Wikidata // was The tragedy of Commons

List overview All Threads
Download

newer

older

VisualEditor office hours

Let Commons do what it knows best...

Fæ

18 Jun 2014 18 Jun '14

1:33 p.m.

On 18/06/2014, Gerard Meijssen <gerard.meijssen(a)gmail.com> wrote: ...

...

A single editor in Commons can tweaks license information as much as it might be in Wikidata. Remember both Commons and Wikidata are wikis. It is likely that a property will be needed to reference an OTRS instance. Reasons why a file has issues will be as much be defined by properties and qualifiers. Really Fae, not much will change except that a Russian may view all this in Russian and a Brazilian in Portuguese. Effectively the number of people who will be able to understand what is expressed will increase. More people will be able to find media files... Remember, that IS the objective isn't it?

With regard to your specific mention of OTRS tickets and identifying different licences for the same image (or derivatives) across projects, can you point to an example, or an agreed plan with a date for implementation on Wikidata? I know that Jarekt has worked hard on Commons licence identification, to my knowledge it is a far more complex issue than people might expect. At the current time, I have yet to see how the Wikidata theoretical use improves finding images on Commons today. After regularly getting emails over the past two years raising expectations of how these problems will be solved through Wikidata, I have yet to be involved or invited to join any Commons based Wikidata project, despite being currently the most active unpaid volunteer uploader for Commons. One of my particular interests was sorting out the Geograph project where I added meaningful place categories to images (around a million images). Last year I deliberately deferred continuing improvements and new uploads based on suggestions of how Wikidata was going to revolutionise the way this worked. As far as I am aware, there is still no published plan to help with identifying Commons categories with places using Wikidata; indeed for projects like our Avionics batch uploads (which rely on Airport categories and their locations) I have been told the opposite, to no longer expect, nor wait, for this to be sorted out and I am likely to revise my Geograph work without any plans to incorporate Wikidata, or work with Wikidata. Fae -- faewik(a)gmail.com https://commons.wikimedia.org/wiki/User:Fae

Show replies by date

David Cuenca

18 Jun 18 Jun

2:32 p.m.

On Wed, Jun 18, 2014 at 3:33 PM, Fæ <faewik(a)gmail.com> wrote:

...

At the current time, I have yet to see how the Wikidata theoretical use improves finding images on Commons today. After regularly getting emails over the past two years raising expectations of how these problems will be solved through Wikidata, I have yet to be involved or invited to join any Commons based Wikidata project, despite being currently the most active unpaid volunteer uploader for Commons.

There is; https://commons.wikimedia.org/wiki/Commons:Wikidata https://commons.wikimedia.org/wiki/Commons:Wikidata_for_media_info https://www.wikidata.org/wiki/Wikidata:Wikimedia_Commons I formally invite you to take a look and join the discussions if you feel like :-)

...

One of my particular interests was sorting out the Geograph project where I added meaningful place categories to images (around a million images). Last year I deliberately deferred continuing improvements and new uploads based on suggestions of how Wikidata was going to revolutionise the way this worked. As far as I am aware, there is still no published plan to help with identifying Commons categories with places using Wikidata; indeed for projects like our Avionics batch uploads (which rely on Airport categories and their locations) I have been told the opposite, to no longer expect, nor wait, for this to be sorted out and I am likely to revise my Geograph work without any plans to incorporate Wikidata, or work with Wikidata.

Magnus Manske

2:58 p.m.

The issue here seems to be "political/community" rather than technology. I could probably set up a MediaWiki installation with WikiBase plugin on Tool Labs within a day or so, then have a bot create/synchronize an item for each file on Commons. Community could start to create properties, and bots could fill in basic statements. That said, I'm not doing that. First, no time. Second, ire of the WMF for making them scramble to re-create my stuff yet again ;-) Cheers, Magnus On Wed, Jun 18, 2014 at 3:32 PM, David Cuenca <dacuetu(a)gmail.com> wrote:

...

