[00:06:05] aude, https://en.wikipedia.org/wiki/User:Yurik/maplink#/maplink/0 [00:06:08] and everyone else :) [00:06:23] a little link that shows a map, on Wikipedia, with the data directly from Wikidata :) [00:06:25] enjoy :) [00:06:30] * Sky2042 looks at his shiny, reused cloak [00:06:47] don't forget to click on states [00:07:58] cc: SMalyshev ^ [00:08:05] \o/ [00:12:26] yurik: yay! [00:12:48] SMalyshev, notice that it is no longer showing the -180 thingy :) [00:12:55] i removed the arg1= hack [00:13:26] yurik: alaska seems to be broken though [00:13:41] SMalyshev, its not broken, its on the outher side of the globe :) [00:14:13] leaflet cannot draw things in multiple spots :( [00:14:14] well it's kind of weird because it displays in a wrong place [00:14:24] yeah I know we've seen that bug already I think [00:14:58] they should either limit zooming or fix the display :) [00:17:23] SMalyshev, its a very tricky thing to fix - ideally anything that is drawn within -180-180 should repeat left and right [00:29:33] yurik: probably could be done if the overlays were done as tiles [00:29:44] as tiles can repeat [00:30:05] * aude realizes this might be more complex to implement [00:30:09] aude, yep, but we don't have that tech enabled yet [00:30:13] yep [00:30:29] technically tiles support all that, and mapbox has that functionality, but there is a lot of stuff to do to get there [00:30:44] * aude has been experimenting with it [08:23:30] question: is the "copy / insert reference" broken? I'm trying to use it and it gets stuck at saving; when I reload the page the reference isn't saved [08:24:03] ESM^: yep [08:24:19] ugh… any chance to have it up again? [08:24:27] there was a announce a few day ago (maybe yesterday?) [08:24:34] people are working on it [08:24:48] so it should be repaired soon [08:24:52] well, I'll wait a bit then, thanks! [08:25:34] I love this gadget too [08:26:02] * harmonia loves many gadgets and tools actually [08:26:21] What gadget is this, which I do not have...? [08:26:53] :) I need it to manually reference some data that can't be introduced via quickstatements [08:27:20] ESM^: yep, I understand [08:27:30] I put a .csv on hold for this same reason [08:27:38] but love it indeed [08:28:09] Josve05a: DuplicateReference [08:28:15] Josve05a: you can copy/paste references on several statements of an item, just open the reference section of any statement and you'll find a link that prompts you to copy the reference [08:28:27] ohh, nice [08:28:31] ESM^: Tpt wanted to create a QuickStatements+ [08:28:46] which could handle multi-properties-references [08:29:00] :D that would be awesome [08:29:06] wonder if he still works on it [08:29:22] he's a little overworked with his thesis :p [08:29:51] * harmonia would love a QuickStatements+ for Wikidatay birthday :p [08:30:09] that's why I'm manually introducing artwork dimensions :| [08:30:53] ESM^: units? [08:31:02] both cm and mm [08:31:19] * harmonia still hadn't finished to add the necessary units on sled dog races items [08:31:23] *sigh* [08:31:28] so many things to do [08:31:34] I got a fellow user code something to solve this but he's MIA until later this month [08:31:46] indeed [08:32:10] ESM^: what's his/her name? we should let him/her in contact with Tpt [08:32:47] it's KRLS, from cawiki [08:32:53] thanks [08:32:57] I'm meeting him soon, will tell him [08:33:15] thanks again :) [08:34:04] * harmonia is always a little disappointed when a constraint violations report shows her hard work *and* a bunch of new errors at the same time [08:34:35] (like they wouldn't even let me a day to admire my no-violation work) [08:34:46] (that's conspiracy :p) [08:57:24] Lydia_WMDE: !!! [08:58:04] :) [09:00:05] yesterday a coworker randomly said he loved Wikidata (I don't work in any related field and they don't know I'm Wikimedian) [09:57:13] Is it possible to see which items has the most eternal identifiers listed on WIkidata? (Like top 10 items) [09:57:21] external* [09:57:26] (most different*) [10:01:01] harmonia: wohooooooooo :D [10:01:08] Amir1: ready for hangout? [10:01:25] yeah [10:01:48] joining? [10:17:46] Can someone make this an external ID? https://www.wikidata.org/wiki/Property:P627 [10:18:53] Lydia_WMDE: What was the link again where the userpage to request such thing? ^ [10:45:09] I will let you know when I see sjoerddebruin and I will deliver that