[02:05:17] Hi [02:09:51] Given a q code, is it possible in a single sparql query to check if this entity is a human and if yes find all entities that have the same country of birth and same occupation ? [02:10:17] e.g. if I enter Katy Perry I should see Nicky Minaj [02:41:54] Given the exemples I see it's most likely possible, but the doc is a bit obscure... [03:15:55] Ok I figur3ed it out :p [04:57:07] Is there a way to only consider the first occupation listen by wikidata ? [04:57:10] in the query [05:19:24] And is there a way to say "people born less than 5 years before or after Katy Perry" [05:43:33] FOr some reason I can't do BIND(concat(STR(year), "-01-01")^^xsd:dateTime as ?test) but I can if I use a string directly [05:58:36] ok figured the date issue out, but I would still like to only consider the first occupation of the list [05:58:41] is there a way in sparql ? [07:04:51] Nazral: I don't think the occupations are listed in order of priority or anything [07:21:12] reosarevok: ah, too bad :( [07:21:14] but good point [07:22:10] In theory, the occupations should be all important enough anyway, but I can imagine some actually aren't [07:22:47] Still, you'd want someone who sings *and* acts (Will Smith or something) to appear on both lists of other singers and other actors, I guess, so restricting it to one is not ideal :) [07:26:56] frickel, Nemo_bis: there is a tool from Magnus for that: https://tools.wmflabs.org/sourcemd/new_resolve_authors.php? [07:29:08] Thanks! I had reached sourcemd but didn't find that specific path :) [07:36:41] It worked! https://www.wikidata.org/w/index.php?title=Special:Contributions&dir=prev&offset=20180605073521&limit=21&target=QuickStatementsBot [07:36:50] Finally I understand what people are doing. ;) [07:38:55] I had not realised that the strings are left in place even after the authors are converted to IDs. Hmm. https://www.wikidata.org/wiki/Q48853238 [08:50:34] reosarevok: and inversely, would there be a way to find entities that match all the instances of ? [08:51:01] e.g. https://www.wikidata.org/wiki/Q11035, a query that would match all 3 instances [08:52:39] Doesn't it work if you just ask for all three? (that's subclass, not instance, but) [08:53:12] So ?item wdt:P279 wd:Q11019; wdt:P279 wd:Qwhatever; etc [08:53:46] The problem is that I don't know them in advance [08:54:45] I could make a query that finds the instances and then build a new query like this though, but I am wondering if it is possible at all in one query [08:55:57] I suspect it might be... but I'm not good enough to tell you [08:56:15] Maybe if Lucas comes around :) Or someone else who's better at hard queries [08:56:32] abian: do you happen to know about that kind of stuff? ^ [08:56:56] thanks for the help anyway [09:02:15] what is the / operator between wdt:P106/wdt:P279 ? [09:04:25] IIRC it means "either of those" [09:13:07] mhh ?statement wikibase:rank ?rank. < this looks like what I want for occupations maybe [09:17:10] nope [09:18:19] I mean, yes technically, but it looks like it's not attributed for many entities [09:21:58] Yeah, you can ignore anything with deprecated rank, but in general I suspect it's not heavily used [09:30:14] My goal is to have queries that help me find the "closest thing to a thing" [09:31:05] for people it's ~easy, I made https://pastebin.com/mudQ0bFd [09:32:27] I made something for companies and things like that [09:32:32] but it quickly becomes hard [09:39:56] Nazral: "?x wdt:P106/wdt:P279 ?z" is a shorter way of writing "?x wdt:P106 ?y . ?y wdt:P279 ?z" [09:42:08] reosarevok: ^ too, for future reference [09:42:47] ok, so it's an and [09:43:14] thank you [09:43:52] nikki: what's the difference between this and brackets ? [09:44:10] I'm not sure what you're referring to [09:44:28] ?x wdt:P106 [ wdt:P279 ?z ] [09:44:39] ah [09:44:41] from https://en.wikibooks.org/wiki/SPARQL/Sentences [09:45:55] you can put multiple things in the brackets (e.g. ?x wdt:P106 [ wdt:P279 ?x ; wdt:Pxx ?a ]) but it's not quite as short as the slash [09:46:01] ok [09:46:05] I don't know if there are any other differences, Lucas_WMDE might know [09:46:26] Nazral: You can't either