On Wed, Jun 18, 2014 at 3:33 PM, Fæ <faewik(a)gmail.com> wrote:

You would need a lua template to do a bit of triangulation: 1. From the linked Commons category to the Wikidata item you fetch the statements 2. One of those statements should be "category's main topic (P301)" 3. From the linked item, you can extract the location data This should be possible once this bug is implemented https://bugzilla.wikimedia.org/show_bug.cgi?id=47930 I don't much about the Geograph project, but perhaps your work might be compatible with a future deployment of "wikidata for media info"? Micru _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

-- undefined

Pierre-Selim

3 p.m.

2014-06-18 16:58 GMT+02:00 Magnus Manske <magnusmanske(a)googlemail.com>om>:

...

And if we say "please" ?

...

Cheers, Magnus On Wed, Jun 18, 2014 at 3:32 PM, David Cuenca <dacuetu(a)gmail.com> wrote:

On Wed, Jun 18, 2014 at 3:33 PM, Fæ <faewik(a)gmail.com> wrote:

-- undefined _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

-- Pierre-Selim

Magnus Manske

3:16 p.m.

It would just have to be ported to Commons once there's a native WIkiBase installation. If it will be ported at all. Is it really worth the fuss, so we can tinker for a few months? (I would genuinely like to know!) On Wed, Jun 18, 2014 at 4:00 PM, Pierre-Selim <pierre-selim(a)huard.info> wrote:

...

2014-06-18 16:58 GMT+02:00 Magnus Manske <magnusmanske(a)googlemail.com>om>:

The issue here seems to be "political/community" rather than technology.

could probably set up a MediaWiki installation with WikiBase plugin on

Tool

Labs within a day or so, then have a bot create/synchronize an item for each file on Commons. Community could start to create properties, and

bots

could fill in basic statements. That said, I'm not doing that. First, no time. Second, ire of the WMF for making them scramble to re-create my stuff yet again ;-)

And if we say "please" ?

Cheers, Magnus On Wed, Jun 18, 2014 at 3:32 PM, David Cuenca <dacuetu(a)gmail.com> wrote: > On Wed, Jun 18, 2014 at 3:33 PM, Fæ <faewik(a)gmail.com> wrote: > > > > At the current time, I have yet to see how the Wikidata theoretical > > use improves finding images on Commons today. After regularly getting > > emails over the past two years raising expectations of how these > > problems will be solved through Wikidata, I have yet to be involved

> > invited to join any Commons based Wikidata project, despite being > > currently the most active unpaid volunteer uploader for Commons. > > > > There is; > https://commons.wikimedia.org/wiki/Commons:Wikidata > https://commons.wikimedia.org/wiki/Commons:Wikidata_for_media_info > https://www.wikidata.org/wiki/Wikidata:Wikimedia_Commons > > I formally invite you to take a look and join the discussions if you

feel

> like :-) > > > > > > One of my particular interests was sorting out the Geograph project > > where I added meaningful place categories to images (around a million > > images). Last year I deliberately deferred continuing improvements

and

> > new uploads based on suggestions of how Wikidata was going to > > revolutionise the way this worked. As far as I am aware, there is > > still no published plan to help with identifying Commons categories > > with places using Wikidata; indeed for projects like our Avionics > > batch uploads (which rely on Airport categories and their locations)

have been told the opposite, to no longer expect, nor wait, for this to be sorted out and I am likely to revise my Geograph work without any plans to incorporate Wikidata, or work with Wikidata.

-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

-- undefined

Fæ

3:43 p.m.

On 18 June 2014 16:16, Magnus Manske <magnusmanske(a)googlemail.com> wrote:

...