message to them [10:45:09] @notify sjoerddebruin Is this persons name Bernand or Bernandus? NL, you be weird... [10:47:23] Josve05a: who is "this person" ? [10:47:41] damn..stupid irc... [10:47:51] * Josve05a goes to MemoServ... [10:47:57] :p [10:49:01] (was talking about https://www.wikidata.org/wiki/Q2898380, but I merged it with already existing item under another name) [10:49:45] Josve05a: https://www.wikidata.org/wiki/Wikidata:Identifier_migration [10:50:19] oh, it's on wikidata ns now :D [10:50:34] Josve05a: I would say, we have sources for boths given names [10:50:40] so we should have both [10:51:12] and if there is a definite one "true" given name, we should then mark the other as deprecated [10:51:56] well, which one shoudl be the "name" on Wikidata, and which should be aliases? [10:52:27] 2 identifiers (P&P and Biografisch Portaal) said Bernandus, and the fr article says Bernand...ugh [10:53:42] Josve05a: oh you mean in label/aliases, I thought you meant in the property [10:54:03] oh :p no, I ment as in label [10:54:39] I used the french one everywhere because the one already in use in most existing labels [10:54:47] but no definite opinion here [11:07:16] Josve05a: people often used a latinized version of their name (with an "us" at the end, or completely translated) when speaking or writing in latin, which they frequently did, just as we now use english. [11:08:03] So my guess is that some bloke called Bernand in Dutch would be known as Bernandus to foreigners, or also in academic circles. [11:08:37] Which name is "real" really depends on context and perspective. I'd just look at what Wikipedia uses as their page title. [11:10:34] (fun fact, Mozart was baptized "Theophilus". He translated the Greek to Latin "Amadeus" - I guess it sounded nicer. I wonder whether he also used the German "Gottlieb"). [11:12:08] old -timey people :p [11:23:17] Josve05a: well, today, Chinese people often have an extra "English Name", and e.g. Russians will have an anglized version of their name ("just call me John"). Same thing. [11:24:42] LeaAuregann_WMDE: want me to just edit the wiktionary model slide? [11:25:13] LeaAuregann_WMDE: do we want to pretend that there is a single structure, though? or would it make sense to show some different structures from different wiktionaries? [11:25:28] Shoudl "Middle names" (if person has 2 names whch aren't surnames) be added as given names? [11:27:58] Josve05a: yes [11:28:12] Josve05a: and you can use "order in the series" as a qualifier [11:28:14] ok, thanks! That has bugged mefor a while, but never asked [11:29:06] because translated to Swedish it is "förnamn" (I think), and that translates back to "first name" [11:29:52] they are many cultures with several given names [11:30:00] and only one or two are used daily [11:30:35] I have one "first name" and two "middle names", btu I guess all three would count as "given names" then... [11:31:04] (and two last names, which some may think is just "one"...) [11:31:22] DanielK_WMDE: hey, one quick question: https://gerrit.wikimedia.org/r/#/c/309551/1/client/WikibaseClient.php [11:31:40] wbentityusagelist is okay? or wbentityusage is enough? [11:31:52] or is it just one last (sur)name, sepearated with a space...hmm...not sure...I always say I have 2... [11:32:29] (also for prefix: wbeul? or wbleu?) [11:32:33] Amir1: just my 2p but would it be consistent to call it wbgetentityusage? [11:32:56] Josve05a: can you use only one in administrative papers ? [11:33:16] addshore: yeah, and about prefix? I think it should be unique [11:33:36] what repo / extension is it being developed in? [11:33:42] client [11:33:51] I add them both in administrative papers with a space in between. But it is in "one" field, so it may be treated as just one surname...not sure how we would do it if I were to become an item... [11:34:00] ahh, its client and a props/list module hmmm [11:34:09] wbceu ? [11:34:21] we have wbeu for the prop [11:34:26] (and merged) [11:34:35] but not deployed yet [11:35:16] looks like it is deployed! [11:35:22] "Jonatan (first name) Pär Magnus (2 middle names) Svensson Glad (2 surnames)" - no dashes, so are seperated names...right? harmonia would you know? xD [11:35:25] https://en.wikipedia.org/w/api.php?action=query&prop=wbentityusage&titles=Main%20Page [11:35:52] oh [11:35:53] Josve05a: when they are no dashes, I usually add them