User talk:Jrm03063

Deleting pages [9 April 2013]

I was notified on my watchlist that you deleted Family:Ambrose Meacham and Ursula Perrie (1) giving a summary "Ancient and unsupported". I am not sure what the justification for this is, as the page is deleted. I do not know what information is on the page, as it is deleted. Personally, I wouldn't call people living in the 1600s "ancient" since there were written records available for that period. Further, while the sources are not what I would call quality sources and may be wrong, there are plenty of sources available that say this page had possibility, including over 50 Ancestral Files and 2 books published in the last 20 years, so "unsupported" is not obvious, and not really much help to a reader wondering about this family.

There were at least two people that contributed to this page (don't know the details since the page was deleted, but their contributions did not say "propagated", so appear to be directly to the page and not merely to one of the children of this marriage, as some of the other contributions were). Since this page was not deleted by Speedy Delete, these people had no chance to voice their concerns, and the page has been deleted solely on your characterization of it as "ancient and unsupported". I do not feel that a good process was followed. It would seem that it would be preferable to merely add a note to the page saying there appears to be no evidence, or to explain why it appears to be an error. It would next appear to be better to use the Speedy Delete process since there were other people who might have been interested in the page to give them a chance to offer their perspective. And I am not sure deleting the page was the right answer, as it perhaps should have been redirected to Ambrose Morgan and Ursula Perrie ([1] matches the marriage date given on various AFNs)? --Jrich 10:08, 3 April 2013 (EDT)

I have deleted close to 6000 pages since Nov 2007, and nearly 1000 in this calendar year alone. I assure you that I am careful with this authority. Indeed, I believe your comment to be the first time anyone has questioned me on the subject. Let me respond to a couple of your remarks and describe my process.

Multiple watchers can mean that the pages arose from different GEDCOM uploads with the same weak content, that was then merged, or that people have automatically added watch/tree pages, and the pages in question were swept onto their watch lists in that manner. Sadly, it doesn't really mean that people are actively watching and participating.

When cleaning about the database, I look for several hallmarks that I've learned suggest content that is weak and untouched since being uploaded years ago. Among these:

* Linking to the place "Burundi" (apparently, at some point, a "Y" in a place field was auto-resolved to Burundi)

* Gray Pratt's <n>th Grand Parent

* Special characters such as "^" in names

* GEDCOM load error dumps in the text body

* Personal "MySource" sources, that correspond to nothing known, and lack any cite-specific content on a per-page basis

* ..and of course, history that consists only of GEDCOM load and the odd redirect

When I find a handful of such pages, my first approach is to just clean them cosmetically. Fix the things that are broken but fixable (like name strings) and remove links to Burundi, meaningless GEDCOM messages, random non-AFN id strings, etc. I'll try to add a source if I can easily find something, but that's not easy in the ancient spaces if you're not looking at nobility. If I find that a weak page is part of a larger contiguous space of similarly weak content, that appears to be a product of the same upload(s), and that doesn't connect in a reasonably short number of "hops" to something substantial or active, then I start thinking delete.

Once I've decided that the area is a candidate for delete, I'll try to find the limits of the space, so that I don't wind up leaving behind orphaned valueless pages. Starting from the oldest part of the content, I'll begin to systematically delete my way down/forward in time. Since I'm dealing with swaths - and not individual pages - I can say that it is beyond impractical, to use "Speedy Delete" for this sort of delete. Indeed, I think that the reason I was originally nominated as an administrator, was because I was swamping the speedy delete list with content that was so obviously in need of delete, that it was easier to give me the authority to just act. A two step process of marking and then, after time, deleting, makes sense for individual pages where there may be active users - does it make sense for large inactive swaths - and if so - who would want to be the lucky contributor who would review the space in question - and then - go back again in order to perform the delete?

As an aside, a great deal of the salvage that I've done in the Medieval Spaces, didn't come as a result of working to save content that was weak. Instead, it came as a purely incidental side effect of improving things where source content was present and could be expanded upon - and then discovering that an unsourced tree fragment represented a known and easily source-able piece of genealogy. While this may potentially be gratifying to people who's pages were thus saved, I generally doubt that they were still around on WeRelate, and I can say with certainty that such salvage didn't help with the content that was being worked. Rather, it simply created a better place to attach/redirect pages that were otherwise in need of delete.

While patches like this sometimes are truly "detached", that's generally not true. Eventually, as I work forward in time, I'll start to find pages that actually contain sources, watchers, and other signs of life. There isn't any perfect way to know how far to go, or when to stop, but hard limits are real sources and/or anything that looks like real activity. I also ask myself with each delete - does this page contain anything that wouldn't be trivial to recreate if even the weakest on-point source were consulted in re[creating] the page?

