin working my way through the new GEDCOM import, I know there's a lot of tidying-up to do, lots of pure housekeeping. And I'm looking at FTE and considering whether it would be easier to do this structurally -- starting at the root person and working my way up and down each branch -- or just alphabetically -- which means starting with someone very minor, whose name just happens to begin with "A". It occurs to me it would be useful to have a worklist/checklist by which I could track my progress and not skip anything. What would you think about adding an option, I suppose in the import process itself, to produce such a list and copy it to the user's talk page? Or maybe create a special "worklist" page as a user subpage, to avoid making a mess? Just a text list of page-names that I could mark up as I go along -- or maybe such a list in linked format, so you can just click on each one. --Mike (mksmith) 10:04, 15 April 2009 (EDT)

I'll add this to my todo list -- you're the second person to ask for it. And I'll email you a way to get a "wiki" list of the pages in your tree.--Dallan 10:45, 15 April 2009 (EDT)

Jillaine's recent GEDCOM import in sandbox [15 April 2009]

Dallan, I received a message that there were problems importing my most recent GEDCOM. When you have a chance, could you look at it (the GEDCOM) and explain why it didn't work? I'm just curious. I had gone through all the steps in the review process (again: SO NICELY PROGRAMMED, DALLAN AND SOLVEIG!), and only received the email message a couple of days after competing the review process. Thanks!

-- Jillaine jillaine 08:05, 15 April 2009 (EDT)

Thanks for letting me know. It looks like there was a problem generating the wiki pages. Nathan, who wrote that part of the program, is out of town until Friday. I've asked him to look into it as soon as he returns.

Regarding the email message, are you saying that you didn't get it until today? It should have been sent on the 13th. If you didn't receive it until today I may have to change how emails are sent from the Sandbox.--Dallan 10:45, 15 April 2009 (EDT)


ooops sorry! [15 April 2009]

Sorry about adding (incorrectly) that Category to the Cite template. DUH! My apologies and thanks for fixing it. jillaine 11:01, 15 April 2009 (EDT)


No problem at all! noinclude is a pretty advanced feature :-) Dallan 11:19, 15 April 2009 (EDT)


Gedcoms ready to import [18 April 2009]

Thanks for the note Dallan, I want to fully review along with my files before I import them. By the way, I really like the new gedcom review process. At first I thought it would be just a time-consuming process, but it has helped me spot some discrepancies in my files that I've been able to clean up before the import process. Hopefully, it will make sure that only "quality information" is added to the site.

Best regards,

Jim:)--Delijim 11:28, 15 April 2009 (EDT)


Hi Dallan, I've reviewed the gedcoms and they're ready to be imported. If you can get to them at your earliest convenience, I'd appreciate it. I still need to go back and add some Augusta County Settler notations and links.

Thanks:)--Delijim 10:00, 18 April 2009 (EDT)

Amelia found a serious bug in the current GEDCOM uploader :-(. They'll be uploaded soon, hopefully later today.--Dallan 16:45, 18 April 2009 (EDT)

Source Discussion [15 April 2009]

Dallan, is it possible for you to stop by my userpage some time and see the discussion on sources. I haven't been here long enough to know the history, and I believe your suggestions would be most helpful. Thank you!--Jennifer (volunteer admin) 15:06, 15 April 2009 (EDT)

I'll look at it first thing in the morning--Dallan 00:19, 16 April 2009 (EDT)

Merging [15 April 2009]

Hi Dallan, Need your opinion on merging a page. See User:Leo Bijl talk page, last entry regarding merging. --Beth 20:09, 15 April 2009 (EDT)

I'll look at it first thing in the morning--Dallan 00:19, 16 April 2009 (EDT)

GEDCOM review [15 April 2009]

Dallan, I wanted to let you know that I didn't intent to review the recently uploaded GEDCOM! I went in to see what the review process looks like - didn't know I would be considered the official reviewer! Sorry--Jennifer (volunteer admin) 20:34, 15 April 2009 (EDT)


The program assumes that if you're the first to look at it, it's yours :-). Don't worry -- this is how I hope things will work eventually, but not right away. Feel free to look at any gedcom's you want -- it's a good idea to "get the hang of it". I'll look at them as well.

This week I'm mainly looking at warnings (we need to relax a number of the things we warn about), place matching (I made some improvements today), and seeing how the family matching algorithm is doing.--Dallan 00:19, 16 April 2009 (EDT)


my gedcom [16 April 2009]

Hi, Dallan.

Thanks for the message. I completely understand your comments about consistency in place names. I emailed your wife earlier in the week to say hello and introduce myself. I'm the newest reference librarian for the Genealogy Center of the Allen County Public Library. I uploaded this small gedcom as an experiment to get familiar with how WeRelate works. I really am enjoying working with the website!

Dawne--Dawne 14:45, 16 April 2009 (EDT)


Hi Dawne, Thank-you for the kind words; I hope you continue to enjoy it!--Dallan 15:01, 16 April 2009 (EDT)


Suggested Addition for Drop-Down "Events and Facts" List [17 April 2009]

I was adding a marriage bond event to a family page and when I check the drop-down list of available events I found that while there are five marriage-related events on the list, there's no "Marriage Bond" -- which is probably the most commonly needed for the 19th century U.S., after the marriage ceremony itself and the marriage license. --Mike (mksmith) 21:32, 16 April 2009 (EDT)

Good point. I added it.--Dallan 16:45, 18 April 2009 (EDT)

Questionable linking page? [17 April 2009]

Dallan, I just came across this while looking at the "Special Projects" list. There's a page under "Louisiana Ancestors" that links to Louisiana_Genealogy_Blogs. I don't know who's running this, but it's been showing up at a lot of genealogy websites out there. Someone's been blanketing the genealogical world with this. I don't know if it's advertising for some commercial thing, but it rather looks like it -- and if it is, it's questionable whether it has any business here. . . . --Mike (mksmith) 15:00, 17 April 2009 (EDT)

Thanks. I don't mind someone promoting their blog on their user page. But it's unnecessary to to promote it on five pages. We've removed the other references.--Dallan 16:45, 18 April 2009 (EDT)

Interesting mysource issues [18 April 2009]

Hi Dallan,

I'm cleaning up my Morrow gedcom today, and I came across what seem to be two consistent bugs: 1) any MySources I left are linked on the pages as Sources (i.e. Source:Amelia.Gerlicher/etc.) even though the MySources were uploaded and pages created; and 2) all of the MySources have been created with a (1) at the end - even if they do not duplicate sources I uploaded previously.

This also raised the issue of the need to be able to merge MySources in the current gedcom with previously uploaded MySources before the update function will work.

--Amelia 15:04, 18 April 2009 (EDT)

THANK YOU for reporting this! I've taken down the GEDCOM uploader until we get it fixed - hopefully later today. I'm sorry for the trouble it must have caused. I have children who are good at repetitive data correction; please send me an email if you'd like them to correct the pages.--Dallan 16:45, 18 April 2009 (EDT)
We figured out how to automatically correct the pages. I'll send an email with more information.--Dallan 18:21, 18 April 2009 (EDT)
I noticed what I think are the same two bugs in cleaning up my GEDCOM import -- MySources (with names like "Mksmith/Frakes Family Bible") that showed "Source" on the drop-down. And they all have "(1)" at the end. The difference is that I didn't realize this was an error! :) --Mike (mksmith) 19:44, 18 April 2009 (EDT)
Yes, it turns out this happened with about six gedcom's, yours included :-(. I sent you an email as well.--Dallan 22:27, 18 April 2009 (EDT)

Difficultly - from MySource to Source [20 April 2009]

Hello Dallan,

Have you made some recent changes?

I used to be able to manually edit easily (clean up) the MySources to Sources.

Now when I manually do so I am stuck in only having access to MySources. I can't seem to switch to Sources.

I just tried editing a MySource and entering #redirect [[Source:title]] in the big text box at the bottom to redirect it to a Source page and it seemed to work ok. Can you tell me more about what kind of problem you're having? I'll try to fix it.


Thanks for the fast response.
My computer code skills are low. I was trying to change them the lazy way. On the person Page I choose the MySource, change it to (text only), then choose Source. Then I go to the pop up screen were you are given chooses. Only MySource come up.
I see what you're saying! Yes, that's a bug. I'll fix it later today or first thing tomorrow.--Dallan 12:13, 19 April 2009 (EDT)
I think that bug has been in there for a long time. I'm glad you reported it. It only appears when you're switching for Source and MySource or vice-versa. Anyway, it's finally fixed now.--Dallan 20:38, 20 April 2009 (EDT)
I will stop for now to change the MySource to Source for now. I will concentrate on entering other Sources for now. I also need to start writing little bios, etc. --DFree 01:44, 19 April 2009 (EDT)


I know you are very busy.

Could we have the ability when manually editing have the page sort by date?

If there is a trick please share, otherwise I have to manually edit by date.

Example would be to move the 1920 census after the 1910 census manually.

Adding buttons so you can move events up and down is on my todo list.


Great!! --DFree 01:44, 19 April 2009 (EDT)

Would it be better to delete my GEDCOM and reinstall? I would prefer not to delete my GEDCOM.


I am pretty connected so I will not delete my GEDCOM. --DFree 01:44, 19 April 2009 (EDT)

That seems to be what everyone is doing. Thanks for all your hard work.

This works if your tree doesn't have anyone else interested in it, but causes some problems if others have linked into your tree already. If the problem you're worried about is cleaning up the MySources, I wouldn't worry about it. Eventually we'll introduce tools to make matching MySources to Sources much easier. (That is, we'll have something like a "Find duplicates" for MySources.)


Great!! Looking forward to the Source Tool --DFree 01:44, 19 April 2009 (EDT)

Debbie Freeman --DFree 23:38, 18 April 2009 (EDT)

I hope things are going well.--Dallan 00:27, 19 April 2009 (EDT)


Thanks for asking. It is slowly getting better.
Now that spring is here I am opening the boxes finally !!

Debbie Freeman --DFree 01:44, 19 April 2009 (EDT)


Searching for Geographic Sources [3 May 2009]

Hello Dallan, I have a question regarding searching for geographic type sources. Say I want to find a census for Windham County, Connecticut. Doing an Exact Match search, I type Windham into the place field, and census schedule into the title field (you get the same results when in the keyword field). No matches are returned. I need to put all the text into the keyword field for it to find what I want. Does this have something to do with how we are searching/entering the title field, (which ties into this conversation)? The rules on the Naming Conventions page says: "In the "Title" field, enter: XXXX U.S. Census Population Schedule".--Volunteer Admin - Jennifer (JBS66) 12:03, 19 April 2009 (EDT)


You can provide the fully-specified place name, i.e., Windham,Connecticut,United States with your original search string of "census schedule" and it appears to work. There are many towns/counties named Windham. --Jrich 13:43, 20 April 2009 (EDT)


Well, very interesting! I would have thought that searching for a more general term of Windham (or even just Connecticut) would have returned more results. Yes, this does work, and so does just Windham, Connecticut. Does that mean we can officially not put the geographic area in the title field??? Also, it might be less confusing if that field were titled 'document title' or something similar. I think users are getting it confused with page title. Thanks for your help!--Jennifer (JBS66) 13:53, 20 April 2009 (EDT)

The search engine tries to figure out which place you're searching for, so it can include lower-level places in the search results (i.e., sources for towns in Windham), and also sources for higher-level places (e.g., all of Connecticut) when you're not doing an exact-match search. So yes, it's best to enter enough levels so that the search engine can figure out which Windham you're talking about.
As for putting the geographic area in the Source page title, we need it in order to make Source page titles unique. I like the idea of changing the name of the "title" field inside the Source page though. What about "Book/Collection title"?

Dallan, I think you should consider fixing the source titles before you leave Beta (my assumption that this is a summer task, and release is tentatively planned for late spring?) Based on what I have learned lately about searching sources, I was going to try and enhance the Help page, but the first example I tried (which I thought was one of the few books that followed the rules) made a liar out of me so I aborted.

Source:Savage, James. A Genealogical Dictionary of the First Settlers of New England has a good page title, but it was done by setting the book title field to exactly that string, i.e., including the author's name in the book title, while the author field remains empty. I expected that when I searched for "Savage, James" in the title field, I would not find this, but I did, while putting Savage, James in the author field does not find it.

The words are there saying that the title criteria for searching is not the page title, but the significance is easy to miss. I was going to add some examples that would hopefully show that search works using the component fields, and not using the page title that gets displayed in the source citations. I was also going to add quick examples showing the use of quoting and wildcards and exact search, all of which solve problems I have been complaining about, but never noticed because there isn't much need for them when searching families and persons. It seems to me that their use is critical for effective searching of sources. Getting back tens of thousands of results is not helpful. If I judge various comments here, there are more than one or two other comments indicating others are mis-interpreting how this works like I did, so it may not be just my unique pattern of thinking. :-)

One of my assumptions is that the first thing new users will do is upload a GEDCOM, which now requires being able to search for sources. So it seemed like it might be worthwhile to add examples to the Help page. But, it seems pointless to write the help when nothing seems to follow any explanation I was going to try and make (assuming I am not just confused, in which case, this is a plea for somebody else to write said help). --Jrich 15:02, 20 April 2009 (EDT)

I agree that we have to fix up sources before coming out of beta. I plan to work on sources over the summer, and don't expect to come out of beta until late in the year. The problem with the Savage source you mention was the James Savage was not listed in the "author" field in the Source, but only listed in the title field. I corrected this, but I agree it's an understandable mistake due to the confusion between the Source page title and the Book/Collection title (or Document/Collection title? or whatever we choose to call it).
What if I make the following changes to make searching sources more straightforward?
  • When searching Sources, we change the "Title" field on the search form to "Book/Collection title", since this is the field being searched (along with subtitle).
  • Source search results include the "Book/Collection title" and "Author" fields.
  • I could also add an option to sort exact-match search results by shortest-title-first, if that would help.
Would you keep pointing out examples where source searching doesn't work? Examples are really helpful.
The new GEDCOM upload doesn't require matching sources; it's just optional. I don't plan on ever requiring people to match their GEDCOM sources to Source page, and I'm also not going to emphasize matching sources until we make source searching better and we add source matching to the GEDCOM upload so that you get a list of likely sources.
--Dallan 23:57, 20 April 2009 (EDT)

Thanks for the update on the schedule. Glad cleaning up sources will get done before release. It will undoubtedly avoid lots of problems.

I agree that strictly speaking, converting to Sources is not required, but the GEDCOM upload process leads people right to it, and it is desirable and should be encouraged when possible. So I feel it would be good to make this process as intuitive as possible.

In my use of WeRelate, I have mostly cited already-created sources, and I mentioned Savage's book just because of the ones I was familiar with, it appeared from the page title to follow the rules. It was of course ironic, perhaps disappointing, then to find out it was still not correctly done. It was not, in and of itself, meant to be an issue, other than as an illustration that almost nothing is done correctly at the current time, so it is very hard to understand how to use the system.

I am not sure sorting by length, per se, is necessary. It seems that this would solve one particular problem. But as a general approach, it seems like the goal should be to have fewer results returned for a given search.

The current search is pretty good if you know just little bits about a source. For example, I have seen manuscripts that cite a source called "Harkness Gen" and in reality mean a source whose title is "Descendants of Adam Harkness". They usually say so somewhere, but somebody might forget to record this. But, I wonder if you could trust someone's identification of a source if all they knew was Harkness? Generally, if you are converting MySource to Source, or citing a source you worked with in a library, you probably have a reasonable approximation of the exact title and probably the author too. You probably know the county a census came from. Etc. You probably don't need a very fuzzy search.

In an example I mentioned previously, I was searching for a book actually titled Gaskill Genealogy (with some subtitle that I didn't record). Entering that in the title field of WeRelate brought back 17 results. (It appears from external observation that this is all the sources that have the word Gaskill in the title and Genealogy was silently ignored because it is too common of a word. So a search criteria of Gaskill Genealogy is entirely identical to a search criteria of Gaskill, notably, that it can only match 1 word and so a title containing both doesn't get ordered to the front of the list.)

Now 17 isn't a bad number, but Smith Genealogy returns 2547 results. A quick scan of the 17 Gaskill results suggests that the ability to require the title contain both of the words Gaskill and Genealogy in any order would limit the results to 4, and the ability to require the title contain the exact phrase "Gaskill Genealogy" would limit the results to 2. Similar reductions on the Smith Genealogy case would be significant.

When I ask for History of Rye in the title, I get several titles. If I put Rye, New Hampshire in the place field, I get none. (I could not remember the county Rye was in, but was trying to distinguish from books about Rye, NY.) Perhaps this place field needs the auto-complete like what we see on the Person pages, so I make sure the place name is in the format expected by the search. (Also see below the discussion on OR versus AND.)

When I ask for title -1900 census- with place of Texas, United States, I also get entries for other states (over 75,000 results returned). If I put exact match, I only get 2 items whose place is exactly the whole state. I thought exact match meant the entry will have all the parts I asked for, but could have more, and I therefore expected all the counties of Texas to be returned to be consistent with how it works on the Title field. An exact match to a title of Gaskill only means Gaskill will be part of the title, not that it will be the whole title. So an exact match to a place of Texas should match all the places in Texas, not just Texas itself? (Of course, since 75,000 returned titles is basically useless, it would just as effective to make the inexact match require the place be in Texas, while the exact match could then mean that it be exactly the state of Texas. But this is not entirely consistent with how it is used with title and other fields.)

If criteria are entered in multiple fields, should they be treated as an OR and an AND operation? With the goal of limiting results to smaller sets, I would argue they should be an AND operation. If I ask for title of Genealogical Dictionary and author of Savage, it seems pretty clear my intent is to find James Savage's Genealogical Dictionary and not a "History of the hundred of Carhampton" (by Savage) and not the "Genealogical Dictionary of Maine and New Hampshire" (by Libby, Noyes, et al., but not by Savage) or the 18,000+ other items that are returned if you actually run this search.

An OR of two criteria will always be a least as many results, if not more, than a single item search, which seems very counter-intuitve: specify more criteria but you get a less discriminating search. (Adding author Savage to the above search caused 100 additional entries to be returned than just searching for title of Genealogical Dictionary.)

How are items ordered if they score the same? If I search for a title of Genealogical Dictionary, there are 18611 items returned. The first 65 contain both words in the title. The rest presumably have one or the other words there. The 66th contains the word genealogical, title starts with I; the 67th contains the word dictionary, title starts with K; the 68th contains the word genealogical, title starts with G; the 69th contains the word dictionary, title starts with H. There is no discernible pattern. If I haven't given up yet paging through this list, I would at that point. It might be nice if the user was offered some control over the sort order (alphabetical, by place, by author)?

--Jrich 14:51, 21 April 2009 (EDT)


Do a little dance! Woo hooo!!!! Rock on!!! Yes! Thanks! YaHOOOO.... (Yes, I'd like the search engine both to search and display NOT by *page* title but by Book/Article title and author separately! Yes!!! Weee!!! LOVE YOU!!! -- Jillaine jillaine 16:16, 21 April 2009 (EDT)

It already does appear to search by the component fields, though the order of display seems only based on the overall number of keyword that match match, with no discernible tie breaker. And it is very hard to tell that search works by the component fields because so many sources do not follow the rules, and do not have the right data in the right place. I have probably put my foot in my mouth many times as I have struggled to learn how the search works, and complained about this or that, having been misled by such inconsistencies in how the source page was created. Thank goodness Dallan is so patient.

The previous posting of mine kind of grew without plan, as I was trying different cases while I wrote it, trying not to put my foot in my mouth again, and as a result, it was undergoing adjustments as I wrote it. Thinking about this, I thought I should restate my conclusion in a more focused manner.

What I was arguing is that the default search, basically a Match Any keyword, is as good as useless and non-intuitive. It returns too many results. Why would anybody type in more keywords than they need? If I don't want just sources that match both Genealogical and Dictionary, why wouldn't I type in just Genealogical or just Dictionary? The only reason to type in both is that I think both are pertinent and I expect it will yield a narrower result than using just one, instead of the opposite case that happens with the default search.

The exact match, which might be better titled Match All keywords, could probably be made into the default search once the titles get cleaned up. At the current time, it is a little funky because author is not always in the author name field where it belongs, and place may be in the document title field where it doesn't belong, etc., etc.

It might be nice to add a more exacting match, what I would call Match Phrase, comparable to using double quotes around a search term in google. If the software could support it reasonably, of course. However, I am not sure this idea makes sense except in the title field.

The criteria of exact or inexact (to use the current language) may actually have semantics that change based on each different search fields and people may want to have the ability to choose between these two modes on each field independently. Building on an example in my previous posting, match inexact on place to accept any place in the state of Texas, but match exactly on title for 1900 census, to end up with a list of all known sources pertaining to the 1900 census in the state of Texas.

When two different criteria are specified in a single search, such as book title and place, it seems intuitive that they should be combined by AND (i.e. the intersection of the two sets) for reasons very similar to the argument why the inexact search is useless. Of course, if feasible, the user could be given a choice between AND and OR, but that would make understanding search more complicated.

--Jrich 19:38, 21 April 2009 (EDT)


Thank-you for your great comments! I'll try to address each point.

  • I'll change the indexer so that it doesn't ignore "genealogy" any more. It takes about three weeks to index all sources, but entering genealogy in the title should work about three weeks from now.
  • The reason you're not seeing results when you enter Rye, NH in the place field is because not all of the sources for Rye have Rye, NH listed in the place field. It's for reasons like this (author missing, place missing in source pages) that I think OR (exact match checkbox not checked) needs to be the default until the sources are in better shape. With OR if a field is missing in the source at least it still comes up in the results. And sources that match all criteria should appear at the top of the list.
  • If the exact match checkbox is not checked, sources are ordered so that sources that match more search criteria appear above sources that match fewer criteria. But if several sources match the same number of criteria they're returned in a random order. It's possible for me to include some other factor to influence ordering in this case, but what would it be? Eventually it ought to be the number of pages that link to the source, similar to what Google does. I'm hoping we can use this sometime next year, once more people start linking to sources. But what should it be in the meantime? Title length? Something else?
  • Recently (2-3 weeks ago) I modified the behavior of the Place field when searching sources with exact match, so that only sources covering exactly the place you specified would be returned. I'm not crazy about this change because it means that the Place field behaves differently in exact-match source sources than in other searches, but it addressed a problem that was brought up earlier: When I'm searching for the 1900 US Census source, I don't want to be inundated with hundreds of county sources. I could solve this problem in a different way by giving users the option to sort exact-match search results by title length (and making that the default sort for exact-match source searches). And I'm open to other suggestions.
  • When you search by author and title, you're currently searching the author and title fields inside the Source page, so that part works like you want. The problem is those fields aren't displayed in the search results. I'll add them. It will take three weeks for the sources to be re-indexed so that you see them for every source, but they'll start showing up tomorrow.
  • Some of what you want to do can be done using advanced search techniques described here. For example, if I want to require that every result contain the phrase "History of Rye", I would enter +Title:"history of rye" into the keywords field. This technique gives you access to plus (required), minus (prohibited), wildcards, and phrase searching. You can do mixed AND/OR searching for example by entering optional criteria into the regular fields and entering required criteria into the keywords field preceded by plus signs.

I hope this addresses the above issues. I'd love to get more thoughts on this, especially how to sort sources in normal and exact search so that the one you intended generally appears at the top.--Dallan 13:39, 24 April 2009 (EDT)


I just updated indexing and added a new search field called "Book/Collection title", which is separate from "Page title". This new field will eventually let you search the contents of the title field on the Source page. (BTW, I've renamed the title field on the Source page to Book/Collection title; I'm open to suggestions for a better name here.) The "Page title" field on the search form will eventually let you search on the contents of the Source page title. And search results will soon display both Author and Book/Collection title. We're in a state of flux for the next week while the source pages get re-indexed. But I've sped up the indexing so that it should be done in one week instead of three.--Dallan 17:25, 24 April 2009 (EDT)


What worries me about the use of "Collection" in the Book title is that collection could come awfully close to Repository. Why not Book/Article Title? Or rather, what do you mean by "Collection"? jillaine 21:06, 24 April 2009 (EDT)
What I'm trying to include in the label are government/church records, like the 1880 US Census, or a vital records collection. I'd be happy to replace "Book/Collection title" with a better label.--Dallan 13:49, 28 April 2009 (EDT)

Dallan, I added some examples of how search seems to be broken as far as finding census records goes. See WeRelate talk:Watercooler#New_Search_.5B29_April_2009.5D--Amelia 00:03, 2 May 2009 (EDT)


Hi Amelia, I read your comments today. I just wanted to let you know that I'll look into it tomorrow or Monday. Thank you for pointing out the problems.--Dallan 02:14, 3 May 2009 (EDT)


Cite Template [15 June 2009]

Beth and I can't figure out the strange behavior of the Cite template that is used on this week's featured page. When either of us try to use it, we get a citation, then it forces a line break. It doesn't stay on one line like on the featured page. Any ideas? If we can't get this to work, we'll need to take the text out of the help page. Thank you!--Jennifer (JBS66) 11:58, 20 April 2009 (EDT)

After reading this, I went and experimented on one of my own pages where I've begun using the template. The forced line feed appears there, too. (I hadn't been aware of the problem because I've been putting the template at the deliberate end of a line, usually leading in to a census listing.) I've re-examined the way the Julia Stuck page is put together and I see nothing different there. No extra spaces, no extraneous characters, nothing. Very strange. --Mike (mksmith) 16:06, 20 April 2009 (EDT)

I figured it out! There was a <noinclude> tag in the template that was forcing a new line. I moved that up, removing any extra spaces. Can someone try it out again to double check?--Jennifer (JBS66) 16:15, 20 April 2009 (EDT)

That was it. I had moved the category that had been added inside the noinclude tags, but I neglected to put the whole thing on one line. Thanks for fixing it!--Dallan 23:57, 20 April 2009 (EDT)

This would probably explain what's happening with the citation needed template as well, which does the same thing. I'll go look and fix if so. jillaine 16:18, 21 April 2009 (EDT)

Beth doesn't have a clue what the noinclude line is but glad y'all fixed it. Thanks.--Beth 20:09, 23 April 2009 (EDT)

Just FYI, noinclude is a rather advanced template tag that surrounds text you don't want included when the template is included in other pages. You can surround a category link with the noinclude tag so that the template goes into the category, but none of the pages that include the template go into the category.--Dallan 13:43, 24 April 2009 (EDT)

Question about this: If the sources are renumbered (say by deleting a useless one, or a merging), does the reference change? And if not, can we perhaps fix that or not encourage the use of this template? Because the reference itself does not point to the source, it has the potential to easily become unhinged from the correct source. <ref>[[#S1|Name of Source]]</ref> is not that much harder and much safer.--Amelia 14:33, 25 April 2009 (EDT)


I just modified the cite template so you can pass an optional second parameter that is the link text; e.g., {{cite|S1|1880 Census}}. Also, cite templates as well as [[#S1|1880 Census]] are both renumbered correctly in merges and when you're editing a page and adding/removing sources. (Auto-renumbering has been in place for several months.) Would someone please update the help pages in the appropriate places to tell people about the new cite parameter?--Dallan 13:44, 28 April 2009 (EDT)


Beth added the cite template info to the Help:Formatting page. I'll go through and edit the specifics now that this is squared away. The only thing I noticed when I tested removing a source cited this way, is that you are left with an unlinked citation of [[#|]] (upon preview)--Jennifer (JBS66) 13:53, 28 April 2009 (EDT)


I forgot about removing references to deleted citations. That's been fixed now, both for merging and editing.--Dallan 14:54, 28 April 2009 (EDT)


Thanks to all for sorting this out; delighted to have this new template. One question, do I need to add the optional second parameter when using this template or will the sources sort correctly after changes made to the sources without using the second parameter? --Beth 21:08, 28 April 2009 (EDT)


You don't have to add the optional second parameter. If you don't add it, the citation number (e.g., S1) appears as the link text. By the way, this template also works for referencing notes and images, and those references get automatically-renumbered as well.--Dallan 22:10, 28 April 2009 (EDT)


Regarding the cite template's use on pages such as Place:Woodend, Birse, Aberdeenshire, Scotland (that don't have true numbered sources), it appears that automatic renumbering doesn't work. Should we eliminate that use from the help page?--Jennifer (JBS66) 06:32, 29 April 2009 (EDT)


The automatic renumbering only works for source citation, image, and note references on Person and Family pages. For references on other pages, you're better off using ref and references tags. Here's more information. If you would update the help page I'd appreciate it - thanks!--Dallan 11:07, 1 May 2009 (EDT)


Att: Dallan - Margaret Williams B 1876 [23 April 2009]

Hello Dallan,

I am asking your permission to edit one of your pages. I think I found an example of a misplacing person.

