Wikidata ontology - Wikidata - lists.wikimedia.org

List overview All Threads
Download

Wikidata ontology

Fwd: [freebase-discuss] So long...

weekly summary #207

Jan Macura

1 May 2016 1 May '16

7:34 a.m.

Hi all I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated? Thanks Jan

Attachments:

attachment.htm (text/html — 528 bytes)

Reply

Show replies by date

Markus Kroetzsch

1 May 1 May

8:35 a.m.

On 01.05.2016 01:34, Jan Macura wrote:

Hi all I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated?

Hi Jan, We have revised the OWL/RDF encoding as part of the work on the SPARQL query service, and the URIs have been changed in the process. Many of the same terms can still be found under the new namespace. The new namespace reflects that the ontology is the same for any site using this software, not just Wikidata. There have also been further modifications in the RDF export, e.g., in relation to how certain values are encoded (e.g., geo coordinates use WKT in Wikidata, but used a custom type with planet in our initial RDF dumps). Another major extension was that simplified data values (encoded as single resources) are now available on every level -- statement, qualifier, reference -- and some new properties had to be introduced for this. Finally, there are also some changes in the URIs used for various RDF properties. All in all, the basic encoding (with statements, references, and complex values represented by own RDF resources) is the same, but the syntactic details changed quite a bit between our original ISWC publication and the launch of the SPARQL service. Cheers, Markus -- Markus Kroetzsch Faculty of Computer Science Technische Universität Dresden +49 351 463 38486 http://korrekt.org/

Reply

Jan Macura

6:59 p.m.

Thank you both! Now I see the point of changing it. But there is another thing that interested me: There have also been further modifications in the RDF export, e.g., in

relation to how certain values are encoded (e.g., geo coordinates use WKT in Wikidata, but used a custom type with planet in our initial RDF dumps).

If geo-coordinates use WKT in Wikidata (which I can't see anywhere there), does it mean, that the original idea of *{latitude, longitude, altitude, precision, globe}* format was abandoned? Oh, and another thing: when I download "wikidata-properties.nt" from WDTK dump files site [1], there is <http://wikidata.org/ontology#> used everywhere. So... is the WB ontology somehow translated to WD ontology? Thanks! Jan [1] http://tools.wmflabs.org/wikidata-exports/rdf/index.php?content=dump_downlo…

Reply

Stas Malyshev

2 May 2 May

3:55 a.m.

Hi!

If geo-coordinates use WKT in Wikidata (which I can't see anywhere there), does it mean, that the original idea of /{latitude, longitude, altitude, precision, globe}/ format was abandoned?

Coordinates are WKT in the RDF output of Wikidata, when represented as single literal. That's not the original representation inside Wikibase (that still has separate elements and can be also seen in the JSON dump) and not the only RDF representation - there's also "full value" representation, described here: https://www.mediawiki.org/wiki/Wikibase/Indexing/RDF_Dump_Format#Globe_coor… This one is harder to index and search for, but it allows you to do lookups that depend on specific part of the coordinate (like, which objects are on Mars, or what is located on the equator)?

Oh, and another thing: when I download "wikidata-properties.nt" from WDTK dump files site [1], there is <http://wikidata.org/ontology#> used everywhere. So... is the WB ontology somehow translated to WD ontology?

Hmm, not sure about that one, Markus should know more about it. -- Stas Malyshev smalyshev(a)wikimedia.org

Reply

Jan Macura

3 May 3 May

3:01 a.m.

Thank you very much! I will try to dig around it a little bit further: 2016-05-01 21:55 GMT+02:00 Stas Malyshev <smalyshev(a)wikimedia.org>rg>:

That's not the original representation inside Wikibase (that still has separate elements and can be also seen in the JSON dump) and not the only RDF representation - there's also "full value" representation, described here: https://www.mediawiki.org/wiki/Wikibase/Indexing/RDF_Dump_Format#Globe_coor…

Maybe you can tell me, why you have dropped the proposed *altitude* value from the datatype? How it exactly suppose to work with different coordinate systems (potentially at different celestial bodies)? And how can one specify the celestial body in WD? I can't see any interface for it there.. Thanks again Jan

Reply

Markus Kroetzsch

2 May 2 May

1:19 p.m.