separately and add the qualifier "order" [11:36:01] \o/ ? [11:36:05] Josve05a: to have them in the right order :) [11:36:20] i dont know wbleu kind of seems a tiny bit nicer to me [11:36:30] wbeul seems to also be fine, :P [11:36:57] harmonia: ok, so you would not create a new item for that "combined" surname then :) good to know [11:38:06] ))) [11:38:49] Josve05a: hmm, without dashes, it wouldn't be my instinct, no [11:39:27] https://en.wikipedia.org/w/api.php?action=query&prop=wbentityusage&titles=Hubble%20Space%20Telescope [11:39:31] lots of aspects :D [12:21:17] Amir1: wbcentityusage? [12:21:28] err wbceu? or wbeu? [12:28:43] :D [12:34:57] Thiemo_WMDE: do you want to join the presentation? [12:43:44] Amir1: the new module lists usages of a given entity, so it's kind of the inverse of the props module, which lists entities used by a specific page, right? [12:43:56] then it should have a different name. [12:43:58] yes [12:44:29] it should have the same prefix though, i think [12:45:07] wbentityusagelist is a bit clunky, but ok for now i think. [12:45:37] okay [13:10:27] DanielK_WMDE: ping [13:10:37] https://phabricator.wikimedia.org/T145138#2621621 [13:48:49] hoo: getLabelTerm -> object? [13:48:59] seems more flexible & future proof [13:49:20] DanielK_WMDE: Yeah, we can also return Term (like) tables, yes [13:49:31] just not as a drop in replacement [13:49:41] Lua doesn't invoke tostring if you concatenate to a table [13:51:06] hoo: i can't really say what the best option would be, since i don't understand the details enough. [13:51:37] I would have loved a drop in replacement, but given how Scribunto handles multiple return values, that's not possible [13:51:47] So I needed to revert my earlier patch [13:51:50] so, scribunto concatenates all values even for internal calls? it should only do that when converting module output to wikitext [13:51:55] maybe that can be changed [13:52:14] DanielK_WMDE: Nah, it only does for the stuff you return to it [13:52:33] so if you just give it return "a", "b" it will make "ab" from that [13:52:53] That's a problem only if the values of mw.wikibase.label etc. are directly returned [13:53:10] so you have Lua code that returns a,b, which is involked from other Lua code, and that other Lua code gets ab? [13:53:41] DanielK_WMDE: No, this works fine as long as Lua is involed [13:53:49] ah, so the problem arises only if i do {{#invoke:mw.wikibase|label|Q1234}} directly? [13:53:58] Basically [13:54:19] in that context, concatenation kind of makes sense, yes... [13:54:26] but obviously also if you pass the two return values through the stack… whenever they end up in Scribunto [13:54:34] returning an object and making scribunto call tostring would make it compatible for this use case [13:54:42] but that wouldn't be compatible for invokation from lua, right? [13:54:55] yeah, taht would work [13:55:04] but would break string concatenation [13:55:17] ...why? [13:55:20] because for some reason lua doesn't call tostring if you concatenate to a table [13:55:38] yes, but we could fix that, no? [13:56:00] * DanielK_WMDE is still blurry on Lua "table" semantics vs "object" or "list". [13:56:44] anyway, a new function returning some kind of term object is probably best [13:56:46] even if it [13:56:50] ah, wait [13:56:52] even if it's not a drop-in [13:56:56] we can overwrite a concat on the table [13:57:03] * concat function [13:57:19] that will probably make us return a table which acts like a plain string, if you want it to [13:57:26] but that sounds a little awry [13:57:53] i' [13:58:07] i'd have to look at the code to fully understand what this means [13:58:17] i'm about to leave to catch my train... [13:58:52] hm... but a concat that would only return the first element would indeed work, and restore compat... [13:59:05] i can't tell what other implications that might have, though [13:59:57] well, the type would change, if manually checked via type() [14:00:01] but I think that's about it [14:01:05] the lenght operator's semantics change, hm [14:57:11] Creating items with mix-n-match :D https://www.wikidata.org/wiki/Q26837519 [15:06:09] Hey! Someone would be available to help me to modify Template:Sister Project Header ? [15:18:18] LeaAuregann_WMDE: I don't have the skills necessary to help but I offer my moral support :-) [17:42:55] evening people ! [17:43:08] Evening [17:43:39] per-wiki