You and User Kquass (seems to be gone now) have a Margaret Williams B 1876 as daughter of Louis Williams and Margaret (1) that looks correct to me.

What does not look correct is that she is also connected to Benjamin Williams B @ 1790 & Rosea Lane B @ 1798 (1).

User Hgosnell (who also seems to be gone) had this Williams/Lane couple, we share Benjamin's father, also named Benjamin Williams Sr.

I am watching Hgosnell Williams pages, and adding sources, etc. Thanks --DFree 22:54, 20 April 2009 (EDT)


Hi, thanks for pointing this out. I removed Williams/Lane as parents from Margaret. Margaret's gedcom was the first or second gedcom uploaded to the website, and it looks like it had some problems. I'll have to go back and see if there are any more incorrect linkages with it.--Dallan 23:11, 20 April 2009 (EDT)

Thanks Dallan!! I am sorry to add to your ever growing list. I would edit it myself, but I can't. I looked a little closer on the Williams/Lane Family Page - the children are pretty mixed up. It is a mess sorry to say. There are definitely incorrect children. --DFree 23:50, 20 April 2009 (EDT)
I'll see if any of the other children in that family belong in the other family instead. (And you're welcome to edit it if you want to; when you say you can't is it a technical problem?) Too bad things are mixed up though; otherwise we'd be related!--Dallan 23:57, 20 April 2009 (EDT)

Thanks for the help. I have started to edit the Williams/Lane family. As for the edit problem. I got a little mixed up. I had in mind that one User could not edit another Users GEDCOM. I also like to be polite. I rarely have a response though. I am sure somehow we are related.

I think there's a between fine line between bold and polite. I'd err on the side of bold. :-)--Dallan 18:16, 23 April 2009 (EDT)
Ok, I think I figured out what happened. It looks as though when you added the marriage information to the family, you changed the titles of the child pages from "... Willams" to "... Williams" (adding the i). The problem is, these page titles were already assigned to children in other families. That's how the children from the other families became linked into the Williams/Lane family. So I changed it back, putting the old "... Willams" titles back for the children.

Oops. Sorry about that. I had not meant to do that.


The best way to change a page title is to first navigate to the page, then click on the Rename link at the upper left corner of the page, then change the name there (leave off the index number - an unused one will be assigned). You'll need to do that for each child if you want to rename the pages. Please let me know if you have any questions.--Dallan 00:26, 21 April 2009 (EDT)

Data entry/editing bug [24 April 2009]

Got a weird one. I'm working (still) on cleaning up my GEDCOM import. One of my Person pages is for a woman who was married four times, and because TMG makes each married name into an "alternate" name, this lady had a primary name and four alternates. I went to click the "Remove" button for each of the alternates, since they serve no purpose here -- but it (the system) would only allow me to perform one "Remove." The "Remove" buttons for the other alternate names didn't respond at all. So I saved the page, reopened the edit, . . . and it allowed me to remove one more alternate name -- but only one. I had to go through this routine twice more to get rid of all the alternate names. I'm pretty sure this behavior is contrary to design. . . . --Mike (mksmith) 21:24, 21 April 2009 (EDT)

Thanks for pointing that out! It's fixed now.--Dallan 13:44, 24 April 2009 (EDT)

Duplicate User [28 April 2009]

Dallan, User 1 and User 2 appear to be the same user. Is there overlap between the two accounts and should we do anything about this? Thank you!--Jennifer (JBS66) 12:11, 26 April 2009 (EDT)


Yes, they're the same individual. I'll send them an email about removing one of their accounts. Thanks for pointing it out.--Dallan 13:44, 28 April 2009 (EDT)


Data entry design thoughts [2 May 2009]

Discussion has been moved to WeRelate talk:Data entry design.


Place matching anomaly [1 May 2009]

I just found a family page that shows a weird error in the place matching algorithm:

http://www.werelate.org/wiki/Family:Corneille_De_Haen_and_Catherina_Van_Havre_(6)

As you can see, both husband and wife were christened in Antwerpen (Belgium), but on the family page that does not refer to the proper country. The husband died in Antwerpen too, and Antwerpen is also the place where they married, and those places do refer to the proper country.

Moreover, when I click on the husband's or wife's name to check their person page, I find the Antwerpen mentioned there as place of christening also refers to the proper country. Isn't that weird?

In my GEDCOM all instances of Antwerpen are just that, i.e. no country specified.

--Enno 12:07, 29 April 2009 (EDT)


I just checked again, and found that where I meant Antwerpen as a town (place), the GEDCOM import linked the name to the province, which has the same name. Maybe this is something that I should have corrected during the import myself, I don't know.

Anyway, it does add to the confusion a bit, since I may have some other towns in my database that duplicate as province names, like Groningen, Utrecht, and Hannover. Is there a default for this? I expect that most genealogists mean to specify a town, unless otherwise noted, but I acknowledge that this a grey area.

--Enno 18:12, 29 April 2009 (EDT)


I looked at Family:Corneille De Haen and Catherina Van Havre (6) and it looks like the problem is that Antwerpen isn't linked properly to a place when it appears as a christening place for a husband or wife (or most likely any family member), but it is getting linked to a place when it appears as a death place for these people. That's a bug, and thank-you for pointing it out. If you go to the corresponding Person page you'll see that it is linked to a place on that page, just not in the person reference on the family page. It's not easy to fix the pages that have already been created with the red links this way, but at some point this year I plan to write a program to correct the information in these family member references so you'll see the blue links then.

The way that place matching works is it assumes that you're entering the highest-level jurisdiction that matches. So "Antwerpen" or "Antwerpen, Belgium" both match the province. If you want to specify the town, you would need to enter "Antwerpen, Antwerpen" or "Antwerpen, Antwerpen, Belgium". If the place matcher were to match "Antwerpen", with the city instead of the province, then it's difficult to know how to match just the province. But if we say that "Antwerpen" matches the province, then people can enter "Antwerpen, Antwerpen" to match the city.

Another assumption is that a place ends with a country or a state/province. So "Lier" won't match anything, but "Lier, Antwerpen", "Lier, Belgium", and "Lier, Antwerpen, Belgium" all match Place:Lier, Antwerpen, Belgium. The reason for this is there are generally too many matching cities unless you have some context. For example, "Lier" could be matched to the Lier in Belgium, Spain, or Norway.

Having said this, you can override whatever place the place matcher matches in the GEDCOM review program so that the correct places are linked to before your wiki pages are created. It's too late to do that for this gedcom, but it's something to consider when we get GEDCOM re-upload working or if you decide to upload another gedcom.--Dallan 12:15, 1 May 2009 (EDT)


OK, I understand. There's a lot of work to do on place consistency in my own database anyway, but I'm glad this helped you to find another little bug. Re-uploading the GEDCOM is something I will probably do later this month anyway, so that's not a problem either.

--Enno 14:07, 1 May 2009 (EDT)


user profile change [3 May 2009]

I need to change my e-mail address and can't find where to do it without registering as a new user.--Mosshammer 16:40, 3 May 2009 (EDT)

My Relate->Preferences->User Profile --Jrich 16:44, 3 May 2009 (EDT)

Gedcom [5 May 2009]

Hi Dallan, I show that this gedcom is still showing "under review by Dallan", is this one OK to release?

http://www.werelate.org/gedcom/index.php?gedcomId=3104

Thanks,

Jim--Delijim 22:36, 3 May 2009 (EDT)

Oops, I missed this one. It was in weird state - still in "Uploader review" but had my name as the reviewer. I'll release it right away.--Dallan 10:45, 4 May 2009 (EDT)
Actually, it looks like there are a half-dozen more families to match in this GEDCOM. I was about to match them, but I wasn't sure if you wanted to update them so I think I'd better let you match them. Could you match them (or mark them not-matched) and mark it ready for import? I'll import it right away. Thanks!--Dallan 10:54, 4 May 2009 (EDT)

I'll take care of it tonight. Thanks:)

It should import shortly.--Dallan 23:11, 5 May 2009 (EDT)

Unable to Upload gedcom [6 May 2009]

Before your GEDCOM is ready to import, you need to: Click on each of the families in the Family Matches tab, decide whether the existing family matches your GEDCOM family, and click on 'Match' or 'Not a Match' in the bottom pane.

This is the message I am getting when I go to upload my gedcom. I have matched everyone through my list and gone back and double checked; what am I doing wrong? Thank you for your help.--Brae57 16:57, 4 May 2009 (EDT)

It looks like you got it imported. I just sent it through the final step. Can you tell me where the confusion was? I'd like to eliminate sources of confusion if possible. Thanks!--Dallan 23:11, 5 May 2009 (EDT)

It had to do with the Susanna Whipkey and John Berkey family match. I ended up excluded them and then I was able to upload. Thank you for your time.--Brae57 08:56, 6 May 2009 (EDT)


I just uploaded your gedcom to the sandbox so I could do some testing on it. I matched Susanna and John's family with one of the matching families in the wiki. It seemed to go alright; I'll keep looking into it. Can you tell me anything more about what happened? Had you matched them and it didn't "take"?--Dallan 15:06, 6 May 2009 (EDT)


Did you get a blank page in the bottom half of the screen when you tried to match this family? I just found (and fixed) a problem that could have caused this.--Dallan 18:10, 6 May 2009 (EDT)


Yes, I believe I had a blank page so I thought I didn't have to match it. So I excluded them and then presto magic :) I was able to go on.--Brae57 18:36, 6 May 2009 (EDT)


GPS coordinates [6 May 2009]

My GPS displays coordinates in DD format 39° 08.785N, 084° 31.720W. Is there a way I can enter these coordinates on a Place page without first having to convert them to another format? --Ajcrow 07:56, 5 May 2009 (EDT)

This is kind of a work-around, . . . but Google Earth displays coordinates in deg-min-sec format -- but it will take input either that way or in decimal degrees. So you could input your decimals and read off the resultant deg-min-sec coordinates. --Mike (mksmith) 14:50, 5 May 2009 (EDT)
Can you tell me what the 08.785 number means? I've seen coordinates in decimal degrees format (e.g., 39.785N) or degrees-minutes-seconds format (e.g., 39° 08' 18.5" N), but I haven't seen this format before. It looks like the 08.785 number might be decimal minutes. I can add this to set of coordinate formats the place page accepts, but I want to make sure that I'm assuming correctly. Can you tell if 08.785 means decimal minutes or something else?--Dallan 23:11, 5 May 2009 (EDT)
It is decimal minutes -- DD MM.MMM. It's the default format for my GPS and the preferred format on Waymarking.com and Geocaching.com. If there is a way to enter this format directly, that would be great! I have quite a few cemeteries that I could add coordinates for, and it would save me having to convert them. --Ajcrow 23:26, 5 May 2009 (EDT)
Yes, you ought to be able to enter it directly, since you can enter decimal degrees -- unless Dallan has a specific converter-machine coded in there somewhere. Decimal minutes is just like decimal degrees, only it slices things even finer. You can even have decimal seconds, which is getting down to inches. Take a look at the bottom of the page for Person:Ale Hatfield (1); the directions to his isolated grave site are in decimal seconds, which work fine when plugged into Google Earth. (Nailed his headstone square on.) I confess, I wasn't able to figure out how to do an external link to Google Earth incorporating those numbers, though there's probably a way to do it. --Mike (mksmith) 12:05, 6 May 2009 (EDT)
The question is: How do I enter it? No matter what syntax I've tried with spaces, colons, decimal points, etc, it keeps being converted to some other format and ends up putting the location somewhere in the next county. --Ajcrow 12:17, 6 May 2009 (EDT)
I've modified the entry field so you can enter decimal minutes now; for example "39 8.785N". Before I was assuming that people would enter 3 numbers, so the 8.785 was incorrectly interpreted as seconds. But I now assume that if you enter 2 numbers the second number is minutes. Please let me know if there are any problems.
You can use the Template:Googlemap to get a link to a google map for a set of coordinates. You can use this template anywhere, but if you use the template in the description field of an event on a person/family page, you'll get a google map link displayed with the event and the pedigree-maps will use the lat & lng in the template for the event location. See for example Person:Ivory Mason (1), edit the page, and take a look at the burial event.
BTW, I've just started geocaching. Going out for the second time this Saturday if it doesn't rain :-)--Dallan 15:06, 6 May 2009 (EDT)
Thanks, Dallan! That worked like a charm! I hope you enjoy geocaching. Just remember -- it's like eating potato chips. You can't stop with just one! --Ajcrow 15:15, 6 May 2009 (EDT)

2 suggestions [14 May 2009]

Dallan, Is there a way to make the unseen changes to watch lists stand out more, like maybe make them black instead of what they are now? These old eyes are having trouble seeing the difference between what looks to me like faint blue and faint purple. It says unseen changes are in bold, but I don't see that. I guess that's what happens when you get cataracks.

I was feeling good that I had no dupes, until I found one that isn't listed as a duplicate. I was going to merge them, but thought I'd tell you how it happens first. It's my fault, but these things happen and I would hope the search for duplicates would be able to notice the similarities.
Family:William Jackson and Susanna Seacriss (1)
Family:William Jackson and Susanah Seacriss (1)--Janiejac 18:08, 13 May 2009 (EDT)

Have you tried viewing your watchlist by selecting Watchlist from the MyRelate menu, then clicking on the Show all pages changed since last visited link to list just the pages that have unseen changes? That link is pretty useful for discovering changes that you may have overlooked or that the system neglected to tell you about (there's a bug where every once in awhile you don't get notified of a change). Does that help?
The new gedcom upload process would have identified these families as probable duplicates when you uploaded your second GEDCOM because it does a much more comprehensive duplicate check than the one that re-generates your duplicate list every day. The more comprehensive duplicate check is too expensive to run every day though; I'll have to think about what I can do to handle situations like the one you mention better.--Dallan 22:48, 14 May 2009 (EDT)

Note & Sources Suggestion [19 May 2009]

Hello Dallan,

I was wondering if it would be possible to add the icon tool bar that has the signature icon, etc to the Notes and Sources box on Family and Person pages? As we are merging these pages it is getting a little difficult to tell who added what to the pages. It would be nice to have the option to know who added the sources and notes at a quick glance since many are a little vague.

Thanks Debbie Freeman --DFree 13:12, 17 May 2009 (EDT)


Can you tell me a little more about what you're looking for? The signature icon would allow you to enter your signature (as an alternative to typing --~~~~, which is the same as pressing the signature icon), but it wouldn't let you specify different people who contributed each source citation or note, and you could use it only on the edit screen, not on the merge screen. Are you thinking that it would be helpful to track who contributed each source citation or note?--Dallan 13:47, 18 May 2009 (EDT)

I think it is more than adding the icon, because the --~~~~ doesn't get processed inside a source or note. You can use the icon in the personal history section, preview it, copy the signature into a note or source, but you have to change the user name part of the signature (i.e., Jrich) back to [[User:Jrich|Jrich]] to make it link to your user page properly. Even if the icon was handy, would it get used? Most histories aren't signed either even though they have this capability.... --Jrich 13:50, 19 May 2009 (EDT)

--- Hello Dallan, Yes I was thinking that it would be helpful to have the option for the User to use on the web page as to who and when (date stamp) contributed each source citation or note? As it is now there are few notes or sources, and many notes and sources are so vague they do not help much. As it is now there is no trail to follow to the User. Debbie Freeman --DFree 15:14, 18 May 2009 (EDT)


I've been wondering about that myself. It seems like a good idea. I'll add this to my todo list.--Dallan 13:22, 19 May 2009 (EDT)


Isn't this available in the history? Not that it is easy to extract, especially if there have been one or more merges, but you can step through the diffs using the Prev Diff link, looking for ones that changed the source you are interested in. It sounds like you are roughly asking for a history for each source separately so it is easier to track that one item? I could see also wanting that on each of the events (birth, death, marriage, name, etc.) rather than all jumbled together in the history of the whole page, but I think in complicated situations, it would still be difficult to decipher. For one thing, on busy pages, each item is likely to be a conglomerate of pieces input by multiple users. I have often added abstracts to bare-bones sources, added page numbers, converted MySources to Sources, etc. Obviously, as a user, I can't comment on how hard this would be to implement, but perhaps some idea of example situations where it would be used, and how often, should be made to see if justified? --Jrich 13:43, 19 May 2009 (EDT)


I was thinking about an optional field for each source or note where you could enter the name of the contributor, along with a button to insert your own name. My thought was that if people could sign their name next to the sources they contributed, perhaps they'd be more likely to add them.--Dallan 15:00, 19 May 2009 (EDT)


Error message on GEDCOM review [20 May 2009]

Hi, Dallan---

We're out here in the RV in the wilds of Pennsylvania (except with WiFi) and I'm spending a little time this evening reviewing my new Gastineau GEDCOM. And I keep getting an error message popping up, quote:

faultCode:Server.Error.Request faultString:'HTTP request error' faultDetail:'Error: [IOErrorEvent type="ioError" bubbles=false cancelable=false eventPhase=2 text="Error #2032"]. URL: http://www.werelate.org/w/index.php'

I click the "OK" button at the bottom of the box and it disappears and everything seems to be working just fine, until the error message pops up again, maybe every 10-15 minutes. Strange. --Mike (mksmith) 18:36, 19 May 2009 (EDT)


I haven't experienced that error before. The error is the gedcom review program, which runs in your browser, saying that it's not able to access the WeRelate web server. My guess is it has something to do with intermittent problems with the wifi (maybe a low signal?). Later this week or early next I'll change the error message to something more understandable, like "there was a problem accessing the server; please try again". Thanks for typing in the message; that really helps me to figure out what's wrong.--Dallan 22:58, 19 May 2009 (EDT)


Yeah, I sort of wondered if that's what it was, except that the wifi signal has been pretty strong. But since I couldn't make head nor tail of the msg, I thought I'd better pass it along.

Thanks for typing in the message -- You're kidding, right? See, there's this thing called "copy and paste". . . .   :)    --Mike (mksmith) 17:50, 20 May 2009 (EDT)


Gedcom upload status [24 May 2009]

Hi Dallan, what is the current status on uploading a gedcom into an existing tree without deleting the original gedcom. May I now upload the gedcom and will WeRelate recognize the duplicate pages?--Beth 19:01, 23 May 2009 (EDT)

Not yet. The fellow who was working on it changed jobs recently and I don't think he'll be able to put much time into it anymore :-(. I'm finishing up GEDCOM export right now, and I'll take over working on re-upload as soon as export is ready.--Dallan 01:02, 24 May 2009 (EDT)
Okay, thanks.--Beth 08:13, 24 May 2009 (EDT)

Family page or Person page or both? [27 May 2009]

I have another example of odd behavior to report (sorry). I'm tidying up my newly imported GASTINEAU tree, and when I filter FTE for "Person," I also get the listing for the following Family page: "James Gastineau and Patience Morrow (1)". And if I filter for "Family," it shows up there, too. I've looked at it closely but I can see nothing that would cause this Family page to think it's also a Person page. --Mike (mksmith) 16:35, 27 May 2009 (EDT)


It looks like it's picking up on the marriage note: "Marriage record taken from personal records". Filtering is pretty simple - it just looks for the word anywhere on the page. As an alternative, you could change the "Namespace" field just above it to "People". That will guarantee to just show people.--Dallan 18:32, 27 May 2009 (EDT)


That never occurred to me; I guess I thought it was reading the heading or something. If I use the namespace drop-down for "People," incidentally, it shows everyone in multiple columns, none of which are wide enough in the present layout to read anything, without having to shove them around each time you open TFE. (I know redoing the layout is on your To-Do list. . . .) --Mike (mksmith) 19:12, 27 May 2009 (EDT)


Place county change on person page does not reflect change; only place link [1 June 2009]

Hi Dallan, Maybe I am missing something but on Person:Newton Coker (3) I had a red link on his burial place location. I had entered Trion, Walker, Georgia, United States. I edited the page and changed the burial place to Trion, Chattooga, Georgia, United States. This now links to the proper place but the person page still shows the burial place in Walker County. --Beth 00:33, 2 June 2009 (EDT)

The issue is that there are actually two parts to an event place: the title of the Place page that is linked to, and the text that is shown. Both parts are in the place field, separated by a '|'. It's easy to change one part without changing the other. When I re-do the Person and Family edit screens later this year, I plan to put the two parts in separate fields so you can see what's going on more easily. I just changed the second part for the burial place so that it displays the correct text now.--Dallan 00:54, 2 June 2009 (EDT)

Is this a Wiki or not? [1 June 2009]

Hi Dallan, Are you considering changing WeRelate from a Wiki to something else? --Beth 00:55, 2 June 2009 (EDT)

No. I've seen how FamilyHistoryLink.com (based upon an blog) works, and how new.familysearch.org works (based upon a model where you think of each person as a folder containing multiple pages; you can edit the information on your page but not anyone else's). I think the wiki model is superior.--Dallan 18:24, 3 June 2009 (EDT)

MySources in the GEDCOM review [16 June 2009]

Hi, Dallan---

I'm presently reviewing a new GEDCOM. I'm aware that all sources imported in the GEDCOM are provisionally MySources until I've matched them up with the appropriate Source, or else created a new one. No problem. But some of these are meant to be MySources. However, I've been trying to regularize my own MySources so I won't have (for example) both "World War I Draft Registration" and "Draft Registration, World War II." Most of these MySources aren't new; I've used them before, with other trees & GEDCOMS. But when I try to do the "Find/Add" thing, it will only search "real" Sources. And the namespace drop-down is grayed out, so I can't change to the MySource list after the first step of the search, either. What this means is, I can't match up the MySources during the review process -- even if they're titled exactly the same. I can't make anything appear in the "Matched Page" column (unless I'm missing something). And if they're not titled the same, I can't change them to the proper title. --Mike (mksmith) 14:12, 2 June 2009 (EDT)

Ah, I never thought that anyone would want to match their MySource's to previously-uploaded MySources. What if I let you edit the title of the MySource page in the review program, so you could change it to the title of your existing MySource? Then during upload it would be replaced by the existing page. Not an ideal solution, but it's quick to implement so I could fix it right away. --Dallan 18:24, 3 June 2009 (EDT)
That would work fine, I think. I keep a reference file of all the Source & Mysource titles I'm using, just so I'll know what they ended up being called. (Retentive, I know.) I copy and paste from the list frequently when I'm tidying up all the pages following an import, so no problem. --Mike (mksmith) 21:32, 3 June 2009 (EDT)
Ok, I'll make this change tomorrow or Saturday.--Dallan 22:53, 4 June 2009 (EDT)
Sorry, I haven't gotten to it yet; it's a bigger job than I expected.--Dallan 17:56, 7 June 2009 (EDT)
It's finally ready. You can change MySource titles by editing the MySource when your review your GEDCOM. If you give it the same title as an existing MySource page, your pages will point to the existing MySource.--Dallan 20:23, 16 June 2009 (EDT)

Wikitable CSS [4 June 2009]

Dallan,

It appears there are a number of tables on WR that use the Wikitable class, but that is not defined in the Common.css. Wikipedia's full CSS is here. Are the table details something that can be added here?--Jennifer (JBS66) 07:06, 4 June 2009 (EDT)

I just added the wikitable class. Let me know if you want anything else from WP's CSS. It's no problem to add.--Dallan 22:53, 4 June 2009 (EDT)

Sandbox: is gedcom import running? [4 June 2009]

I have a teeny gedcom that's been "waiting for analysis" since last night. --Judy (jlanoux) 08:16, 4 June 2009 (EDT)

Oops sorry - I restarted the gedcom upload process on the sandbox.--Dallan 22:53, 4 June 2009 (EDT)

Network and removing unwatched pages [7 June 2009]

Dallan, From my network page, I unwatched several pages, however some of these unwatched pages still appear on the list. See User:Suzyq, I am not watching any of the pages on the list for Suzyq and Suzyq is still in my network. --Beth 08:34, 5 June 2009 (EDT)

Hi Beth, I just checked and it looks like you and Suzyq have 21 pages in common. I checked three of these of these pages and it looks like you're still watching them. Are you sure you unwatched these pages, or did these pages get watched since you left your message?

