[09:48:03] Hi, is something wrong today with wikidata api access? Last week it gave some timeouts, but today it doesn't work at all, even for reading. [09:48:33] I'm using pywikibot [11:18:40] SothoTalKer: yes it is a bit annoying [11:19:15] scientific publishing has become a bit spammy [11:19:21] and now you have preprint servers also [11:19:26] and some acceptence for this [11:19:28] and citing of this [11:44:31] CustosLimen_: Pre-prints have an advantage - they can be commented on and improved. Then the improved version is published. These predatory publishers will pretend to do peer review (there is no way to ensure that it was actually done) and then publish any rotgut someone pays for. [11:44:59] I agree with that [11:45:12] but I still think academic publishing tends to be a bit spammy [11:45:31] I don't think preprint servers should be shut down either [11:45:42] I get the idea behind it [11:45:43] and it is good [11:46:29] when it comes to notability criteria though, I think the only solution there is to relax it [11:46:43] and build capacity to deal with more spammy stuff like academic publishing [11:46:46] and even preprint [12:44:52] Added counter examples now: I don [12:44:56] https://www.wikidata.org/wiki/Wikidata:Property_proposal/Goodreads_work_ID [12:46:14] I honestly don't know what I can do to make it clearer [13:23:26] https://www.wikidata.org/wiki/Wikidata:Project_chat#Wikidata_COVID-19_recorded_deaths_pass_1,000_today [13:23:40] Is GerardM griefing? [13:24:05] why even make that comment [13:24:09] they can just add other items [13:24:15] free encyclopedia [13:24:18] as in speech and beer [13:24:23] why go griefing [14:07:21] seems very passive agressive to say "because we can" [14:07:31] if there is someone who is doing something wrong - why not say it [14:11:56] is there some way to mark a statement as "needs attention" [14:12:05] or "incorrect" [17:55:39] CustosLimen_: i don' [17:55:48] t know what GerardM wanted to say [17:56:13] I don't get it either [17:56:24] 🤷 [17:57:34] was probably on some drugs [17:58:25] CustosLimen_: to mark a statement as "wrong", you can give it the deprecated status with a "reason for deprecation" qualifier. [17:58:36] I see [17:58:47] Okay, thanks, will consider it [17:59:46] https://www.wikidata.org/wiki/Help:Deprecation [18:01:31] funny thing is that this is also used for authority control IDs when it's a redirect. but then there is a certain bot that replaces those deprecated redirects with the new value [18:14:49] so you got the same value twice... once as deprecated with reason "redirect" :] [18:15:28] example: https://www.wikidata.org/wiki/Q1237844 [20:59:56] 100k =D [21:04:43] annoying [21:13:53] CustosLimen_: what exactly? :) [21:14:19] the bots creating dupe statement after it was deprecated [21:14:59] I have somewhat of a similar issue with wikibase-edit npm package, does not have a way to create a statement only if it does not exists, have to implement that yourself [21:15:09] seems like a pretty common usage pattern [21:16:06] there are no guidelines how those values should be handled [21:16:31] some say "keep and mark as deprecated", others say "delete" [21:16:43] i personally do delete them [21:17:19] Well I think if something is clearly wrong, deleting makes sense [21:17:21] and I do do that [21:17:37] but sometimes I just don't care enough, or it is just weird [21:18:23] like this: https://www.wikidata.org/wiki/Q6437542 [21:18:30] the value is not wrong, but if we keep redirects, some bot could just scrape all possible values and add the redirects [21:18:55] it has edition number of "Original edition" and "Full Text edition" [21:18:59] which is all kinds of wrong [21:19:10] but maybe it made sense to someone [21:19:20] especially for volatile databases like VIAF, which do tons of changes [21:23:06] CustosLimen_: people add stuff that makes no sense sometimes :] [21:23:54] Well maybe making those deprecated (as I did now) will help some future editor when they come along, but otherwise I guess it is just confirmation for someone that they can be deleted [21:46:59] that property does not even belong to a "written work"?