User talk:Dallan

Old topics archived: 2007, 2008, 2009, 2010, 2011, 2012

Topics


Migrating to Wikimedia [13 January 2013]

Hi, I noticed you discussed extensively about migrating WeRelate to Wikimedia on this page, and you offered to bring it up at the watercooler. Has there been any news since then? -- Ypnypn 12:43, 31 December 2012 (EST)

I'm going to presume to comment, since Dallan can be a little hard to reach and I've got an opinion anyway! I just skimmed that page - and it looked to me like getting the WR code base to the point of being a reasonable open-source project is a practical pre-requisite. To my knowledge, Dallan is still working on that. Further, I'm not aware of a wider discussion of doing something like this yet - though such a discussion might be confined to the oversight committee. As far as I can see - the primary benefits would be prestige and/or credibility. I wouldn't expect any immediate up-side from the perspective of most users. Is there a particular benefit that you see from such an affiliation? --jrm03063 14:32, 31 December 2012 (EST)
I've been on a full-time consulting contract and don't have a lot of free time right now, but the plan is to bring this up on the watercooler to get reactions to the idea. Yes, it requires me to migrate to the latest version of MediaWiki, and I don't see any particular change that it would make from the perspective of most users other than being able to remove the google ads. I'm going to start taking Friday's off this year, so hopefully I'll be able to make more progress in migrating the codebase.--Dallan 07:59, 13 January 2013 (EST)
Have you noticed that, in spite of the huge increase in productivity brought about by contemporary tools (relative to what I presume you and certainly I started with), there are still not enough hours in the day to accomplish what needs to be done? I sometimes feel like there's a perverse and sadistic variant of Moore's law at work here. Research appropriate for an Ig Nobel I should think!  :) --jrm03063 11:10, 13 January 2013 (EST)

Propagating GEDCOM data? [23 January 2013]

Yesterday a page I happened to be on the watchlist for was changed by a GEDCOM upload. The GEDCOM added a marriage to a person who had previously only a birth. When I looked at the Family page for the marriage, the person's information was different than what I saw on his Person page. Among other differences: the family page showed the surname capitalized, the double/dating of the birth date and location of birth were removed, there was a death date added. I.e., the information shown on the Family page looked like what I imagine was in the GEDCOM, even though the actual Person page had no changes other than the addition of the marriage. In the History for the Family page, look at the older versions to see this, while the history of the Person page shows no version that had these changes. I assume the Family page is different because the GEDCOM version of the Person page was propagated to the Family page? Even though the GEDCOM person page wasn't or hasn't yet been stored. Finally I edited the Person page to make some other changes and everything is again consistent.

Now the GEDCOM upload process has multiple steps and may take a user several days to complete, so I am wondering if the observed differences portended pending updates already stored, or if the user will do their final merge of Person pages using a dynamically-loaded realtime version of the page? Further, what effect will my edits will have on the whole process? Will the changes I make cause an error because the page no longer matches? Will it confuse the user because the page doesn't look the way it did when he matched the Families? Are the user's changes stored pending completion of the process, when their updates are applied in bulk, wiping out the edits I did today? On top of those worries, this could be (i.e., was) confusing, to see information on a Family page that can't be found, so that one can't do edits to fix it or see the attached sources, etc.

Don't know if there is any need to mark a page as being involved in a GEDCOM upload, as a warning to other users. I just get notified of changes when the page is saved, and it is hard to tell if more is coming or not. Of course, there is the issue of how long it can take, and whether it gets abandoned. --Jrich 12:22, 23 January 2013 (EST)


Jonas Ullberg 1706-1770. [24 February 2013]

Hello Dallan, I am new on werelate so I do not understand from the set up of this system if you are interested in Jonas Ullberg or you are a watcher to all what is happening on werelate. In any case I ask you to superwise me on the use of werelate. See my article about Jonas Ullberg wchich should be of interest to his family. What will be the best way to make it possible for them to find this information and search and check for swedish history and swedish churchbooksfacts? I live in Norway and are not related to Ullberg, but I found the misspelling of the names "hellberg" og "ullberg" so interesting that I wished the right persons to be known with it. I think that only a genealogist would be able to put together historical and genealogical material so a 250 years old misspelling of a name would come to the surface. Is a point like that within your specialities? --Frank Burmann--Frank Burmann 16:10, 17 February 2013 (EST)

Hi Frank. I looked at Person:Jonas Ullberg (1) and clicked on the History link to see that the only contributor is User:Dlongmore. There are a couple of things you could do to tell people about your research.
  • Leave the information on Person:Jonas Ullberg (1), or leave a URL that points to your article if you've already posted it on a different website.
  • Go to User_talk:Dlongmore and leave a message, telling them about your article.
--Dallan 22:02, 24 February 2013 (EST)

Main Page [8 March 2013]

Hi Dallan,

The main page recognition article doesn't appear to have changed for a while. Is this something you are open to anyone doing with the direction of the nominated pages and the guidelines that are there?