Having said this, one possible explanation is that calculating the network is expensive, especially for someone with a large watch list like you, so I save the result for about 20 minutes so that as you're paging down through the list it doesn't have to be re-calculated. But what I'm not doing is updating the saved result in case you change your watch list during those 20 minutes. I need to do this. Could that explain the problem?--Dallan 08:52, 7 June 2009 (EDT)

Dallan, I can only say that I tried to unwatch all of the pages. I have since unwatched those pages again I think so all of the pages for this user should be unwatched and no longer there. I waited 20 minutes but 18 still remain. Does it have something to do with the Redirect perhaps? Suzy is not the only one with problems, just the one I used as an example. I do wish one could use the control shift in the Wiki to select all and unwatch. --Beth 19:25, 7 June 2009 (EDT)

Sandbox: Export Testing [13 June 2009]

Is it possible to refresh Sandbox without messing everyone else up or taking too much effort. Sandbox hasn't refreshed since I joined, so pages that I have created and nicely formatted aren't available to test export. I'm very interested in what's going to happen to them on the return trip.

My first export attempt looks very promising. The gedcom uploaded in Sandbox and never touched looks very good coming back. --Judy (jlanoux) 11:04, 5 June 2009 (EDT)


Is it possible to put line breaks (CR or LF) in the exported GEDCOM? I know they aren't required, but they're all but impossible to read without them.

I'll do this.

Is "build a tree" going to be ready soon? It seems that an export user would need to build the tree before they can export. It would be good to test them together. --Judy (jlanoux) 13:57, 5 June 2009 (EDT)

It's coming soon - either this week or next.

Export Result - Yikes! I built a tree with one person who has one event (birth) and One Source on Ancestry and exported it. The resulting gedcom was nice and compact. I imported into Sandbox, linked the source to the 1850 US Census source page on sandbox and tried the new export. It sorta blew up in size! It's hard to tell with no line breaks in the gedcom (thus above request) but it looks like you included all of the text on the source page, which is rather excessive. I created a user page User:Jlanoux/Gedcom Export Testing to keep track of the tests so we don't get too confused. --Judy (jlanoux) 18:16, 5 June 2009 (EDT)

I do include the full text on the source page (I'll remove this), but the blow-up is probably due to the entire text of the GFDL being included in the exported GEDCOM. Including the full text of the license is a requirement of the GFDL. We can avoid this requirement by switching over to CC-BY-SA, which I'd like to ask people to vote on later this month. See the bottom part of the topic on Copyright at the end of the WeRelate talk:Watercooler.
I understood the license was in there. Is it going to
I think I'll post the GEDCOM export onto the main website later today and call it experimental. Then we can test it there as well. Thank-you for your testing!--Dallan 08:52, 7 June 2009 (EDT)

I'm looking at what happened to the WERELATEELLIOTTTAYLOR2.GED, a TMG file, that was in sandbox and now exported and brought back into TMG. As I said, it looks good overall in that nothing got me screaming. I am familiar with the leaky nature of a Gedcom round trip so was expecting some changes. I got an "Unknown Gedcom tag" for each person. Example: "Unknown GEDCOM tag: _UID WeRelate:Person:Oscar_Rhodes_(1)" is in the memo for a tag called "GEDCOM". I think that was the TMG way of handling data it didn't know what to do with so that it wasn't lost. But isn't there a standard way to transport a UID? Is it just that the WR version is too long or something? I'm worried that it won't serve the purpose because someone would just delete it. I also got a new tag NOTE with the words "Family Source" in the memo and WeRelate.org as the source with other stuff in the citation. Then a new tag Person-Sour with "Person Source" in the memo and WeRelate.org as the source. No problems there. Event Sources are a MAJOR problem...

This database was imported into sandbox and never touched (no cleanup). All of the events had your MySources created for them. Nothing came back except the citations!!! There's no source. The citations are all attached to source 0 (nonexistent). I'll have to go study the gedcoms to see what you put in there to analyze the problem (maybe I'll wait for those line breaks<g>). For me, this would be a showstopper.--Judy (jlanoux) 16:03, 7 June 2009 (EDT)

There isn't a standard way to transport a universal identifier (UID). Some software programs use a tag name of "_UID", but not all. Maybe it would be best for me to not add a "_UID" tag after all. If an exported GEDCOM is ever re-imported, I can use the source citation to determine which WeRelate page to match it to.

I actually like the idea of having the WR page name in my database. I've been mulling over the best way to do this. I just wish it wasn't so ugly. Maybe it could be a note? Either way, I'd rather have it than not. But to be able to recognize your stuff coming right back at you, the labels have to be hidden better. OTOH, I swear that the gedcommers never even look at what they have been importing judging from what I see them upload.--Judy (jlanoux) 21:51, 7 June 2009 (EDT)
In addition to the _UID tag there should be a source citation attached to every person, family, and source, with links to the corresponding WeRelate page. Given the source citation would still be there, let me know if you still want the _UID tag; otherwise I'll remove it.--Dallan 11:41, 13 June 2009 (EDT)

Thanks for finding the bug with the source citations! I'm looking at your GEDCOM file and I can see the problem. I don't know why it didn't show up previously, but I'll try to fix it (and the other things) tomorrow.--Dallan 19:38, 7 June 2009 (EDT)

OK, drop a note when you're ready for a test. I'm also working on a smaller file in Sandbox to play export with. --Judy (jlanoux) 21:51, 7 June 2009 (EDT)
I've been under the weather this week, but I'm really hoping to get all of the fixes to GEDCOM export that we've talked about in by this weekend. I'll let you know when it's ready for the next round of testing. Thanks!--Dallan 11:41, 13 June 2009 (EDT)

Rename during import [11 June 2009]

The option to rename a page during the Gedcom Review has disappeared. Was it causing a problem? It sure was handy. --Judy (jlanoux) 21:55, 7 June 2009 (EDT)

Another anomoly in Gedcom review: When editing the person page, the Save button has a hover message saying press Alt-S to save. I've tried that from several places on the page and it just doesn't work. If it can be fixed it would save wear and tear on my shoulder. --Judy (jlanoux) 10:18, 8 June 2009 (EDT)

Mike tells me that this is because FireFox captures the Alt S to display history (I dunno why). So please excuse the ring.--Judy (jlanoux) 00:03, 11 June 2009 (EDT)
Actually, I said ALT-S is the Firefox keyboard shortcut for "History" on the menu, just as ALT-F drops down the File menu. :) Dallan, you might want to pick a slightly less obvious keystroke combination for saving, like CTRL-SHIFT-S, if the media-wiki software allows it to be changed. (You can even do that one one-handed.) --Mike (mksmith) 17:29, 11 June 2009 (EDT)
I know this is kind of annoying for firefox users (including myself). I'll see how much trouble it would be to allow both key combinations.--Dallan 11:41, 13 June 2009 (EDT)

Merge problem - Users name not transferred to watch page [13 June 2009]

Hi Dallan,

When I merge page 2 by userB with page 1 by userA; only userA is watching the page. --Beth 09:06, 11 June 2009 (EDT)

I'm pretty sure that both users are watching the page, but there's a bug that a new watcher isn't always listed in the watcher list. I need to track that bug down and fix it.--Dallan 11:41, 13 June 2009 (EDT)

Deletion of Duplicate account with We Relete [11 June 2009]

Please could you delete an account called welshmorts & it's no longer required--Colin Madge 12:29, 11 June 2009 (EDT)

Will do.--Dallan 11:41, 13 June 2009 (EDT)

GEDCOM Export Ready [14 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


GEDCOM Export Ready [14 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


PAF [17 June 2009]

If you wanted a report of how it works when imported to various desktop programs, here's how it looks in my PAF software:

  1. On the whole it looks great!
  2. I got a lot of error msgs that PAF cannot store more than one name per individual. But that's not your problem, it's PAF.
  3. There are two individuals in the database who have no date of death. If I were to upload them somewhere, my PAF would assume they are living since they have no d/o/d. But one was born in 1720 so is surely deceased; the other one is prob deceased judging from dates of the parents, but that's not positive. Hmm, well perhaps PAF would know about the one born 1720 but when I'm not sure I always put in 'unknown' to force it to assume deceased. I don't know how others handle this.
  4. There are two folks whose surnames are in caps.
  5. Ella Boone Grey is duplicated - she is in there twice with dif RIN numbers; one time with a spouse and one time without. Martin Hawkins Gibson is also duplicated.
  6. The sources didn't fare so well. Several came as only a number, as in '#6', many were duplicates and none had any reference as to where these souces might be found.
  7. The one note I checked had two URLs referring back to WeRelate. They aren't links but I think my PAF doesn't do links in notes. But the URLs are there to copy and paste if one wants.

Looking good Dallan! Better than a lot of GEDCOMs folks have sent me! --Janiejac 23:59, 14 June 2009 (EDT)


Thanks for the feedback Janiejac. I'll try to figure out why Ella and Martin are duplicated, and what's going on with the sources.--Dallan 01:07, 17 June 2009 (EDT)

I fixed the duplication problem and also the source problems. Hopefully sources will come down a lot better now.--Dallan 17:29, 17 June 2009 (EDT)

(Very) Old version of FTM [17 June 2009]

General impression is that it is very functional and works.

file: testa.ged

Line 11: TAG: CHAR, character set: UTF-8 is not supported, ANSEL assumed. Line 22: Tag: TYPE, subordinated to wrong item, line ignored [repeated for lines 293,219,290,530, 563, 650, 812, 816, 887] Line 2262: Tab: SOUR, has an invalid CRI: , line ignored.

Place aliases, i.e., [[Place:Orange,California,United States|Orange County, CA]] created two facts, but I suspect this was designed that way. Personal preference would be to just use the actual Place Page name and ignore the alias so no alternate facts are created unless there is a real alternate.

The layout of the sources once I displayed them in FTM seemed a little strange (don't know how this translates into GED structure), but it works, so no real problem, just fyi here.

  • The source has no title (missing a marketing opportunity: something like "Cheerfully provided by WeRelate"? :-) ).
  • The import date shows up in the Citation Text field of FTM with no label to tell what that date was, but the rest of download information (license and where it came from) was in the Footnote field.

[My experience with FTM (which may not carry over to newer versions) is that it is best to put data in Citation Text field, then make sure the "Include citation text in footnote" checkbox is checked to let FTM build the formatted footnote that prints in the formatted reports automatically. Otherwise, if the user presses the "Restore Footnote" button, it wipes out any data manually entered in the Footnote field. Again, I don't know if this check box translates to GED or is just a FTM "feature", possibly in conflict with other genealogy software. Again just an fyi, not unworkable.]

--Jrich 09:48, 15 June 2009 (EDT)


Thanks. I'd like to keep UTF-8 for the character set (finding something that will write ANSEL nowadays is nearly impossible). I'm going to look at the other warnings though. And the place should have created just a single fact, using the user-entered "Orange County, CA" as the place (I could make it an option to use the Place page name instead?), so I'll look into that. And I'll also check the source layout. One issue with sources is that GEDCOM has multiple fields for storing source text, and some programs seem to expect it in one field and some in another. I originally had it in the citation section but recently switched it to a note attached to the source because that's where PAF seemed to want it. I'll look through some of the uploaded GEDCOM's to see which is most popular.--Dallan 01:07, 17 June 2009 (EDT)

Can you tell me which GEDCOM file you're looking at? I can't find it. The TYPE problem on line 22 is expected (it's not a big deal - it's throwing away the type of the alternate name). But I'd like to look into the problem on line 2262 further. The missing source title and putting the source text in the citation should be fixed now.--Dallan 17:29, 17 June 2009 (EDT)

Refreshing deleted GEDCOM's doesn't work anymore [16 June 2009]

In the old days, I could refresh a tree by removing it and uploading a new GEDCOM afterwards, but today that doesn't really work any more. Here's why:

When I review a GEDCOM that contains data that was previously deleted, the merge tab shows hundreds of families that should be merged, but 99 % of those can't be merged at all, because the right half of each potential match is not found. That's quite logical, because it was deleted earlier. Nonetheless, the software wants me to mark each one of them as no match, so the merging process becomes a very daunting task.

I'm pulling back all GEDCOM's for the moment, and I really hope that you can find a solution for this.

--Enno 16:52, 15 June 2009 (EDT)


I was wondering if the answer is to wait a day or two and let the deleted names get flushed out of the search indexes? One would think the process shouldn't match deleted pages, but probably the indexes may contain data up to a day old making it appear to the search process as if they are still there. I am only guessing. --Jrich 16:46, 16 June 2009 (EDT)


When I tried earlier, I made the upload shortly after deleting the old tree, and came back next day to see if those duplicates were gone. When they were still there, I gave it another day, and still found the same amount of possible matches.

Today, I made a new upload, and found no matches at all, so it looks like you're right. Just delaying the review probably doesn't work because the list of possible matches is created immediately after the upload, and is not refreshed on a daily basis like the normal duplicate list. --Enno 18:34, 16 June 2009 (EDT)


Yes, if you wait an hour or two between deleting a tree and uploading a new one, the deleted names will get flushed out of the search indexes. I remember reviewing your GEDCOM and seeing all of those red links - now I know how that happened. I should do a double-check to verify that the page still exists before showing the potential match. I'll add that to my todo list.--Dallan 01:07, 17 June 2009 (EDT)


We Relate Tree [16 June 2009]

was referred here by a magazine where they have obviously never worked on the site. It is clumsy, non-intuitive, obstructive .... seems to be only geared towards accepting Gedcoms. Adding by hand is....difficult.

 Does not accept children. If it won't accept the children, how is it going to accept the siblings?
 Does not catch simple errors - born in 1937, married in 1856. It was a typo, but I left it to see what the system would do. Absolutely nothing.

I have panels clogging my screen all over the place with no way to turn them off - yes I can expand or contract them, but that doesn't clean the trash off hte screen.

I really don't have either the time or inclination to spend a couple of days digging around the basement of your site trying to figure out how to add things in - and if it doesn't catch simple errors, how is it going to catch duplicates and other errors in that gedcome import you are so despearte to have?

I may come back later and play with it some more - but at first blush, this site will be thrown in the trash bin. And what is so sad about this whole thing (I have a fairly large tree over at geni.com without any of the problems here, although that is far from an ideal site) I LIVE in Ft WAyne, Allen, Indiana. And go to the Genealogy department at the library. Sad.

What could you do ti make me more intersted in ever coming back? When I click ADD to tree, I expect to find the info IN THE TREE.--Dragonasbreath 15:31, 16 June 2009 (EDT)


You may want to view the tutorial videos. As you can see on the recent changes screen, hundreds of pages are added/edited by hand every day. And hundreds more are added via gedcom upload (which aren't shown in recent changes).--Dallan 01:07, 17 June 2009 (EDT)


GEDCOM Export Ready [17 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


GEDCOM Export Ready [17 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


GEDCOM Export Ready [17 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


GEDCOM Export Ready [17 June 2009]

The GEDCOM for tree testa is ready to download. Click here.


Review of GEDCOM - too small [19 June 2009]

Hello Dallan,

User Loulehmann is requesting help.

User wrote:

Hello, I'm new to WeRelate. Wow! This text is so small but I don't see how to enlarge it. Anyway I uploaded a gedcom and started to review it but the list of names on the extreme left of my screen is so far from the scroll bar on the extreme right that I can't see the list while I use the scroll bar. The window needs to be wider or the contents need to be smaller but I don't know how to fix this problem. The material seems to be displayed with Adobe Flash Player of which I know nothing. What can I do?

Lou Lehmann--Loulehmann 13:46, 19 June 2009 (EDT)--DFree 13:51, 19 June 2009 (EDT)

I'll leave a message on his talk page. Thanks!--Dallan 18:50, 19 June 2009 (EDT)

Living people in gedcom upload [20 June 2009]

Hi Dallan, could we eliminate the living in the gedcom upload process? See Family: Living Burdette and Living Unknown (1) Living Burdette and Living Unknown (2) in the B's in the duplicate review project. These pages were recently uploaded. --Beth 00:41, 20 June 2009 (EDT)

The only living people who are imported during a gedcom upload are the root of the tree and its ancestors unless the uploader specifically marks others to include. I don't want to prevent this until we have the other approach to handling living people that I mentioned to you via email implemented. There's a lot to do, but I'm hoping to get to this project within the next 9 months.--Dallan 10:01, 20 June 2009 (EDT)
How about if I modify the duplicates lists in the meantime to not include living people?--Dallan 10:13, 20 June 2009 (EDT)

Hi Dallan, that is a great idea. We can eliminate the duplicate "living" people pages when you initiate the other approach.--Beth 16:32, 20 June 2009 (EDT)


Upload GEDCOM review - Sources - [20 June 2009]

Loulehmann asks

Thank you for your kind comments regarding my sources. I would like very much to add them to a unified tree for the two matched families. However before I read your message about my sources and how to add them, I mistakenly did an update review of the two families with a couple of "keep" notations but did nothing about the sources as you suggested as I then didn't know how. Now how can I merge my sources before importing?

Lou Lehmann--Loulehmann 12:05, 20 June 2009 (EDT)--DFree 12:19, 20 June 2009 (EDT)

Answered on User talk:Loulehmann.--Dallan 14:45, 20 June 2009 (EDT)

Overlapping GEDCOM file [27 July 2009]

Dallan, can you override the feature that tells me I have to delete my old tree before the new GEDCOM I just tried to upload can be uploaded? I know some overlap but I'll take care of that in the review process. There is new info I want to add to that tree. This new file is a very small addition - only abt 20 people; file named James Strother. --Janiejac 18:56, 22 June 2009 (EDT)

No problem. I just sent it through.--Dallan 19:48, 22 June 2009 (EDT)

Please can you override this so I can upload the file Sam and Mary Farrow into the tree GWJackson by Marda. I'll take care of the duplicates. I thought we were supposed to be able to import into older trees now. And if so, there probably will always be a little overlap. Even is there is a large overlap, that would be a good way to update the older info. But I sure wouldn't want to delete the old info in order to do that. Perhaps that's a programming impossibility?? --Janiejac 23:25, 5 July 2009 (EDT)

I sent it through. Currently, if you upload a GEDCOM that contains some of the same people that you've previously uploaded, you have to do just what you did and ask me to override it, and you have to merge the people yourself during gedcom upload. In the future you won't have to do this, but the fellow working on this project started a new job recently and work on this project has stalled. It looks like I'll have to finish it, which I'll try to do in the next few months.--Dallan 00:13, 6 July 2009 (EDT)

Please can you override this so I can upload the small GEDCOM file named GeorgeW. I'll take care of the duplicates. Just want to add a few children to this tree. --Janiejac 21:09, 25 July 2009 (EDT)

It should be going through now.--Dallan 10:59, 27 July 2009 (EDT)

Merging duplicate pages - regarding surnames beginning with De and semi-protected pages [24 June 2009]

Hi Dallan, I know that you are extemely busy but I have 2 requests. The surnames beginning with De should be matched by the last portion of the name; i. e. De Grandison should be on the "G" page. I have found that there are more matches for the De Grandison pages under Grandison. Now I can recheck all of the De ___ that I already merged for duplicates under the last portion of the name.

It would be helpful to me if you could move the semi-protected notice to the merge review page. On some pages it does not matter and on other pages I may have spent 10 minutes and then get to the actual merge page and find the pages are semi-protected. I am not merging the semi-protected pages at this time.

I am not a programmer so if my requests would take too long to implement don't worry about it. Thanks. --Beth 19:01, 22 June 2009 (EDT)


I really appreciate all the help you've been to the merge project. I'll make these changes as quickly as I can. I'll make the first change hopefully tomorrow when I remove living pages from the duplicates list. (Expect the list to be fixed Wednesday morning.) And I'll make the second change later this week.--Dallan 19:43, 22 June 2009 (EDT)


Thanks Dallan, worked like a charm, but then the A, B, and C list that we finished suddenly had more entries. I have now finished the A and B list "once again" and am working on the C list.
I noticed that too :-). BTW, Taylor has started merging the "easy" ones (where it's fairly clear whether the families are duplicates) starting with the Z's.
Thanks Taylor, I suppose that you will be leaving for college soon or are you already there? --Beth 23:42, 24 June 2009 (EDT)

One question; I know of two familes that I would like to merge, but I have not found the "other" family using the compare and search feature. These families are on a person page as the main parent family and alternate parent family. I stopped after 150 or 200 results because I was tired of searching. --Beth 22:58, 24 June 2009 (EDT)

You can enter the family titles in directly to the Special:Compare screen if you want. Alternatively, if you go to the Person page there should be a Compare parents option in the More menu when the person has multiple sets of parents.--Dallan 23:06, 24 June 2009 (EDT)
Yep, it was there alright; I need to expand my field of vision. Thanks worked fine. --Beth 23:42, 24 June 2009 (EDT)

GEDCOM Extract/Export [24 June 2009]

Tried out the GEDCOM export. Pretty slick, but I noticed that the wikipedia sources and links retain WR link syntax ("[[<stuff]]" instead of "http//www...."). Maybe that's helpful if you know that a GEDCOM going to be offline edited and then returned, but it probably isn't what most people are looking for. If you want to see some examples of this, export my default tree.--Jrm03063 10:16, 24 June 2009 (EDT)

Good point. I can't do much with the other wiki markup, but I could at least convert wiki links to URLs. I'll add that to my todo list.--Dallan 23:06, 24 June 2009 (EDT)

google map template [25 June 2009]

Something is a little weird with the google map template these days. The title seems to be dropped in next to the coordinates in a search field, without actually going to the specified location.--Jrm03063 21:30, 24 June 2009 (EDT)

I just tried it and it seemed to work ok for the example I tried. Can you point me to a page where it's not working? Thanks.--Dallan 23:06, 24 June 2009 (EDT)

Thanks [27 June 2009]

Hello Dallan,

Thank You. I highly appreciate your help/input on the new Category. Debbie Freeman --DFree 18:22, 27 June 2009 (EDT)

Hope you don't mind me asking, what new category? --Beth 18:39, 27 June 2009 (EDT)
Check out Category:American Revolutionary War Figures and Category:Daughters of the American Revolution--Dallan 13:57, 29 June 2009 (EDT)

Newly added categories [29 June 2009]

Hi Dallan, Can your bot print a list of the newly created categories for the month and place on the home page or a link from the home page to the list? --Beth 20:28, 27 June 2009 (EDT)

You can go to Special:Newpages (it's one of the special pages listed under "Special pages" in the Admin menu) and get a list of newly-added categories, but this only shows categories added during the past 5-7 days. Is that good enough? Otherwise I can query the database to get a list of categories added during the past month and add them to a page somewhere.--Dallan 13:54, 29 June 2009 (EDT)
Another alternative would be a sort option on the Search page, that would allow you to search all pages in the Category namespace and sort them chronologically with most recent first. (And I nearly choked when I saw there were 1691 categories!!!!) jillaine 14:08, 29 June 2009 (EDT)
One thing you can do is select the Categories namespace on the Search screen, then check the "Exact match" box, then check "Sort by date modified". This sorts the categories by when they were last edited, which of course isn't the same as date created, but category pages don't usually get edited very often after they're initially created, so it's not too bad.--Dallan 14:21, 29 June 2009 (EDT)

User subpage [29 June 2009]

Hello Dallan,

I have a question about User:Leo Bijl's Userpage (he's asked for my assistance). He has a subpage that has been deleted, but continues to show up on his userpage. [6].

I tried to restore it and then delete it again - that didn't work. I tried purging the page, that didn't work :-)

If you notice, the URL has Roerbodehttp://, while the page title has Roerbodehttp:/ (single slash). Your help would be appreciated!--Jennifer (JBS66) 16:20, 29 June 2009 (EDT)


That was sure strange. The two //'s in the URL got interpreted as a single / for some reason. When I clicked on the "Edit" link, I ended up with a URL that had just a single /. When I manually added a second / to this URL, I got to the page that needed to be deleted. I could then click on the delete link to remove it. The MediaWiki software does some URL rewriting on the nice-looking "/wiki/<page title>" URLs to turn them into the real "/w/index.php?title=<page title>" URLs; the rewriting logic must accidentally convert two //'s into a single / as a byproduct. Thanks for letting me know.--Dallan 16:53, 29 June 2009 (EDT)


Thanks for your help!--Jennifer (JBS66) 16:58, 29 June 2009 (EDT)


Trees Problem [3 July 2009]

I see you added "Build a Tree". I was so excited that I raced off to create a new tree. Whoops, there is a problem with creating a tree on the Manage Trees page.

I selected "Create a new family tree" at the bottom
I got a screen asking for the name of the new tree
After entering the name I was returned to Manage Trees, but the new tree wasn't there. I tried "refresh" and the browser warned me about resending data.
I tried again, to make sure I hadn't just done it wrong, created, Testtree. Yep, it brings you back to Manage but the new tree isn't visible. I reloaded the page and it appeared. I tried to delete Testtree and got an error "Error deleting testtree" in red above the table.

Just thought you should know. Now I'll go play with creating a tree. --Judy (jlanoux) 17:50, 1 July 2009 (EDT)

Thank-you for letting me know! I found the problem and fixed it.--Dallan 17:56, 3 July 2009 (EDT)

FTE Problem?

I must have done something bad. I went to FTE and opened my new empty tree "Lanoux".

It asked me for my parents, etc. Blithely assuming there was going to be a search, I entered the names. When I hit "Finish" button, it says it created new people. Oops, not what I intended. Did I overlook something on the page where it asked for all the names? Don't worry, I'll clean up any duplicate people I created.

