Page MenuHomePhabricator

Get Tora on the SPARQL Federation whitelist
Closed, DeclinedPublic


Task created 2 oct 2019: Would like to do SPARQL Federation Wikidata -> Swedish National Archive TORA i.e. we need to get TORA SPARQL Endpoint on the white list

Issue: Looks like noone is managing the Wikidata:SPARQL federation input since Stas left the building

Would like to compare our sources on a map see example and use Listeria as we do for T200668: Set up Nobel Data as federated search with Wikidata

Event Timeline

Hi @Gehel just an indication if we could plan that this will be working or not would be of great help..

I hope to focus on better connecting the Swedish National Archive with Wikidata at the end of this week see T237950: Feedback errors in Wikidata <-> TORA as I feel Wikidata has a better SPARQL engine compared to the TORA project I feel getting TORA on the whitelist will help us a lot of getting better data quality....

Sorry, we've really not been good at keeping track of incoming requests lately.

At the moment, we're having a lot of trouble with WDQS, so it's not really a good time to add more changes (even if fairly minor). Now that this ticket is on our radar, we'll get back to it once the situation is more under control.

Addshore added a subscriber: Addshore.

@Gehel how are things with the wdqs no? any chance we could add this one?

Addshore triaged this task as Medium priority.Jan 16 2020, 11:20 PM

ping @Gehel any good news for adding this to the whitelist ;-)

Sorry again for the delay. We are focusing on stabilizing WDQS at the moment. We don't have a good understanding of the impact of federation on the stability of the service, especially since this 'SPARQL engine they have is "shaky"'. So it is unlikely we will had more complexity until we are in a much more stable situation.

@Gehel Ok we have SPARQL federation Tora -> Wikidata link but for cleaning the data I feel the SPARQL engine at WIkidata is better so that is the main reason..... As TORA is a project at the Swedish National Archives it has high visibility in Sweden for cultural institutions so we will appreciate when we can test it.... and maybe get more culture data available using SPARQL in Sweden

OT: @Lydia_Pintscher I met today the CEO of Neo4j one of the best graph db:s Emil Eifrem and asked him about scaling and Neo4J and if they are an option for being the engine for Wikidata instead of Blazegraph... he told me that if you (Wikidata) are interested in evaluating Neo4j then send him an email and he will get you in touch with the right people. His guess is that scaling is not a problem ... but you never know

Emils contact info:

you can mention that he had a small chat with Magnus Sälgö in Stockholm today. He mentioned that they have been part of things like the panama papers and are interested in also new challenges

@Salgo60 Please remember that Phabricator is a public space, and personal data of other people should probably not be exposed there.
As for Neo4j, we had a look at it but it does not match with our product at the moment, as it's not handling triple endpoint and sparql.