Jeffrey--JeffreyRLehrer 21:30, 4 March 2013 (EST)

User:Delijim is currently volunteering to change the featured page, but I'm sure he'd love to have some help! I'll leave him a message on his talk page. Perhaps you'd like to leave him a message as well?--Dallan 11:58, 8 March 2013 (EST)

Congratulations on your consulting [20 April 2013]

I was glad to see you say that you had secured a full-time consulting position with a company which was occupying your time. Good to hear someone secure full employment in the current economic environment. --ceyockey 05:43, 16 April 2013 (EDT)

Thanks! Unfortunately we're behind a bit in development - trying to get a beta out soon - so I haven't been able to spend as much time on WeRelate as I would like lately. I'll let you know when we have something to show though; I think it will be pretty interesting.--Dallan 21:44, 20 April 2013 (EDT)

Rating the suggestions [6 July 2013]

I guess counting the number of folks who watch a suggestion is one way to rate which suggestions might need your first attention; but I'll bet you could get a pretty good picture if you'd start a "Pet Peeve" page. I'm thinking of little things that cause big frustrations. Like missing a button here or there. Nothing major, but little things can cause the frustration level to go pretty high.

While I have your attention, I will add that by removing accomplished suggestions, you are losing the psychological impact that a crossed off list of items accomplished would give. Let us see what is being accomplished so we don't give up hope. Right now my bucket of hope is pretty low. --janiejac 22:36, 20 April 2013 (EDT)

Not removing accomplished suggestions is a great idea. You've probably noticed that suggestions aren't getting implemented right now. The problem is that I'm working more than full-time for a genealogy start-up, trying to help them get launched later this year. Once I help them get launched, I'll be able to spend more time implementing suggestions on WeRelate.--Dallan 10:44, 21 April 2013 (EDT)
Hi Dallan, not sure if you've seen my question elsewhere but do you know what it would cost for the site to employ a developer to work on the suggestions list? AndrewRT 15:35, 5 July 2013 (EDT)
I believe developers cost anywhere from $10-15/hour for a college student to $100+ for an experienced professional. The website makes just enough to cover hosting costs; we would need additional money to hire a developer.
On a related note, I'm in the midst of making the WeRelate code open-source. It's not quite ready for others to use it yet (lack of documentation), but you can see the PHP and javascript code at https://github.com/DallanQ/werelate-wiki.--Dallan 23:42, 6 July 2013 (EDT)

Raw HTML appears on WeRelate pages [16 May 2013]

Dallan,

I have been using WeRelate just a short time but started noticing raw HTML code in some of the drop-down menues. I am using a MAC-book computer and have made a screen shot of a page that exhibits this. I can email you the screen shot if you wish.

Best regards,

Richard -Rcampbell14 10:53, 15 May 2013 (EDT)

I have also noticed this from time to time, and it goes away on its own (maybe after I close IE and open it again - can't really remember). I think I figured out once that it might be related to an update occurring to WeRelate since I had logged on, but I could be wrong. I use a PC and IE 9.--DataAnalyst 21:42, 15 May 2013 (EDT)
This is an intermittent issue that I believe is caused by the WeRelate system being busy. It can be ignored. --Jrich 11:17, 16 May 2013 (EDT)

WeRelate agent [5 July 2013]

Hi Dallan, do you have any information anywhere as to the tasks that the WeRelate_agent is programmed to do? I think it may be useful to add this information (or a link to it) from the page WeRelate:Bots_for_page_maintenance to avoid duplication. AndrewRT 19:31, 21 June 2013 (EDT)

Good point - I'll do that now.--Dallan 11:30, 3 July 2013 (EDT)
Many thanks AndrewRT 15:33, 5 July 2013 (EDT)

Problem with biz coaching,info/link [7 July 2013]

Hi Dallan,

Having a problem with this site: [1]

I am using Google Chrome. I cleared my browsing info forever and it still is a problem when I select various pages on Werelate. Even when I clicked on your name on the home page just now to send you this message a new window opened for this Biz site. What should I do to eliminate this problem? I also ran a full virus scan and found no problems. Clicked on this message to add that you should not open the site link because I have not and a new window opened for the same site. So now I have 3 windows open for this site. Thanks, --Beth 21:50, 4 July 2013 (EDT)

Hi Dallan, I think this problem is now resolved. When I reviewed this page this Template Template:Vrtranscriptpage came up. I have deleted the template. It was attributed to jrm but doubt that he did it. --Beth 22:00, 4 July 2013 (EDT)

That did not fix it. The template is supposed to be some kind of Vital Records header but somehow this spam is connected to it. Help. --Beth 08:54, 5 July 2013 (EDT)
Hi Beth, I don't think the problem is related to the WR template, but to your computer in general. Is this happening when you open other sites or do a Google search? There is some information about this problem here. In Chrome type the following into the address bar:
chrome://extensions/
Is there something there called Bizcoaching.info? If there is, click on the garbage can to the right of it. --Jennifer (JBS66) 09:03, 5 July 2013 (EDT)
Hi Jennifer, thanks. This Bizcoaching is unique to WeRelate on my laptop. I did not find this program in the Chrome extensions or in my installed programs. I ran the trial version of the Malware Removal program but the trial version leaves some registry errors and Bizcoaching Malware still exists when I use Google Chrome to access WeRelate. Will have to try another program.--Beth 11:02, 5 July 2013 (EDT)
I just googled "bizcoaching.info malware" and found several results with instructions for removing it. You might want to try one of those.--Dallan 23:48, 6 July 2013 (EDT)
Thanks Dallan, finally removed the malware yesterday.--Beth 08:21, 7 July 2013 (EDT)