So when I'm doing a delete, I am certainly not representing that the content was non-factual. Instead, I'm saying that the content is part of a large, very weak swath, where the facts alleged are systematically devoid of support, and the cosmetics are embarrassing. I am also saying that the content present does not appear to be significantly more helpful than knowing nothing whatsoever, if one were seriously interested in working on the ancestry in question.

I am sure, that over my history of cleaning and deleting, there are pages where I didn't choose the right comment to indicate why the delete should occur, or perhaps that there are pages that could have been salvaged if I had put in somewhat more diligence. Still, I am very sure that on balance, I've hit this about right. Others may disagree, but so far at least, they havn't.

--jrm03063 14:26, 3 April 2013 (EDT)

What I get out of this is a big "trust me" without any attempt to provide an evidence-based basis for your deletion. --Jrich 12:34, 9 April 2013 (EDT)

GEDCOM Export Ready [5 April 2013]

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

GEDCOM Export Ready [5 April 2013]

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

History of Parliament [5 April 2013]

SOunds interesting! Would like to know more and how I could help? AndrewRT 15:21, 5 April 2013 (EDT)

GEDCOM Export Ready [6 April 2013]

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

GEDCOM Export Ready [6 April 2013]

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

WeRelate growth [14 April 2013]

Your following comment:

"I think the real risks for the site lie in doing what its been doing - and hovering around 2.4M person pages with very little upward movement"

reminded me to upload a graph I've been working on for a while on this topic - I've started a discussion at WeRelate_talk:Watercooler#WeRelate_growth_.5B6_April_2013.5D. Looks interesting! AndrewRT 17:51, 6 April 2013 (EDT)

That's a little more optimistic than my gut feel for the matter. My sense has been that the genealogical domain wasn't expanding with any rapidity, but that the quality of content to be expected for any given page seemed to me like it has been on the way up. But I suppose that my perspective is too narrow to be useful for drawing conclusions. --jrm03063 21:56, 6 April 2013 (EDT)

It's a little encouraging although in my mind still much too slow. Bearing in mind ancestry.com is in the hundreds of millions, it puts WeRelate well above other wikis but well behind the commercial alternatives. AndrewRT 06:51, 7 April 2013 (EDT)

I agree. While I don't begrudge the larger commercial sites their size and the opportunities that a lot of money offer - I'm constantly frustrated that the WeRelate community doesn't embrace our own unique opportunities, for example:

When I first joined, new users would often had a perception that WeRelate was sort of the "wikipedia" of genealogy. Instead of embracing that, and saying to folks - sure, we use media wiki software, we have a compatible licensing arrangement, and we're non-profit. They would instead CORRECT people and focus instead on the differences. While I wouldn't ever want to appropriate anything from wikipedia unfairly, why attack a positive association that lets people relate to and understand what you're doing? Of course it even goes beyond this - and gets us to a recent re-vote on whether WP extracts should even be allowed on pages that didn't have any other content. The idea was voted down - but the fact that it is even still entertained as a measurable minority point of view strikes me as appalling.

What we should really be doing is making common cause with the Wikipedia biography project. Instead of me having to keep my eyes open and pick up biographies wherever I can, in ones and twos by hand (14 new ones this week alone! it's only taken me 4 years to get us up to 20,000 of the 600K-1M biography pages that exist on WP - at this rate, I'll get them all in 106-196 years!). That is crazy. We should automatically have new biography pages added over here. Not only that - but when a page is voted off the WP island because it's only reason for existence is genealogical - we should just hoover up the entire body, history and all - and the page delete log on WP should indicate - "transferred to WeRelate".

The History of Parliament effort is a little like this (though I don't think pages will be voted out of the project!), since I wouldn't think that a commercial entity would have quite as much "license" (at least in an informal sense) to make use of the content without abridging copyright. Hopefully, it will be better received by the community because it doesn't have the genealogical negative connotations of WP, but I view the effort as not fundamentally different from what would be done w/WP.

Oh well, I guess that's as much belly-aching as I want to do today.... :) --jrm03063 12:52, 7 April 2013 (EDT)

Amen! I have likewise been surprised at how anti-Wikipedia many people here have been. I think that a Wikipedia to WeRelate project would be awesome - I would be interested in helping out on the software side. -- Jdfoote1 12:43, 9 April 2013 (EDT)

Great to hear at least one other vote for sanity! Once, I would have thought that software to do some of these things would be the hard part. Now that I've seen what a wiki-culture is like, I see that software would be the easy part.

There are a few things I can imagine doing to start - figuring out how to identify biography pages that are in line to be deleted by Wikipedia would be one. I know there are sites that simply contain deleted WP content, so they must do it somehow. Maybe they just look at pages that exist in adjacent dumps of the database, and look for any that are deleted.

