I'm not quite sure what is meant by the remarks that GEDCOM structures - imposed on a Wiki - lead to something clunky. Are we talking about UI design, code implementation, maybe both?
If you want your genealogy system to be useful in the genealogy community - it really has to be able to load and export GEDCOM - without loss of meaning or structure. It could be MORE capable and descriptive than GEDCOM - but it has to at least be able to start with GEDCOM and export a result that is no less meaningful.
Presumably, you want a UI that helps to make sure user edits enhance and retain the GEDCOM structure, so the UI would have to look quite a bit like what you get from WeRelate.
I can well imagine that a Semantic MediaWiki might be a superior place to implement something like WeRelate - were the effort being started today (I don't have enough knowledge to say with certainty). That would be particularly appropriate if the Semantic MediaWiki provided for property references that are reciprocal/doubly-linked - so that a reference of type T, on object A, to object B - automatically results in a reference type T', on object B, to object A. I've done a few searches to see if that concept exists for the Semantic MediaWiki - but haven't found anything. If that's not present in the Semantic MediaWiki - it seems like a concept of general utility - appropriate for a lower level implementation.
So maybe a semantic MediaWiki basis would allow for WeRelate to be implemented without reliance on hacking the base code? Thereby less clunky?
???
-jrm
On 2017-09-08 08:09, imackbaxter@gmail.com wrote:
Tell us more about CRUD data interfaces. Mack
Sent from my HTC ----- Reply message ----- From: wikimedia-genealogy-request@lists.wikimedia.org To: wikimedia-genealogy@lists.wikimedia.org Subject: Wikimedia-genealogy Digest, Vol 5, Issue 5 Date: Fri, Sep 8, 2017 8:00 AM
Send Wikimedia-genealogy mailing list submissions to wikimedia-genealogy@lists.wikimedia.org
To subscribe or unsubscribe via the World Wide Web, visit https://lists.wikimedia.org/mailman/listinfo/wikimedia-genealogy or, via email, send a message with subject or body 'help' to wikimedia-genealogy-request@lists.wikimedia.org
You can reach the person managing the list at wikimedia-genealogy-owner@lists.wikimedia.org
When replying, please edit your Subject line so it is more specific than "Re: Contents of Wikimedia-genealogy digest..."
Today's Topics:
- Re: Is the delivery of software fundamental to this project?
(Amgine) 2. Re: Is the delivery of software fundamental to this project? (Sam Wilson) 3. tiny URL (Sam Wilson)
Message: 1 Date: Thu, 7 Sep 2017 17:30:43 -0700 From: Amgine amgine@wikimedians.ca To: wikimedia-genealogy@lists.wikimedia.org Subject: Re: [Wikimedia-genealogy] Is the delivery of software fundamental to this project? Message-ID: 8219628e-99cb-f046-5250-c4289f381c91@wikimedians.ca Content-Type: text/plain; charset=utf-8
Thanks for the response Sam! Again I am replying to a digest, and apologize...
Date: Thu, 07 Sep 2017 15:37:20 +0800 From: Sam Wilson
I certainly agree with you about the NIH syndrome within the Wikimedia world. (I think it's getting better though, and I think a lot of it is part of the general PHP/web-dev community too, and not specific to MediaWiki.) I really don't think we need yet another software solution for genealogy! However... :-) I think I basically take as my starting point "base MediaWiki". As in: there's a great flexibility in a website that is basically just freeform text boxes into which you can put whatever. At its heart, a wiki is free and open and really easy to just jump into and start putting content up. That's why we love 'em! And I think it's a good platform for genealogy: we can write whatever we need to, and collaborate with others, and it's not constrained by any software-imposed structure.
Genealogy data is not free-form. It is extremely structured, rather like Wiktionary's data. To represent that data in wiki syntax will require extensive templating and modules, resulting in the kind of professional-only-contributors you find at Wiktionary - and the unpleasant work-arounds required as pages bump up against the limitations of Mediawiki (e.g. [[wikt:en:water]].) I've been working with that project for a dozen+ years, and it is now so beginner-hostile I do not feel qualified to make more than the most-minor edits, and most of those assisted by js gadgets.
I would hate to see a genealogy project go down that path.
Since most genealogy practice works with CRUD data interfaces, and this has been extremely successful in helping people of all ages and technical experience begin their personal genealogies on their desktops, I think we should focus on that for the actual genealogy work. We can leverage Wikisource and commons for documents/sources; for example transcription of government census, voter rolls. Other elements might also better be 'outsourced', like geolocation names in temporal context.
But just in case you are missing the free-form text box, WebTrees allows hand-editing of GED textual representation, either of a whole entry or any single object. Each edit pop-up and record includes a link to edit as Raw GED. (I do not believe WebTrees actually store entries in GED format, so your edits will round-trip into database representations much as Parsoid does for MW syntax.)
