[01:44:52] Has anyone here used dgraph? [01:57:29] How much RAM does the wikipedia blazegraph use? [11:50:25] Amir1: You're making Nick very happy ;-) [13:44:07] multichill: :D [13:44:22] At first I thought you mean NickServe [13:44:36] but now I get it :D [15:15:40] Amir1: I wonder what would make a Nickserv happy ;-) [15:18:24] by changing nick names too much :D [15:25:25] moin moin, I have a question about IUPAC names, for chemical compounds... this has been in hold since 2003 because of the need of "multilingual type" ... see https://www.wikidata.org/wiki/Wikidata:Property_proposal/Pending [15:26:28] IUPAC names are always in a single language... but, yes, 1-propane@en == 1-propaan@nl [15:26:59] but that is not really different from trivial names, I think... "acetic acid"@en == "azijnzuur"@nl [15:27:18] comments, suggestions, and advice is most welcome [15:40:35] Lydia_WMDE: You might see a slight increase in entity usage on Commons. What was the graph for that again? :-) [15:42:17] multichill: https://grafana.wikimedia.org/dashboard/db/wikidata-entity-usage?refresh=5m&orgId=1 this one? [15:42:38] (the drop is not an actual drop but instead us changing how we handle label fallback tracking there) [15:42:56] there is indeed a little spike in commons [15:42:58] \o/ [15:43:01] what is it? [15:47:46] https://grafana.wikimedia.org/dashboard/db/wikidata-entity-usage-project?orgId=1&var-project=commonswiki gives some details... [15:59:11] Lydia_WMDE: https://commons.wikimedia.org/wiki/Special:Contributions/BotMultichill [16:03:04] ah cool [16:06:52] Modified the template so it links to the articles. Was planning to spend some time on this for a while [17:37:21] egonw: as far as I know, the multilingual datatype is not even being worked on, and these days we seem to just add monolingual text properties [17:38:08] nikki: so, what would you recommend, repropose the property? [17:38:19] yeah, that's probably the best idea [17:41:14] I don't know anything about it, if there's only one name per language, we could have a complex constraint for that. if there can be multiple, how easy would it be to fix the data later if we ever do get the multilingual type? [17:43:57] IUPAC names are not even unique... [17:44:09] one compound can have multiple valid IUPAC names... [17:44:16] one IUPAC name, however, is always one compound [17:44:48] so, there can be more than one for each language... [17:44:53] but they are special... [17:45:00] (well, for chemists :) [17:46:09] ok, thanks... I will take it up with Wikidata:WikiProject Chemistry