Wikipedia has some templates for creating family tree displays of related ancestor pages. To my mind, every such instance is potentially a little GEDCOM. Since we already have a lot of wikipedia genealogical content, we'ld need to figure out how to avoid re-reloading pages. Or maybe - we could just look at each such fragment on wikipedia - and see if we have corresponding pages for all the WP biographies noted in the tree display template.

It would really be nice if WeRelate had something other than the current implicit correspondence between WeRelate pages and wikipedia pages - at least for Person pages. Maybe the anti-wikipedians would go along with this - because it would mean that they would be free to leave wikipedia more or less entirely off of a WR biography page.

Anyway, an effort of this sort will have my support, but I'm not sure how many more times I want to have the "but every genealogist knows - WP is crap!" discussion. Or the more recent and subtle variant - where you get stony silence while doing four years of the most odious grunt work - interrupted every so often by a complaint that you're spamming people's inboxes with change mail - followed by a vigorous and personal attack when you dare to try making changes to further improve the content that was formerly unusable and embarrassing crap! (and by the way, since you've changed from sending e-mail with each change, to not sending e-mail, your recent changes will be labelled as an effort to work BEHIND the community's back). --jrm03063 14:54, 9 April 2013 (EDT)

I can definitely be guilty of trying to solve everything with technology, but I wonder if there might be a way to make everyone happy. Here are a few (crazy?) ideas:

1. Import people from WP, but auto-add a template that explains that it's from WP, and asks people to add sources, check for duplicates, etc. 2. Create a page that's just a list of WP people, that can be used just to keep track of which ones have already been added to WR (e.g., they would be deleted by people when they have been added). 3. Create a true sync, where data moves back and forth between the two systems.

-- Jdfoote1 15:05, 14 April 2013 (EDT)

Not crazy - just be ready for slings and arrows that are not rational - and don't be surprised if some separate idea or practice of yours - is attacked with a vigor that is inexplicably disproportionate.

So, having fully warned you...

* "1. Import people from WP..." - I formerly avoided this by the certain (I thought) defense of adding pages that connected with existing trees. However, you might do a small preemptive load of the 200 articles indicated as "Core", or even, the 1000 (or so) indicated as "FA" (featured article?) quality. One approach might be something like what I'm thinking of doing with History of Parliament. Create a degenerate GEDCOM consisting only of the people you're trying to add, and no genealogical connections whatsoever. Create the GEDCOM in such a manner, that it contains a proper cite of the corresponding WP page, and a category that labels the page as corresponding to a WP Biography, and use the name of that WP biography as the category sort key. I would also suggest creating a legitimate meat puppet (is there such a thing?) that exists for the purpose of doing the uploads and "owning" any tree(s) thus created. After the needs for being able to log in to that account pass, presumably it can be blocked from further logins, until or unless it is needed again.

* 1A - pages that are biographical, but are facing deletion on grounds of not meeting notability standards, seem like a place where a semi-heroic effort is called for. Still, I'm a little perplexed on how this could be detected with certainty. It seems like the threat to delete is relatively common - but I havn't yet figured out when a WP biography is actually at the moment of being deleted. Perhaps this is somehow detectable by looking at pages in consecutive drops of the WP database (and noticing that the page in one no longer exists in the next).

* "2. Create a page that's just a list of WP people,..." - probably the least controversial idea. Perhaps the most valuable thing that could be done here, is to try to cross reference some subset of WP biographies with the existing WP-referenced set on WeRelate. While 21,000 is small, compared against the 600K/1M claimed biographies on WP, it can be enough to be a bit of a problem when trying to pre-emptively add.

* "3. Create a true sync..." - now, you are dreaming. Still, I can't resist thinking about it myself. Obviously a long term goal, and one where it's hard to be sure what it might look like, but laying groundwork isn't out of the question. For example (as implied for "1."), a category that explicitly marks a "Person" page as corresponding to a particular WP Biography page. --jrm03063 18:08, 14 April 2013 (EDT)

GEDCOM Export Ready [10 April 2013]

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

[9 May 2013]

Hey, Jrm03063, I'm inputting a bunch of Welsh pages to connect a couple of lines and I can't remember if the talk page you created on the subject suggested that the "ap" and "verch" go in the fore or sur name fields. Consensus?--New.incarnation 20:02, 8 May 2013 (EDT)

The consensus was that I was an arrogant idiot for daring to make a proposal on the matter - and even worse being counter-wiki for actually attempting to try out my ideas on a thousand or so pages I'ld been editing (my promises to restore things to whatever standard the community adopts not withstanding). I refer you to the various folks who attacked that proposal yet refused to offer anything within that discussion thread (though, in fairness, one member of the oversight committee did offer to appropriate the matter in favor of her own, unstated preferences).

For informed discussion, I suggest you seek out Werebear. For villagers with pitchforks, you can see this.

Good Luck! --jrm03063 16:58, 9 May 2013 (EDT)