Can't upload gedcom [28 September 2013]

Hi Dallan,

I have tried several times to upload my gedcom file, but keep getting an error message saying the file is too big or has the wrong file extension:

  "This file appears to be larger than the maximum allowable size of 5,000 people. Another reason you might be seeing this message is if you're trying to upload a file that is larger than 12 megabytes, which generally means that it is not a GEDCOM. See Help:GEDCOM for more information."

The file in question is:

  28_Sep_2013_all.ged, 2.7MB in size, less than 700 people

Help please!

Thanks, Ruth Stephens--Marytennessee 21:39, 28 September 2013 (UTC)

I moved WeRelate to a faster server last week and one of the settings was mis-configured so that all GEDCOMs over 2Mb in size got rejected. It's fixed now. Thank-you for telling me about the problem.--Dallan 20:27, 30 September 2013 (UTC)

Update of 1 October 2013 [2 October 2013]

I opened my email this morning (Tuesday) to see this latest upload of wiki sources. Usually this happens on Sundays.

The first place I saw was [[Place:Little Smeaton (near Pontefract), West Riding of Yorkshire, England|Little Smeaton (near Pontefract)]] to which I made another amendment yesterday about 6pm UK time (1pm EDT). I can't see those changes in the History and I can't be sure what they were--I may alter 50 places a day and it's too much to make paper notes beyond a "completed" tick. I thought I changed the name--replacing Pontefract with Selby, to which it is better linked.

The upload procedure is probably a two-operation thing. First finding the places to be updated and then fixing them. If there is a break in time between one operation and the other at your end, this kind of problem can occur. Is there a phrase you can put in the email to indicate when the first run happened, so that we can be aware of double-fixes like this which may not mesh with each other?

Many thanks --Goldenoldie 08:48, 1 October 2013 (UTC)

The wikipedia update process has been having some problems since we switched over to the new hardware last week. I tested it yesterday afternoon and it stopped about half-way through. I re-started it last night and it finished this morning. I'm going to keep monitoring it, but generally I would expect the time between the two phases to be fairly short -- less than an hour. You can see the changes that it's made by selecting "Recent changes" from the "Admin" menu, then clicking on "Show bots", then selecting either the "Template" or the "Place" namespace. (Templates are updated first, then places.) I know this is not what you're asking for above, but it wouldn't be that easy to add a phrase in the email. Alternatively, I could create a special wiki page that the wikipedia update process would update as it was running, that could include when it started the template update phase, when it started the places & persons update phase, and when it finished.--Dallan 13:43, 1 October 2013 (UTC)

Hi Dallan Thanks for your explanations as to how the wikipedia update works and where to find the progress. Now, where in my mental or paper or electronic filing cabinet of numerous instructions am I going to put it so that I can use it again? <smile> What you suggest in your last sentence appears to be pretty well a re-hash of what is in Admin/Show bots/Template or Space.

Although AndrewRT has been mentioning for several weeks there was new software which we might take on, your message above is the first that this lowly but keen WR member knew that it had been instigated. Perhaps the Home Page needs a box with news of WR technical updates?

Regards --Goldenoldie 08:51, 2 October 2013 (UTC)

I believe AndrewRT is talking about updating the software to the latest version of the MediaWiki software. That hasn't started yet. What I did last week was update the hardware to faster hardware. Not much visible has happened with the software recently unfortunately. I'm making the code open-source and making it easy for developers to install and run, to encourage other developers to contribute. Once that's done, which should be later this week, I'll make an announcement, ask for volunteers to help with the coding, and we'll update to the latest version of MediaWiki.--Dallan 19:46, 2 October 2013 (UTC)

Sharing an article [14 October 2013]

Hello Dallan,

I thought I would share this article I compiled

http://www.werelate.org/wiki/Combined_use_of_WeRelate_and_Wikitree

Thank you for WeRelate

Regards Jeffrey--JeffreyRLehrer 01:08, 12 October 2013 (UTC)

Nice article - thanks!--Dallan 22:31, 14 October 2013 (UTC)

Duplicate Pages Patrol [10 December 2013]

Hello, I would like to volunteer for the Duplicate Pages Patrol. Run4Fun--Run4Fun 16:56, 8 December 2013 (UTC)

Thank you! I've added you to the list. The instructions, which I'm sure you've seen, are here. There's not a lot of activity currently, but there are several hundred potential duplicates that need to be reviewed. Let me know if you run into any questions. Thanks again!--Dallan 18:33, 10 December 2013 (UTC)