reporting tool for items pointing to redirects. this week's reports : https://tools.wmflabs.org/wikidata-redirects-conflicts-reports/reports/2016-36/ [17:44:15] interface will be better over time, and will include graphic representation of evolution over time :) [17:44:39] hopefully we'll be able to bring everything to 0 ^^ [17:55:23] Alphos: for mtwiki, Q3915462 is listed, but the sitelink for it is not a redirect [17:55:41] that's odd [17:57:25] same problem for Q13553871 it seems too [17:58:18] hold on, let me bring up a mysql-client to see what's what [18:01:03] this may take a while, labs are sluggish :x [18:02:03] that's ok :) it's not exactly urgent [18:02:23] still weird though [18:02:45] picking up the data straight from the db, so might be an issue in mtwiki_p [18:03:01] all three for cdowiki have the same issue too [18:08:26] also the one for jamwiki, which was only edited a couple of months ago :/ [18:10:01] i'd love to debug, but not really in the capability of doing so right now ^^' [18:13:12] yeah, understood [18:13:16] was just giving you some more examples [18:14:42] thanks :) no such thing as too many examples when debugging ^^ [18:52:23] =o [18:53:14] nikki ok, so the issue is it WAS used https://www.wikidata.org/w/index.php?title=Q3915462&diff=149779052&oldid=149648778 [18:54:15] it still appears with eu_aspect X (which tbh i didn't take into account) anyway, which makes discriminating it from the legitimate [[Lingwa_tas-Sinjali_Ħawsija]] pretty darn difficult [18:54:16] 10[2] 04https://www.wikidata.org/wiki/Lingwa_tas%2DSinjali_%C4%A6awsija [18:54:25] but that's over two years old D: [18:54:52] i know :/ [18:54:59] still, nothing i could do there ^^' [18:55:48] what's eu_aspect? [18:56:15] basically determines what "aspect" of the page was used by wikidata or sister projects [18:57:40] SMalyshev: thanks for the merge :) [18:58:07] Do you think it's appropriate to say Futaba chan (Q1191246) is inspired by 2channel (Q222909) considering what's said in the wikipedia page it being a refuge for 2channel users? https://en.wikipedia.org/wiki/Futaba_Channel [18:58:17] addshore: np :) [18:58:37] nikki : S is sitelinks, the old kind, residing in wikicode ; L means the label was used, and i'm not sure in what situation that aspect is used ; T means the local page name was used, and also not quite sure what it's for, O is for "anything else" (same thing) ; and X is for "ever aspect" https://github.com/wikimedia/mediawiki-extensions-Wikidata/blob/master/extensions/Wikibase/client/includes/Usage/EntityUsage.php [18:59:01] Alphos: what are you trying to work out? [18:59:53] nikki for Q3915462, wbc_entity_usage has these three rows http://pastebin.com/R0CSYp4J 22490 is for "Hawsija", 22499 is for the double-barred version [19:00:11] addshore a kink on mtwiki_p.wbc_entity_usage [19:00:23] not sure if the kink is there or in my understanding of it, tbj [19:00:24] tbh [19:00:35] entity usage is a confusing area ;) [19:00:44] that much is obvious >_> [19:01:08] so basically, i have an item Q3915462 which has 3 rows in mtwiki_p.wbc_entity_usage [19:01:18] yup [19:01:33] one of these has aspect S, so i'm guessing someone linked it from another wiki the old way of interwikifying things [19:02:00] hmm, no, that shouldnt cause a usage flage (I dont think) [19:02:33] two of them have aspect X, so i'm guessing someone linked them both from wikidata, they were both used for their labels, and both of them had their local page name used *somewhere* [19:02:52] the issue is wbc_entity_usage doesn't seem to reset values that were used but aren't anymore [19:03:13] one thing that is important to note is that an entity usage flag doesnt 100% mean that data is actually used, it just means that it could be used [19:03:16] and the funny part is one page (a redirect, btw) has two rows in wbc_entity_usage, one with S, one with X [19:03:19] / looks like it could be used [19:03:50] oh dear ^^' maybe relying on it wasn't such a great idea after all :D [19:03:58] I think X may be there for most cases. [19:04:18] Usage flag indicating that any and all aspects of the entity were (or may have been) used. [19:04:49] earlier my script used a db-spanning JOIN with an index-breaking REPLACE(), took far too long [19:05:12] so i switched to wbceu, but it seems it returns not-so-valid data :D [19:05:36] what are you trying to