Yes, that was put in there when I thought that people would want to use the FTE to start a brand-new tree.
It is handy to be able to start a new tree that way. Maybe a note to move to the end if you don't want to create people?--Judy (jlanoux) 18:18, 3 July 2009 (EDT)
I'm going to re-do the FTE in a major way later this year. Let's talk more then.

Now I've tried to add people to the new tree. I selected an ancestor and clicked "Trees" on the person page. It looks like it intends to add ancestors and descendants to all of the trees the person belongs to, not just the open tree. Does it have to do that? I tried unchecking the others long enough to finish the operation and then went back and added them. That seemed to work. It looks like now I have a bunch of people in the new tree. Great! I'll go play with adding more.--Judy (jlanoux) 18:12, 1 July 2009 (EDT)

Sorry - I'm not sure how to easily do that without complicating the UI. For now you'll have to uncheck and then re-check the other trees.--Dallan 17:56, 3 July 2009 (EDT)
Ok, Since I'm able to return to the person and fix it immediately, it isn't a big deal.
Question: Is it safe to leave FTE open while it's creating the tree. I was using FTE to find the people I wanted to add ancestors for.--Judy (jlanoux) 18:18, 3 July 2009 (EDT)
That should be ok. Sometimes when you do things in the wiki the FTE application doesn't get notified and it gets "out of sync". To re-sync the FTE you have to close the tree in the FTE and re-open it. That's one of the big reasons that I want to re-do it.
Question: I'm wanting to create my own little Pando (one big tree). Do I have to leave the old trees because they have gedcoms attached? I never did understand the relation between gedcoms and trees.--Judy (jlanoux) 18:18, 3 July 2009 (EDT)
There's not much of a relationship. I was going to make a stronger relationship to make GEDCOM re-upload easier to code (i.e., you would have to upload your GEDCOM into the same tree), but nobody liked this idea so I'm not going down that road. The main uses for trees now are for searching, browsing, and exporting some collection of people that you're interested in. Just make sure that you don't end up watching too many people (adding too many people to your tree). You can end up getting a lot of email notifications that way as pages are changed :-).--Dallan 19:02, 3 July 2009 (EDT)
If anyone ever edited a person I'm interested in, I would be delighted!

I got a huge list of duplicates yesterday. I would have thought it was because of adding ancestors to the new tree, but they (the duplicates) aren't in that tree. Many of them have been out there for a while too. Did you change the duplicate-finding algorithm?

I haven't changed the algorithm. Several people are going through the master list of possibly-duplicate families and merging them. When that happens, relatives of those merged families that now look like they're also possible duplicates start showing up on people's duplicate lists. That's probably what's going on.

One strange thing is that when I do the manual look for duplicates (from the more menu), the matching person is not first up even though his name is identical. I have to wade through all the siblings to get to the person I need. Doesn't name count? --Judy (jlanoux) 19:12, 3 July 2009 (EDT)

I've noticed this as well. The weights I'm using for ordering the possible-duplicates aren't quite right. It's on my list of things to look at.--Dallan 19:19, 3 July 2009 (EDT)

Ongoing gedcom upload problem [7 July 2009]

It appears that a user is going through and merging her gedcom with families that I'm watching. Some was done before I woke up this morning, and I checked them when I got the emails. In a few cases, I made minor cleanups, generally turning MySources into Sources, removing them in favor of the same Source already there, or merging the narrative with what was there. But then over the last couple hours, I've gotten the same notification emails again, and the user has basically added back the changes originally made, on top of what I did. See, e.g., Person:Abraham Doolittle (3), which I've reverted. Or Person:Thomas Bingham (6). What's up? And could we fix it by not having these half-baked changes go live until the user is done? The red mysource links mean I will ideally revisit all of these to check if the similarly named sources are the same ones already there.--Amelia 13:59, 4 July 2009 (EDT) I too noticed this morning that user Loulehmann has confused information between Richard Church of Hadley and Richard Church of Plymouth, undoing corrections made over a year ago. I just got back home and saw your post here. Have you tried contacting him/her?--Scot 18:25, 4 July 2009 (EDT)

No, but now that Dallan's probably on vacation, I will. What they added to the pages I'm watching wasn't wrong (so far), it was just a little redundant (to be expected from a new user in the interface we currently have.)--Amelia 19:44, 4 July 2009 (EDT)
Please go ahead and contact them.
Figuring out what to do with edits made during a gedcom review is a difficult problem. Making them immediately is simplest from a programming point of view. If I hold off changing the pages until the user uploads their tree, I have problems dealing with the case where another person has edited the page in the meantime. And if someone is doing a lot of editing, I think it might be nice to be notified of their edits as they make them so people can contact them about poor edits rather than have the system store up the edits and apply them all at once. But I can see how the red mysource links are a problem. What if we addressed just that problem? For example, I could immediately create the MySource pages for gedcom sources that were referenced on edited pages. This raises a question if the user subsequently edits the MySource and then marks the GEDCOM ready to import: should the edited MySource override the one just created? But maybe that's a relatively minor issue.--Dallan 00:13, 6 July 2009 (EDT)
I contacted him and he says he doesn't remember making those changes. Which presents the problem that either the system screwed up without human input, or it made it so appear that the changes hadn't been done yet he didn't even notice that he'd done them yet. Either could be a mess if we can't change the 'live as they are merged' issue, especially with someone merging over days like this guy.--Amelia 23:20, 6 July 2009 (EDT)
I just went to the GEDCOMs list and reviewed the GEDCOM that User:Loulehmann is trying to import. He/she definitely edited the pages.
I'm not sure that deferring all edits until the GEDCOM is imported is the right solution to the problem though. If I waited to apply all of the edits at once when the GEDCOM was ready to import, there would be an even longer time between the time that the edits were made and the time that you were notified. So more time for the uploader to not remember what edits had been made, and no window of time for you to contact the uploader and ask them to stop making these kinds of edits before they had edited possibly a lot more pages. And it still has the problem of what to do about other edits that were made to the page during the time between when the uploader edited the page and the GEDCOM was finally uploaded. I'd rather address the problem from another angle.
I looked at the edits that were made to Person:Abraham Doolittle (3). The edits look like normal page edits except:
  • The same content is appended multiple times.
  • The MySource links are red
  • The same page is edited multiple times by the same user.
I could try to address these problems by:
  • Check to see if the content being appended already appears somewhere in the Person/Family page, and not allow it to be appended if it does.
  • Creating MySource pages immediately when the MySources are referenced from edited Person/Family pages.
  • In this case the same person appeared in 3 different families being merged, so the uploader was given the opportunity of editing it multiple times (one for each family). I could allow the uploader to edit the same person page only once during a gedcom upload.
--Dallan 00:10, 7 July 2009 (EDT)

If it's the same person (i.e not a dup), then I think it makes sense to only permit (require) a user to deal with that person once. That would kind of drive me nuts if I had to merge Henry VIII 8 times because of all his marriages. That's a mess that's just asking for inconsistency. I think that would fix most of it. If it's not a big deal to do the content check, that would stop not only this, but actual file duplicates, and probably instances where people have copied from what's already there. I'm not so fussed about the red MySource links for the moment - users will hopefully usually merge them themselves, and keeping them red stops me from doing that work for them... ;-)--Amelia 00:31, 7 July 2009 (EDT)


I was going to go in and clean up these recently messed up Doolittle pages, but should I wait until y'all figure out what the problem is? jillaine 12:12, 7 July 2009 (EDT)

No, go ahead. The content of the pages doesn't affect it.--Amelia 12:34, 7 July 2009 (EDT)

I'll do the content check and require the person to be updated once only as soon as I get back from vacation. Both of those are pretty easy; creating the MySources immediately would take a bit longer. Let me know though if that becomes more of a problem and I'll implement that later.--Dallan 13:18, 9 July 2009 (EDT)


Gedcom Importer is stuck [6 July 2009]

It's been chewing on delijim's little file all afternoon and others are piling up behind it.--Judy (jlanoux) 20:13, 5 July 2009 (EDT)

Yes, the gedcom uploader crashed. I just restarted it.--Dallan 00:13, 6 July 2009 (EDT)
Did I read something about Dallas and vacation? You know, don't you, that systems ALWAYS crash as soon as the IT person goes on vacation, right? :-) jillaine 15:57, 6 July 2009 (EDT)
Actually, it's Florida (Orlando - we're going to see a space shuttle launch on Saturday!). And you're absolutely right about systems always crashing when the IT person goes on vacation ;-).--Dallan 22:02, 6 July 2009 (EDT)

Error msg during review of GEDCOM [6 July 2009]

Rec'd this error msg when trying to click on Samuel Jackson & Mary Farrow family pg during upload review.

faultCode:Server.Error.Request faultString:'HTTP request error' faultDetail:'Error: [IOErrorEvent type="ioError" bubbles=false cancelable=false eventPhase=2 text="Error #2032"]. URL: http://www.werelate.org/w/index.php'

But I was able to go ahead and do the merging dupes and updating; and now it's ready for review. --Janiejac 02:06, 6 July 2009 (EDT)

Thanks for letting me know. It sounds like a temporary glitch communicating with the server. I'll look into it when I return from vacation.--User:Dallan 11:02, 6 July 2009 (EDT)
My favorite phrase is "bubbles=false". No champagne on vacation? jillaine 15:58, 6 July 2009 (EDT)

Thank you for all your help [9 July 2009]

Dallan, Thank you for all your helpful feedback. Regretfully I'm finding the WeRelate process a bit too problematic as the following message to Scot will illustrate. Good luck to everybody at WeRelate. It is a very good concept and a good way to collaborate. The merging/matching process may need some tweaking.

Lou Lehmann


Scott,

I've reviewed all of your messages. I think some serious problems are arising from merging. None of the problems you mentioned correspond to the material in the tree which I originally uploaded. Unless there is some system problem, I'm guessing that I didn't scrutinize material enough to notice the errors in other trees - although I must wonder why those errors in the other trees weren't caught during the import process. For the record, I did not use and did not upload any of the following three sources noted in this dialogue: ...."From: Ancestors of Douglas Christian LUSTY http://freepages.genealogy.rootsweb.com/~lustypetersen/d5705.htm ", "From: Family Trees - Throop. monica@odi.ca <monica@odi.ca> ", "S1. New York Genealogical & Biographical Society, "The Record", 36/123, Jan 1905 1 ". Nor did I submit any birthdate at all for Lydia Wills. Of course a 1618 birthdate is absurd but it did not originate with me. Nor did I use any Ancestral Files or World Family Trees as sources. I am puzzled by your allegation that I did not use The "Chapman Genealogy" as a source for the family of Ralph Chapman and Lydia Wills. Here is a copy of my notes about Ralph Chapman. Please note that the last two items are referenced to The Chapman Genealogy, which is also listed in my sources along with "Thirty-one English Emigrants Who Came to New England by 1662" and "Genealogical notes of Barnstable Families"

"Ralph Chapman became a ship carpenter and moved to Southwark, Surrey, across the river from London. On the 13th of April 1635 he sailed on the 'Elizabeth de Lo' (Elizabeth of London) for Massachusetts. His age on the list of passengers was 20 years, which is not quite right according to the date of baptism in the church register. All those sailing from Southwark 'brought cert: from the Minister of St. Saviors Southwark of their conformitie' Ralph settled first in Duxbury and continued his trade as a ship carpenter there. He bought land the 8th of October 1639 and again in 1645. He married in Marshield the 23rd of November 1642 Lydia Wells/Willis who was in all probability the daughter of Isaac and Margaret Wells of Marshfield. She is almost surely the Lydia Wells who came to New England in 1634, apparently as a servant to William Hatch. Her period of servitude would, most likely, have ended in 1641. Isaac and Margaret Wells later moved to Barnstable, Ralph and Lydia Chapman remained in Marshfield.......Ralph Chapman made his willl the 28th of November 1671 and as he did not mention his wife, we can assume that Lydia Chapman died before that date. The will was 'exhibted to the court held att Plymouth 4 June 1672.. .He had a long terminal illness and was unable to sign his will because 'of weakness of body and lamness and swelling of his hands..' " From: "Thirty-one English Emigrants Who Came to New England by 1662" by Dorothy C. and Gerald E. Knoff Gateway Press 1989 pages 52-56

Ralph Champan appeared before the General Court at New Plymouth (4 March 1651) for striking Herman Haddon.. (The Chapman Genealogy by John Harvey Chapman - p. 9)

Ralph Chapman's will was not signed - causing the Court to question his capacities. Witnesses testified that his body and and hands were too swollen to hold a pen or antything else. (The Chapman Genealogy by John Harvey Chapman - p. 12-13)

   I really do appreciate all of the reviews by you folks at WeRelate and I do think that the Wiki concept is a very good one.   I guess I didn't realize how very accountable I would be for someone else's mistakes if I missed them during the review and matching process.  (I still wonder how those trees got imported with the errors which were were attributed to me, apparently as a result of the merging). I think all this is getting to be a bit too much for me so I will just take my tree off of WeRelate.   But I do want to express my thanks for all the very helpful feedback I received during this process.   I certainly don't regret it as the whole experience has helped me to clean up my notes and sources.  Please share my appreciation to everybody who was so helpful and patient during my time with WeRelate.  (I'll send a copy of this to a few of them)

Lou Lehmann--Loulehmann 18:16, 8 July 2009 (EDT)


Thank you for responding. Perhaps I have sounded too harsh, if so please forgive me. If you did not submit the errors, my comments are for whomever did. None of these pages are my family so I had not looked at them before. I got tied into them through Richard Church's page which was merged with another Richard Church that is of my line a year ago. I do not mean to blame you for the errors, but as you do merges or review them as I have done it pays to critique them and attempt to correct errors add sources if possible. I am pointing out errors called to my attention by your edits even though they may have already been there. Hopefully someone among the watchers will correct their own submissions. I did find the New York Record on line and added it as a source and it does not contain the errors. We need concientious researchers as you appear to be here. So I hope you will reconsider your decision. I'm not sure what you mean about the Chapman Genealogy. I was trying to see if I could convert what appear to be sources, but came through as my sources. If they are not under copy right, you can quote as you have done above to clarify what came from them. Seeing mindless errors and bogus sources just push my buttons because they don't add value to the site. The longer you are here the more you get a sense of community and ownership.--Scot 19:15, 8 July 2009 (EDT)

Scot, the idea behind a wiki -- any wiki -- is cooperation and collegiality. When you see an error on a page uploaded by someone else (and possibly edited by several other people since then), and you have better information (especially if you have a good source), then just go ahead and fix it. The biggest mental adjustment folks generally have to make in dealing with wikis, is "ownership." When you upload a page, it ceases to be "your" page. Everyone has access to it now, and everyone is expected to contribute when they have something useful to add to it. Nobody gets graded on the quality of their original uploads and everyone benefits. This is especially true of a very large community like Wikipedia; I've done the original work on a number of articles there, and all of them have been enhanced and expanded by other people since. Those aren't "my" pages -- certainly not anymore. WeRelate is a much smaller community but the principal -- and the mindset -- need to be the same. --Mike (mksmith) 12:32, 9 July 2009 (EDT)

The 1618 birthdate seems to have been the result of merging (3) and (2) all into (1) of this page. However, it was 1621 before that through several modifications, and always set in Marshfield, MA. So the work all along has been sloppy and certainly does not appear to be a result of anything Loulehmann did. The person merging probably did not know anything about this family, and 1618 does not look patently absurd compared to Ralph's birthdate of 1615, nor does a marriage at age 24 suggest it is wrong. So to be more precise, it is most likely the birth location that is patently absurd. However, this is a common error on WeRelate, probably contributed to by some people's practice of putting "Of so--and-so" in lieu of an unknown birth location. --Jrich 21:00, 8 July 2009 (EDT)


I agree with User:Jrich. As best as I can tell, the issues pre-date Lou's upload. From what I've seen, there are issues like this on a fair number of pages. The whole point of a wiki is that we have an environment where eventually these kinds of errors can be corrected. I like what User:Mksmith said - I hope that we make this wiki as cooperative and collegial as possible. Collaboration on genealogical research is a relatively new concept, and email/wiki make it easy for sentiments to come across different from what was intended. Because of this, we need to be extra careful to invite and encourage in our communication.--Dallan 13:18, 9 July 2009 (EDT)


10k GEDCOM waiting [18 July 2009]

Dallan,

Awhile ago, I'd mentioned that I have a GEDCOM of near 10k people. This is a town genealogy I've been working on for many years. It's predominantly German, but also tracks to some extent the emigrating descendants who came to the states.

You can read about it here:

Schwenningen GEDCOM

Back then you'd asked me to hold off until the merge-upon-upload tool was done.

I'm wondering-- especially given the problems that Loulehmann went through-- when/if at all I should upload this beast?

Thoughts?

(Is there a way we might use this example as a test of some sort?)

jillaine 21:43, 9 July 2009 (EDT)

Do you know how much you overlap existing pages? You might want to try uploading the gedcom and browsing the Family Match tab to see how much impact you'll have on existing pages. The gedcom could be withdrawn if you decide to wait till later. --Judy (jlanoux) 21:57, 9 July 2009 (EDT)


There is some overlap with my existing Jillaine Smith GEDCOM / tree because my Schwenningen ancestors are a subset of this tree. But I'd say about 15%. that's a complete guestimate. Supposed it could be as high as 30% jillaine 22:12, 9 July 2009 (EDT)

We have another issue with large gedcom's, which I found recently when someone uploaded a 5K gedcom: our GEDCOM import tool uses a lot of memory to process large gedcom's. The current importer maxes out at around 5-6K. We can process a 10K gedcom, but I'll have to run a special version of the GEDCOM import tool and give it additional memory. Fixing the importer to handle GEDCOM's larger than 5K is kind of low on the priority list right now, but it would be a good test of the difficulties associated with reviewing and importing a large GEDCOM, and would help us determine whether we wanted to encourage large uploads (probably limited to non-newbie users). Also, if it's too onerous you could always delete it before finalizing the import. Could you email me the GEDCOM so that I can upload it when I return from vacation? I'll also fix the multiple-update bug that Lou uncovered before uploading it.--Dallan 09:47, 13 July 2009 (EDT)


Scot offered a suggestion for breaking my GEDCOM into smaller chunks for easier uploading, but the interconnections in this town genealogy are so twisted, I'm not sure how best to select specific chunks.
In the meantime, I've been TRYING to clean it up. Due to this GEDCOM having travelled through several different family tree programs and two different operating it systems, it now has a bunch of junk in it. USERIDs in places where they shouldn't be. Stange GEDCOM codes that look like @cn12345@ appearing in fact and event descriptions. I tried going directly into the GEDCOM with a text editor, but there are so many @cn... instances that it will take me FOREVER (with my skills) to get rid of them, and basically, I'm not sure I SHOULD be. I just don't like how they're showing up in my file.
I have a feeling that there are other ways to clean up this file that are beyond my technical skills. Either way this file has to get cleaned up--before upload or after, and I'd rather do it before. I'm pretty discouraged... :-( jillaine 15:57, 13 July 2009 (EDT)
While I can't do much with a GedCom per se, if you can send me a text file of some descrption I may be able to remove data elements such as @cn12345@. As long as there's a consistent and unique pattern that can be recognized, its usually possible to split out things like this. no promises, but I'll see what I can do. Q 17:03, 13 July 2009 (EDT)

Thanks, Bill, for the offer. I've already tried this. Unfortunately, the numbers following "@cn" are not consistent. I can see no pattern in them. I wish GEDCOMs could be imported into an excel file, with one column per field, then I could simply and easily delete/clear the offending fields.--jillaine 17:32, 13 July 2009 (EDT)

I could be wrong, but the ampersand (especially a set of pair ampersands, should be sufficient to cue a grep search and replace with a tab. That can often be used effectively in combination with a spreadsheet to quickly break things open, get similar fields aligned in a column, and dump the column. The irregularity of what's in between shouldn't be an issue. Its a bit tricky as sometimes the character combinations defy any sensible search and destroy, but usually it works okay. I do a fair bit of this, and am set up to do it routinely. With the right combination of software, its not that hard. You usually need something more than a spreadsheet word processor combination for an editor. But in a pinch Excell-Word are sufficient in many cases. You need Word to search and replace ampersands with a tab, and excell to align the fields based on tab-delimited files. Q 19:15, 13 July 2009 (EDT)

Dallan,

I've been working on getting my town genealogy file ready for sending you. I've also drafted a "home page" for it here:

Our Schwenningen Ancestors

Please take a look when you have some time. As I was putting that page together, I'm wondering if this upload may be more trouble than it's worth, although, frankly, if you really want WeRelate to be a big genealogy wiki, I think we'll need to face such challenges.

Basically, this is a gedcom of a small town in Germany where there were a limited number of both surnames and given names. What I realized is that this means that WeRelate-- and specifically the merge tool-- are going to think that a LOT of people are merge candidates. It will also be confusing for humans not familiar with the town who are busy going around de-duping. There is a very good chance that the system and humans will (reasonably but not accurately) believe records should be merged that should not be.

Thoughts?

-- jillaine 16:41, 17 July 2009 (EDT)


I like the Our Schwenningen Ancestors article! And I agree that we need to be able to handle submissions like this. Regarding the incorrect-merge issue, matches internal to the GEDCOM could be ignored during the GEDCOM review process, but after the GEDCOM is uploaded there might be matches that would be incorrectly-merged in the duplicate review project. What if you were to hold off on uploading the GEDCOM until after we finish the duplicate review project? I think once this project is over we can be a lot more careful about what gets merged going forward. What do you think?--Dallan 12:35, 18 July 2009 (EDT)


okey dokey; there's always more I can keep doing to get the gedcom into better shape. Meanwhile, is there some target date for the dupe review project to be "done"? I figured it would never really be done. jillaine 14:21, 18 July 2009 (EDT)
We have just over 11,000 possible-duplicate families left. This number has been decreasing by almost 2,000 per month due to what you and others are doing. Now that duplicate checking is part of GEDCOM import, we get very few new duplicates, so the project is primarily de-duping old GEDCOM's. I'm hoping that we can finish the existing duplicates by the end of this year.--Dallan 23:29, 18 July 2009 (EDT)

Auto-Generated Category on Uploaded GEDCOMs? [15 July 2009]

Related to the upcoming upload of my 10k person GEDCOM, I'm wondering if there is a way to automatically add a category onto each person page of said upload?

This particular GEDCOM is a part of an evolving (but already large) town genealogy for the town of Schwenningen. I'd like to mark all person pages within it (not sure if family pages should also have this) with a category such as

Schwenningen Town Genealogy Project

or

Our Schwenningen Ancestors

(or whatever I end up naming this project)

The resulting category page could then be the "home" of the project.

-- jillaine 10:34, 15 July 2009 (EDT)


This question is related to the "trees" discussion awhile back, where we talked about the idea of making trees categories. If we were to allow everyone to add categories during GEDCOM upload, we could end up with some pages having a lot of categories, so I'm reluctant to do this. We currently display the users watching a page. I've thought about changing this to also include the trees that the page is in, in which case your tree would show up in the "Watchers" list.

Alternatively, what would you think about creating a new user called "Schwenningen Town Genealogy Project" and uploading the tree as that user? Then that user name would appear in the "Watchers" list.--Dallan 11:23, 15 July 2009 (EDT)


I like your last suggestion; I will do that when ready.

I'm busy cleaning up the gedcom before uploading.

Perhaps this weekend it will be ready for me to send it to you.

jillaine 11:37, 15 July 2009 (EDT)


I'm not at all a fan of having tree names in the Watcher list. My tree names mean absolutely nothing, and I watch thousands of pages not in trees. I've learned a number of user names, and changing them to tree names will remove all that knowledge. (But then, I really have found no use for the tree concept, so take the comment in that vein.)--Amelia 12:00, 15 July 2009 (EDT)

Amelia, the "new user" account would not be a tree name as much as a project name, and an account I would use while working on that project.
Of course if there was some other way to tag all pages generated by this Schwenningen GEDCOM, I'd happily use that as well. I'm just trying to avoid adding 10,000 pages manually to the "Our Schwenningen Ancestors" project. jillaine 17:31, 15 July 2009 (EDT)
I know - I was responding to his idea to change "watchers" to "trees.--Amelia 00:12, 16 July 2009 (EDT)"

Thinking ahead, when everything becomes part of Pando, if you set things so that you were watching a tree, exactly what would you be watching? Pando? If so, would this be meaningful? Q 14:36, 15 July 2009 (EDT)

A "tree" relative to Pando is a section of Pando-- a set of particular roots of Pando. So yes, it's still meaningful. jillaine 17:31, 15 July 2009 (EDT)

Right, a tree is just a set of pages that you can treat as a group in search, GEDCOM export, deletion, and a few other areas. In the end it's nothing more than a set.