On 01.05.2016 12:59, Jan Macura wrote: ...

Oh, and another thing: when I download "wikidata-properties.nt" from WDTK dump files site [1], there is <http://wikidata.org/ontology#> used everywhere. So... is the WB ontology somehow translated to WD ontology?

I have some pull requests in the pipeline for WDTK to update our exports to be aligned with the official format. This includes the change of the ontology URIs. To be honest, I had not expected this to be of much importance for many users at the time, but if there is interest in us updating our dumps to the new format, we can certainly do this in the next couple of weeks. Even with the official dumps in place now, it is probably useful to have our modular dumps of smaller sizes. Best regards, Markus -- Markus Kroetzsch Faculty of Computer Science Technische Universität Dresden +49 351 463 38486 http://korrekt.org/

Reply

Jan Macura

3 May 3 May

2:26 a.m.

Thanks for reply. 2016-05-02 7:19 GMT+02:00 Markus Kroetzsch <markus.kroetzsch(a)tu-dresden.de>de>:

To be honest, I had not expected this to be of much importance for many users at the time, but if there is interest in us updating our dumps to the new format, we can certainly do this in the next couple of weeks.

Well, the point is, I was using the thirt party SPARQL enpoints of Universidad de Chile ([1], dead now) and Openlink ([2]), where the queries with this ontology worked fine, and now when I discovered, that WMF already launched its own endpoint, the queries don't work.. Little confused, so I am asking.. The other thing is, my primary sources were available scientific articles, namely [3]. Now it seems that *a lot* has changed since. [1] http://milenio.dcc.uchile.cl/sparql [2] http://lod.openlinksw.com/ [3] http://korrekt.org/papers/Wikidata-RDF-export-2014.pdf Thanks Jan

Reply

Markus Kroetzsch

3:03 a.m.

On 02.05.2016 20:26, Jan Macura wrote:

Thanks for reply. 2016-05-02 7:19 GMT+02:00 Markus Kroetzsch <markus.kroetzsch(a)tu-dresden.de <mailto:markus.kroetzsch@tu-dresden.de>>: To be honest, I had not expected this to be of much importance for many users at the time, but if there is interest in us updating our dumps to the new format, we can certainly do this in the next couple of weeks. Well, the point is, I was using the thirt party SPARQL enpoints of Universidad de Chile ([1], dead now) and Openlink ([2]), where the queries with this ontology worked fine, and now when I discovered, that WMF already launched its own endpoint, the queries don't work.. Little confused, so I am asking.. The other thing is, my primary sources were available scientific articles, namely [3]. Now it seems that *a lot* has changed since.

Yes, we should publish an updated journal version of that paper to ensure that written records are up-to-date again (research and application each have their own pace that is hard to synchronise). It is not that much work to update queries to the new format, but I can see that it takes a while to look for IDs to replace the old ones. Maybe the easiest way is to look at the example queries provided in the query.wikidata.org UI. The use of the new ontology URIs is really almost the same as for the old ones. The Chilean endpoint was provided as an early testbed before the official service was life. We have stopped running it, for obvious reasons. A big advantage of the official Wikimedia endpoint is that it is updated with live data in near real-time (subject to availability ;-). It is also fairly reliable, since there are normally two servers answering queries. Best regards, Markus

[1] http://milenio.dcc.uchile.cl/sparql [2] http://lod.openlinksw.com/ [3] http://korrekt.org/papers/Wikidata-RDF-export-2014.pdf Thanks Jan _______________________________________________ Wikidata mailing list Wikidata(a)lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata

-- Markus Kroetzsch Faculty of Computer Science Technische Universität Dresden +49 351 463 38486 http://korrekt.org/

Reply

Jan Macura

3:43 a.m.

Thanks for your answers, Markus. One last question (I hope): What advantage brings the new reification process (using different namespaces instead of {s, v, c, ..} suffixes)? Thanks Jan

Reply

Markus Kroetzsch

3:52 a.m.