figure out? [19:06:00] wikidata items that link to wikipedia pages that are redirects, and the target of said redirects [19:06:28] (if possible, with the target item, but i dropped that one as it really took far too long) [19:07:04] Alphos: have you seen the wb_entity_per_page table on the wikidatawiki db? [19:07:21] no i haven't, but i sure am going to ! [19:07:21] oh wait, no, thats not the right one [19:07:27] darn >_< [19:07:44] don't worry, you said it yourself, [21:00:33] entity usage is a confusing area ;) [19:07:45] :p [19:07:55] I mean the item_per_site table [19:08:07] | ips_row_id | ips_item_id | ips_site_id | ips_site_page [19:10:12] hm [19:10:22] ERROR 1146 (42S02): Table 'wikidatawiki_p.item_per_site' doesn't exist [19:10:54] oh, wb_item_per_site [19:12:13] my help you :) [19:12:16] *might [19:12:24] darn, it was what i initially used [19:12:31] https://github.com/alaefin/wikidata-redirects-conflicts-reports/blob/dea669e48b20f1a0c32e2cf88fbdc8a4aa8be5f0/sprintf.sql ON REPLACE( source.ips_site_page, " " , "_" ) = sourcepage.page_title -- identifying the redirecting wikipage locally [19:12:36] breaks indexes :-( [19:13:45] current version takes less than 15 minutes to walk through all wikipedias [19:14:04] that old version took 70 minutes for arwiki alone, and that's the biggest wiki i used it on [19:14:20] didn't even try with frwiki, dewiki, nlwiki or enwiki [19:14:29] 70 minutes was already quite a bit much [19:17:16] and quite frankly i don't expect removing the last JOIN to get the target item to make much of a dent in the time it takes [19:17:49] (removing the last JOIN *which* gets the target item) [19:19:00] otoh, i could see if performing it in pure sql would be faster than using php [19:24:54] whaddaya know, it is actually quite a bit faster indeed [19:27:15] darn, should've done this earlier >_> [19:27:43] addshore i don't thank you for making me feel stupid, but i thank you for making me realize i could better my software :D [19:27:47] nikki : thanks a bunch as well :) [19:38:20] oh wait, maybe i wasn't so stupid, was running the wrong sql -_-" [19:38:53] i'll let it run for 15 minutes for arwiki, but if it hits that limit, it's a bad sign [19:56:26] addshore i'm afraid it's a nope : takes more than 15 minutes for arwiki alone (mid-range in size) using wikidatawiki_p.wb_items_per_site, compared to 12-13 minutes for all of them when relying on wbc_entity_usage [19:57:16] nikki : i'm sorry, but there doesn't seem to be a decent workaround [19:57:44] :/ [20:00:29] in an ideal world, it wouldn't take this long. but i managed to work out the time taken per wiki when relying on ips was roughly quadratic on the number of pages on said wiki [20:01:35] arwiki is about 450k pages [20:02:08] enwiki is give or take ten times that, which means it would take about 100 times the time arwiki took, which was roughly 70 minutes [20:03:21] 7000 minutes is an awful lot for ONE wiki, nearly 5 days ^^' [20:05:07] well, one workaround would be to have wb_items_per_site on each wiki, with page titles using underscores instead of spaces, so as to not break the indexes [20:05:24] but that's really out of my hands ^^' [20:42:09] Josve05a, you were looking for me? [20:58:13] Alphos: what were you trying to do with wbc_entity_usage? [20:58:53] DanielK_WMDE trying to get a list of items pointing to pages on a given wiki that happen to be redirects, for all wikis [20:59:13] results are https://tools.wmflabs.org/wikidata-redirects-conflicts-reports/reports/ for the past two weeks [20:59:39] ni.kki pointed out some results were bogus, for instance Q3915462 on mtwiki [21:00:58] listed as pointing to https://mt.wikipedia.org/wiki/Lingwa_tas-Sinjali_Hawsija?redirect=no when in fact it points (legitimately) to https://mt.wikipedia.org/wiki/Lingwa_tas-Sinjali_%C4%A6awsija [21:01:17] (page_id respectively 22490 and 22499 [21:02:10] Alphos: so you want "give me items that point to redirects on xywiki"? [21:02:22] the "solution" would be to use wikidatawiki_p.wb_items_per_site, but that would mean a cross-db JOIN ON REPLACE() [21:02:25] DanielK_WMDE yup [21:02:51] yea... cross-db isn't a problem, but cross-servre is. [21:02:52] won't work [21:03:03] it does [21:03:07] it just takes eons to complete [21:03:51] how do you join across servers without federated tables? ah, but maybe they are all on the same server on labes? i