I was thinking that I could display the tree names next to the user names, so something like "Dallan/My tree", or "Dallan/Watchlist" if I'm watching the page but don't have it in any tree. But if this isn't a widespread need, then creating a new user with the project name and continuing to display watching users only without tree names, works for me.--Dallan 00:23, 16 July 2009 (EDT)


Sandbox Down [23 July 2009]

Dallan, I know you're on vacation, but I just wanted to let you know the Sandbox site has been down for a few days.--Jennifer (JBS66) 09:50, 13 July 2009 (EDT)

I should have mentioned this earlier. My router at home failed a few days ago, and since the sandbox runs on one of my home computers, with the router gone it's not possible to access it. I'll fix it when I get back from vacation early next week.
(The real site runs on some machines at Amazon, which has been terrific.)--Dallan 10:01, 13 July 2009 (EDT)

The sandbox is back up again.--Dallan 17:54, 23 July 2009 (EDT)


What does "vacation" mean to Dallan and Solveig? [15 July 2009]

Need I say more? ;-) jillaine 10:17, 13 July 2009 (EDT)

Very funny Jillaine :-).
Yesterday was an interesting day. We spent several hours waiting to see a shuttle launch. Ten minutes before launch they scrubbed it because it of rain. There must have been an accident or something on the freeway because it took us 3 1/2 hours to get back to our hotel when it took 1 1/2 hours to drive out there. Why they built the shuttle launch pad in Florida, where it seems to rain a bit nearly every day as far as I can tell, I don't know :-). They've re-scheduled the launch again tonight, so we're headed back out tonight after spending the afternoon at Seaworld.--Dallan 10:25, 13 July 2009 (EDT)
It was a safety issue early on in the program. By placeing it in Florida (and at Cape Canaveral, umm Kennedy in particular), you minimize the potential impact to the civilian population if something unforward happens (e.g., Challenger disaster). Lots of nice unpopulated water nearby for things to land on. Also, originally they were better equipped for at sea landings, so their were technological advantages as well. Q 11:06, 13 July 2009 (EDT)
Ok, that makes sense.--Dallan 11:23, 15 July 2009 (EDT)

Bummer! just heard the news. They're postponing launch until Wednesday. How long will you be there? jillaine 19:04, 13 July 2009 (EDT)


We're going to go to the beach today and will take some time out to go to Space Park in Titusville and try to watch it again. The people at SpaceFlightNow have a Twitter account that lets me get launch updates on my phone, even at the beach. :-) I'm just amazed at how accessible information has become in the past 15 years.--Dallan 11:23, 15 July 2009 (EDT)

We finally saw the shuttle launch today! The weather was a no-go until about an hour before the scheduled launch, and then it cleared up. From the distance we were viewing (10-15 miles), we first saw a big cloud of steam from the rockets vaporizing 300,000 gallons of water that are poured on the launcher to suppress sound vibrations at lift-off, then we saw the shuttle rise slowly and gracefully into the air. About a minute later, when we could no longer see the shuttle, we heard the roar of the engines. It was incredible.--Dallan 00:23, 16 July 2009 (EDT)

Still have WP linking bug [15 July 2009]

(See [7])

Category:U.S. Secretaries of State is still a mess. Did it get fixed and get to be a mess again with a refresh? Or did it just not get fixed?--Amelia 00:56, 14 July 2009 (EDT)

It hasn't been fixed yet. It requires updating the existing templates, rather than just creating templates for newly-added source-wikipedia tags. I thought I would wait until the next WP database dump was available, since that also requires updating the existing templates. I just checked the download site and it looks like the next dump is in progress, so I'll apply it when I get back from vacation. And the categories will be fixed at the same time.--Dallan 11:23, 15 July 2009 (EDT)

Jillaine's test GEDCOM [18 July 2009]

Dallan,

I would have done this on the sandbox, but it's down. I just uploaded a very small gedcom that I'm testing something with. We can happily delete it later, but I want to try something and see if it works.

Thanks.

-- Jillaine--jillaine 15:23, 16 July 2009 (EDT)

No problem, I sent it through.--Dallan 12:31, 17 July 2009 (EDT)

Thanks, Dallan. My test worked. Take a look if you want. By editing the raw GEDCOM (shudder), I managed to add [Category:Our Schwenningen Ancestors] to each person's record (in the memo/notes field). Therefore, I *could* do the same thing to my 10,000 person gedcom (shudder shudder), obtaining the results I seek once uploaded. But um. I figure maybe I should ask permission for that first. In the meantime, still cleaning up my file. (See questions above about this, though...) What say you? jillaine 16:36, 17 July 2009 (EDT)
That's a novel approach! I wouldn't have thought of that. I wonder - it seems like some genealogy programs let you add a note (or is it a source, which wouldn't work?) to each person when you import a GEDCOM. Could you export a GEDCOM and then re-import it into the same genealogy program, adding a note to everyone during import, then export it again with the notes in place?
I think if you want to add a category like this it's fine. I just don't want to make it a standard feature of GEDCOM upload because unlike yours, most GEDCOM's don't warrant their own category.--Dallan 12:35, 18 July 2009 (EDT)

Gedcom export [30 July 2009]

Hi Dallan, I would like to work with you on the gedcom export feature; maybe after this week. Have too much stuff to do this week. I have kind of fine tuned my census source citations, but I think I need another field label added. I think once upon a time it was stated somewhere that it did not matter what you entered where, but it does for the gedcom export. --Beth 19:49, 19 July 2009 (EDT)


Also I am trying to design a method to confine my data entry to WeRelate and export to my Legacy program. I am attempting to figure out if the Intellishare feature on Legacy will work for this. I figure I can share the data with myself although it is designed to share with a group of researchers.

I know that you are extremely busy but just wondering if some point in the future you could add a feature to the export program so one could export only a selected portion of the tree. I know that Janie is working the other way; she is uploading gedcoms and merging with her existing trees on WeRelate. I would like to determine which one is the best method for me. --Beth 21:59, 29 July 2009 (EDT)


I don't believe that I can convert this Intellishare feature for the above stated purpose. I will probably just use Roots Magic because I much prefer the split screen view and drag and drop over Legacy's merge feature. Also I don't think that you need to change gedcom export. I can do that with my genie program. Thanks. --Beth 00:39, 30 July 2009 (EDT)


It looks like you don't need it anymore, but if you wanted to export selected individuals, you could create a new tree, then add the ancestors and/or descendants of selected people to it.--Dallan 10:50, 30 July 2009 (EDT)


Deleting image generated doozie error message [23 July 2009]

I was attempting to delete Image:

Image:OurSchwenningenAncestorsBanner.jpg

and got the following error message.

FileStore::copyFile failed to copy '/mnt/images/d/d8/OurSchwenningenAncestorsBanner.jpg' to '/mnt/deleted-images/0/e/x/0exbhqmju25z2fj6vucv2xmvbg3xg1fz.jpg'

Backtrace:

  1. 0 /var/www/htdocs/w/includes/FileStore.php(86): FileStore->copyFile('/mnt/images/d/d...', '/mnt/deleted-im...', 1)
  2. 1 /var/www/htdocs/w/includes/Image.php(2007): FileStore->insert('0exbhqmju25z2fj...', '/mnt/images/d/d...', 1)
  3. 2 /var/www/htdocs/w/includes/Image.php(1951): Image->prepareDeleteVersion('/mnt/images/d/d...', 'Going to replac...', 'image', Array, Array, 'Image::prepareD...')
  4. 3 /var/www/htdocs/w/includes/Image.php(1849): Image->prepareDeleteCurrent('Going to replac...')
  5. 4 /var/www/htdocs/w/includes/ImagePage.php(539): Image->delete('Going to replac...')
  6. 5 /var/www/htdocs/w/includes/ImagePage.php(497): ImagePage->doDelete()
  7. 6 /var/www/htdocs/w/includes/Wiki.php(341): ImagePage->delete()
  8. 7 /var/www/htdocs/w/includes/Wiki.php(50): MediaWiki->performAction(Object(OutputPage), Object(ImagePage), Object(Title), Object(User), Object(WebRequest))
  9. 8 /var/www/htdocs/w/index.php(123): MediaWiki->initialize(Object(Title), Object(OutputPage), Object(User), Object(WebRequest))
  10. 9 {main}

jillaine 12:44, 20 July 2009 (EDT)


Thanks for letting me know. I fixed the problem and deleted the image :-) --Dallan 10:50, 23 July 2009 (EDT)


Free-form Sources [27 July 2009]

Dallan, please visit Help_talk:GEDCOM#Sources_that_use_.22Free-Form_Text.22_.5B20_July_2009.5D. Your guidance is needed. Thanks. jillaine 19:22, 20 July 2009 (EDT)

Works now.--Dallan 10:59, 27 July 2009 (EDT)

Too many emails... I think [27 July 2009]

So I just got about 20 emails from a user "adding data from GEDCOM". But for each and every one I click on, there is no change. What's going on and why am I getting the emails? See, e.g., Person:Jonathan Hatch (6).--Amelia 00:25, 26 July 2009 (EDT)


Amelia, just happened upon your question while checking my watched list. See the link for the difference in versions for Person:Jonathan Hatch (6). There appears to be a change by a user; says that child of family was added or am I confused? Link [8]--Beth 12:17, 26 July 2009 (EDT)


That's not actually a change, at least to a human. The same family is listed as the parent before and after.--Amelia 12:23, 26 July 2009 (EDT)


I understand now Amelia. Appears to be a bug. If I check the page history back to the original page, it appears that the family was added in the latest update. But if I actually select to view page for the first and subsequent revisions; you are correct the same family is listed on every revision. --Beth 12:42, 26 July 2009 (EDT)


Family was not just added in latest revision, was also subtracted from previous version. So no net change, but order of the XML items changed (as displayed in the diff) so not identical if straightforward comparison is done by the system? Person probably matched to this page during GEDCOM upload, or was it a merge with nothing new kept during the merge, the page got saved, perhaps to add the person to the watchlist? During the save, the order of the XML changed, creating the appearance of a net change when there was none?? I've gotten some of these, too, as well as similar ones where it only appears a blank line is removed from an otherwise empty narrative, nothing else. --Jrich 18:09, 26 July 2009 (EDT)


Thank-you for pointing out the problem. The problem is due to propagation not adding the family element in the right order, so when the page is involved in a gedcom update (which puts the family element in the right order), it looks like the page has changed when it hasn't. The same thing would happen if you were to edit the page online and just press "save" -- the family would be re-ordered and it would appear that the page had been changed when it hadn't. It looks like there's be a similar problem with blank lines in an empty narrative. What I need to do is fix how family propagation works so that the family element is put in the right place to begin with, which will take about a day. Before doing this I'd like to get the new wikipedia refresh (with the category fix), renaming the source pages, and the UI items under discussion implemented, unless people would really like to see the propagation fix right away.--Dallan 10:59, 27 July 2009 (EDT)


How to refer in text to a category? [29 July 2009]

I'm drafting an explanatory page for the First Families of Louisiana program -- I'm going to be adding pages for Louisiana colonial ancestors who have been qualified by the program -- and I'm setting up a category where the names will appear. (And there's a banner involved, but I know how to do that.) When I try to refer to the category in the text, however, I'm seeing puzzling behavior. Obviously, I'm not understanding something about the way categories work. The relevant line is near the bottom of the page. When you have a minute (assuming you ever do . . .), could you take a look at that? Under "Edit," you can see what I'm trying to do. I expected it to behave the same way as a Person or other page, with a link to the Category page, and the alternate wording following a pipe, and all that. Instead, it's simply invisible. Is this perhaps because there isn't really a category "page," since the information is created dynamically? --Mike (mksmith) 16:07, 29 July 2009 (EDT)

Mike, to make that work you need to use the form [[:Page Name]]. Don't include the url, just the WeRelate page name, PLUS the ":" before the page name. I don't know for certain, but I assume that the url address is preventing this from being picked up as a category. But then you have a second problem. If you don't include the ":", you'll simply be telling the system to include the page as an item in the category, and no link will be created except at the very bottom of the page. You want the link embedded in the surrounding text, so you need the ":" Q 16:18, 29 July 2009 (EDT)
Hmmm. Well, that simply makes the words appear as bolded text. No link at all. And how would the system know what namespace to find the Category page under? I know if there's no namespace included -- and no colon -- the search for the link defaults to the "Article" namespace. But that doesn't work for categories. --Mike (mksmith) 17:12, 29 July 2009 (EDT)

Hi Mike, I believe that I fixed the link for you. Check out the page. It didn't work because one failed to include the namespace after the colon. --Beth 17:33, 29 July 2009 (EDT)

Yes, I normally think of "namespace" as part of the page title. Should have been more explicit. Q 18:14, 29 July 2009 (EDT)

Please import my gedcom CaryWR.ged [1 August 2009]

There may be a few duplicates which I will merge. But I need a new tree for these people so I can send to some new collaborators that I am trying to entice to join WR. Most of these are new people. --Judy (jlanoux) 13:51, 1 August 2009 (EDT)

Done.--Dallan 21:37, 4 August 2009 (EDT)

My GEDCOM is a little larger [1 August 2009]

I just registered......and, then found out about the 5,000 people ceiling on GEDCOMs. Mine has a little over 7,500. How long do I have to wait, without a family tree on here, in order to be able to upload my GEDCOM?

--Account removed at user's request 17:14, 1 August 2009 (EDT)


Refreshing links [5 August 2009]

A wishlist item...Is it feasible to have incoming links (the 'what links here' items) refreshed when a page is renamed or merged? I have been using a lot of links which is a big advantage of a wiki. But I've noticed that there is a lot of churn as we work on improvements and pages get renamed or folded into other pages. Also could the watchlist be notified when these links are changed? --Judy (jlanoux) 08:28, 3 August 2009 (EDT)


When a person/family page is renamed, I automatically update the relatives that linked to the page to link to the new title, so they should appear in the what-links-here list. In all other cases of page renaming, the what-links-here list includes the originally-named page, and all of the pages that link to the originally-named page should be listed under it on the what-links-here list, so we should be covered in all cases for renaming. When a page is merged, I update the people/families that linked to the merging page to now link to the merge target, so they should also appear in the what-links-here list for the merged page.

When a page is renamed, people on the watchlist aren't notified, but when a page is merged, they should be notified.

Can you tell me more about what you're looking for, or if you've noticed a problem?

We did have a problem Sunday to Monday noon where change notification emails weren't being sent. This is fixed now.--Dallan 21:37, 4 August 2009 (EDT)


I guess that wasn't clear - I'll try again. The problem is when a page is changed 'indirectly' by a change of title on a linked to page.

When I link to a source page, then later someone changes the title of the source page (there's a lot of that going on), my page isn't updated with the new title and I am never notified that the title has been changed. Yet when someone looking at my page clicks on the source, they see the new title. It's possible that the new title isn't what I intended when I made the link, so I want to know when this is done so I could do any fixes needed on my page.
I don't want to automatically notify people when source titles change because the changes are generally just to standardize the title and I don't think most people would care. Once we start displaying the full source citation on person/family pages in place of the source page title, I think it will be more important for people to be notified when the citation information is changed on the source page. This could be done by encouraging people to watch the source pages that they reference, or maybe automatically-watching referenced source pages could be a system preference.
I haven't tested Person links, but from discussions I read as a newbie (a few months ago) I understood that if I have a link to another Person page in the body of the page somewhere (say a census listing transcript) and that Person page title is later changed, that my link would 'turn red', but I don't think that I would be notified.
The link in the body of the page should still be blue and should automatically redirect you to the person page with the new title.
So my request is that the originating page (the one at the other end of the link) get updated when a page title is changed. In other words, I would like every page on the 'what links here' list for the changed page to be updated with the new title and the watch list for that page be notified.
I'm starting to realize that one changed source could kick off a thousand emails. Maybe I have to keep all my sources on the watch list. And my request would be modified to simply "refresh the 'linked from' page with the new page name".--Judy (jlanoux) 10:47, 5 August 2009 (EDT)
Given that clicking on the old page title on the 'linked from' page automatically takes you to the new page title, do the 'linked from' pages need to be changed to have the new page title? I'm hoping not. Perhaps an "auto-watch referenced source pages" system preference would take care of the need.--Dallan 12:04, 5 August 2009 (EDT)
I think you're saying that broken links is not a problem, I must have misunderstood. So I'll just put the sources I care about on the watch list. That was a quick fix - thanks! --Judy (jlanoux) 13:36, 5 August 2009 (EDT)

Audio [10 August 2009]

Hi Dallan,

Is their a way to upload audio to WeRelate? Not that I have something now but this is a what if question. Suppose I interview a 106 old person still living on the estate of someone who died in 1928 and this person has stories from memories about the estate and the person who died in 1928. Is their a method in place on WeRelate to preserve the interview. --Beth 22:13, 5 August 2009 (EDT)

Beth, there are probably other ways to do this, but the way I found would work is illustrated at Grey_Wolf. This involves placing the soundfile into the digital library in this case as an MP3 file, the creating a link to that file. Clicking the link to that file will open a window in which it will play. It would be good to have a way to play the soundfile without going off the page. There's probably a way to do that, but I haven't explored that further. It may be that you can save the soundfile as an image or something like that so that it will play as you view the page. Q 08:07, 6 August 2009 (EDT)
I'm reluctant to store audio files in the wiki because they have the potential to occupy a lot of space, and if a lot of people were uploading audio files that would become a problem. So Q's suggestion, or uploading the audio file to another website, seem to be the best approaches.--Dallan 12:48, 7 August 2009 (EDT)
At some point I understand the DL will be taken out of Beta, and mainstreamed into WeRelate. Since I've used it to store some sound files (e.g., Appalachian Music, etc) for use in conjunction with the Southwest Virginia Project, will mainstreaming the DL interfer with use of these files? And you know, it would be really cool if you could open up the main page on Southwest Virginia Project and hear St. Anne's Reel playing in the background while you viewed the page. I do understand about the space limitations...and some of the risks. Maybe uploads by management only on special request? Q 13:09, 7 August 2009 (EDT)
Okay, if I obtain an interview I will place in the digital library with a link. --Beth 18:13, 7 August 2009 (EDT)
Uploads of large files by special request seems like a good idea. I don't mind large files; I just wouldn't want it to get abused.--Dallan 22:37, 7 August 2009 (EDT)
Go to The Tapestry, there's a link under the Banner and welcome marque. Click the link and you'll be taken to a new window where you can play the tune. If I create a new tab in my browser and return to the Tapestry, the song continues to play in the background. I'd like to by-pass creating the tab dodge, and just let the tune play automatically when the link was clicked. This is the most I envision doing with sound files in the way of a special request. What do you think? Q 20:35, 8 August 2009 (EDT)
Guys, if you're seriously thinking about installing a music clip that plays automatically, I urge you to reconsider. Websites that play clips when you log onto them, especially with no way to turn them off, are the bane of librarians and teachers. Nobody wants to hear music blaring from a public-access computer in a library or a school language lab. Seriously. It's far worse than blinking text. --Mike (mksmith) 20:54, 8 August 2009 (EDT)
What was requested was for "the tune play automatically when the link was clicked" That is automatically in the sense that you didn't have to a) go off the page to play the tune if you wanted to, and b) didn't have to use the additional tab dodge. 21:53, 8 August 2009 (EDT)
I agree - sound should play only when the user clicks on something. I'll put this on my todo list - being able to add a tag to a wiki page that would play a sound stored in the digital library.--Dallan 23:25, 9 August 2009 (EDT)
Thank you! Q 08:02, 10 August 2009 (EDT)

Kilmersdon Somerset England Place page [1 September 2009]

Hi Dallan

I was hoping to access the transcriptions listed for Kilmersdon that are listed on the place page but unfortunately only the first one for baptisms comes up. All the others come up with an error message. Is this a problem that can be fixed or have the details been deleted? It is wonderful to find this great site and to actually find transcripts for Kilmersdon are available only to find I can't access them. Can you help?

Kind regards

Siowiel--Siowiel 10:30, 11 August 2009 (EDT)

It looks like you can get to the records here: http://kilmersdon-radstock.org.uk/kilsom.htm . If you wouldn't mind, perhaps you could edit Place:Kilmersdon, Somerset, England and replace all of the broken links with this url?--Dallan 13:53, 11 August 2009 (EDT)

All done and I have added the new transcripts now completed. I have checked all the links work.--Siowiel 09:47, 12 August 2009 (EDT)

Thanks!--Dallan 23:26, 12 August 2009 (EDT)

Dallan

The server link seems to be broken on Kilmersdon again and none of the links are working. regards Siowiel--Siowiel 08:41, 23 August 2009 (EDT)

Sorry, it appears that the entire server is down this time. I don't think there's anything we can do.--Dallan 13:58, 1 September 2009 (EDT)

Place editing [17 August 2009]

Hi Dallan,

What is the difference in this edit? [9] --Beth 20:58, 17 August 2009 (EDT)

It looks the same to me too. Probably a difference in spacing that's invisible to everyone but the computer (maybe a hard space being converted to a normal space).--Dallan 00:35, 18 August 2009 (EDT)

Is this a bug? or did the missing comma cause it? [18 August 2009]

Dallan the WeRelate agent just changed the area covered for this source: Source:Burgner, Goldine Fillers. Carter County Tennessee Marriage Records 1796-1870. Would you compare the 14 June version to the 17 August version to see the change. The 14 June version correctly had the state to be Tennessee; WeRelate agent changed it on 17 August to be Montana! I have changed it back to Tennessee. But I'm glad I checked to see what change had been made. Wondering how this happened. Perhaps my lack of a comma between county and state confused it. --Janiejac 23:13, 17 August 2009 (EDT)

Thank-you for pointing this out. It was the extra s in "Tennesssee" that confused the agent. (The place was listed as "Carter, Tennesssee, United States") The agent couldn't understand Tennesssee and it incorrectly standardized the place to Carter county in Montana. I just checked and it looks like roughly 5 other sources had similar problems. I'll fix the others.--Dallan 00:30, 18 August 2009 (EDT)

I think I have one, too. On the Source:United States, New York, Kings. 1920 U.S. Census Population Schedule page, the place-name box used to show this: "Kings, New York, United States|Kings County (Brooklyn), New York". Now, with the bot changes, it shows this: "Kings County (brooklyn), New York, United States". Besides the mis-capitalization (which may be systemic, which is why I mention it), that doesn't look like what I understood was going to be the result of the change. --Mike (mksmith) 14:03, 18 August 2009 (EDT)

Thanks for pointing out the problem. When I re-standardized the source places, I took what was listed after the bar: "Kings County (Brooklyn), New York" and standardized it. What I should have done in this case is kept what you had before the bar, instead of standardizing what you had after the bar (which lowercased Brooklyn). This appears to be the only source that with "Kings County (Brooklyn), New York" that was mis-standardized. I went ahead and fixed it.
There might have been some other sources with |'s in the places where I should have taken the text before the bar instead of trying to standardize the text after the bar, but I added the place drop-down to sources fairly recently so the vast majority of source places don't have |'s. I'll be posting a list of all human-edited or linked-to sources and how they'll be renamed for people to review soon - hopefully later today.--Dallan 16:05, 18 August 2009 (EDT)
Well, here's another with a similar problem: Source:United States, New York, New York. 1920 U.S. Census Population Schedule; "manhattan" in lowercase. I've corrected this one, but there may others. I kind of didn't think there was supposed to be anything at all in parentheses, though. --Mike (mksmith) 16:18, 18 August 2009 (EDT)
---> Now that's downright strange. I capitalized "Manhattan" and saved the page . . . and it didn't take. I tried it a couple more times, and it (the system, I guess) won't let me change the letter. It's still lowercase. Is there some sort of protection in effect while the bot is doing its thing? --Mike (mksmith) 16:23, 18 August 2009 (EDT)
It's trying to standardize the place you enter by matching it to an existing place. (I added standardizing source places, just like we currently standardize event places on Person/Family pages, about a month ago.) It doesn't recognize "New York County (Manhattan)", so you get a red link and Manhattan is lowercased. The place wiki treats New York boroughs as sub-jurisdictions of the counties (according to Wikipedia), so I edited the page and standardized Manhattan to "Manhattan, New York, New York, United States". In the upcoming rename, this Source page will be renamed to "Manhattan, New York, New York, United States. 1920 U.S. Census Population Schedule".--Dallan 18:08, 18 August 2009 (EDT)

Blockquotes [18 August 2009]

I can't seem to get block quotes to work properly. The text is indented from the left margin, but not the right. When using block quotes for lengthy quotes, the text should appear inset from both sides. Wikipedia says <blockquote> tag works this way. But it doesn't seem to work this way here. Example: Person:James Cary (10) has a quote right at the top of the text box.--Judy (jlanoux) 10:57, 18 August 2009 (EDT)

I've put this on my todo list for when the source renaming project is over.--Dallan 16:05, 18 August 2009 (EDT)
Thanks. The ads seem to be causing the problem. It does indent on the right once you get below.--Judy (jlanoux) 18:16, 18 August 2009 (EDT)

Source page-place issued getting fixed? [19 August 2009]

Can we have the Place Issued field on the Source page excluded from the place standardization efforts? This field is for the purpose of helping to identify the book, not so I can go visit the publisher. It should read the way the title page reads. "London" should stay "London". I would prefer to see this as a plain text field and not linked to the place database at all. I'm always having to fight the dropdown when I'm typing. --Judy (jlanoux) 08:45, 19 August 2009 (EDT)


I second that emotion jillaine 09:06, 19 August 2009 (EDT)
What a coincidence. I had the same thought yesterday while reviewing source renames and turned off the drop-down list and standardization for place issued. And it shouldn't be linked to the place wiki; it should display as plain text. If it's still happening, please let me know.--Dallan 10:36, 19 August 2009 (EDT)



A couple of non-urgent suggestions [1 September 2009]

Two things kind of came to mind in the source-renaming project -- but aren't actually part of that, so I'll put them here:

  • There's presently no way to properly indicate that a book-type source is multi-volume. It makes a difference in the citation, esp. when each volume starts over with Page 1. I've been putting this info down in the text box for lack of better: "Published in 5 vols." I would suggest an additional field just before or just after the "Periodical/Series name" field called "Pages/no. of vols." You might add "(optional)" after that, so page-creators won't feel pressured to count pages in a book. In auto-constructing a footnote-type citation, the number of volumes wouldn't enter into it -- but if you have in mind at some point auto-constructing a bibliography-type listing, it does: Blow, Joe. History of My Life. 20v. Peoria: Narcissism Press, 2020.
I'll add this field next week.--Dallan 14:07, 1 September 2009 (EDT)
  • On Person pages, the "Events and Facts" set of fields includes "Christening." This is sort of denomination-specific, isn't it? Most of the birth-substitute records like this that I've come up with are baptism records, not christenings. Yet they're equivalent in utility as being close to date-of-birth, . . . as long as you're dealing with a denomination that practices infant baptism, which is why I don't personally feel comfortable sticking a baptism date in the "christening" field and pretending it's the same thing. You might consider changing the label to "Baptism/Christening" (or an abbreviated version so the fields don't get pushed over too much). Or you might even add a separate "Baptism" field, though that probably isn't necessary. --Mike (mksmith) 12:40, 21 August 2009 (EDT)

There is an other event type for Baptism. According to what I remember of the GEDCOM standard, the description of Christening (CHR) and Baptism (BAPM) are nearly identical. There is a separate tag CHRA for adult christening. The question is, does Baptism get used in lieu of birth the way Christening does when there is no birth date to display? I am guessing it does not and have never used Baptism for that reason. --Jrich 13:53, 21 August 2009 (EDT)

