[00:31:37] SMalyshev: Here? [00:31:47] yes [00:32:06] What do you think about Markus' remark about not using claim as the prop name in RDF? [00:32:15] "One more thing: it should be called "statements" instead of "claims". This will also better match the terminology used elsewhere in RDF (I don't think the notion of "claims" [=statements without the references] occurs in RDF at all)." [00:33:08] can be changed, it's just config. Probably even better since the entity itself is called wikibase:Statement [00:34:36] ah, it's already statements now [00:34:36] just not in commit message [00:35:01] ah cool [00:35:12] in that case, I think it's ready to merge [00:35:34] Given that Markus is ok with the output and the code looks good to me now [00:39:13] ok, so someone needs to just +2 it then I guess :) [00:45:54] Removed a miss placed tab [00:46:03] I guess it's fine now… tested it manually [00:46:59] I wonder how it's going to affect the run time of the dumpers… I guess (and hope) it's not to bad [00:58:56] SMalyshev: Little follow up to add docs https://gerrit.wikimedia.org/r/307666 [00:59:20] ah, forgot about that, thanks! [01:02:06] SMalyshev: How long will it take you to include this in production? [01:02:38] I guess we wont be able to get this deployed before the September 12 dump run [01:02:55] hoo|busy: well, that depends on deployment schedules [01:03:13] I imagine couple of weeks at least... thats's what it usually takes [01:04:52] Ok [01:05:02] So I put it in the weekly summary and said "soon" [01:05:30] I hope we can get it deployed next week at least [01:06:40] yeah hopefully :) [01:10:05] Calling it a day now [03:47:32] * hare wonders if anyone is still awake [04:10:40] \o/ [04:17:01] http://tinyurl.com/gr2vm36 << i'm trying to filter out use of P2860 that don't have a reference of P248: Q229883 (stated in pubmed central). But I don't think it's working? [04:17:01] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:17:02] Basically, I want a list of all PMCIDs, plus all statements of item-x-cites-item-y. But I don't want to include such statements *unless* the reference is "stated in PubMed Central" [04:18:18] hmm can you get me the query ? [04:18:53] http://hastebin.com/raw/acogataroq [04:19:39] thx i look at it [04:20:43] One idea I had was to split it up into two separate queries but I'd rather not do that if I can avoid it. [04:21:37] But I think even if I did that, the underlying logic isn't working. [04:21:48] So it would not really make a difference. [04:22:21] hmm [04:23:23] what do you want if a P2860 has 2 refs : one P248: Q229883 and another [04:23:23] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:23:58] I don't know how many P2860 statements have two refs, but I say it should be included in the result set anyway [04:23:59] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:24:03] * hare pets stashbot [04:24:12] :) [04:25:00] I have a script that fetches the citation graph from PubMed Central. But I don't want it to tell me to generate a statement in Quick Statements if the statement already exists *and* is attributed to PubMed Central. But *if* it exists but *isn't* attributed to PubMed Central, then PubMed Central can be added as a reference. [04:25:37] well then I don't think you need to bother with filter [04:27:56] What strategy do you recommend instead? [04:28:54] just a sec, i try something [04:32:53] damn i really should bookmark useful pages ^^ [04:35:14] I tried http://hastebin.com/raw/ucovosiwum but it only returns unique items, not individual statements [04:37:59] i have been checking the usage of wasDerivedFrom [04:38:09] seems you need to do it on statement [04:38:25] p:P2860 [04:38:25] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:38:35] instead of wdt: [04:40:16] So what would the full query be? [04:40:33] http://tinyurl.com/zrss2v5 [04:40:42] i think [04:41:16] that still only returns unique items, not individual statements [04:41:27] hmm [04:42:53] seems like it returns especially the items with no cite claim [04:43:14] (that's the meaning of OPTIONAL) [04:43:47] But it also does return items with cite claims, but just doesn't include them in the results [04:44:23] hmm ok i see [04:45:33] ok first, you want items with no cite claims ? [04:46:33] Yes [04:46:49] ok so we keep optional [04:47:46] Might I be better off having two separate queries: (a) all items with PMCID [that's easy] and (b) all cite claims with "stated in PubMed Central" as a reference ? [04:49:00] sometimes it's quicker yes [04:50:15] i wrote something but .. timeout ^^ [04:53:28] http://tinyurl.com/z9sqx4m << for what it's worth this returns 0 results [04:54:24] when it should return... at least 370,000, if it did what we wanted it to do [04:55:36] hmm you want all empty P2860 statements + all P2860 statements that don't have any P248/Q229883 [04:55:37] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:56:24] for that query, i want the opposite: all P2860 statements that have P248/Q229883 as a reference [04:56:25] P2860 cu_changes deletion on beta - https://phabricator.wikimedia.org/P2860 [04:56:42] ahh sorry [04:56:53] i get confused at this time of the night [05:01:01] i'll try with a subquery [05:01:20] and if not then i'll try with a union :) [05:02:02] A union might make sense [05:02:15] could be easier right [05:02:28] { [05:02:29] A [05:02:29] } UNION { [05:02:29] B [05:02:31] } [05:02:42] oops sorry fucking c/p [05:02:58] * Cameron11598 beware of Sigyn I've been accidentally klined before due to that... [05:12:09] ahhh [05:12:21] { [05:12:21] A [05:12:21] } UNION { [05:12:21] B [05:12:25] }fuck [05:13:06] can i say i hate firefox ? :D [05:13:11] melderick: Sigyn is really gonna love you lol [05:13:31] Cameron11598: why ? [05:13:53] Sigyn: Has klined people for copy paste accidents before [05:14:00] <-- Victim of such an incident [05:14:04] oh [05:14:34] i could understand if it's more than a page of text [05:14:49] I got it for 8 lines [05:15:03] that's bad [05:15:04] "Cameron11598 was killed by Sigyn )Spam is off topic in free node [05:15:20] And I was like o F* [05:15:32] no wonder :) [05:15:53] a simple kick should be enough [05:16:06] Sigyn: is one of Freenodes "services" [05:16:28] ah [05:17:42] hare: you sure you want ?ref ? [05:17:54] melderick: I don't need it; I included it for debug purposes [05:17:54] because it looks like wdref:94cecb4679dfd59834590547c974afc78d255645 [05:18:02] ok [05:19:03] hare: your error is ?ref pr:P248 wd:Qxxxx [05:19:10] pr: [05:20:12] So what would the correct query look like? [05:20:34] working on it [05:24:55] i get 402100 [05:25:28] That sounds right I think [05:25:48] i run it with an order by [05:26:06] http://tinyurl.com/j4tvsbh [05:28:08] and this gets me all PMCIDs and all cite statements if they are attributed to pubmed central [05:28:57] yes [05:29:16] excellent, thank you [05:29:19] have a good night ! [05:29:27] thx :) [05:29:35] you too if you're going to bed [05:31:27] http://tinyurl.com/go8nodv [05:31:45] this one slightly different and get a few more results [05:31:53] but i don't know why ^^ [05:32:00] 404814 [07:26:44] sjoerddebruin: You are awesome. I just wanted to say that. Thanks for always being around and always being nice and helpful. :-) [07:27:13] Ah, what a lovely way to start the day. Thanks, you too! [08:22:28] wikidata is down? [08:22:59] nope [08:23:03] works fine here [08:24:58] ok. works again here [08:25:29] someone in the Wikimedia Indonesia office reported it was down, I checked and also saw a "maintenance" error message [08:29:44] jayvdb: well then it was only a few minutes? [08:29:55] no idea, I didn't see it [08:30:45] yea, she came to me immediately, and had been using it previously, so only