Only if the end result is usable for the *majority* of active Commonsists without having to break new ice all the time.[1] If the workflow is: 1. Become a Wikidata volunteer, have long debates with other Wikidata volunteers and fix missing "big stuff" such as mapping to OTRS tickets, licence types or working out from scratch the differences between Wikimedia Commons categories for UK counties, historic county boundaries, UK vs. EU vs. unitary authority defined county boundaries ... and that's just the "level 2" location boundaries, 2. Learn to create Lua templates from scratch, 3. Get the WMF to fix some fundamental blocking bugs, 4. While you are there, gain a consensus on Commons for translation formats across all categories, templates and image pages, 5. Probably several other stumbling blocks I have not thought of, having never done this before and having no real case studies to compare to... Then this looks like a lot of barriers to success, considering that I believe this has had significant/serious funding, and been intensively discussed at many international conferences and on email lists for a period of years now, without these barriers being removed. Every time I read these discussions (I tend to save time by doing something more productive) they seem to be talking hypothetically about future plans which are always a year or two away, before we can see anything I could realistically and systematically incorporate into a batch upload, without it being considered highly experimental. I think it's already too late to do anything meaningful to bring Wikidata to Commons in 2014, as I don't even see a community agreed plan being presented at Wikimania rather than more of the same promotional and hypothetical presentations about what Wikidata might be able to do; if I have missed a plan with committed dates, perhaps someone could provide a link in this thread. Links 1. Which I tend to define minimally as this list <https://commons.wikimedia.org/wiki/User:F%C3%A6/Userlist>, perhaps unfairly as it excludes folks like admins with fewer than 10,000 edits. ;-) Fae -- faewik(a)gmail.com https://commons.wikimedia.org/wiki/User:Fae

Lydia Pintscher

4:28 p.m.

On Wed, Jun 18, 2014 at 5:43 PM, Fæ <faewik(a)gmail.com> wrote:

...

Every time I read these discussions (I tend to save time by doing something more productive) they seem to be talking hypothetically about future plans which are always a year or two away, before we can see anything I could realistically and systematically incorporate into a batch upload, without it being considered highly experimental. I think it's already too late to do anything meaningful to bring Wikidata to Commons in 2014, as I don't even see a community agreed plan being presented at Wikimania rather than more of the same promotional and hypothetical presentations about what Wikidata might be able to do; if I have missed a plan with committed dates, perhaps someone could provide a link in this thread.

I understand you've not seen much until now. But look at it from the other side. Over the last two years we've created the base with Wikidata. Over the next year we'll be expanding it to Commons. There is still a lot of groundwork to do over the next months for us before you can see something visible. But the point is you will. Over the next week the Wikidata and the Multimedia team will publish more plans. What is published so far has already been linked in this thread by David. Cheers Lydia -- Lydia Pintscher - http://about.me/lydia.pintscher Product Manager for Wikidata Wikimedia Deutschland e.V. Tempelhofer Ufer 23-24 10963 Berlin www.wikimedia.de Wikimedia Deutschland - Gesellschaft zur Förderung Freien Wissens e. V. Eingetragen im Vereinsregister des Amtsgerichts Berlin-Charlottenburg unter der Nummer 23855 Nz. Als gemeinnützig anerkannt durch das Finanzamt für Körperschaften I Berlin, Steuernummer 27/681/51985.

Fæ

4:58 p.m.

On 18/06/2014, Lydia Pintscher <lydia.pintscher(a)wikimedia.de> wrote:

...

Gerard Meijssen

7:46 p.m.

Hoi, If you care to a long time ago in the days when Wikidata was years in the future, I have created a proof of concept based on the OmegaWiki technology. It demonstrated how multilingual support would improve the usability of Commons. Wikidata is a different beast but not so different that the lessons learnt are not relevant. In the meantime it is a really bad idea to wish Wikidata away and wish away the existing experience and vision. Working case studies have been produced and thinking has to become current. So pull up your socks and start experiencing into Wikidata. Thanks, GerardM PS you will find that I blogged about it somewhere... On 18 June 2014 18:58, Fæ <faewik(a)gmail.com> wrote:

...

On 18/06/2014, Lydia Pintscher <lydia.pintscher(a)wikimedia.de> wrote:

Cool, I look forward to hearing about it at Wikimania and reflecting on a committed plan with dates and outcomes defined from the end user perspective. Hopefully some of the outcomes can be implemented on Commons in late 2014 rather than in 2015. In the meantime, it might be an idea for folks to avoid claiming that any issue that pops up on Commons can be solved in Wikidata, unless they can produce a working case study rather than discussion about proposals. Fae -- faewik(a)gmail.com https://commons.wikimedia.org/wiki/User:Fae _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

3621

days inactive

3621

days old

wikimedia-l@lists.wikimedia.org

Manage subscription

8 comments

6 participants

tags (0)

participants (6)

David Cuenca
Fæ
Gerard Meijssen
Lydia Pintscher
Magnus Manske
Pierre-Selim