Re: sameAs.org

Hi Hugh,

Very relevant to me, as I wanted to use it for live query federation with Linked Data Fragments (http://linkeddatafragments.org). I was thinking of building a proxy, but something native would be better of course.

I suggest to simply publish the dataset as a Triple Pattern Fragments API (all software is already available, see http://fragments.dbpedia.org/en ) or have a hydra (http://www.hydra-cg.com/spec/latest/core/ <http://www.hydra-cg.com/spec/latest/core/>) description of your new service to aid LDF clients.  If you’re interested, let me know.

In any case, an available RDF data dump would be very useful.

Best,

Miel






> On 03 Feb 2017, at 15:55, Hugh Glaser <hugh@glasers.org> wrote:
> 
> Hi Tim,
> Thanks.
> Can I just warn you that the API for it may change in the near future! (That is, it will become more REST-like modern.)
> You may be interested in some of the sub-stores that are national library-based:
> http://sameas.org/store
> And there is also a store that is drawn from solely library sources:
> http://sameas.org/store/kelle/
> Best
> Hugh
> 
>> On 3 Feb 2017, at 14:44, Tim Thompson <timathom@gmail.com> wrote:
>> 
>> Hugh,
>> 
>> Like Andy, I haven't used sameAs.org before, but it looks like a valuable resource indeed. My library is currently taking steps to reconcile our legacy data and add URIs, and I can see us making use of the service.
>> 
>> Thanks for your work!
>> 
>> Tim
>> 
>> --
>> Tim A. Thompson
>> Metadata Librarian (Spanish/Portuguese Specialty)
>> Princeton University Library
>> 
>> www.linkedin.com/in/timathompson
>> tat2@princeton.edu
>> 
>> On Fri, Feb 3, 2017 at 9:11 AM, Hugh Glaser <hugh@glasers.org> wrote:
>> Hi Andy,
>> Welcome!
>> 
>> Yes, I've been meaning to add the wikidata data for a while.
>> But sort of putting it off because the current machine is past its last legs., and the software it runs on is no longer supported.
>> 
>> I'm very happy to work more closely with wikidata - there will be a lot of stuff I have that isn't there.
>> Partially because we have been around for so long, and a lot of URIs have died; see, for example
>> http://sameas.org/html?uri=http%3A%2F%2Fdata.kasabi.com%2Fdataset%2Fchembl-rdf%2Fchemblid%2FCHEMBL13296
>> I consider dead URIs are still useful, certainly fro lookup.
>> Also, I cast the net quite wide (high recall, low precision) for the main sameas.org store, since it is a search mechanism.
>> I suspect that this might cause problems for putting stuff into wikidata.
>> 
>> Thanks for your message - I feel encouraged to do the work :-)
>> 
>> Hugh
>> 
>>> On 3 Feb 2017, at 11:25, Andy Mabbett <andy@pigsonthewing.org.uk> wrote:
>>> 
>>>> Do you ever use sameAs.org?
>>> 
>>> [First post - hello, everyone!]
>>> 
>>> I hadn't seen sameAs before, but it looks useful, and I will do in future.
>>> 
>>> That said:
>>> 
>>> 1) It doest seem to know about Wikidata. For example, given
>>> 
>>>     https://www.wikidata.org/wiki/Q42
>>> 
>>>     (or https://www.wikidata.org/enity/Q42 )
>>> 
>>> it doesn't return anything other than the input; and given:
>>> 
>>>   http://sameas.org/html?uri=http://viaf.org/viaf/113230702
>>> 
>>> it doesn't include the above Wikidata item in the values returned.
>>> 
>>> 2) It may be that all of sameAs' functionality can be provided by
>>> Wikidata, now or in the future. This may be done indirectly - see, for
>>> example
>>> 
>>>   https://tools.wmflabs.org/wikidata-todo/resolver.php
>>> 
>>> as explained at
>>> 
>>>   http://pigsonthewing.org.uk/finding-orcid-wikidata-wikipedia/
>>> 
>>> and
>>> 
>>>    https://tools.wmflabs.org/wikidata-todo/beacon.php
>>> 
>>> as explained at
>>> 
>>>     http://pigsonthewing.org.uk/wikidata-beacon-match-orcid-authority-control-identifiers/
>>> 
>>> 3) You may have data that is not yet in Wikidata. It would be good to
>>> look at how this could be imported there.
>>> 
>>> 4) Similarly, you are welcome - and encouraged - to use Wikidata's API
>>> and/or dumps; it has a CC0 licence.
>>> 
>>> --
>>> Andy Mabbett
>>> @pigsonthewing
>>> http://pigsonthewing.org.uk
>>> 
>> 
>> 
>> 
> 
> 

Received on Monday, 6 February 2017 08:56:26 UTC