write things like wdt:P106/wdt:P279* or wdt:P106/wdt:P279+ [09:46:43] abian: I can't ? [09:47:23] couldn't you write "wdt:P107 [ wdt:P279* ?z ]"? [09:47:53] And do you know if there is a way to rank items by (for exemple) how many [P31 or P279] they share ? [09:49:04] e.g. wd:Q_myrefitem wdt:P31 ?item and then list the ?item by how many common "instance of" they have with the reference item [09:53:48] it should be possible somehow, but I'm not sure how... [09:54:37] I'm not sure how many results you would get if you search for items where the item and the instance of both have the same instance of statements though... that sounds like it shouldn't happen [09:55:48] Yeah, or very rarely, but classifying them by # of shared instances/subclass of that would be even better [11:32:45] PROBLEM - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is CRITICAL: HTTP CRITICAL: HTTP/1.1 200 OK - pattern not found - 1946 bytes in 0.080 second response time [11:37:45] RECOVERY - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is OK: HTTP OK: HTTP/1.1 200 OK - 1942 bytes in 0.096 second response time [13:34:46] hi, what property do I use for where an item is from? ie https://www.wikidata.org/wiki/Q4207259 is from "korea" (also for a non-contry like the whole korean peninsula what Q to use?) [13:34:57] non-country* [13:43:55] sorry about that [13:44:05] (i'm kassotsimine) [13:49:53] you can add multiple "country of origin" statements to indicate multiple countries, there's also "indigenous to" [13:50:04] yea. but which to use? :D [13:52:23] no idea [13:54:01] yea, that's the problem :) [14:06:15] sjoerddebruin: Ping? :) [15:24:49] hoo: yes? [15:38:50] sjoerddebruin: Shall we kill the commons media gadget? [15:38:57] I love killing things. [15:39:39] I'll remove it from MediaWiki:Gadgets-definition first and see if anything breaks, okay? [15:58:54] addshore: during the cloud services team meeting the idea came up to combine the general-k8s, wbaas, and wikibox project proposals into one project [15:59:44] hare: sounds possible, although I guess it would need increased instance size then? [16:00:13] i think that's basically fine since one project with larger quota would probably still be smaller than three separate projects [16:00:26] I would want to put up ~3-4 instances for wbaas probably so I can continue work, and also have ~3-4 to try and setup the full general k8s bit [16:00:35] ack [16:01:48] hi, what is exactly the issue with the statement constraint on https://www.wikidata.org/wiki/Q846109 ? (bamboo musical instrument) it says "value requires statement constraint [16:01:48] Bamboo musical instruments should have a statement subclass of." [16:02:01] but what does that mean about the danso? [16:02:11] addshore: for what it's worth i'd be happy to have wikibox as a customer to the general-k8s project. in any case figure out how much the quota should be and then we can combine the three proposals together [16:02:26] hare: okay [16:02:29] *thinks* [16:07:57] sjoerddebruin: YES! [16:09:25] hoo: please try [16:09:37] HSOWA: it means that the linked item (bamboo musical instruments) should also have a "subclass of" statement [16:10:21] thumbnails seem wider? [16:12:09] in this case, "subclass of: musical instrument" would make the most sense (plus another statement "material used: bamboo") [16:13:07] aha! (i didn't understand it becasue the (!) was shown on a (unrelated from this) instrument danso. i'll go fix then. cheers nikki) [16:13:44] sjoerddebruin: Shouldn't be the case [16:13:58] hmmm ok [16:15:29] Maybe purge the page? Some might have an earlier version of the thumbnails in cache [16:18:23] mmm, if there is a wide image all images will be big okay [16:26:34] HSOWA: well, an alternative interpretation would have been that “Bamboo musical instruments” is not an item that should be used in “subclass of” statements [16:26:42] imagine if someone added “subclass of: Donald Trump” to some item [16:26:55] then the solution would not be to add a “subclass of” statement to Donald Trump :) [17:34:58] PROBLEM - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is CRITICAL: HTTP CRITICAL: HTTP/1.1 200 OK - pattern not found - 1966 bytes in 0.117 second response time [18:12:47] sjoerddebruin: Do you have an example? [18:20:56] can I get only the Q number of an item in sparql? [18:27:38] Lydia_WMDE: About? https://phabricator.wikimedia.org/T196471 [18:28:19] Reedy: hey [18:28:21] looking [18:28:30] Thanks [18:29:09] There seems to be quite a few of those errors today... ~650 [18:29:17] over 1000 yesterday [18:30:02] Reedy: *nod* I put it on the board of the people responsible for looking at this [18:33:53] hoo: https://www.wikidata.org/wiki/Q22151 [18:34:37] Lydia_WMDE: Can you get raz_WMDE to add a description so the rest of us know what the campsite is? :P [18:35:27] Reedy: *lol* yes. the short version: there is always a part of the team in the "camp" to take care of maintenance etc while some other parts of the team go on "hikes" to develop larger features [18:35:40] this board is the campers' kanban board [18:35:48] greg-g: ^ [18:35:50] :P [18:36:08] RECOVERY - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is OK: HTTP OK: HTTP/1.1 200 OK - 1948 bytes in 0.105 second response time [18:36:26] hah, nice! [18:38:28] raz_WMDE: once we tried this for a bit we should probably publish something about it [18:39:26] (this is pretty new. we started properly on Monday. so still figuring some things out and wiggling them into place.) [18:40:45] :) [18:41:01] Just a short description on the phab tag is more than enough for now [18:41:23] There's something about things burning too in there somewhere [18:41:30] :D [18:53:37] PROBLEM - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is CRITICAL: HTTP CRITICAL: HTTP/1.1 200 OK - pattern not found - 1971 bytes in 0.095 second response time [18:54:46] is there a reason why there is no ASIN (Amazon Standard Identification Number) in wikidata? [18:54:59] for products and stuff [18:55:30] wait [18:55:37] I should use catalog (P972) , right? [18:58:47] RECOVERY - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is OK: HTTP OK: HTTP/1.1 200 OK - 1958 bytes in 0.062 second response time [19:00:02] there's been at least one proposal before, https://www.wikidata.org/wiki/Wikidata:Property_proposal/Amazon_Standard_Identification_Number [19:01:27] I guess the main reason is because we don't tend to have items that correspond to a single asin other than editions of books and amazon uses the isbn for those [19:02:08] I am tring to import editions and I have plenty of ASIN numbers attached to my data [19:02:48] they all should correspond to a single amazon item [19:04:41] oh, I could use `described at URL (P973)` [19:05:50] nikki: what do you think? [19:09:33] I would probably ask on project chat first [19:09:55] I'm not sure whether people would like the idea of linking to shops [19:12:29] ah I thought there was another proposal - https://www.wikidata.org/wiki/Wikidata:Property_proposal/Archive/5#Amazon_Standard_Identification_Number_/_Amazon_Standard-Identifikationsnummer_/_numéro_d'identification_standard_Amazon [19:12:32] that one's a lot older [19:17:31] but they all assume its used for works rather than editions [19:26:07] PROBLEM - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is CRITICAL: HTTP CRITICAL: HTTP/1.1 200 OK - pattern not found - 1965 bytes in 0.072 second response time [19:37:16] ASINs can be different for the same item and ASINs are not global. [19:41:27] RECOVERY - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is OK: HTTP OK: HTTP/1.1 200 OK - 1961 bytes in 0.069 second response time [20:01:36] so. https://www.wikidata.org/wiki/Q749247 the english now is redirection to a section of a page. is that good still or should be removed? [20:04:21] depends who you ask [20:15:56] how should I add the format of a book e.g. hardcover, softcover e-book etc? [20:16:06] instance of? [20:16:55] distribution (P437) ? [22:26:17] PROBLEM - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is CRITICAL: HTTP CRITICAL: HTTP/1.1 200 OK - pattern not found - 1947 bytes in 0.063 second response time [22:41:37] RECOVERY - wikidata.org dispatch lag is higher than 300s on www.wikidata.org is OK: HTTP OK: HTTP/1.1 200 OK - 1972 bytes in 0.075 second response time