Things about the issue work/edition are simpler for projects like fr.source, bn.source and other projects with a very high proofread/total ratio, since there's usually an unique Index page linked with an unique Commons djvu/pdf page and very often an unique source of that image. Things are much more confused with "naked" works. I suggest - for simplicity - that by now any effort should be focused on proofread works/editions ignoring the case of naked works.
I don't know in detail the work flow into de.source, German language being a hard obstacle - what's a pity!
Alex
2017-10-23 14:04 GMT+02:00 Thomas Pellissier Tanon < thomas@pellissier-tanon.fr>:
Hi,
I am also going to be at Wikidata Con.
In the French Wikisource we started a Wikidata project. https://fr.wikisource.org/wiki/Aide:Wikidata We plan to do a small hackathon soon to start uploading our book data to Wikidata (I already have a prototype of tools that extract data from Wikisource to upload them to Wikidata but it still needs some work to do the job well).
Still my biggest issues/hurdles for good data are • capture of information from WS to WD — it just is hard work, WEF
tool is still not sufficiently aligned
Yes, we need some specific tools.
• the ever problematic inability to link WP book to WS editionthrough Wikidata What we could do is use the new Wikisource MediaWiki extension to add a piece of code to add links to Wikipedia from Wikisource and to the other Wikisources. We could start prototyping it using lua modules.
• that cannot capture information for Wikidata at archive.org,and relate that through to the file at Commons, and then the edition at Wikisource (or pick another starting point and interrelate0
In Wikidata you can point to the commons file and to IA [1]
• the inability to create an edition from a book/work, theinability to create a work from an edition
Yes, we should create a UI on top of Wikidata to do such task.
Thomas
[1] https://www.wikidata.org/wiki/Property:P724
Le 23 oct. 2017 à 13:48, Gerard Meijssen gerard.meijssen@gmail.com a
écrit :
Hoi, A Wikipedia matra is be bold and another is that things are a work in
progress. In my opinion, what we need is the name of a book, its author and the fact that people can read it. All the other stuff like what "version" is a particular book pales in comparison. We should not let the quest for perfection be the enemy of the good.
Also Archive.org and Open Library are two different entities. Both the
Open Library and the Internet Archive have their own identifiers for authors and they are not necessarily linked. We are talking about books from the Open Library and they are available as an E-book or a PDF.
My problem is not with Open Library, my problem is that we do not know
what is available from Wikisource as a finished good ready for reading. In the end what we advertise is the author the book, versions are secondary.
Thanks, GerardM
On 23 October 2017 at 12:36, billinghurst billinghurstwiki@gmail.com
wrote:
Hi Nicolas,
Still my biggest issues/hurdles for good data are • capture of information from WS to WD — it just is hard work, WEF
tool is still not sufficiently aligned
• the ever problematic inability to link WP book to WS editionthrough Wikidata
• that cannot capture information for Wikidata at archive.org,and relate that through to the file at Commons, and then the edition at Wikisource (or pick another starting point and interrelate0
• the inability to create an edition from a book/work, theinability to create a work from an edition
Maybe you can even ask what we need to improve to get bots to run
through and autocapture, is our meta-data in headers not suitable? What is it that is problematic?
Thanks for asking.
-- billinghurst (being so remote for the action <sigh>)
------ Original Message ------ From: "Nicolas VIGNERON" vigneron.nicolas@gmail.com To: "discussion list for Wikisource, the free library" <
wikisource-l@lists.wikimedia.org>
Sent: 23/10/2017 7:30:44 PM Subject: [Wikisource-l] WikidataCon 2017
Hi all,
For information, the WikidataCon is this week-end in Berlin. While
there is no talk nominatively around Wikisource, there is some intervention on relation subjects (inventaire.io, WikiCite, German National Library, FRBR, and so on).
The event is sold out, but you can follow remotely some of the
presentation (link will be added here : https://www.wikidata.org/wiki/ Wikidata:WikidataCon_2017/Program/Remote ).
I'll be there and I'll be happy to talk about Wikisource, who else will
be there?
Cdlt, ~nicolas
Wikisource-l mailing list Wikisource-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikisource-l
Wikisource-l mailing list Wikisource-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikisource-l
Wikisource-l mailing list Wikisource-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikisource-l