Well, maybe it's because so much of the research we do is in Louisiana, which means sacramental records, which means baptismal (not birth) records. And among all my frontier Protestant ancestors, there's not even any such rite as "christening." I know baptism is on the "Other events" drop-down -- but that means it's at the end of the events list, not the beginning, where it almost always ought to be. If the events ever get properly chronologized [?], maybe it won't be a problem. --Mike (mksmith) 16:47, 21 August 2009 (EDT)
In Gedcom, CHR="The religious event (not LDS) of baptizing and/or naming a child". Sounds like using this for baptisms is perfectly appropriate, though others may take the two events to mean different things? My original understanding is very much what you are saying, but I adjusted based on the functionality of WeRelate which uses the two events differently. It will copy a Christening date onto the family page when there is no Birth date, but it does not do this with Baptism. I have no problem changing labels, but I have been entering the baptisms from English parish registers into the Christening field so they will act as a proxy birth, on many, many pages that I doubt I could begin to go find and change. Should one of CHR or BAPM be allowed, but not both, and whichever one is used, displays the same, or does WeRelate need to support both? It doesn't seem to offer an event type for Adult Christening (Robert Duvall in Tender Mercies)? --Jrich 17:25, 21 August 2009 (EDT)
I wasn't aware of that definition of christening in the GEDCOM specs (which I haven't read through in a very long time). I think they're technically wrong but this is genealogy, not anthropology or theology, so I'll regard the two as functionally equivalent. And since I can blame it on GEDCOM, I won't lose sleep over it! :-) (I spent nearly all of my adult life in north Texas, surrounded by fifty flavors of Protestants -- and then twelve years ago, I moved to south Louisiana, where most of the population is default Catholic. It's been a "learning experience". . . .) --Mike (mksmith) 18:59, 21 August 2009 (EDT)
This conversation has been a learning experience for me. I figured that everyone called infant baptisms "christenings". What would you two suggest? Should we merge CHR and BAPM into the same field and label it something like "Christening/Baptism" (or maybe just "Christening")? I'd prefer not to try to copy both Christening and Baptism separately to the Family pages if we can help it. I do expect to display all events chronologically when I re-do the Person/Family page display later this year.--Dallan 14:07, 1 September 2009 (EDT)

already imported tree??? [1 September 2009]

I know I'm getting forgetful, but I sure don't remember trying to upload this tree before: John Jackson of Loudoun Virginia. But in my list of trees there does seem to be now 3 trees listed in blue saying import a GEDCOM. What does this signify? What is the name of the tree that John Jackson of Loudoun Virginia is attempting to duplicate? --Janiejac 22:58, 23 August 2009 (EDT)


My upload of John Jackson of Loudoun Co overlaps a previous upload. Lets upload mine anyway. So I can see where the duplicates are. I'll exclude or merge whatever needs it. --Janiejac 07:38, 26 August 2009 (EDT)

I took care of this (finally).--Dallan 14:28, 1 September 2009 (EDT)

Link problem, still [1 September 2009]

I hate to bring it up, but while the links to the three misc. source lists now work, the first one and third one both take you to "MiscOnePlace". Sorry. --Mike (mksmith) 18:32, 25 August 2009 (EDT)

Yes, thanks for pointing it out. I can't believe I've had such difficulty with that.--Dallan 14:28, 1 September 2009 (EDT)

Gedcom upload [1 September 2009]

Hi Dallan, in the duplicate pages review, there are many pages where the generations are somehow confused and the sons are married to their mothers etc. Is this problem resolved with your newest gedcom upload and review? --Beth 19:46, 27 August 2009 (EDT)

Now we analyze the gedcom and list warnings during the upload process. The uploader has to click on each warning one at a time to review it. GEDCOM's with too many warnings aren't supposed to pass the review process. I don't currently have a specific warning for "married his mother" but there are warnings about people who get married too old or too young which I think will pick up most of these problems.

Adminstrators group on Yahoo [27 August 2009]

When you get time please read the messages on our Yahoo group? --Beth 19:47, 27 August 2009 (EDT)


"Losing my religion" (well, not mine actually, but some may be missing?) [1 September 2009]

Dallan, you may want to add Unitarian Universalist to the drop-down list of religions; just looked at a book source to update it and couldn't find that religion in the drop down box. Also Swedenborgian. Haven't seen any sources with that yet, but there is a Swedenborgian church near me here in Bath that has a significant role in our local history; it might come up...

Brenda --kennebec1 09:00, 29 August 2009 (EDT)

done.--Dallan 14:28, 1 September 2009 (EDT)

User Name [31 August 2009]

Dallan, how do I (or can you) set up my signature to show Brenda and Kennebec1? I see for example that Judy and Mike have their names and their user name integrated into their signature. That would save me a bit of typing "Brenda" every time before I click the signature box... Not urgent, obviously. Thanks, Brenda --kennebec1 09:25, 29 August 2009 (EDT)

Brenda, I just checked to see how I did that. Under preferences, user profile I have my nickname set to "Judy (jlanoux)". I think it uses that automatically in a signature. --Judy (jlanoux) 11:13, 29 August 2009 (EDT)

Dallan, judy's correct and I fixed it myself. Thanks! --Brenda (kennebec1) 10:04, 31 August 2009 (EDT)


Pando and always free [2 September 2009]

Hi Dallan,

New users will never even find the duplicate project page; Pando is not on the home page that I can find and I really do believe that we are asking for more headaches unless you spell out how it works in plain talk. I also recommend removing that it will always be free. Just leave it that is free now. Reminds me of another site that I was on that would always be free but then had to charge a minimum amount to keep the site going. That was not a popular choice. If you remove the last statement it really doesn't change anything. Reminds me of the "no new taxes" phrase. --Beth 00:03, 1 September 2009 (EDT)

I don't really want new users to find the duplicate project page, but I agree that we need to talk about Pando on the main page. I'll add it shortly.
I really do intend for WeRelate to always be free. One nice difference between now and 10 years ago is that the cost of website hosting has dropped significantly. The ads and donations just about cover the cost of hosting.--Dallan 14:28, 1 September 2009 (EDT)

Um... how attached to Pando are you, Dallan? Coming from a communications/marketing background, I can't help but encourage you to reconsider. Especially as you've recently worded the opening paragraph on the main page. "We call this free public resource pando for genealogy." I thought the name of this free public resource was WeRelate? You're creating brand competition within your own business. What do you want to be known as? WeRelate or Pando? I think you've got to choose. My sense is that you are in love with the imagery of Pando-- it really is quite great. But adding that word-- which most people are not familiar with-- confuses your audience and blurs your brand.
I won't bill you for this. ;-) jillaine 16:10, 1 September 2009 (EDT)
I'm always excited to get good advice! :-) I added the message because of the flak Beth's been getting lately where people don't realize that they're contributing to a shared genealogy space. I wondered about featuring the word pando too prominently, but didn't think too much about it obviously. What would you suggest? Should we drop the second sentence and make the "unified tree" a link to the pando page? Should we do something else entirely?--Dallan 16:28, 1 September 2009 (EDT)
I suggest you draw from some of the language here ;-) jillaine 20:56, 1 September 2009 (EDT)
I just took another stab at it. I don't want to make it any longer (shorter would be ok). Feel free to modify it if you like; I'm not a marketing person :-).--Dallan 22:01, 1 September 2009 (EDT)
Much better! jillaine 07:39, 2 September 2009 (EDT)

Duplicate review project [6 September 2009]

Actually I was going to finish it by tomorrow except for a few crumbs until I overloaded the mail. Anyway it will be finished in 2 weeks. --Beth 00:07, 1 September 2009 (EDT)

You deserve a huge THANK-YOU for this!--14:28, 1 September 2009 (EDT)
I have pronounced this project finished. Thanks to the many people who have worked on it. Update all of the pages and you will see that there are only a few pages remaining for various reasons. --Beth 00:21, 3 September 2009 (EDT)
That's amazing Beth! You've done an incredible amount of work on this project. Thank you!!!--Dallan 06:24, 6 September 2009 (EDT)

Gedcom upload and surname unknown [2 September 2009]

Dallan,

If possible could you convert LNU to Unknown. LNU is last name unknown. This will be consistent with WeRelate and possibly eliminate some duplicate pages. --Beth 00:01, 2 September 2009 (EDT)

I didn't know that. I'll do that when I return from FGS. It looks like we have about 6500 people with either LNU or FNU.--Dallan 14:52, 2 September 2009 (EDT)

I've also seen variations on the theme - Maiden Name Unknown (MNU).--Jrm03063 16:44, 2 September 2009 (EDT)

Merging duplicates and the letter U [2 September 2009]

Hi Dallan,

The site is slower than usual; but still working; not as bad as it was last time. Anyway this long string beginning with Family:Nn Unknown and Nn Wife (1). I don't know how to check these. I cannot search for duplicates because the name cannot be parsed. --Beth 19:49, 2 September 2009 (EDT)

Also the long string starting with Family:Nålevende Unknown and Nålevende Unknown (1), checking the first seven it is impossible to tell so I don't think this string is worth the trouble. --Beth 19:54, 2 September 2009 (EDT)

Where are the maps and gazetteers? [8 September 2009]

The video that I watched, that induced me to get in here (along with the name cachet of the Allen County Public Library), showed that there are maps we can use and historical gazetteers we can consult to look for out-of-date place names.

Where are they? I cannot find how to find them so I can use them to look up a place name!

Help!--Bitbucket001@comcast.net 11:29, 6 September 2009 (EDT)

WeRelate doesn't contain many maps/gazetteers itself. To find out more about a historical place you have two options:
Select "WeRelate" from the Search menu, then change the "Namespace" from "All" to "Place", and enter the name of the place you're looking for. This will search the place wiki pages on both current and alternate (historical) names of that place. For places that have latitude and longitude coordinates set, you can see where the place is on a google map by clicking on the place in the search results list. The google map appears on the left-hand side of the place wiki page. The place wiki pages include places from FamilySearch, Getty, and Wikipedia. It's certainly not comprehensive, and you are encouraged to add additional places that you know about.
Alternatively, you can search for maps & gazetteers covering a place by selecting "WeRelate" from the Search menu, changing the Namespace to "Source", setting the subject to Maps/Gazetteers, and setting the Place to the place you are interested. A search of maps/gazetteers for Germany yields 423 results for example. Most of them are microfilms that you'd have to order through a family history center.
I'm going to change the "Search" menu later today to include Places, Sources, People, etc. as options directly on the Search menu, so if you read this message after I've done that, instead of selecting "WeRelate" you'd select Places or Sources from the "Search" menu respectively.--Dallan 14:37, 8 September 2009 (EDT)

GEDCOM review for existing GEDCOMS? [9 September 2009]

An idea for consideration:

Enable the possibility to take a GEDCOM that is already online (i.e., uploaded before the gedcom review tool was completed) and run it through the GEDCOM review tool.

That would be totally cool

-- jillaine 10:36, 9 September 2009 (EDT)

Once we get gedcom re-upload working (hopefully before the end of this year), we'll be able to support this.--Dallan 10:40, 9 September 2009 (EDT)

People in my network [10 September 2009]

Hi Dallan, we talked about this eons ago but I don't think I am watching any of these pages and they are still there. [10] --Beth 20:36, 9 September 2009 (EDT)

I fixed the earlier issue, but the issue now is that the pages you share in common are redirects. You're not watching the target of the redirect, but you're watching the redirected page. You can unwatch the redirected pages by clicking on the links to the pages you share, then clicking on the redirected from page title link just under the page title, and clicking "Unwatch" on the redirected page. Now that we're the duplicate project is complete so we're merging a lot less often, watching redirected pages won't be such a common occurrence going forward.--Dallan 14:25, 11 September 2009 (EDT)

Home page changes - removal of access to Dallan [10 September 2009]

I guess I need to move your address to some other place. I always accessed your talk page from the home page and cannot find it anymore. --Beth 20:40, 9 September 2009 (EDT)

I moved it off when I added the "Support" forum link, because I wanted to encourage people to post questions to the forum. I've decided to go ahead and put in a simple "bookmarks" feature later this year. Hopefully that will help.--Dallan 14:25, 11 September 2009 (EDT)

Articles in a tree and gedcom export [15 September 2009]

Dallan, are articles in a tree (not the person or family pages) exported in the gedcom. If so to what tags are they routed? --Beth 19:49, 10 September 2009 (EDT)

GREAT question beth. I want to know what happens to articles that are tagged as part of a tree, too. By tags, do you mean GEDCOM field tags? jillaine 20:42, 10 September 2009 (EDT)
Exactly, Jillaine, gedcom field tags. --Beth 21:23, 10 September 2009 (EDT)
Articles, images, and other types of pages are currently not included an exported gedcom because gedcom doesn't seem to have a good place for them (and it doesn't store images). I'm open to suggestions on how to incorporate them. How would you like to see them represented in the gedcom? As sources?--Dallan 14:25, 11 September 2009 (EDT)
Well, what kind of articles do you have Jillaine? The 2 I created recently are a worksheet and misc. notes. Maybe a note field; but how to determine which one? I am not really interested in downloading the images. The few that I may be interested in I could do so on a one by one basis. --Beth 14:38, 11 September 2009 (EDT)
Actually, I have no idea. I know that I've periodically attached pages to a tree, but now that you ask, I realize I have no way of knowing which ones those are. jillaine 21:19, 15 September 2009 (EDT)

Jillaine, go and view your tree and select articles and find out.--Beth 21:26, 15 September 2009 (EDT)


Changes to Home Page [15 September 2009]

Dallan,

In your effort to draw new users away from asking questions directly of you to the WeRelate_talk:Support page (which, btw, I think is a very good idea), I fear you've inadvertently misdirected folks to the Community portal talk page. (A new user recently posted a support question there.) This user saw and clicked on "Start Collaborating" before they saw -- "below the fold" -- the link to the support talk page.

This situation had me review the layout of the home page. Home pages are always tough. You want to convey as much as you can "above the fold" (I.e., without requiring users to scroll.)

I took a look at the code and attempted to create an alternative over on my sandbox page, but the very thing I want to change appears not to be changeable. I could not find the header "Welcome to the World's Largest Genealogy Wiki!" and figure it must be the page title or something. Can you suppress the page title's appearance on a page so that you can have more control over the home page. Here are my recommendations:

  1. Center "Welcome to the World's Largest Genealogy Wiki!" over the MIDDLE column, so that you can bring up the right-hand column to just below "Article, Talk, etc..."
  2. Keep content in the News box to no more than half of what's currently in there-- i.e., no more than 5-6 lines.
  3. Move "twitter" and "blog" to the bottom of the far-left column.

If we keep "Did you know?" content to three lines, the above changes *should* bring the "Questions & Comments" box up high enough for people to see.

Is #1 possible?

-- jillaine 08:26, 14 September 2009 (EDT)

I think there probably needs to be a fourth big button in that middle column for asking questions at Watercooler. And you can set the length of the "News" (in the style sheet) to limit the number of lines before getting to "Read more..." -- but I believe Dallan said he knew the current news was longer than usual. And I would definitely move "Did You Know" to the bottom of the stack. Also, I think he said he was contemplating simplifying & reducing the height of the header so everything would crawl upwards a bit. OTOH, I have a wide screen and almost everything is above the fold for me. I don't know if Dallan (or anyone) has made a habit of checking some of the key pages at WeRelate in different browers, or at different resolutions, but it would be a good idea. --Mike (mksmith) 11:03, 15 September 2009 (EDT)

Number 1 isn't easy because of the way that page content is laid out. The page title appears in a box above the page text. Like Mike says, I want to redesign the header so that its height is reduced on all pages, so that should help. I'll start working on page redesign as soon as the source renaming is complete.

The current news is twice as long as most news items because I wanted to highlight the people who had been working on the duplicate project. In the future I can keep it to 5-6 lines. I'd like to keep the twitter and blog links under the news item since they're "news-y" links. But I can re-order the "Questions & Comments" and "Did you know" sections. I'll do that now.--Dallan 16:13, 15 September 2009 (EDT)

I also shortened the current news item a bit. Comments and Questions is above the fold now.--Dallan 16:19, 15 September 2009 (EDT)

Need your attention on Talk:Support [15 September 2009]

Dallan,

In general, I will try NOT to bother you with support questions, but I also don't want to provide incorrect information. Could you please either a) visit WeRelate_talk:Support#preparing_a_gedcom_-_names_.5B14_September_2009.5D and answer Amelia J's questions or b) point me to where I can learn what the GEDCOM upload process automatically changes and what it doesn't? Specifically, I'm curious about what if any automatic changes happen to uploaded GEDCOMs that:

  • have a name like this: John Smith Gilcrest (2). What will the resulting page TITLE look like? I know that the page title will drop the middle name, but will it also drop anything that appears in parentheses behind the last name? If not, won't she have a page title that looks like John Gilcrest (2)(2)? (She also has a John Smith Gilcrest without any number behind it.)
  • have a question mark (?) in any of the name fields. Will the upload process change "?" to "Unknown"?
  • have initials such as "J.N." (without a space between the initials) in the first name field? (I presume that if it was "J. N." -- i.e., a space after the first period -- that the system will compute "N." as a middle name and drop it.)
  • have a suffix in the last name field; e.g., John Smith, Jr. where "Smith, Jr." is in the last name field (instead of in the Suffix field).

Thanks, Dallan! jillaine 08:50, 14 September 2009 (EDT)

I just uploaded a test gedcom to see what would happen in these situations. I'll report what I found on WeRelate talk:Support--Dallan 16:32, 15 September 2009 (EDT)

Change in how WatchList works? [15 September 2009]

Is it possible (and desirable?) to change how WatchList works? When I visit my WatchList, I prefer to use, in most cases, the "Diff" link. Most of the time, reading that page is sufficient for me. However, the WatchList only appears to update itself if I click on the actual page. I'd prefer it if the watchlist recognized that I'd visited the page when I click on "Diff" instead of the main title. I realize this might not be desirable to other people, though (although I can't think of why). jillaine 09:53, 14 September 2009 (EDT)

Using the diff link does clear the watchlist. I do this all the time by clicking the links in the email. However, some active pages may have changed again while you were reading so it appears that it didn't clear. This is especially true of the Source renaming talk when there are several topics under discussion. --Judy (jlanoux) 12:04, 15 September 2009 (EDT)

Delete power? [15 September 2009]

Dallan, I'm about to delete a bunch of pages as part of the duplicate project, unless someone objects. I've already added quite a bit to the Speedy Delete list when I took on the Inventoire sources, so perhaps you should temporarily give me delete power? I'm fine with continuing to use speedy delete if you prefer, but I don't want to hold up the dupe project. Seems like we're *almost* done... --Brenda (kennebec1) 17:47, 15 September 2009 (EDT)

I'm glad that we're almost done! :-) I've given you temporary admin privileges during this project so you can go ahead and delete the pages. Thank you!!--Dallan 18:59, 15 September 2009 (EDT)

How I added places [19 September 2009]

Hi,

Saint Pierre is the "Parish" or Church probably...a lot of the French Canadian place names you pick up in "sources" are written that way...it can be changed to just be Sorel without the parish. Please feel free to correct or change any that "appear" to be incorrect.

When I uploaded the ged.com last night I went to review each of the little tabs to see corrections, problems, people, places etc. Under the places one it said if it was "blank" on the right side a place wasn't found, so I clicked on that place & it took me into a page to be able to select one that you had or it said I could add it so I just selected that option. I hope this makes sense & I didn't make a mess for you.--Suepcard 21:12, 19 September 2009 (EDT)

Thank-you for letting me know! Now I know what to fix. By the way, one or two of the places you added were already in the wiki under slightly-different names. The system might not have shown them in the list of possibly-matching places (I didn't check), but if you find an existing place that matches the place you're trying to add, be sure to select that place instead of creating a new one. Thanks again!--Dallan 21:25, 19 September 2009 (EDT)

Place Redirects [21 September 2009]

Dallan, I see that you (or some automatic system under your name) is redirecting a number of Places. Specifically, I'm looking at Germany, and more specifically these changes:

  1. (diff) (hist) . . m! Place:Talheim, Tuttlingen‎; 13:11 . . Dallan (Talk | contribs | block) (Redirecting to Place:Talheim (OA. Tuttlingen), Württemberg, Baden-Württemberg)
  2. (diff) (hist) . . Place:Altensteig, Schwarzwaldkreis, Wuerttemberg‎; 13:10 . . Dallan (Talk | contribs | block) (Redirecting to Place:Altensteig, Württemberg, Baden-Württemberg)
  3. (diff) (hist) . . m! Place:Feilbingert, Rheinpfalz, Germany‎; 13:07 . . Dallan (Talk | contribs | block) (Redirecting to Place:Feilbingert, Bavaria)
  4. (diff) (hist) . . m! Place:Unterreichenbach, Schwarzwaldkreis, Wuerttemberg, Germany‎; 13:00 . . Dallan (Talk | contribs | block) (Redirecting to Place:Unterreichenbach, Baden-Württemberg)

I thought we'd agreed that the format would be:

Town, Amt (where we know it), Stadt, Germany

Translating to (for the above)

  1. Talheim, Tuttlingen, Baden-Württemberg, Germany
  2. Altensteig, Baden-Württemberg, Germany
  3. Feilbingert, Rheinpfalz, Germany
  4. Unterreichenbach, Schwarzwald, Baden-Württemberg, Germany

So I'm confused by your recent redirects. Also in the case of Altensteig, your redirect resulted in a double redirect. I haven't checked the others yet.

Please advise. Thanks. jillaine 11:00, 21 September 2009 (EDT)

All of these pages were redirect pages. What I was doing in that particular batch of edits was simply correcting the form of the redirect. A few redirects had a space before the Place: or a bar (|) in the redirect. Although this is ok from the wiki's point of view (the wiki ignores the text after the bar), it caused problems for my program that checks place pages for consistency. I updated my program to handle bar's and extra spaces, but then I thought I ought to remove the spaces and bars in case other programs that I had written also didn't expect them. So all I did in the edits you mentioned is remove the text after the bar. I didn't try to fix double-redirects. We have tons of double-redirects in the place wiki. I looked at the four places you mentioned and saw that they all eventually redirect to a place with the correct title. After sources get renamed I'm going to delete unused place and source redirect pages and fix double-redirects automatically.--Dallan 12:20, 21 September 2009 (EDT)

I just uploaded a small GEDCOM and went through the review process. I was very surprised to find that New Orleans, LA is considered an ALT name for Orleans, LA. Is that right? Sounds backwards to me. --Janiejac 13:16, 21 September 2009 (EDT)

The issue is that Place:Orleans, Louisiana, United States is the parish and Place:New Orleans, Orleans, Louisiana, United States is the city. I removed "New Orleans" as an alternate name for the parish, since that's causing the confusion I think.--Dallan 13:33, 21 September 2009 (EDT)

Adam and Eve persist [20 October 2009]

Dallan,

While continuing on the second phase of merging dupes, I came across Adam and Eve again. So I went looking for the resolution that we came up with (it's in the 2009 water cooler archives) and found this from you:

Having said this, I just looked at User:Socrtwo. I don't think we need trees from someone who's primary objective is to upload trees of Biblical and mythical genealogies. We'll email this user about deleting their trees.--Dallan 14:20, 9 February 2009 (EST)

Socrtwo's mythology project is online. Can you please delete it? Thanks. jillaine 13:08, 22 September 2009 (EDT)

Socrtwo's tree was removed. The two pages that I found now: Family:Adam God and Eve God (1) and Family:Adam God and Eve God (1) appear to have originated in a tree uploaded by User:Catcall. I'm reluctant to delete Catcall's tree because it has a lot of non-biblical people it and Catcall was active earlier this year when we asked people to help out with the merging. Also, multiple people are watching some of these pages so they wouldn't be deleted even if we did delete Catcall's tree.
The only thing I can think of is to delete the person and family pages involved one at a time by hand. Deleting them is in accordance with our policy of no ancient genealogies, but I don't think it's critical that it be done right away. Would you or someone else mind deleting them sometime?--Dallan 17:02, 27 September 2009 (EDT)

I received a notice today of an edit of a Charlemagne page by user Luc1961. Looking at the history, I also see a multitude of edits by user OrphanAccount which have garbled things immensely. For example OrphanAccount has merged the families of Charlemagne and Himiltrude with Charlemagne and Hildegard with Charlemagne and Amaltrud and consequently Pepin the Hunchback with his half brother Pepin of Italy. Looking at both users contributions, I find biblical and mythical lines and medieval royal and noble lines, unsourced and apparently no contemporary people. Neither user has created a user page, OrphanAccount's edits appear to number in the hundreds dating back to mid January. Both appear to contaminate what should be protected pages, nullifying much of the good work performed by JRM. --Scot 15:37, 18 October 2009 (EDT)

*sigh* This is why genealogical novices who only collect other people's name-lists and do no research themselves -- not even any reading, usually -- ought not to be "allowed" to meddle in anything medieval. I have a reasonably good historical/academic background in this area, but even I, when I lurk on alt.genealogy.medieval, have generally learned to keep my mouth shut! Practically speaking, I don't think Charlemagne's ancestry is going to be extended, verified, or enhanced on WeRelate. But, on principle, I'm cautious about the notion of protecting "ordinary" pages. I think probably that only "official" pages -- which speak for the whole site, so to say -- ought to be protected. --Mike 16:05, 18 October 2009 (EDT)

I just checked and Luc1961 uploaded a gedcom this morning. They've worked about half of their Family Matches. Much of the file seems to have pre-700 warnings. What do we want to do? --Judy (jlanoux) 16:13, 18 October 2009 (EDT)---- We have discussed in the past protecting pages for people living prior to a certain date, say 1500 requiring administrative revue for any edits of them. Many of the medieval people we are discussing have accepted data, but separating it from all the drivel requires a lot of knowledge. Mike, as far as Soc.genealogy.medieval is concerned, it used to be a valuable forum, but after years of abuse by a certain user it has degenerated to a senseless flame war and most serious genealogists have left. There are no proven descents from antiquity, including from the Merovinian dynasty. As far as I know, the earliest person with known descendants today is St. Arnulf, Bishop of Metz, the ggg-grandfather of Chalemagne, b 582 in Heristal, Belgium, parents unknown.--Scot 22:28, 18 October 2009 (EDT)

Not to quibble (realizing the current WeRelate pages involve largely European-based genealogy) but I believe that there are authenticated Chinese genealogies that predate the Christian era. Confucius (551-479 BC) is reported to have two million plus descendants. Interestingly, a Chinese genetic center has launched a DNA project to confirm one's descent(though this approach was rejected by the committee that compiles the Confucius genealogy records).