Draw attention to question on Naming Conventions talk page [13 December 2013]

I'm not sure how carefully you read every change to every talk page, so I thought I would draw your attention to the fact that I have asked you to weigh in on a discussion (regarding the effort and advantage of making a system change) on the Naming conventions talk page. The question is in section: Norwegian Names on WeRelate. Thanks.--DataAnalyst 15:28, 13 December 2013 (UTC)


Drinkwater not a gedcom file [29 December 2013]

Thanks for putting up with me as I learn to use this website. I need to do a major housecleaning job on the computer. Will try this again later when I am better organized. EKBDVA--EKBDVA 19:38, 29 December 2013 (UTC)


Drinkwater sources [12 January 2014]

I submitted a GEDCOM for the Drinkwater family and when I reviewed it (after it was accepted and imported) I discovered that the sources/references are all screwed up especially Samuel's. Is this the robot at work? Or am I a bigger idiot than I realized. I put a query on "talk" about it. Please advise. EKBDVA--EKBDVA 21:55, 12 January 2014 (UTC)


Drinkwater sources [13 January 2014]

I submitted a GEDCOM for the Drinkwater family and when I reviewed it (after it was accepted and imported) I discovered that the sources/references are all screwed up especially Samuel's. Is this the robot at work? Or am I a bigger idiot than I realized. I put a query on "talk" about it. Please advise. EKBDVA--EKBDVA 21:55, 12 January 2014 (UTC)

Sometimes the sources in GEDCOM files coming from certain record managers are difficult to import well. Can you add a link to Samuel's page here? I'll take a look at it.--Dallan 15:20, 13 January 2014 (UTC)
Hi Dallan, here is Samuel's page: Person talk:Samuel Drinkwater (7) --Jennifer (JBS66) 15:28, 13 January 2014 (UTC)

Poor quality GEDCOM - how to handle [28 January 2014]

Is there a recommended way to address a recent GEDCOM upload that appears to have very poor quality data? I have only looked at small parts of this GEDCOM (from User:KellysFamilyTree), and what I have seen are:

  • a person not connected to anyone else: Person:Elizabeth Ditty (3) - I merged her record with an existing record for presumably the same person
  • a couple not connected to anyone else, which was a duplicate of a pre-existing WeRelate page: Family:Michael Feasel and Elizabeth Ettinger (2) (I merged them)
  • a person whose family tree appears to be a mess: Person:John Dixon (82) was connected to parents married a hundred years before his birth (I unlinked him); his source is an 1841 census record that shows him with a different wife and children than are shown in the GEDCOM; his son Thomas's record appears to be a compilation of two of more different Thomas Dixons, etc.

I started trying to fix John Dixon and decided that fixing his record would involve chopping it up into separate trees and then adding the correct information - basically redoing this person's family tree.

I think that the best thing to do is to remove this GEDCOM, but I don't know what the protocol is. I'd like to be tactful - it appears that this part of the tree was picked up from an Ancestry.com Public Member Tree that has the same mess. I have not reviewed other parts of this tree, so there may be parts that are perfectly good, but I don't know how much effort I want to put into validating it. Suggestions?--DataAnalyst 01:49, 24 January 2014 (UTC)

If this were an inactive user who had uploaded their tree a long time ago and hadn't been back since, I'd simply delete the tree. But this is a new user who is currently active on the site. I'm going to ask User:JBS66 what she would suggest.--Dallan 04:33, 28 January 2014 (UTC)
We tend to consider deleting older trees from inactive users (especially those uploaded in 2007) because, at that time, WR did not have GEDCOM review standards. The thought is that users dumped their files and never came back to fix any of the multiple problems. With this file, the user is new, the GEDCOM went through a check where if there had been extensive problems it would have been rejected, and the user is uploading items like photos and newspaper clippings. I would not suggest deleting this file. Rather than taking on the task of cleaning up their errors, you may want to leave a message on the talk pages where you see problems, alerting this user (and others looking at the page) of the discrepancies. --Jennifer (JBS66) 12:41, 28 January 2014 (UTC)

Thanks for the advice. Will do.--DataAnalyst 02:34, 29 January 2014 (UTC)


All sources not showing on person pg [1 February 2014]

I thought I'd move this from the support pg to your talk page. Perhaps this is something pertaining to my computer but I don't think so. So I want to run it by you.

Would someone have a look at Person:Joseph Jackson (98). There are a total of 5 sources there, two of them census records for 1850 and 1860. But look at the edit page and see that I've input census source for 1850, 1860, 1870 and 1880. The last two do not show up on the person page. I first thought it was because I did not use S1, S1 etc but that's not it; so what could be the reason all the sources do not appear on the person page? --janiejac 03:43, 24 January 2014 (UTC)

I see 7 sources. S1 and S2 appear to be duplicates.

Jillaine 09:28, 24 January 2014 (UTC)

   That is so strange! I see only 5 sources. And yes, S1 and S2 are the same source. I'm using Mozilla firefox. 

