Hi Magnus,

Magnus Manske schreef op 8-1-2015 om 22:07:
Hi Marteen,

they have evaluated, and rejected (rightly so) WDQ as one alternative. They also seem to have all my query commands covered, AFAICT. See the notes Nik linked to.
I don't see WDQ as an alternative. Maybe I wasn't clear enough, I'm talking about the *functionality* of WikidataQuery. So for example WDQ has the feature that I can get a list of items for which certain claims are true. That's a feature.
Personally, I will be relieved once running such a vital service is not on my part-time shoulders anymore :-) WDQ has been challenging and fun, but it should really be an official WMF service offering a Wikidata query API, just like Wikidata proper is.
Of course, but if the new service is lacking features or just not better, people will continue using WDQ. It would be like the Toolserver -> Toollabs experience all over again :-(

Maarten

Cheers,
Magnus

On Thu, Jan 8, 2015 at 9:51 PM, Maarten Dammers <[email protected] <mailto:[email protected]>> wrote:

    Hi Nik!

    Nikolas Everett schreef op 7-1-2015 om 0:01:

        tl/dr: we're going to move forward building a query service
        against Titan/Cassandra rather than OrientDB or ArangoDB or
        anything else.

    Great to hear that you're making progress! On thing you didn't
    mention in your email or in the spreadsheet is that we already
    have a query service prototype at http://wdq.wmflabs.org/ . You do
    know about that, right?

    Whatever you come up with will be compared with WikidataQuery. You
    need to manage expectations here and communicate. WikidataQuery is
    your baseline. If you lack certain features that are in
    WikidataQuery, the new query service sucks from a user
    perspective. If on the other hand you have things that are better
    (reliability? speed?), the new query service is a good next step
    from a user perspective.

    I would strongly suggest you make a comparison of WikidataQuery
    and the new query service so the users know what to expect and can
    comment in an early stage. If you don't do this you risk ending up
    with a technical outstanding service that nobody is using.

    Maarten


    _______________________________________________
    Wikidata-tech mailing list
    [email protected]
    <mailto:[email protected]>
    https://lists.wikimedia.org/mailman/listinfo/wikidata-tech




_______________________________________________
Wikidata-tech mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-tech

_______________________________________________
Wikidata-tech mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-tech

Reply via email to