OrphanAccount was a very active user that was removed. I agree that Person:Charlemagne (1) and his family needs work. I looked at it awhile ago and decided I wasn't the person to do it. His page (and his family) is semi-protected because of the number of watchers and the wikipedia template and therefore can't be edited during GEDCOM upload. But I have the same reluctance as Mike to completely protect all medieval pages from editing.

As far as I can tell, the changes that Luc1961 has been making to these semi-protected pages during the gedcom upload involve either adding children, which we haven't disallowed, or updating links to the place pages to point to the final redirect target, which happens automatically when a page is saved. So I don't think any harm is being done except for possibly adding fictitious children (see below).

Going forward, should we say that everyone born before 1500 is semi-protected? It seems like we should.

Regarding the Luc1961 gedcom, wow. If they hadn't already done so many family matches, I'd say we should toss the gedcom and ask them to re-upload one without the pre-700 people. The problem is that although we issue a warning in the gedcom review program that WeRelate doesn't accept pre-700 people, I just checked and it looks like the system creates the pages anyway if the gedcom is uploaded. Judy, could you leave a message for User talk:Luc1961 and let them know that we won't be creating pages for their pre-700 people? Also, if you would let me know when their GEDCOM is ready to import, I will have one of my children go through and manually exclude all of the pre-700 people in it before importing it. I'll add to my todo list automatically excluding all pre-700 people in gedcom uploads. Until then, please let me know about gedcoms with pre-700 people and I'll have my children exclude them manually.

And yes, I'm aware of the Confucius genealogy. If we ever have someone upload a gedcom with Chinese ancestry, we'll make an exception :-).--Dallan 11:25, 20 October 2009 (EDT)


I think the cut-off of 1500 for semi-protecting pages is a great one. I was going to suggest it. Anyone actually doing research is going to want to hand-craft it anyway. What's coming through gedcom is mostly junk picked up from one of the 'world trees' or other gedcoms. I did not realize we weren't excluding the pre-700 people. <oops> I do see those frequently. Maybe we could treat them like we do the living - exclude and don't let them uncheck it. I will let you know in the future when I see them. Your kids have a high tolerance for grunt work. --Judy (jlanoux) 13:04, 20 October 2009 (EDT)

Right. That's what I think too - exclude them and don't let the user include them. I thought we were excluding pre-700 people, but looking at the code it turns out that we're not. Thanks for letting me know about them until I get that working.--Dallan 14:08, 20 October 2009 (EDT)

Search Engine Question [22 September 2009]

Hello Dallan,

Claudia--LutheranChickadee asked me (DFree) this question. I am not sure what the exact answer is so I am forwarding this to you.

Debbie, I have a question. How can other people find our information if they are not a "member" of Werelate? Are the names of our ancestors hooked up to search engines such as google? Or do I just have to wait for someone with the same ancestors to find Werelate to hook up with them.

Thanks for your help! Claudia--LutheranChickadee 17:01, 22 September 2009 (EDT)

Debbie Freeman --DFree 17:19, 22 September 2009 (EDT)

Debbie, I just answered her question on her page. I didn't realize you had moved here. --Judy (jlanoux) 19:00, 22 September 2009 (EDT)

BB software [27 September 2009]

I'm presently trying to redesign our state society's web page (which is truly dreadful at the moment), and in my research I came across a very interesting open source bulletin board package called phpBB. I seem to recall you had mentioned something about a possible redesign/conversion of WR's talk pages to a more forum-like format, and this looks it might be worth looking at. (Although, since this is your profession, you probably are already familiar with it.) And the price is right! --Mike (mksmith) 21:47, 26 September 2009 (EDT)

I agree it's pretty good. I've set it up for a currently inactive forum here. It's pretty easy to set up, and it has good functionality. The things that I don't like about it are: (1) you can't use wiki tags in forum postings, so linking to wiki pages would be a bit more difficult, (2) since the forum and the wiki are two different pieces of software, I would have to write some code to automatically log you onto the forum if you were logged onto the wiki and vice-versa, and (3) it is a bit simpler to use, but since it's a different system once you've learned how to use it you still have to learn how to use the wiki, and (4) searching the wiki would be a different function than searching the forums - you wouldn't be able to search both at once. I originally thought that I would use phpbb for WeRelate support, but now I'm thinking that it would be better for me to improve the functionality of our current talk pages - making them easier to navigate, add new posts to, and reply to existing posts, which would probably take a similar amount of time and benefit all of the talk pages.--Dallan 16:41, 27 September 2009 (EDT)

Gedcom export of source citations not pretty and data entry design modifications [15 October 2009]

Hi Dallan; I created a test tree and copied one of my families and descendants to the tree and exported the tree. I am sending you copies of two reports via email created by Jillaine via a Mac using the Reunion genealogy program and a source report created by me using Windows and Legacy 7. The placement of the license in every citation really is very annoying. Could we have one general citation per person or something? Also the HTML formatting is not removed. There are probably other problems also. We also need to consider these results when we decide on the modification of the source citation fields in the new data entry design. As it stands now if a person chooses to export an entire gedcom there will be alot of clean up necessary.

Also wondering if WeRelate could somehow create a Family Group Sheet and Register report or something similar to a register report. The U.S. generally uses the NGSQ format or the Register format. There are differences which need to be considered. But perhaps there is an international version not sure. --Beth 17:54, 27 September 2009 (EDT)


Hi Beth,

Thank-you for the feedback on the reports. I agree that they don't look good. I removed the license links from the sources except for the main WeRelate source. That should help. I can't remove the links to WeRelate because I believe they're required to give proper attribution to the authors of the Source pages.

The HTML codes you mention are really wiki tags that it looks like you added to your pages. If you don't like them, you could remove them. If I were to remove them automatically during export, it would cause problems once we started to support re-importing an exported file, so I'm reluctant to do that.

In looking at the reports that you emailed I didn't notice any other problems. Can you if you notice any other problems?

I'm not sure what you don't like about the source citation fields. Can you give me some examples?

I don't plan to get into the business of generating custom reports at WeRelate. It would be a nearly endless task to come up with all the different variations of reports and charts that people would want to generate. That's one of the reasons that I added GEDCOM export - so you could use your desktop genealogy program to generate reports.--Dallan 00:07, 29 September 2009 (EDT)

Sorry that there is no plan for custom reports. I was not recommending an endless number of reports. Rather something like this that includes a group sheet, register report, ancestor chart and abnentafel that could be printed [11]--Beth 19:04, 29 September 2009 (EDT)
But we do have a plan for custom reports. Export a GEDCOM file. :-) Seriously, the desktop genealogy programs out there (several of them free) do a much better job providing reports than I could do in the forsee-able future. But if anyone watching this page is willing to write some code to generate genealogy reports in PDF format from a GEDCOM file, I will happily install it.--Dallan 16:46, 2 October 2009 (EDT)
Thanks Dallan, busy tomorrow but will do another test and check out the changes soon.--Beth 01:07, 29 September 2009 (EDT)
Dallan, my second gedcom export test never appeared. Sorry but I am not brilliant and don't know any code.<g> You can just give me some reports for my 65th birthday in Aug 2015. I wasn't referring to reports from the gedcom; but some type of template so one could create these reports on their pages.--Beth 01:37, 3 October 2009 (EDT)
I'm not sure why it didn't appear. How about if you try it again and see if it shows up.--Dallan 21:32, 6 October 2009 (EDT)
Gedcom export finally showed up. Thanks for the improvements. Exactly how do I remove the wiki tags? I did not add them; they are in the source title. --Beth 22:26, 14 October 2009 (EDT)
Sorry, I'm not following you. Do you mean the URL links from the sources back to the WeRelate pages? They are required for the "BY" (attribution) part of the CC-BY-SA license.--Dallan 18:49, 15 October 2009 (EDT)

Dallan, Just a report on my recent export of just 18 pages which I imported to my PAF program. It generally looks good. Checking my sources I see two odd things. One is that a new 'source' is created that says just 'WeRelate' and the date exported. WeRelate is where it was stored when I downloaded it but cannot be considered a real source any more than me saying my source was the encyclopedia. Is this what you meant when you said the license required some reference back to WeRelate? It is not all that bothersome, just unnecessary from my point of view.

Right, there's a new source that just says 'WeRelate'. Every person and family in your GEDCOM should cite this source. The citation should include the date the export was created as well as links to the WeRelate page and history in order to comply with the "BY" part of the license. The links should be in the "text" portion of the citation. In looking at the gedcom you exported I see that the links are in the gedcom so they should have been imported into your desktop program. Source citations in GEDOM's can have both "notes" and "text", and different desktop programs display notes vs. text differently. PAF for example doesn't display the text unless you click on the "Actual text" button. Hopefully you'll see the links somewhere.

The next odd thing I noticed was that it listed in the source list, a source I did not use for this tree. It is a MySource that I created - but it doesn't go with this tree. The only way I can figure that it got linked to this tree was by that default check mark when editing a page. I may have missed removing it or something. I have no other idea - the MySource:Janiejac/Will of Robert R. Duncan, 7 June 1788, of St. Mark's Parish, Culpeper, Virginia shouldn't link to Tree:Benjamin Jax by Gay Nix that I input manually. If that default check mark is always 'on', maybe we should rethink that.

I agree it's a problem -- random pages get added to trees because your first tree is checked by default when you edit pages. But if no tree is checked by default then pages don't get added to trees when they probably should be. I'm not sure how best to solve it. I could only export Sources/MySources in your tree that were actually linked to from people/families in the tree. That would be a pretty easy change.

When I used the feature on my PAF to see what individuals used that source, the program froze up and I had difficulty even using 'end task'. I went to WeRelate to see what links to this source, and the persons linking to it are not in this 'Benjamin Jax by Gay Nix' tree.

On the whole tho, everything else looked good.--Janiejac 19:25, 15 October 2009 (EDT)

Great!--Dallan 20:17, 15 October 2009 (EDT)

Need larger Gedcom File [28 September 2009]

Got an error message uploading gedcom file to the internet

Import Error

This file is larger than the maximum allowable size for a GEDCOM.

Bruce O'Brien--Snake-2000 07:35, 28 September 2009 (EDT)

One of two things is happening: either (a) you're trying to upload a file that's not a gedcom, or (b) you're uploading a file with more than 5,000 people in it. Make sure that the file you're uploading ends with ".ged", and that it doesn't contain more than 5,000 people. Let me know if you have any questions.--Dallan 19:05, 28 September 2009 (EDT)

Source title Help pages [28 September 2009]

Dallan, just so you'll know, I've begun rewriting the Help:Source page titles page and (mostly) updating it to the new style, as well as simply recasting some of the sentences. (I had referred someone there and then realized it was still giving now-incorrect instructions.) I haven't finished yet, but I'll come back to it in a little while. I also have to add a section on draft cards, and perhaps a couple of others. Please check for accuracy when you have the chance. --Mike (mksmith) 10:43, 28 September 2009 (EDT)

Thank you! I'll take a look.--Dallan 13:03, 28 September 2009 (EDT)
Ah! You beat me to it. I've finished now (I think I have, anyway), except for a couple of entirely new sections. Generally, in addition to rewriting instructions to follow the new style, I've also edited the sentences to make them, frankly, a bit less wishy-washy, and perhaps more reader-friendly. I've replaced "suggestions" with "standard," and so on. (Since we know that many people won't really follow the instructions anyway, it doesn't hurt anything to encourage them a little more assertively, right up-front.) I've also added invisible comments to a couple of areas that I suspect can be entirely deleted -- if not now, then pretty soon, as the project wraps up. --Mike (mksmith) 14:04, 28 September 2009 (EDT)

Assisting in administering KEMP surname and Australian place names [7 October 2009]

Hi All,

After using WeRelate for some time and managing other Wiki's, I would like to offer my assistance in administering the KEMP related pages and Australian place names.

Could someone advise how I go about doing this and complying with existing standards? Thanks.

--nastrond 23:45, 28 September 2009 (EDT)

Wonderful! With regard to the Kemp surname, I suggest that you get involved with the folks building the one-name studies portal. I see that Kemp is already on the list, due to an edit you made back in January of this year :-).
With regard to Australia, first take a look at Portal:Place, especially the Netherlands and Quebec projects in the "Helping out" section. You could add "Australia" to that list. Then go to Place:Australia and look especially at the links under "All places in Australia". These lists are refreshed every morning based upon the changes that were made the previous day. Currently towns in Australia appear directly under the state that they're located in. There isn't a "county-level" subdivision between town and state like there is for places in the US. I don't know if it's really important that there be a county-level subdivision. That's probably the first question. I'm watching Place talk:Australia now, so if you have any ideas about how to improve Australian place names, that's the place to post them.
Thanks!--Dallan 00:21, 29 September 2009 (EDT)
Hi Dallan, glad to help where I can. I welcome all feedback and suggestions --nastrond 00:03, 7 October 2009 (EDT)

Need volunteers to adopt a gedcom [6 October 2009]

We have two gedcoms holding over 7,000 people that I think are worth saving. Anybody with admin rights want to volunteer? LAYMAN-MOOMAW file has 3349 people, lots of sources, but 44 Family Matches apparently discouraged the uploaded. KAPatterson file has 170 Family Matches and the uploaded is back in school and can't manage to get to it. Just go to Admin, Gedcom and click on the file name and do what you can. If we all pitch in maybe we won't have to delete these nice files. --Judy (jlanoux) 15:15, 3 October 2009 (EDT)

Hello Judy, I will adopt the LAYMAN-MOOMAW file. Debbie Freeman --DFree 18:17, 3 October 2009 (EDT)
Thank you Debbie. I will update the log to show you are working on it. This one only had 44 matches, but there were a few dup families on the warning page.--Judy (jlanoux) 08:47, 4 October 2009 (EDT)

The KAPatterson file has disappeared from Gedcom Admin. Did someone delete it? There's no message on the User Talk to say it was imported.--Judy (jlanoux) 08:47, 4 October 2009 (EDT)

It looks like the user removed it.--Dallan 21:34, 6 October 2009 (EDT)
Interesting, she said not. C'est la vie. Did tell her I was recruiting volunteers to help.--Judy (jlanoux) 21:51, 6 October 2009 (EDT)
Judy, I forgot to ask how much time I have to edit the Layman-Moomaw file? I can't seem to locate this information, Do you know? Thanks Debbie Freeman --DFree 14:53, 5 October 2009 (EDT)
Take all the time you need. Just do the Family Matches. Look at the admin menu, Gedcom review. You have to be an admin to open someone else's file. --Judy (jlanoux) 15:14, 5 October 2009 (EDT)
Judy Thank you for the information. I am cleaning up as I go. I will keep you in the loop. Debbie Freeman --DFree 16:15, 5 October 2009 (EDT)
You might want to let the user know that you're matching the families for them so that they don't delete it while you're in the middle of matching.--Dallan 21:34, 6 October 2009 (EDT)

Drop Down Menus [7 October 2009]

Hi Dallan, Using IE 8 and the drop down menus are not working. Can't access my dashboard or anything else. Tried the compatibility view and that did not help.--Beth 17:13, 6 October 2009 (EDT)

Nevermind; husband who hates laptops is using mine more than I am and he introduced some weird popup blocker that blocked the menu dropdowns.--Beth 20:41, 6 October 2009 (EDT)
tsk tsk tsk... you'd never do that to HIS computer! ;-) jillaine 08:32, 7 October 2009 (EDT)

Odd behavior . . . [15 October 2009]

'Morning, Dallan---

I seem to be getting some strange behavior when I do "Find duplicates." Example: Going to Person:Henry Stout (7), I clicked on the "Find duplicates" link on the drop-down and got this apparent error message, in red: Cannot parse 'PersonSurname:Stout PersonGivenname:"Henry B." Person. So I clicked on the Search button again, and got this:

[COMPARE] checked pages with Person:Henry Stout (7)
[COMPARE] checked pages with Person:Henry Stout (7)

Two identical lines, that is. And note that the Person page is the same as the one I was searching from. Out of curiosity, I clicked on the "Compare" button, and got the usual screen to compare pages -- but with only a single column, for Person:Henry Stout (7). I guess it wanted me to compare Henry to himself. Twice. I've been going through people from my recently uploaded GEDCOM last night and this morning, looking for possible dups, and this keeps happening. (And I was checking for dups in the first place because I've found several that apparently weren't caught in the GEDCOM upload and review process. . . .) --Mike (mksmith) 10:26, 8 October 2009 (EDT)

Thanks for reporting this. It's working now.--Dallan 18:49, 15 October 2009 (EDT)

is there a way to find out what my user name is? [13 October 2009]

Since I"m receiving notification that my pages have changed, I assume I must already have a username, but I've tried everything I can think of and none of them will fly.--Thetherapydog 20:26, 12 October 2009 (EDT)


I'm gathering you created a new user id to post this message?
The email you got about changes should start with a salutation to your username. The links in the email would take you to the page, or just the differences between the page before and how it looks now (diffs).
Once you login as your username, you can user the Menu MyRelate->Watchlist to see what changed. It may be that the welcome message was posted to your user talk page.
The more specifics you give, the easier it is to provide pertinent help. Don't be afraid to post the email if this still isn't clear. --Jrich 22:53, 12 October 2009 (EDT)
In case Jrich's ideas didn't work for some reason, your other user name is User:Jeri.--Dallan 18:49, 15 October 2009 (EDT)

Duplicate review project [15 October 2009]

Hi Dallan,

Trying to figure out how to edit some of the pages to have them removed from the duplicate list. Here is an example: Family: Anselan McCausland and Anselan Mc Causland (1) Anselan McCausland and Anselan Mc Causland (2) .--Beth 20:34, 12 October 2009 (EDT)

Well, first off, change the names of the "wives". But won't adding the "do not merge" instruction keep them off the dupes list? Seems like it *should*... ;-) jillaine 23:52, 12 October 2009 (EDT)
I agree; either approach should work.--Dallan 18:49, 15 October 2009 (EDT)

GEDCOM export & your vision for WeRelate [15 October 2009]

Dallan, is there a place for discussing GEDCOM export (other than the watercooler)? I can't help but think that it might be appropriate to review the status, as well as get a handle on user expectations. (See my recent email exchange with Beth that you were cc'ed on.) And I guess I'd also like to know how you and Solveig think of WeRelate. Is it a wikipedia for genealogy or is it a genealogy program? I think you have a divided audience on this, and I think the GEDCOM export is calling into question just what WeRelate is or is hoping to be. There appear to be a (growing?) number of people who are trying to use WeRelate as a genie program, and they're getting frustrated. But if you really are trying to create a hybrid that is both wiki and a genie program, then I think it may deserve some really focused thinking and community discussion about how to do it. (I'm sure you are doing such thinking, but it might help if you shared your thinking more with the community.) Thanks for considering my request. Jillaine 08:30, 15 October 2009 (EDT)

It's a great question. I'll create WeRelate talk:WeRelate vision and answer it there.--Dallan 18:49, 15 October 2009 (EDT)

a way to a girl's heart... [15 October 2009]

... gel refrigerant, a re-usable transport for perishables, and Sees chocolates (the latter of which one cannot obtain on the east coast!!!) Hugs! Jillaine 11:48, 15 October 2009 (EDT)

Update: approaching 10pm eastern and um, I've got a NASTY choco-sugar rush happening which means the crash is going to be bad.... but it sure was nice getting here. Jillaine 21:55, 15 October 2009 (EDT)

repository window on source pages [20 October 2009]

I've been adding a few sources and notice that the drop down menu for repositories shows the names available and is handy for knowing how to spell it right but you cannot get it into the window by clicking on it.

Also when adding respositories, I often know the name of the repository, but don't remember the details of the URL or location. Couldn't the URL/location come up automatically when the name of the repository is typed in? Sure a bother to have to leave the page to go look it up. I hate to leave it blank. --Janiejac 14:29, 17 October 2009 (EDT)

Janie, are you on a Mac? Your description of trying to click (unsuccessfully) on something in a pull-down menu matches my experience. (I use a macbook and Firefox). I've found a solution that works, instead of clicking on something from the drop-down menu, use your arrow key to highlight it, then simply click the tab key on your keyboard. That works. It took me awhile to figure that out.
Also a trick for moving between windows, instead of leaving the page you're on, is to open a new tab in your browser window. In Firefox that's command-t (for tab)-- assuming you're on a mac. Jillaine 19:59, 17 October 2009 (EDT)
No, I don't use a Mac. I use both Mozilla and IE to get access to more tabs. If it had been just one instance, I won't have mentioned it, but I was adding the same respositories to several sources, so the lack of clicking on the name and also the lack of automatic url/location info was very noticable and frustrating. I've got to go back to all of them to add the url/locations. That info is on the repository page but doesn't show up automatically when I type the repository name on the source page. So it has to be typed in over and over. --Janiejac 21:13, 17 October 2009 (EDT)
I lost most of the links to sources when someone else & I were editing changes at the same time and I couldn't figure out how to merge our changes as suggested on the edit conflicts pages. So I suggested at the watercooler that more specific instructions were needed for that edit conflict page! I've got to redo the links. mutter, mutter .... --Janiejac 21:39, 17 October 2009 (EDT)

The drop-down list has some known bugs, sorry. It's on the todo list.--Dallan 11:28, 20 October 2009 (EDT)


Pre-Merge Discussion [20 October 2009]

I recently uploaded a GEDCOM. You have provided a list of people who presumably share an interest in one or more persons whose file I uploaded.

I have read the instructions on merging two files and I believe I undersatand them. My understanding is that the result will be a single merged file (which I believe is a good thing. My thought is to leave a message with the other person(s) prior to merging to let them know my intention and determine if they feel comfortable with the merge. While I underatand that there is a process to undo meeges and resolve conflict, I believe this is a courteous and helpful approach. How do you feel about adding this step into the merge process?--Ckamp3 12:47, 18 October 2009 (EDT)

You are welcome to try this. It has been my (and others) experience that you won't get a response most of the time unfortunately. One thing you could do is click on the user in question to go to their user page, then click on "Contributions" in the "More" menu to see if they've been active recently.--Dallan 11:30, 20 October 2009 (EDT)

Heads-up on sources NOT renamed [27 October 2009]

Dallan---

I've been scanning through a considerable number of Source pages (in the process of reviewing a newly uploaded GEDCOM), which is the first time I've looked at them since the auto-rename. Out of 100+ Source page titles (so far), I've found maybe 20% with problems. These are mostly (1) multiples of the same actual book, but with different parenthetical FHL catalog numbers following the title, (2) dups of the same title -- often three instances: one as a "book" (with full bibliographical info), one from Ancestry, and one from the FHL, and (3) dups where the author part of the book title differs -- usually with one including a middle name and the other not. Some of these (like #3) are probably points that the robot's programming missed. Others (like #1 & #2) should have been caught in all that manual un-duping we did. I've also found several dups where one version had an author but the other was still title-only in the old style. And I've found several where the stated author was dropped in favor of a geographical-style artificial title -- but the result was so lengthy, the real title was truncated. I've been fixing and merging and renaming as I go along, but I really wasn't expecting to have to do so much of it at this late date. I don't know how many of these could be fixed with programming and how many need personal attention, but I thought I'd better mention it. --Mike 19:33, 18 October 2009 (EDT)

(1) was purposely allowed because there were thousands of (I think around 5000) FHLC entries with the same title in addition to the ones that we looked at (which also matched an Ancestry title). Of the FHLC-only duplicates that I looked at, many were the same book, but not all. Sometimes it was a different record collection, even though the first author and title were listed the same. The duplicate list was so intimidating already, it didn't seem worth it (to me) for us to manually de-dupe several thousand more FHLC items, so I left the FHLC ids as part of the titles in those cases. I'm not sure about (2). I could get you a list of the duplicate titles for case (1) if you wanted to review them, or I could go ahead and merge them automatically, figuring that we'd probably merge a few that shouldn't be merged, but the majority would be merged correctly.--Dallan 11:43, 20 October 2009 (EDT)
Dallan, I'd be happy to do some more duplicate reviews. maybe not 5,000 more, but I'd be happy to plug away at a list, if you want to provide one before automating more mergers for these titles. I too noticed the many many FHLC titles with the same name but different contents. Once you delve into them, sometimes there is an obvious solution that provides disambiguation. Sometimes not so much... --Brenda (kennebec1) 18:26, 26 October 2009 (EDT)
Thanks! I'll construct a list in the next couple of days and post a link to it.--Dallan 19:12, 26 October 2009 (EDT)
There are a lot of duplicates that may or may not be findable programmatically, e.g.,

The first leaves off the year range, the second two differ by a comma after Massachusetts, but otherwise all three appear to be the same.