So I went back again today January 30th. The problem of missing sources still exists when viewing the page with Mozilla firefox; the 1870 and 1880 census sources don't show. But the page shows all sources and looks great when viewing with Internet Explorer. But when viewing with Chrome, the page is so wide the siblings and family chart don't even show without scrolling over but the 1870 and 1880 census sources do show. If this is something that doesn't show this way on your computer, it must be something pertaining to my computer. But thought you should know just in case. --janiejac 23:24, 30 January 2014 (UTC)

I use Mozilla Firefox and I have no problem. Can see two possible problems: 1) you need to scroll because they're simply off the screen. Assuming it's not that simple (probably, but had to check), then the other item that looks funny is the link to ancestry, because it is so long (though it doesn't cause me problems). I would suggest enclosing the link to ancestry inside brackets, as in [http://www.ancestry.com....]. This will mean the link doesn't take up so much screen space, just in case its excessive length is causing the browser to stop working when it hits the first long link (which happens to be in source 5, i.e., the last one you can see any part of).
I didn't want to change the page and have it behave differently and nobody would know why. So I offer here links to the census pages on archive.org (free and usable by people like me that don't have ancestry): 1850, 1860, 1870, 1880. I would recommend adding a link to the Find a Grave citation instead of giving the memorial number. If you use the standard template, it looks like this: {{fgravemem|45153349|Joseph T. Jackson}}. I would also recommend adding an estimated marriage date to Mary's marriage so it shows up at the top of the page with Louisa's marriage and so the infoboxes show in the right order (bef 1863, based on birth of eldest known child). --Jrich 00:27, 31 January 2014 (UTC)
I think it has to do with the length of the links in the notes. Try putting the links inside brackets, like this.--Dallan 16:46, 1 February 2014 (UTC)
Always a learning experience! I had not noticed earlier that yes, those 'missing' source references were clear at the bottom of the page because the length of the URL forced them below the family table! So I've learned how to shorten that long URL, used the shorter version, and now it shows just fine.
And yes, it would be much more convenient to those who don't have access to ancestry.com to use a different source for census records but finding them elsewhere would be a major undertaking for me. The same for changing my method of references to Find-a-Grave would entail going back to each person uploaded to edit the reference. Folks can search by memorial number. Hmm, perhaps I can start putting the URL of the memorial page in the notes. That will help for future uploads. Thanks everyone for helping me work though this!

Andrew Cowan (10) [14 February 2014]

Andrew Cowan is noted in your changes as being married to Nancy Rutledge. This Andrew has birth date of his father b. 1744 Robert Cowan?? I think?? Nancy Rutledge married Robert Cowan, not Andrew. Andrew Cowan, the son, is buried in Cowan-Delaney Cemetery near or next to his parents. I have noted this incorrect birthdate before and it is still there. Robert Cowan was b. 1744 and lived in Sullivan Co. I don't know anything else about this Andrew but he didn't marry Nancy Rutledge, sister of General George Rutledge, father George Rutledge (maybe William George).

Thanks

da--DAtkinson 15:58, 3 February 2014 (UTC)

I just looked up the page for Person:Andrew Cowan (10) and it appears to have been deleted in 2009. Perhaps you are referring to a different page? I'm pretty sure I haven't edited any pages like that recently; perhaps you could make a note on the page or on the talk page about the incorrect information, or even better correct it? Thanks--Dallan 02:31, 14 February 2014 (UTC)
Sorry, Dallan, I should have posted a follow up here. The message above was meant for me and I followed up on DAtkinson's talk page. The errors have been taken care of.--DataAnalyst 12:44, 14 February 2014 (UTC)

Could we have a Place Pages--Wikipedia update please? [1 March 2014]

Now that the where do we go from here questionnaire has been analysed, would you be kind enough to do another update to the Place Pages. This is promised at once a week and it's been far from that for quite a while.

I seem to have nominated myself to update British place pages and it's a big job. I depend on Wikipedia as well as some British sources that I am familiar with. And I hate to see the "the following text is copied from an article in Wikipedia" notice without the article in question. But so many of the Wikipedia entries include distances from place A to place B and, as you know, these don't convert. Any suggestions as to how to work around this without copying and pasting the WP text and marking it up again would be appreciated. (Mind you, the copying and pasting often leads to dropping sentences or paragraphs of material that have nothing to do with genealogy--such as what kind of supermarket exists in the village or who their present Member of Parliament is.)

Another problem I encountered a couple of weeks ago: I was working on a little place named Melbourne in the East Riding of Yorkshire. I decided to investigate "What links here" and discovered about 50 people whose link was not to Yorkshire, but to Melbourne, Victoria, Australia. Every time I corrected one the search engine pulled up Melbourne, Yorkshire, England again. Each correction had to be done twice. The task took most of an afternoon. To the technically uneducated, there is something wrong here. I wonder if Melbourne, Derbyshire, England holds as many would-be Oz natives as Melbourne in the East Riding did?

