I'm at the Crossref LIVE 16 event
<https://www.eventbrite.com/e/crossref-live16-registration-25928526922> in
London where I just gave a presentation
<https://dx.doi.org/10.6084/m9.figshare.4175343.v2> on WikiCite and
Wikidata targeted at scholarly publishers.
Beside Crossref and Datacite people, I talked to a bunch of folks
interested in collaborating on Wikidata integration, particularly from
PLOS, Hindawi and Springer Nature. I started an interesting discussion with
Andrew Smeall, who runs strategic projects at Hindawi, and I wanted to open
it up to everyone on the lists.
Andrew asked me if – aside from efforts like ContentMine and StrepHit –
there are any recommendations for publishers (especially OA publishers) to
mark up their contents and facilitate information extraction and entity
matching or even push triples to Wikidata to be considered for ingestion.
I don't think we have a recommended workflow for data providers for
facilitating triple suggestions to Wikidata, other than leveraging the Primary
Sources Tool <https://www.wikidata.org/wiki/Wikidata:Primary_sources_tool>.
However, aligning keywords and terms with the corresponding Wikidata items
via ID mapping sounds like a good first step. I pointed Andrew to
Mix'n'Match <https://meta.wikimedia.org/wiki/Mix%27n%27match> as a handy
way of mapping identifiers, but if you have other ideas on how to best
support 2-way integration of Wikidata with scholarly contents, please chime
in.
Dario
--
*Dario Taraborelli *Head of Research, Wikimedia Foundation
wikimediafoundation.org • nitens.org • @readermeter
<http://twitter.com/readermeter>
Héllo,
I am developping a graph database using Python. The point in that
is to have an easy to setup database that can handle bigger than
RAM dataset like wikidata. Right now there is no such database
available except in the Java world via Noe4j embedded.
AjguDB [0] is a graphdb library I've been a working on for some
time that allows to store graphical data on disk a query the data
using a query language similar to Tinkerpop's gremlin. There is
NO SPARQL query engine, yet.
I am looking for tasks to solve using wikidata that can help
demonstrate the use of the library. Otherwise said, what do
you use wikidata for using your favorite tool? I'd like to
replicate that work using AjguDB and see whether it's up to
the task.
Thanks in advance!
PS: I have written a similar library in Scheme.
[0] https://github.com/amirouche/AjguDB
--
Amirouche ~ amz3 ~ http://www.hyperdev.fr
Those interested in translations of world places, please give your feedback
on this discussion [1]. There is a dataset of professionally translated
list of world places [2] that has already been through a round of
independent review by native language speakers and is open to be used to
improve Wikidata.
331 places were translated into 8 languages. Of the 2,648 translations that
were received, 1148 are new or differ from existing label translations,
while 1508 match perfectly with existing labels.
>From the discussion so far it seems like the next steps would be to get the
translators to update the Wikidata entries from their own accounts with a
paid contribution disclaimer as per [3]. The other option is to get
community members in each language to once again do a review and update the
corresponding labels. Any suggestions of which is the recommended way to go
about this?
[1]
https://www.wikidata.org/wiki/Wikidata:Project_chat#Data_donation_of_transl…
[2]
https://docs.google.com/spreadsheets/d/1SKVi9PZ_1ebxwWJAGshLQvkrJ5esOUwPHEk…
[3] https://en.wikipedia.org/wiki/Wikipedia:Paid-contribution_disclosure
The Wikidocumentaries project is planning to create a wiki that will have
articles about microhistory - places and people. The article topics will be
created based on entries in Wikidata, and the content will be mashed up
from content in Wikimedia projects and other open repositories. The content
would be enriched with additional, more personal material that is not
suited for Wikimedia projects. Whenever suitable, the wiki would also work
the other way, providing data and content to Wikimedia projects.
The wiki itself is going to be a hub for citizen history, offering tools to
work with documents (transcription, identification, georeferencing etc) and
user generated remixes (storymaps, slideshows etc.).
I am willing to open this idea for discussion and ask your advice. I would
like to know if you have experiences with Semantic MediaWiki to evaluate if
it is suitable for being the database for indexing linked content, or how
would Wikibase play out.
The project is being prepared by a small working group and a group of
Finnish project partners. There's not much to link to yet, but I will be
happy to tell as much more as is possible.