never checked. [21:03:58] https://tools.wmflabs.org/wikidata-redirects-conflicts-reports/reportstimescatter.html that's the time it took when i run it using wb_items_per_site [21:04:04] DanielK_WMDE : "slowly" [21:04:15] Alphos: what syntax would you use? I'm confused. [21:04:23] also, did you try to do this with a single query? [21:04:33] that's probably not going to work well [21:04:59] https://github.com/alaefin/wikidata-redirects-conflicts-reports/blob/dea669e48b20f1a0c32e2cf88fbdc8a4aa8be5f0/sprintf.sql [21:05:07] (the fact that items_per_site has spaces in the titles is rather... ugh) [21:05:15] technically it's merely cross-db [21:05:37] but you can technically use server.db.table with mysql [21:05:44] it's just awfully slow [21:06:07] (been there, done that, for my job, a few years back) [21:06:23] heh [21:06:27] didn't know that [21:06:36] https://github.com/alaefin/wikidata-redirects-conflicts-reports/blob/dea669e48b20f1a0c32e2cf88fbdc8a4aa8be5f0/sprintf.sql is the old query. %1$s would be replaced with "arwiki" [21:06:40] (for instance) [21:07:24] current query is https://github.com/alaefin/wikidata-redirects-conflicts-reports/blob/69e968543cf3b571c15f0afd4cc8ec6b0a4c3712/wikidata-redirects-conflicts-reports.sql , passed to every *wiki_p [21:07:27] hm... one problem is that you don't have namespace IDs. [21:07:43] you just assume all sitelinks go to the main namespace, right? [21:07:48] no i don't [21:07:49] not all do... [21:08:11] i do however not care with the current query :p it only takes 15 minutes to do all wikipedias ^^ [21:08:42] (not each, all. really. all. all of them. the entirety of them. all together. 15 minutes in total) [21:09:41] there's the issue of non-X, non-T aspects, and there's the issue of "once used", "could be used" which i absolutely don't understand ^^' [21:09:58] oh, that sounds pretty good! what what the one that would take 7 days, then? [21:10:17] but i'd say the tradeoff being orders of magnitude is sufficient to prevent me from being too sad :p [21:10:39] i just rechecked, with the current query arwiki takes 7 seconds. compare with 70 minutes. [21:10:57] (arwiki was the biggest wiki i tried with the old query) [21:11:12] old query took quadratic time on pagecount [21:11:22] new query, i didn't even bother to check [21:12:44] i'd probably build a table with all the titles of redirects with the namespace expanded and "_" replaced by " ", and then use that to compare to ips. you could use a temp table on the wikidatawiki db, or at least build the table on the same host. [21:13:07] Not sure how long that will take for enwiki though... maybe it's slower. [21:13:15] but you'd get the namespaces right [21:17:47] i do get the namespaces. by id, granted, but still ! :p [21:18:36] and i don't really dare crate a temporary table [21:18:40] sjoerddebruin: I also sent a MemoServ :) [21:19:24] Josve05a: ugh, those are pretty hidden [21:19:51] okay, okay... [21:39:26] Alphos: Database queries are fun! https://commons.wikimedia.org/wiki/User:Multichill/Zandbak only took 4 seconds to generate last time I ran it ;-) [21:41:11] And doing some first steps on Commons to reuse some more data from Wikidata: https://commons.wikimedia.org/wiki/Category:Artworks_with_accession_number_from_Wikidata [21:47:31] multichill of course they're fun :) [21:57:20] matanya: YAY! [21:58:08] sjoerddebruin: I'm in an editwar with succo. https://www.wikidata.org/w/index.php?title=Q19088&action=history time to block me? [21:58:40] succu* [21:58:45] I'm not going to involve with taxon wars... [22:03:32] https://www.wikidata.org/wiki/Talk:Q19088#iNaturalist :/ ugh [22:12:22] not so much a taxon war, more of a "we should / should not have IDs which do not link to anything, even if existing" [22:29:34] * DanielK_WMDE just learned that an Entity (Q35120) consists of Former Entities (Q15893266), which is a subclass of Entity (Q35120) and of Nothing (Q154242)! [22:29:53] ...and Q154242 is used as the concept, but defined to be the noun "nothing". ugh. [22:29:55] ontology. [23:31:22] DanielK_WMDE: I believe at one time, Wikidata internal stuff was an instance of Wikidata internal stuff [23:31:35] xD [23:31:48] Which is a perfectly valid statement [23:32:35] Incidentally, any admins who want to process https://www.wikidata.org/wiki/Wikidata:Property_proposal/Authority_control#OpenCitations_bibliographic_resource_ID