/cheers --Goldenoldie 07:52, 25 February 2014 (UTC)

The wikipedia update needs to be started manually and now and periodically monitored, which takes extra time, so I no longer do it unless people ask. I'll start it up right now.
If the place text is simply "Melbourne" with no country, the place matcher doesn't know which place to link to and chooses the place with the shortest page title, which is the one in Australia. If the place text were "Melbourne, England" it would likely link to Melbourne in Derbyshire, since that's the page with the shortest title in England. If the place text were "Melbourne, Yorkshire, England" then it should link to the Melbourne in East Riding of Yorkshire. Choosing the page with the shortest title isn't always a great choice, I know, but when there are multiple matching places to choose from, it's difficult for the place matcher to know which one to choose.--Dallan 16:19, 25 February 2014 (UTC)

Many thanks for the update. --Goldenoldie 17:10, 25 February 2014 (UTC)


Dallan

I think something went wrong with the update. I have only checked two places so far, but on both all there is is a red-lined template reference, not the text from Wikipedia. Take a look at [[Place:Howgrave, West Riding of Yorkshire, England|Howgrave]]. {{source-wikipedia|Howgrave}} was correct, wasn't it?

I won't have time to check any more for 3-4 hours.

Regards --Goldenoldie 08:12, 26 February 2014 (UTC)

Yes, I'm re-running it now. It's started crashing periodically; I'm not sure why. I'll continue to monitor it over the next couple of days and reply here when it completes successfully.--Dallan 16:16, 26 February 2014 (UTC)

Sorry to hear about the glitch. Look forward to seeing it right.

--Goldenoldie 17:11, 26 February 2014 (UTC)

All pages should be updated now.--Dallan 15:24, 1 March 2014 (UTC)

Thank you. Until the list started filtering into my emailbox, I had not idea how long ago some of the pages had been worked on. I use {{moreinfo}} a lot more now. That should save you some trouble from here on in. Regards --Goldenoldie 17:11, 1 March 2014 (UTC)


Simplest Idea for Living People [26 March 2014]

Here's the simplest idea I can think of - it does require some effort from you - but maybe not much.

Find a web hosting service that will work with you - craft a complete install of the WR server and environment in a virtual machine - and let the hosting service rent people access to whatever processing, storage and other resources they needed on a month-to-month basis. Only change is to document how to change whether non-logged in users can display pages or not. Maybe they wouldn't need a VM, but instead, just a list of installation instructions from tar balls. Not sure what's involved.

This would also work for groups who want to set up their own server for whatever purpose - DAR chapter, historical society, etc. They might want to let their versions show pages in the clear.

As time permits, in piecemeal form, try to evolve features into the "sub domain" server, that make it simpler to make use of and exchange information with the primary WM machines.

It's almost zero code, and even if the private instances of the server don't interact with the public shared server at all - people would still be using a package that looks and works in every way like the "real thing". --jrm03063 20:49, 27 February 2014 (UTC)

Oh - and a PS on this - when this idea was tossed around on the BIG THREAD - someone suggested that the private side would be a nice way and place for people to work out the issues with their GEDCOMs before bringing them over to WR. With the right sorts of support and tools - maybe there would be no need to do an enormous transfer of the entire GEDCOM - but instead, just page by page as they're ready. Not my idea - but could be cool. If cheap enough - might even be a replacement for our current GEDCOM upload process. --jrm03063 20:57, 27 February 2014 (UTC)
It's a possibility, but it's not zero code. The challenge is that WeRelate wasn't designed from the start to work in a multi-tenant environment. We'd have to figure out how to handle gedcom uploads into the private space, how to handle searching, how to link to place pages and source pages that are in the public space, and a few other things. Every user would need to have their own database. It would work, I'm just not sure how great of a solution it would be long-term. It would be a very sandbox-like experience unless we made a few changes.--Dallan 22:04, 27 February 2014 (UTC)
I was thinking that it's one server per machine. Different instances live on their own (presumably virtual) machines. --jrm03063 22:10, 27 February 2014 (UTC)
A micro server at Amazon costs $0.02/hr or roughly $15/month. Do you think people would be willing to pay that?--Dallan 22:14, 27 February 2014 (UTC)
I would - and remember - it's not just you, but you and the other trusted members of your family group. When you travel to a cemetery or other places where you have wireless web access - you're still able to reach your site. I think there are other good reasons too - lots and lots of family photos, scanned to somewhere - then lost because no one really has a backup policy they adhere to? I agree it's not an entirely free approach - but I like the idea of keeping Allen County and FOLG at arms length from what people do with "living" data. --jrm03063 22:41, 27 February 2014 (UTC)
That's an interesting idea.--Dallan 16:52, 2 March 2014 (UTC)
I realize my $15/$20 month commitment isn't all that compelling, but I will work with you and commit to trying this for a year if we work out how to do it. I think it's a "product" that doesn't exist anywhere else - other genealogy "site software" seems just to be producing reports of a PC's database in HTML. You could argue that wikitree is trying to provide collaboration on a customized scale, but there are lots of problems with what they're doing and certainly the way they do it. I realize the target market is tiny compared to anything that would run directly on a Windows PC - but there's nothing else in that market. I did a quick google search for "cloud genealogy" - and came up with a talk that someone gave 3-4 years ago - talking about "find a grave" and other things being sure to follow. I would argue that find a grave really isn't a cloud anything - but I think this genuinely would be. Choose your group, decide on your rules, buy your storage as you find you need it, and off you go.
Were we to do this, I think we just start by trying to put together the software as it exists on a microserver. Then, try to figure out how to intelligently make use of the backing store at Allen County. Maybe "PLACE", "SOURCE" and "REPOSITORY" would just come from there.
BTW, thanks for the WP update and looking at the login problem, which obviously, seems not to be an issue today. --jrm03063 18:38, 2 March 2014 (UTC)