Amgine
Message: 2 Date: Fri, 08 Sep 2017 08:46:43 +0800 From: Sam Wilson sam@samwilson.id.au To: wikimedia-genealogy@lists.wikimedia.org Subject: Re: [Wikimedia-genealogy] Is the delivery of software fundamental to this project? Message-ID: 1504831603.958760.1099016000.0B14F984@webmail.messagingengine.com Content-Type: text/plain; charset="utf-8"
Hm, yes, I really do see the multiple sides to this! :-) It's very interesting. Thank you for going into it all.
I'm not sure I agree that genealogical research is *uniquely* structured. It's no more sturctured than, say, writing histories of companies, or political parties, or railways... I mean that there are always requirements for strucutred data in any research, but that we don't bother with bespoke tools for most of them. I think primarily because the ultimate desired output is readable, linear prose, with images, figures etc. -- I think this is my usual goal with genealogy too. Perhaps that's where I'm understanding things wrong.
Wikipedia might be a pain to edit (although, I think it's getting easier) but it *is* easy to read. I think it's worth keeping the audiences in mind when talking about different approaches to a genealogy project.
We could look at setting up a demo Webtrees site too, if we want. :-)
The other thing, of WeRelate's approach of forcing Gedcom structures into MediaWiki, I still feel is a bit clunky... I'm very open to being convinced though! I have the beginnings of some code here that was about syncing trees off werelate into a modern WeRelate extension; it could be resurrected.
--sam
On Fri, 8 Sep 2017, at 08:30 AM, Amgine wrote:
Thanks for the response Sam! Again I am replying to a digest, and apologize...
Date: Thu, 07 Sep 2017 15:37:20 +0800 From: Sam Wilson
I certainly agree with you about the NIH syndrome within the Wikimedia world. (I think it's getting better though, and I think a lot of it is part of the general PHP/web-dev community too, and not specific to MediaWiki.) I really don't think we need yet another software solution for genealogy! However... :-) I think I basically take as my starting point "base MediaWiki". As in: there's a great flexibility in a website that is basically just freeform text boxes into which you can put whatever. At its heart, a wiki is free and open and really easy to just jump into and start putting content up. That's why we love 'em! And I think it's a good platform for genealogy: we can write whatever we need to, and collaborate with others, and it's not constrained by any software-imposed structure.
Genealogy data is not free-form. It is extremely structured, rather like Wiktionary's data. To represent that data in wiki syntax will require extensive templating and modules, resulting in the kind of professional-only-contributors you find at Wiktionary - and the unpleasant work-arounds required as pages bump up against the limitations of Mediawiki (e.g. [[wikt:en:water]].) I've been working with that project for a dozen+ years, and it is now so beginner-hostile I do not feel qualified to make more than the most-minor edits, and most of those assisted by js gadgets.
I would hate to see a genealogy project go down that path.
Since most genealogy practice works with CRUD data interfaces, and this has been extremely successful in helping people of all ages and technical experience begin their personal genealogies on their desktops, I think we should focus on that for the actual genealogy work. We can leverage Wikisource and commons for documents/sources; for example transcription of government census, voter rolls. Other elements might also better be 'outsourced', like geolocation names in temporal context.
But just in case you are missing the free-form text box, WebTrees allows hand-editing of GED textual representation, either of a whole entry or any single object. Each edit pop-up and record includes a link to edit as Raw GED. (I do not believe WebTrees actually store entries in GED format, so your edits will round-trip into database representations much as Parsoid does for MW syntax.)
Amgine
Wikimedia-genealogy mailing list Wikimedia-genealogy@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikimedia-genealogy
Message: 3 Date: Fri, 08 Sep 2017 15:23:19 +0800 From: Sam Wilson sam@samwilson.id.au To: Wikimedia Genealogy wikimedia-genealogy@lists.wikimedia.org Subject: [Wikimedia-genealogy] tiny URL Message-ID: 1504855399.1836036.1099264352.238D7AF9@webmail.messagingengine.com Content-Type: text/plain; charset="utf-8"
I made a tiny URL for the Meta-Wiki page: http://tinyurl.com/wikigenealogy -- just recording the fact here in case anyone else needs such a thing. (I know tiny URLs are bad, but I wanted a thing to put on a piece of paper! :-) )
One day we'll be able to use https://w.wiki but I dunno when...
Subject: Digest Footer
Wikimedia-genealogy mailing list Wikimedia-genealogy@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikimedia-genealogy
End of Wikimedia-genealogy Digest, Vol 5, Issue 5
Wikimedia-genealogy mailing list Wikimedia-genealogy@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikimedia-genealogy