[03:17:27] <icinga-wm>	 PROBLEM - WDQS SPARQL on wdqs1013 is CRITICAL: CRITICAL - Socket timeout after 10 seconds https://wikitech.wikimedia.org/wiki/Wikidata_query_service/Runbook
[03:24:47] <icinga-wm>	 RECOVERY - WDQS SPARQL on wdqs1013 is OK: HTTP OK: HTTP/1.1 200 OK - 689 bytes in 1.061 second response time https://wikitech.wikimedia.org/wiki/Wikidata_query_service/Runbook
[08:33:22] <Wessel>	 Hi, small question: is it possible for an entity/item to have more than one P31 claim? I.e., can an item be an instance of more than one thing?
[08:34:11] <hoo>	 Sure
[08:34:39] <Wessel>	 Ok good to know, thanks!
[09:26:14] <Myst>	 Hello, I am currently developing a tool to update Wikidata, but I would need data that need to be updated weekly, would anyone have an idea about this kind of data? It's mainly to test my code (under supervision)
[10:27:30] <sdesalcala>	 What would be an efficient way that given a Wikidata item, you find all the relations between that item and other items? Basically, all triples (Item1, Property, Item2) in which your chosen item is either Item1 or Item2. You can do something like that with SPARQL, https://paste.debian.net/1193308/, but it gives a timeout error for any item that has
[10:27:30] <sdesalcala>	 a non-trivial amount of relations.
[12:04:21] <wmchat-bridge>	 [mattermost] <Lucas> sdesalcala: you can get all the incoming links like this: https://w.wiki/3BHB
[12:04:40] <wmchat-bridge>	 [mattermost] <Lucas> or all the outgoing ones, by changing ?object to ?subject in line 4
[12:04:46] <wmchat-bridge>	 [mattermost] <Lucas> combining both into one query doesn’t seem to work for some reason
[12:05:37] <sdesalcala>	 Thank you. Is this more efficient than the way I showed?
[12:05:51] <wmchat-bridge>	 [mattermost] <Lucas> I would assume so? since it doesn’t time out ^^
[12:05:59] <wmchat-bridge>	 [mattermost] <Lucas> or maybe I misunderstood
[12:07:57] <wmchat-bridge>	 [mattermost] <Lucas> hm, it does time out for Q183
[12:08:24] <sdesalcala>	 Yes, you're right, it doesn't time out. I guess what I meant to ask is why is this more efficient? Also, how come the labels are not shown?
[12:08:45] <wmchat-bridge>	 [mattermost] <Lucas> the labels aren’t shown because I disabled (commented out) the label service
[12:08:50] <wmchat-bridge>	 [mattermost] <Lucas> which is also part of why it’s more efficient
[12:08:56] <wmchat-bridge>	 [mattermost] <Lucas> the label service is super expensive
[12:09:20] <wmchat-bridge>	 [mattermost] <Lucas> here, this version still works for Q183 (1.2 million results), at the cost of only counting “best-rank” statements https://w.wiki/3BHE
[12:10:03] <wmchat-bridge>	 [mattermost] <Lucas> but that kind of result size is completely out of the question with the label service
[12:10:14] <wmchat-bridge>	 [mattermost] <Lucas> I think four-digit or five-digit results counts are more or less the limit of what you can get labels for
[12:10:19] <wmchat-bridge>	 [mattermost] <Lucas> before it’s just an inevitable timeout
[12:12:25] <wmchat-bridge>	 [mattermost] <Lucas> (I also removed the qualifier part, since it didn’t seem relevant to the main part of your question – I assumed you meant only main statement triples)
[12:18:48] <sdesalcala>	 I see. If I also wanted to get the labels, I should just call the Wikibase API?
[12:19:56] <sdesalcala>	 And also the property itself
[12:28:06] <wmchat-bridge>	 [mattermost] <Lucas> probably better to get the labels from the API, yes
[12:33:24] <sdesalcala>	 I see, thanks for the info!