The challenges are I would have to make sure that the servers stayed running, and we'd still have the somewhat difficult-to-use wiki interface that many people say is challenging. An alternative I've been thinking about is an interface for online genealogy that was as easy to use as desktop genealogy - similar to how many people are moving from desktop Outlook to online gmail or yahoo mail, and then integrating this with WeRelate. I'll post some comments on WeRelate talk:Next Steps.--Dallan 16:48, 3 March 2014 (UTC)

I guess I don't see why that would be your responsibility. Is WR proper that flakey?
As to where you're going with this - I'll just have to see. I know I'm not the target market, but for my purposes, I'm sure I'm not interested in a desktop app - and definitely not something that's less capable than what I'm used to. --jrm03063 19:39, 3 March 2014 (UTC)
The three servers running WeRelate aren't flakey. But 100 servers, one for each of 100 people, would require some up-keep. Also, I'm not talking about a desktop app with the living thing. I'm talking about a cloud app that's as good as or better than a desktop app, like gmail is as good as or better than outlook.--Dallan 13:22, 5 March 2014 (UTC)
Dallan, when I think about priorities that you mentioned earlier, I think that this app is more important than a multi-lingual UI. Many European users are quite comfortable with the English language, and WikiTree is English only too. An app that allows for living people, and gradual transfer of selected information to the wiki part may have a much better effect right now. And I think that this gradual transfer is needed, because the current GEDCOM upload procedure put me off long ago, and still does that. I have many places that are not converted properly, even when I select The Netherlands as default, and that makes uploading here a no go for me. There are other aspects involved, like Fred Bergman moving to Geni, which made WeRelate less atractive to me, and for the local part of my research that is a gap that's quite impossible to cross. Another aspect that an app can hopefully improve is the social part. I know that people like to get in touch, which is exactly what remote cousins do on WikiTree. And that works through subtle things like the ability to send a private message to a profile manager, which probably looks more inviting than the wiki lingo "Talk". I think that many people that don't edit anything on wikipedia have no idea what that "Talk" means, and even those who do, may not use it, because it is not private. Many cousins that contact me on other sites like to tell who they are, when they were born, who their (live) parents are, etc., and that's something that definitely does not work here. --Enno 22:41, 12 March 2014 (UTC)
I'm starting to work on this app in parallel with making WeRelate multi-lingual. Actually, my son Taylor is preparing WeRelate to be multi-lingual (he spends a few hours each Saturday doing this - should have an initial cut ready to show in a few weeks), while I'm working on a "minimum viable product" (rough draft) of a site to handle living and support gradual syncing with WeRelate. I hope to have something very rough to show to solicit comments in a few months.--Dallan 18:20, 15 March 2014 (UTC)
Thanks. I'm replying here, because the next steps page is so large that I get lost there. Can you give a hint on what you're heading at? I can imagine that there will be pages for the same objects as we're used to, but I'm curious about possible support for rich content, like xhtml. I mention that, because I sometimes use Evernote to save pages that I find, and would really like to see a genealogical workbench that works a bit like that, or like Google docs or sites. Such technologies could be a real advantage over database driven raw text desktop programs that I use today. --Enno 11:52, 20 March 2014 (UTC)

That's interesting. I was thinking about an evidence-based model where you would create entries about the people and facts/events that you had found on a particular source, and then link the people listed in the entry to person pages. Each entry could contain unstructured text, pictures, links, and structured facts. You could create entries without linking them to anyone initially, and you (and others) could add comments to the entries. A possible metaphor is facebook, where you'd have a person profile for each person in your tree, a bunch of posts (entries), and the people listed in the entries would be "tagged" to the person profile pages.