On 02.05.2016 21:43, Jan Macura wrote:

Thanks for your answers, Markus. One last question (I hope): What advantage brings the new reification process (using different namespaces instead of {s, v, c, ..} suffixes)?

Technically, these two variants are just a case of "different URIs": RDF graphs do not have a notion of "suffix" or "namespace". The reason why we put the modifying part of the URI before the Wikidata entity ID is that it is easier for people to write in the interface. The reason for the exact shape of the URIs (with several, sometimes nested "directories") has been decided mostly based on technical considerations that I don't recall in detail. Cheers, Markus -- Markus Kroetzsch Faculty of Computer Science Technische Universität Dresden +49 351 463 38486 http://korrekt.org/

Reply

Stas Malyshev

1 May 1 May

10:47 a.m.

Hi!

I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated?

The items in wikiba.se/ontology hierarchy are those that would be relevant and would have the same meaning for any Wikibase install - i.e. individual statements, ranks, properties, etc., basic data type predicates. They do not depend on particular Wikibase instance, thus they were moved to wikiba.se. If you have your own Wikibase install, your items and properties would carry your domain name, since they are not Wikidata items. But Wikibase ontology would stay the same. Thus the difference. And yes, the old form of the ontology URLs will not be used anymore on Wikidata or Wikibase code. -- Stas Malyshev smalyshev(a)wikimedia.org

Reply

Stas Malyshev

10:47 a.m.

Hi!

I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated?

The items in wikiba.se/ontology hierarchy are those that would be relevant and would have the same meaning for any Wikibase install - i.e. individual statements, ranks, properties, etc., basic data type predicates. They do not depend on particular Wikibase instance, thus they were moved to wikiba.se. If you have your own Wikibase install, your items and properties would carry your domain name, since they are not Wikidata items. But Wikibase ontology would stay the same. Thus the difference. And yes, the old form of the ontology URLs will not be used anymore on Wikidata or Wikibase code. -- Stas Malyshev smalyshev(a)wikimedia.org

Reply

Tom Morris

11:25 p.m.

On Sat, Apr 30, 2016 at 7:34 PM, Jan Macura <macurajan(a)gmail.com> wrote:

I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated?

The original is currently a 404 which isn't very cool URI <https://www.w3.org/TR/cooluris/>-ish. Shouldn't the ontology still be available for those who have used it in the past? Are there additional URIs for non-Swedish versions of the new ontology? Tom

Reply

Markus Kroetzsch

2 May 2 May

1:16 p.m.

On 01.05.2016 17:25, Tom Morris wrote:

On Sat, Apr 30, 2016 at 7:34 PM, Jan Macura <macurajan(a)gmail.com <mailto:macurajan@gmail.com>> wrote: I've been using the <http://wikidata.org/ontology#> namespace for datatype properties for some time (more than a year). Now I can see everywhere only the <http://wikiba.se/ontology#> ns. Was there some reason for change? Are these two somehow compatible? Will the first one be deprecated? The original is currently a 404 which isn't very cool URI <https://www.w3.org/TR/cooluris/>-ish. Shouldn't the ontology still be available for those who have used it in the past?

I believe the ontology has never been available under these URLs. It had been supplied in a different document at the time when our work on RDF was still a proposal from a research paper rather than the official implementation by Wikidata. We didn't know that our work had been so widely adopted in the short time between its publication in late 2014 and the roll out of the current solution in mid 2015. If there are any linked data users who need the old ontology under this old location, it would be interesting to hear about their use cases, but I would rather suggest us all to switch to the new URIs.

Are there additional URIs for non-Swedish versions of the new ontology?

No. :-) Markus -- Markus Kroetzsch Faculty of Computer Science Technische Universität Dresden +49 351 463 38486 http://korrekt.org/

Reply

2779

days inactive

2781

days old

wikidata@lists.wikimedia.org

Manage subscription

13 comments

4 participants

tags (0)

participants (4)

Jan Macura
Markus Kroetzsch
Stas Malyshev
Tom Morris