Best,
Susanna Ånäs
Forwarding.
Pine
---------- Forwarded message ----------
From: Sage Ross <ragesoss+wikipedia(a)gmail.com>
Date: Fri, Nov 11, 2016 at 12:42 PM
Subject: Re: [Wikimedia-l] Pebble smartwatch tool for finding Wikipedia
photo opportunities
To: Wikimedia Mailing List <wikimedia-l(a)lists.wikimedia.org>
Thanks everyone for the feedback!
I just published version 1.3; Albin added Wikidata support to the
wmflabs API, and Diderot now queries Wikidata if no nearby
unillustrated articles are found on Wikipedia.
(You can also now customize the color of the time text.)
-Sage
On Mon, Nov 7, 2016 at 5:22 AM, Magnus Manske
<magnusmanske(a)googlemail.com> wrote:
> Looks like a cool thing indeed! Any chance to expand that to Wikidata?
It's
> a superset of all Wikipedias, plus things like UK National Heritage sites,
> so it will give you many more candidates in some regions.
>
> On Sat, Nov 5, 2016 at 10:30 PM Sage Ross <ragesoss+wikipedia(a)gmail.com>
> wrote:
>
>> Hi folks!
>>
>> I made something that I think is pretty cool: a watchface for Pebble
>> smartwatches that shows you the nearest unphotographed Wikipedia
>> article. I've been using it for about a month, and I'm really happy
>> with how it's turned out. I've taken a lot of photos for articles that
>> I wouldn't have otherwise.
>>
>> It's called "Diderot". If you have a Pebble, check it out:
>> https://apps.getpebble.com/en_US/application/57dc94602a6ea665510000f0
>>
>> A few neat things about it:
>>
>> * You can configure it for a number of different language versions of
>> Wikipedia
>> * It uses a wmflabs API by Albin Larrson which filters out articles
>> that have only a png or svg illustration, so you still see the
>> articles that have a map or logo but lack a real photograph.
>>
>> -Sage
>>
>> _______________________________________________
>> Wikimedia-l mailing list, guidelines at:
>> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
>> New messages to: Wikimedia-l(a)lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
> _______________________________________________
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines
> New messages to: Wikimedia-l(a)lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines
New messages to: Wikimedia-l(a)lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
Dear Wikidata Community,
We are happy to announce today *a new status indicator icon about the
relative completeness of pages*, available via our Recoin
<https://www.wikidata.org/wiki/User:Ls1g/Recoin> user script. The user
script adds a color-coded progress bar icon to the status bar, which
expresses how extensive the information on the page is in comparison with
pages about similar entities. Currently, the script works for all humans
that have an occupation, by comparing the typical properties of entities
having that occupation.
It can be enabled by adding the following line to your *common.js*:
*importScript( 'User:Ls1g/recoin-core.js' );*
Further information on how the values are computed is on the tool's page
<https://www.wikidata.org/wiki/User:Ls1g/Recoin>.
We are looking forward to your opinions and feedback, especially how useful
you find such an indicator in general, and regarding the accuracy of the
computed values.
Best wishes
Simon Razniewski
On November 15, 2016, csarasua(a)uni-koblenz.de wrote
> It would be very interesting to see the status of the interlinking to
> external sources through Recoin. I guess that the claims related to the
> interlinking (e.g. authority control properties
> <https://www.wikidata.org/wiki/Wikidata:List_of_properties/
> Generic#Authority_control>)
> will already appear in the explanations of Recoin, but it might be
> useful to show the status of the completeness of the interlinking
> separately. Maybe as a separate analogous indicator, or in the same
> indicator but with separate explanations.
>
Hi Cristina,
In fact, our focus was first on the "regular" properties, thus,
unfortunately, we are filtering out exactly the ID properties at the moment.
But it is no technical problem to set up a second indicator specifically
for authority control properties, if that is of interest we can think about
that. Would the same approach as for the facts, i.e., computing the color
using a count-based on comparison with what other, similar entities have,
make sense for claims about interlinking?
>
> Is the source code of Recoin available?
>
It should become available, in fact, it is a pretty straightforward
count-based comparison with the facts that entities of the same occupation
have, I can send you the code immediately if you want, though
albin.ahmeti(a)gmail.com could tell you more about the technical setup.
Cheers,
Simon