Can you tell me more about what you're looking for?--Dallan 18:28, 20 March 2014 (UTC)

I'm quite confused by your use of the entries term. It suggests something quite specific, linked to a source, while my research process looks more chaotic, and is based on notes. Notes are all sorts of texts left behind by my late father, in Word, and emails and notes that I collect myself, partly in Evernote, others in LibreOffice Writer, or in Gramps. Some of these qualify as sources, clipped from archive sites, or trancribed from other documents, others are just thoughts and theories, or parts of other people's genealogies. They can contain everything you describe, and my first task would be archiving them in a way that works for me, like Evernote. And like Evernote, I like the app to store them without forcing me to create titles, or perform other bureaucratic tasks. Once archived, I like to add tags or categories and such, and to be able search them too, of course, again like Evernote, or OneNote, whatever you are acquainted with. Import from Evernote would be a real bonus here.
For notes that qualify as a source, I like to add attributes like author, title, date, volume/page, etc., so that, when I link those to persons and events, a proper reference (citation string) is shown there. Same for notes that describe places, buildings, their architects, whatever, quite wiki like, actually, but without having to know much wiki markup, if that's possible.
Structured facts are nice, but not a priority for me as yet. I spend a lot of time on Ancestry these days, and I see that they have a feature that allows one to create or update persons in the tree with data from the entries in an indexed source record. That is very nice, because they add the source references too, but right now, I don't see myself adding structured facts to the sources that I have. They could be derived by letting an app read the texts that I clipped from FamilySearch, or local archive sites, but that would mean interpreting html pasted from sites with lots of different lay-outs, and languages, because not every site I use has proper micro data yet.
I see these notes, which start as unlinked entries in your terms, as the 1st phase in an evidence-based workflow, and I really like to put them on a page of their own, like they appear on Evernote, not posts that are created on a person's page, or a group's, so in that way, I don't think that the Facebook model works for me. What I do like is their ease of use, but not the chaos that I feel when accessing such a page.
Notes are top level entities in Gramps, and I think they can be transferred via GEDCOM too, although the xhtml used by Evernote may be a bit harder to embed in there. That's more GedcomX like, where it is a valid document type, I think. I like to put notes first, and add attributes from the source and citation to them where available, also when importing from GEDCOM files, because that's where my information is. Source and citation attributes don't count as evidence for me. It is the actual text or structured data that I like to concentrate on, and that is in the note, or entry in your terms. --Enno 22:23, 22 March 2014 (UTC)
Maybe WordPress is a better metaphor. It allows posts and comments, and has a decent editor, and posts can be marked private where needed. It might be a bit weird though, to have MediaWiki and WordPress on the same site. --Enno 13:00, 23 March 2014 (UTC)
There are a lot of good ideas here! I'm thinking an "entry" (post) can have unstructured text, links, and structured-data where you can enter names, facts, and relationships about people (technically personas in an evidence-based model), but it wouldn't have to have structured data at all. Some entries would correspond to sources, others would correspond to notes, others would correspond to discussions. I agree that entries should also allow you to enter bibliographic data like author, title, volume/page, etc. I hadn't thought about adding tags to entries, but I like the idea.
I think an advanced genealogist like you wouldn't use entries that correspond to sources much, but a beginning genealogist probably would, and having names+facts from the entry flow onto the person profile when the entry is tagged to someone in your tree, similar to how it flows in ancestry, would be convenient for them. I don't intend to write parsers to automatically extract names+facts from HTML pages for every site out there, but I think if I wrote parsers for FamilySearch and Ancestry and maybe one or two more it wouldn't be too onerous.
Entries (posts) will be first-class citizens. You'll be able to see all of the entries you have created in one big list, and you'll be able to sort, filter, and search them. Does that address your concern about the facebook model? How do you organize your entries in Evernote? Do you put them in separate folders or just use tags?
I like the wordpress metaphor as well, but I like facebook a little better because you can have the same facebook post tagged to multiple individuals so it can show up on multiple profile pages. Facebook posts can't be assigned tags though and you can't see a single list of all posts, so maybe the right way to think is using a bit of both.
I'm not thinking of merging this idea with MediaWiki. This is for a personal trees; we'll have to figure out how to integrate personal trees into the WeRelate shared tree in the future.--Dallan 01:52, 26 March 2014 (UTC)
I'm still having problems with that "entry" term, but otherwise I think we're on the same track, and I like that. I like to create documents/notes/pages, whatever you call it in the end, and hope that this can be done from notes that I have on Evernote, and notes that I already have in Gramps as texts in sources and citations in such a way that each text is shown with the attributes of the source/citation that it belongs to, where citation refers to the SOURCE_CITATION structure as defined by GEDCOM 5.5, not the formatted text which is shown in source references.
You may not have expected this, but a large portion of the texts that I have in Gramps are indeed sources copied from local sites or FamilySearch. They are not very pretty, formatting wise, but I keep those to see what names appeared in records, hoping that one day personas will be properly implemented. And with rich text documents, maybe xhtml like in Evernote and GedcomX, I hope that some time I can make them look better, and sync them with my Gramps database.
The notes that I have in Evernote now are not well organized, meaning that they are all in one folder now, and don't have many tags either. I didn't spend time on that, because I soon figured out that without a stable site to put my tree on, it would not make much sense to create links from notes to persons and families. That's why I hope to be able to transfer them here, and the number of notes is not so large that I would really need an Evernote import feature here. That's just nice to have, but copy paste will probably work for me too, provided that your support of RTF or xhtml is OK. Most important for me is to have a workbench that concentrates around the entities that I prefer to work with, which are evidence (entries), families, and individuals. --Enno 22:14, 26 March 2014 (UTC)