I'm not sure what to do about those. Maybe we should have a "find duplicates" link in the "More" menu on source pages to make it easier for people to find and merge duplicate sources.--Dallan 19:27, 27 October 2009 (EDT)
On a different, but related, topic, I do all my entry by hand, and I rely heavily on the drop-down auto-complete list for sources (or whatever it is called). For example I type "Cutter, William" in the title field, wait a second or two, and then choose from the list of all his books that I am presented with. This way, I get the spelling, punctuation, etc., exactly right.
The old source names seem to still be populating the auto-complete list. For example, if I type, "Concord, Mass", I get a choice of Source:Concord, Massachusetts, births, marriages, and deaths, 1635-1850. But I believe this is the obsolete name, and the current title is Source:Concord, Middlesex, Massachusetts, United States. Births, Marriages, and Deaths, 1635-1850. If I search for a source page in the regular search function, I cannot get the old title to come up at all, so the regular search seems to be skipping the old names.
I'll start removing them later this week. There are now a million "old" titles to remove, so it will take a couple of weeks.
I remember a discussion about editors versus authors, but not the decision. Anyway, somebody had changed the author to "Under the editorial supervision of William Richard Cutter". Needless to say this made the title of the page so long the significant part of the title ("Middlesex County") that identified that book from his others was not even in the title of the page anymore, which means it is not in the auto-complete list either. I changed it back to "Cutter, William Richard". Perhaps it is because I rely on the auto-complete list, but having the title start with "Cutter" makes a lot of sense to me. If it is important to mark him as an editor, isn't it usually done by adding "(ed.)" after his name?
As far as I'm aware, the decision was to treat the "Author" part of the page title as equivalent to "creator" or "person primarily responsible" -- but to not put any sort of personal qualification in the page title itself. Put "(ed.)" or "(comp.)" or whatever in the Author field on the page itself, so it shows up when you view the page. That will cover it both ways. The "editorial supervision" example you cited, however, is absurd. It may actually have been phrased that way on the title page, but that sort of pomposity is never repeated in a bibliography (or anywhere else). --Mike 23:12, 26 October 2009 (EDT)
Also, when adding a source, if you put "Cutter, William Richard (editor)" in the author field, the system should omit the (editor) reference when it generates the source page title for you.
The Vital Records of Cambridge are under the compiler's name: Source:Baldwin, Thomas W. Vital Records of Cambridge, Massachusetts, to the Year of 1850. To find other vital records, I type the name of the town, and perhaps county, and pick from a list, e.g., I would type "Newton, Middlesex, Massachusetts" and the list would be populated with all the place-named items pertaining to Newton, usually vital records show up near the end because they are alphabetized. I'm sure there was a compiler of the Newton records, but the Newton VRs aren't listed under his name. Shouldn't the Cambridge ones be named using the place-oriented method? --Jrich 20:47, 26 October 2009 (EDT)
Yes, they should. For what it's worth, someday soon I'm going to change the drop-down list for sources so that author is displayed on the line under the source page title, just like place "type" is listed on the line underneath place page title today. Also, I'll make government/church records findable by typing either the place name or the first author. I'm hoping this will help alleviate the some of the current issues around finding government/church records in the drop-down list.--Dallan 19:25, 27 October 2009 (EDT)

Creating a "Periodical" Source Page [20 October 2009]

Okay, here's another one. I went just now to create a Source page for the "Germanna Record," an annual periodical put out by the Germanna Foundation (which is the short version of a much longer legal name I didn't want to have to put in a page title). Clicked "ADD," selected "Source," chose "Periodical" as Type. I typed in the title of the periodical under "Title" and the name of the foundation under "Publisher," and hit "Add Page." Which took me to the Search form -- where I noted the "Publisher" field was no longer available. When I clicked "Add Page" again, it created a page the title of which was just "Source: Germanna Record" -- when it should have been "Source:Germanna Record (Germanna Foundation)." I backed up and tried it again a couple of different ways, but with the same result. I finally had to accept the offered title and then rename it by hand to match the new style. --Mike 20:41, 20 October 2009 (EDT)

Thanks for letting me know. Even though the publisher field isn't available on the search form, it's supposed to remember the publisher you entered on the add-page form. I'll fix it.--Dallan 21:32, 20 October 2009 (EDT)

Souces - search and add [26 October 2009]

I recently wanted to add several sources, so I went to the blue bar, clicked search to see if it was already in the system. At that point I already had typed in the info and the search said no hits. It seems that would have been a good place to have a button for 'add' source! But no, I had to leave there and go to 'add' where I typed the info in AGAIN and clicked add. I know now I could have avoided the extra typing step by going directly to blue bar 'add' but why not make it easy to add something after searching from the blue bar? --Janiejac 01:21, 21 October 2009 (EDT)

Good idea. I'll add it to my todo list. In the meantime, if you think you'll be adding some sources, try clicking on "add" in the blue bar. You still start out with a search as before, but you get an "add" button on the search results screen.--Dallan 16:14, 26 October 2009 (EDT)

adding persons from edit person page [26 October 2009]

Concerning ease of use for a newbie: On editing a person page there is that window with 'find/add' beside it. It is frustrating to (1) type the info in that window and then it takes you to another (2) 'add' window where you type the info in AGAIN and then it takes you to another window (3) where you search and if not there, hit the add button. The info put in the window on the edit page should carry over to the second step, and even the second step seems redundant because you're going to do the seach/add in the (3rd step) following window. So it's a 3 step process to add anything and that second step seems really unnecessary.

Since we are so used to our desktop where we type the info in once and its done; a three step process can be very putting-off! It works but it is clumsy. Since it has to be typed over anyway, perhaps the edit page could have just a find/add link with no window. The final 3rd step search/add window is where we will be including more info anyway, so that little window on the edit page seems unnecessary. Seems like the clickable link going directly to the find/add page would be all that is needed. Hmm. I see the need for a window on the edit page to show the result of what has been added tho, so maybe just rearrangement or re-label to indicate what that window is for.

We need to keep the (2) add page for use when someone clicks the blue bar add, but I don't see the purpose of it when trying to add from the edit person page. Perhaps there are good programming purposes I don't know about. But anything you can do to streamline the process will help! --Janiejac 01:38, 21 October 2009 (EDT)

I'm planning to have find/add on sources take you directly to a search screen. I'll look into streamlining find/add for people as well.--Dallan 16:14, 26 October 2009 (EDT)

Erratic "fact" tags [4 December 2009]

Scot has begun to upload pieces of the so-called mature trees from the Family Search site. For many of the people that are coming in and/or being merged, we're picking up fact-like labels and content on the page narrative body. From the GEDCOM fragment that Scot showed me, I doubt that the importer could do anything about this in terms of recognizing these items and automatically putting them on the fact list, so we'll need to decide what the best practice is going to be for manually cleaning them up.

The tags uniformly appear in all UC. Names seen include: KINSHIP, OFFICE, TITLES, HONORS, PROPERTY, PROBATE, ASSIGNMENTS, INHERITANCE, MILITARY, CONDITION, RESIDENCE, DEATH, BIRTH. Some of these, of course, correspond to WR known tags (probate, military, residence, property) and fit very nicely. Others tend to offer descriptions of the information (DEATH & BIRTH) such that the information sometimes fits best as description on the death or birth fact, others as a note that is attached to the appropriate birth or death fact.

The hardest to handle are those that don't directly correspond to anything already in WR (KINSHIP, OFFICE, TITLES, HONORS, CONDITION). At present, I'm dropping them in as "other" type events, leaving the label as the first part of the fact description. This is a little unsatisfying though, being as the UC label is a screen real-estate hog, appearing under the already useless "other" label. I think that new tags for these "facts" need to be added, even if not supported by GEDCOM. Having actual tags available will simplify things as users bring data in from these large public-access trees. Also, I should think that the person/family page display code may eventually take advantage of the event types to present facts in something like an appropriate sequence.

Since I understand that WR already carries a superset of tags beyond the GEDCOM standard, I'm assuming here that you already know what to do on export with such tags. Whatever you do, I trust you can recognize such forms if/when they subsequently turn up in a GEDCOM re-import.

--Jrm03063 18:44, 31 October 2009 (EDT)

I simultaneously got notice of this page and one of Scot's edits. Based on that content, I don't think these belong in the fact fields. They're too long, and they don't seem to contain the type of information that one looks to those fields for. Even DEATH in this one is a cause of death, which could be a note to the event, but shouldn't be dumped into Alt Death.--Amelia 21:23, 31 October 2009 (EDT)
There's lots or space in the main text box. . . . --Mike 23:01, 31 October 2009 (EDT)


There are reasons not to leave this material floating around in the narrative area.
  • Unless carefully merged with existing content, it looks like crap. Considering the small number of people trying to impose sanity on this relatively huge space, that's just a hugely unfair and unreasonable expectation. We need simpler rules than that.
  • Cause of death is an existing WR fact, as are many of these items showing up in the body. If the content associated with an item exceeds a phrase or two, then it can certainly be broken into a summary assertion and an attached note (or two). To be sure, "KINSHIP", "OFFICE", "TITLES" and the rest of these can be nicely summarized for appearance on a fact list.
  • All of us who have done a lot of merging have wound up with pages where the narrative body content was a hopeless mish-mash of stuff. We should organize our materials so that those problems are less apt to occur in the future.
--Jrm03063 10:06, 1 November 2009 (EST)
Well, I've put together a couple thousand pages on WeRelate so far on people of personal interest -- not just "stranger" pages I come across, that is. When I import and merge pages, I go through and clean the suckers up. When I import a GEDCOM, in fact, I sit down with FTE and clean up every single new page. And several hundred of the pages I've done are largely or wholly hand-built. So when I have accumulated a whole series of bits of information, taken from multiple sources, I make the effort to write it up in coherent fashion in the narrative text box, with the {{cite|S-}} template to tie statements to their sources. What's so difficult about that?
Of course every page uploaded via GEDCOM has to be visited and reviewed. But we're not talking about working through a few hundred or even a few thousand pages - we're going into the large tens of thousands territory, and there are a demonstrably small number of people involved. We need to have some simple practices that let us get through this material in a reasonable amount of time and with reasonable precision. --Jrm03063 21:16, 1 November 2009 (EST)

A couple of thousand out of more than a million, too bad everybody doesn't take the care you do. The challenge here is not to produce a few pretty family lines. There are plenty of vanity sites out there already. The goal here is to create a comprehensive and credible body of knowledge. While it may not difficult to clean up a page or even a couple of thousand, we will eventually hope to end up with millions of entries requiring years of effort by everybody involved if done manually. Dallan has taken on a monumental task to create software to automate the process, we need all the help we can get.--Scot 17:18, 1 November 2009 (EST)

Frankly, I think having a whole string of disconnected little factoids dangling down the lefthand side of the page, often out of chronological order, looks much crappier. I look on the reporting of the results of family research as "telling a story," not merely stacking up a bunch of bullet points, with standardized labels that often don't quite fit. --Mike 16:29, 1 November 2009 (EST)
Hence the suggestion that some of these be done as a fact and an attached note. I realize that the existing page display doesn't do much with respect to organization and display of the facts, but I presume that an appropriate sort sequence will be provided eventually. Also, as I mentioned, facts maintain their integrity across GEDCOM export/import cycles. When that stuff is cast adrift in narrative however... --Jrm03063 21:16, 1 November 2009 (EST)

Facts are facts and are best presented simply. Frankly one of my pet peeves is wading through reports generated by genealogy programs that insert them into per-canned narrative. It may look pretty but it reads crappy and the narrative is useless. Save the text box for well thought out narrative or text from Wikipedia and such.--Scot 17:18, 1 November 2009 (EST)

Scot, it's facts floating around without context that are "useless." Without the structure of a narrative, facts are just a box of loose parts. They don't mean anything. And I've never imported a canned/generated report in my life. I've been writing and editing for part of my living for several decades. The narrative is what makes it "family history". And, frankly, my personal peeve is the constant importing of Wikipedia text into WR pages -- as if stuff from Wikipedia is somehow superior, as if the people here are incapable of writing a narrative with a family-history slant. Much of what is imported/copied from Wikipedia is pretty pointless, genealogically. The more famous the person, the more that's the case. --Mike 08:10, 12 November 2009 (EST)

Scot, since you're the one doing the uploading I'll leave it up to you. It's easy enough for me to add a few more event types to the list. But it may be more trouble than it's worth to move the information over.--Dallan 17:34, 4 December 2009 (EST)


Another FTE "to-do" [4 December 2009]

Something else for your ever-lengthening list: I was in FTE on one tree a while ago, doing some clean-up on a family's related pages, and I went to switch to a different tree. My mind was elsewhere, I guess, and instead of closing the first tree, I just opened the next one -- assuming, I suppose, that the first tree in FTE would automatically close itself. Of course, it didn't. I ended up with three panels: The old FTE, the new FTE, and the "work" screen for the page selected in the new FTE. Oops, I thought, and closed the old FTE. Which unexpectedly closed both of them. You might consider at least having a pop-up saying "Do you want to close the previous FTE? Y/N" -- or having the old FTE close automatically when you open a new one (though I would prefer being asked). --Mike 07:55, 12 November 2009 (EST)

---> Hmmm. Just went back and tried it again. This time, it closed the old tree and opened the new one -- the behavior I was expecting. So I tried it a third time. This time, it gave me the three panels again. Something strange here. . . . --Mike 08:14, 12 November 2009 (EST)


Having the FTE pop up in a frame has causes problems. When I get back to working on the FTE again I think it would be better to give it a full screen and to have the Person pages open up in pop-up windows. How does that sound?--Dallan 17:34, 4 December 2009 (EST)


watch page loads much faster [15 November 2009]

I don't know if it is something you did or not, but my watch page is loading much faster now!! Oh, so much better. Even the 'show duplicates page' is loading faster. Just wish some duplicates would show up so I could merge them! --Janiejac 15:34, 15 November 2009 (EST)


Feature Request: Invite People to Watch a Page [4 December 2009]

I'd love to "invite" people to watch a page. Kind of like how Facebook invites people to sign up for a cause. It would be an option on any given page that would send either a WeRelate email or Internet email that says something like:

Jillaine Smith thinks you'd be interested in and invites you to watch Help:Sources.

Accept | Ignore | Contact [Jillaine]

It would also have the option to add a personal message.

-- Jillaine 18:39, 19 November 2009 (EST)

Why add a feature when there is already "Email this User"? And if you want to broadcast it, put a message on Watercooler or elsewhere. It seems to be what's been done before. --Jrich 18:58, 19 November 2009 (EST)
I believe that email this user only invites one to watch a tree not a particular page. Broadcasting a page for everyone to watch on the watercooler would only address those registered users watching the watercooler. --Beth 19:11, 19 November 2009 (EST)
I'm seeking something that can be done from any page; to email a particular user, you have to go to their User page. I want to share a page from a given page. And I'd like the option to include non-registered users as well. Jillaine 19:22, 19 November 2009 (EST)

For what it's worth, there's an "Email this page" option in the "More" link of any page. You can enter either WeRelate user names or anyone's email address. It emails them a link to the page and gives you an option to send them a personal message. So it does most of what you want, but doesn't have links for them to watch the page, which I agree would be nice.

The "More" menu hasn't worked out very well. Too many things are hidden in it. I need to figure out how to re-work it.--Dallan 17:34, 4 December 2009 (EST)


Feature Request: Indented numbering uses outline formatting [4 December 2009]

I don't even know if this is possible, but currently use of indented # signs at the beginning of the line results in the following

  1. First level out
    1. Second level in
    2. Second level in
      1. Third level in
      2. Third level in
    3. Second level in
    4. Second level in
  2. First level out
  3. First level out

I'd like the use of indented pound signs to display the following:

1. First level

a. Second level in
b. Second level in
(i) Third level in
(ii) third level in
c. Second level in
d. Second level in

2. First level out
3. First level out

Thanks!

-- Jillaine 19:06, 19 November 2009 (EST)

Hi Jillaine, this may not be possible in wiki code. Here is some HTML that accomplishes what you are looking for:

  1. First level
    1. Second level in
    2. Second level in
      1. Third level in
      2. Third level in
    3. Second level in
    4. Second level in
  2. First level out
  3. First level out

--Jennifer (JBS66) 20:09, 19 November 2009 (EST)

Wow, that's cool! I didn't know you could do that :-) --Dallan 17:34, 4 December 2009 (EST)

"Infant" children [21 November 2009]

Hi Dallan, I usually enter 'infant' children in my database as 'infant'; but I searched the internet and found a different approach; but this does not work on WeRelate. I tried to create a page titled "Infant of Mr. and Mrs. Charles W." but the result was "Infant with no closed quote. The name I entered is on the page but not in the title. I like this approach because it would identify the infant and father's first name when one is searching and also like it for my database search engine. Why is not working? Do I need to use apostrophes or something else? --Beth 18:37, 20 November 2009 (EST) And why is my topic indented? Does this have something to do with the quotation marks? --Beth 18:38, 20 November 2009 (EST)

Never mind; Legacy doesn't seem to approve of any of these data entries and alleges that I entered all of these weird characters. I am changing the name to infant no name.--Beth 18:54, 20 November 2009 (EST)

I like to use "son" and "daughter" if you know the sex. Otherwise, why not "infant". I doubt anyone will confuse that for a real name, so it kind of communicates the situation?

I think the indenting is probably left over from the previous topic. Haven't looked at it to see what is going on.

--Jrich 19:14, 20 November 2009 (EST)

Well, infant is on the title page so I can change the name on the person page and remove no name. I doubt that there are many individuals named "infant" but I am sure there are some. Years ago when there were 1 or 2 television stations, no fast food restaurants, no internet, etc.; my husband enjoyed searching the phone directory for unusual first names. There were many. The one I remember is E. Pluribus Unum.--Beth 19:55, 20 November 2009 (EST)
There was a Civil War diary published a couple decades ago, the author of which was named "Decimus et Ultimus Barziza." I used to wonder if his mother chose the name as an ultimatum to the father. --Mike 10:36, 21 November 2009 (EST)

I think the page should be titled "Infant Jones". Then in the name field on the page, you can add "child of whoever" in the Title suffix field. I've been doing this with wives whose surname is unknown and it works nicely. It helps to differentiate the thousand Elizabeth Unknown pages. Search is done on the name field and not the page title.
When creating a page title, everything after the first space is assumed to be the surname.
Your topic was indented because of a typo in the last close of the outline above. I fixed it. --Judy (jlanoux) 20:59, 20 November 2009 (EST)

Long ago on WeRelate the request was to enter only the given name but if someone entered two or more names in that field then they were entered in the title field. At some point it was decided that it was okay to enter middle names etc. but then we reverted to the previous format. So I did not remember that one could no longer manipulate the system. Thanks for the reminder. How does your suffix usage appear in your reports? Are you happy with the output in reports? --Beth 21:42, 20 November 2009 (EST)
Using the suffix has worked surprisingly well. It avoids the problem of having women with unknown surnames float in a sea of identical people. Even in my own database it help to identify the person and it is better than using married names as ids.
With unnamed children who died as infants, I would consider not creating a page for them. The text box on the Family page can include mention of them. A Person page wouldn't offer much. --Judy (jlanoux) 09:07, 21 November 2009 (EST)

Letitia.ged appears to overlap a previously-imported GEDCOM [27 November 2009]

Dallan, as you know by now I overlap on purpose and then merge. Please release the gedcom. Thanks, Howard--HLJ411 13:50, 27 November 2009 (EST)


Large gedcom [3 December 2009]

My file has 5773 individuals in it and the website will only accept up to 5000 without permission. How would I go about uploading all of my individuals? Thank you. Glen Williams--Gwill65074 11:42, 2 December 2009 (EST)

Are you willing to review 5,700 Person pages at one time in the upload procedure? Plus the Family pages? And checking all the possible merges? If that's too much to handle, I suggest you break your database down into several smaller GEDCOMs -- one for each great-grandparent, for instance. (That's assuming these are your own family and not a "community" project or something.) --Mike 11:30, 3 December 2009 (EST)

I'll break it down.--Gwill65074 12:27, 3 December 2009 (EST)


Ref: Nov 30 - Data files do not appear to be gedcoms [4 December 2009]

You advise my exports on George D. Kiefer and John Hart families "do not appear to be GEDCOM". These rescued files were made years ago on Family Tree 3.4 and Windows 95 or 98. Any suggestion on problem appreciated. Burley1--Burley1 15:40, 2 December 2009 (EST)


Did you first export a GEDCOM of the file from your genealogy software? If not, please do so and upload the gedcom file not the original. The export option should be under your genealogy program's File menu.--Jillaine 16:21, 2 December 2009 (EST)


I suggest that you try to open the files with a desktop program. If you don't have one, there are several free downloads that will import a file. Something like PAF may be able to read the files. Then you can export from there to create a new gedcom for upload. --Judy (jlanoux) 18:46, 2 December 2009 (EST)


The files are both FamilyTreeMaker files. To create a GEDCOM file from FamilyTreeMaker, follow these instructions.--Dallan


Alternate names [4 December 2009]

Thank you for adding Baptism to the Alt name types. However, when you added Baptism, we lost Birth. Is there a limit to how many alternates types we can have? I have found Birth useful is several cases and miss it now that it's gone. If there is a limit, I could live without Religious. The Alt type would take care of the people I have who joined convents. Thoughts from others? --Judy (jlanoux) 09:22, 4 December 2009 (EST)


I would like a choice for alternate spelling. Because of the great variability in Colonial spelling, one person's name is often recorded with widely varying spellings. I use Alt Name, but not sure that is the intent. I am also not sure how much I should be worrying about alternate spellings.

I assume that at some point in time, a person will be searching, or uploading a GEDCOM, using any one of the spellings seen in contemporary records, plus possibly whatever modernized form may be in common use. For some colonial persons, there could be half a dozen, or more, ways of spelling a name. My concern is that somebody will be searching with one of these authentic spellings, but that it does not match what is currently on the WeRelate page. I assume if an alternate spelling for the Surname is listed on the Surname page, e.g., Jocelyn and Joslin, then that will be handled by WeRelate searching? But I am not sure about given names. Is there a list somewhere of equivalents for given names? Would Elisabeth and Elizabeth be treated as equivalent? If not, should each form be entered as an alternate name to help people who might be searching for that page (ugh)? --Jrich 10:14, 4 December 2009 (EST)

I use alternate names within TMG for spelling variations (I have a bunch of Myers/Mayer/Meier folks) and for identification in a specific document or source (a young widow, "Mary A.", with son "John W." in one census, who shows up in the next census as "Alice M." with son "Wellington J."), and just variants of an unusual name ("Permelia" is often construed very inventively by court clerks and enumerators) -- and these are all brought into WeRelate via GEDCOM as Alt Names. The fact that this may not be quite what Dallan intended it to mean doesn't change the mechanics of it. I want those variations to be searchable, but I'm not sure what I would do with them if I didn't use Alt Name. (I also wouldn't want to have to search them all out and chage them, frankly.) On your other question, "Elizabeth" and "Elisabeth" are not only equivalent, they're identical as far as I'm concerned. So are "Willie" and "William," etc. The issue is a little different with "Mary" vs. "Maria" (I have a couple of families that include two daughters, one with each name), and it's a lot different with "Elizabeth" vs. "Isabel" -- which are etymologically the same but come from different languages & cultures and are not obvious to a novice. A girl who was Isabel in the Olde Country might be Elizabeth when her family comes to Philadelphia in 1750 -- and if you're not familiar with the language, you might assume they're two different people. So I think I would tend to standardize a casual spelling like "Elisabeth" to the 'Z' version, but I wouldn't conflate the other two examples; those are true variations. --Mike 13:29, 4 December 2009 (EST)
It's not really about what to enter - I agree with your approach on that score. I worried about other people searching. I'm sure some people would enter Elisabeth, and then I am assuming they would also search for Elisabeth. Would they match Elizabeth? Would it match on an exact search? or do you have to do a non-exact search and wade through lots of results? On the inexact search would Elizabeth score higher, when searching for Elisabeth, than Joseph (assuming both have the same surname), or could the Elizabeth page end up on the third or fourth page of results? --Jrich 14:12, 4 December 2009 (EST)
The surname pages provide a place for alternate spellings that is used for Search. I prefer to use the Surname pages for alt spellings. It saves a lot of clutter on the Person page - especially when you consider how many people you would need it for. The advantage of using the Surname page is that search will find the page no matter what variant is searched and you don't have to remember to enter all of them. Also it helps you find pages that other people have entered under a variant.
Dallan had indicated that he might be able to modify search to match if two variants appear as alternates on the same page. I'm not sure if that was implemented yet. If so, all variations could be recorded on the same surname page making it much easier to keep track of. And we would need fewer surname pages while making each more useful. It would be a big improvement.
I just added "Thacher" as an related name for "Thatcher", and vice versa, but it didn't seem to make a difference on a search. Even on a search that was not marked exact match only. (Perhaps there is some caching issues. I will try again tomorrow.) In this case, there are several Peter Thachers in colonial New England, but at least one was originally created as Peter Thatcher, and wasn't being returned in searches. This obviously could lead to duplicates being created. --Jrich 14:12, 4 December 2009 (EST)
Here's the help. --Judy (jlanoux) 12:18, 4 December 2009 (EST)

There are pages for given names as well as surnames. Just check them any time you want to make sure search will turn up your page. I have tested and it will turn up a match regardless of the sequence of given names. You do have to wait for the search index to be updated when making changes. It seems to take about an hour. But I'm sure there are times when it could be longer as some days everything seems to work slower. --Judy (jlanoux) 15:31, 4 December 2009 (EST)


We can have as many different name types as we want. I removed birth name because at the time it seemed to be causing confusion. Some people were wanting to use a person's stage or married name as the primary name and to use the "birth name" alternate to record the name the person was born with. The presence of a "birth name" type seemed to be saying that the primary name shouldn't be a person's birth name. So I removed birth name in order to remove the confusion, since I believe the general practice is to use a person's birth name as their primary name.

Exactly right on how alternate names are used. And it takes 1-3 hours for a change to the givenname or surname pages to take effect in searches due to caching previous results. When you do a search for Elizabeth and don't check the "Exact" box, then people named "Elizabeth" will be ranked highest, then people named "Elisabeth, "Eliza", "Elise", etc. (the variants listed on Givenname:Elizabeth), then people with the same surname that you searched for but different given names.

The problem mentioned above is, suppose you do a search for a name that (1) does not have a surname or given name page of its own, but (2) appears as a related name on other names' surname or given name pages. Then searches for that name will search just the name that you entered and the names in the titles of the given name and surname pages it appears on. The variants of those given names or surnames will not be searched. That needs to be fixed.

At some point I plan to take the related names out of the given name and surname pages and put them on separate "related names" pages where you can see lists of the related names for all names that start with each letter of the alphabet. I think that will be easier to maintain in the long run.--Dallan 17:34, 4 December 2009 (EST)