a few minutes [09:17:42] LeaAuregann_WMDE: would it be a good idea to attend QTcon to try to meet Free Software people interested in wikidata ? It seems a bit out of topic but ... [09:19:31] dachary: yes, good idea! there will be a presentation of Wikidata during QTcon. Also, we have free tickets for the ViewSource event in Berlin in september, if you want one! [09:21:32] LeaAuregann_WMDE: ouch, https://qtcon.org/about : I'm afraid it's too expensive. ViewSource sounds like a more affordable alternative :-) [09:22:59] dachary: I send you a mail about ViewSource :) [09:23:07] thanks ! [10:48:52] aude :) [10:49:54] You here? [10:50:18] hi [10:50:29] only for maybe 20 min [10:50:37] I'm just curious about when we deploy next [10:50:46] next week? [10:51:16] That sounds good to me… sadly I will be busy with exams, so I can only work a bit every day [10:51:38] :( [10:51:38] i can only minimally monitor irc though [10:52:01] Tuesday and Wednesday I can probably at least monitor IRC [10:52:05] but things should be fine [10:52:05] Thursday not sure [10:52:09] yeah :) [10:52:17] We even have more reliable browser tests now [10:52:43] i want us to get to a point that we don't have to as actively be on irc at deploy time [10:52:54] :) [10:53:00] I would also like weekly deploys [10:53:08] bi-weekly-ish is a mess [10:53:13] (e.g. we test more with test.wikidata) [10:53:20] yeah [10:53:40] Greg also wished for that… and I think we're getting there [10:53:45] yeah [10:53:59] sadly browser tests are still flaky, but I guess we can get that done soonish [10:54:12] :/ [11:03:45] https://integration.wikimedia.org/ci/view/Selenium/job/selenium-Wikidata/ :( [11:04:45] Once they are reliable, we can switch on email notifications in case of failures [11:05:03] so that relevant people can get notifications if things on test.wikidata seems awry post-deploy [11:05:10] * seem [11:06:17] :/ [11:06:44] i usually check the tests and see if the failures are legit or not [11:07:35] hi all, my UreomiczBot has bot rights now, but when using the API, I still get the commit fail because of suspect URL spamming... the only URLs I have are in 'reference URL' and in the calendarModel for the date of 'retrieved'... only if I remove both, I can make a commit... that doesn't sound right to me... suggestions? [11:08:01] Will check the abusefilter now for the requirements [11:08:11] sjoerddebruin: very much appreciated! [11:08:33] aude: That works, but makes the whole thing manual again [11:08:41] sjoerddebruin: I can put the JSON the bot is sending the API in a gist or so, if you prefer [11:08:43] no automated reports, but manually checking the results [11:09:20] egonw: please [11:10:12] sjoerddebruin: https://gist.github.com/egonw/7f262bcc8dfee923f72e1d690e9b205b [11:10:23] :/ [11:10:32] sjoerddebruin: line 26 is one of the calendarmodel URLs [11:10:46] editing monolingual text is a little bit strange [11:10:46] sjoerddebruin: line 34 an example of the 'reference URL' [11:11:32] sjoerddebruin: when I leave out both these statements from the "Reference", then it works [11:11:51] Does it give you the abusefilter number or something? [11:12:48] sjoerddebruin: see the 2nd file I just added to the same gist [11:12:59] "URLs within label/description/alias" [11:13:22] right [11:13:35] that's the message basically [11:13:57] lin 251-253, however, look OK to me, and don't have URLs [11:14:28] so, the error message is not matching the test, it seems to me [11:14:45] The abusefilter isn't adjusted to Wikidata, it seems to check the whole JSON [11:15:04] so, how are other bots setting 'reference URL' then? [11:15:13] are those whitelisted from the filter? [11:15:28] are are none using this P for provenance? [11:15:31] or are none using this P for provenance? [11:15:50] I really don't know, you should message Matěj Suchánek [11:16:14] ok, thanks! [11:16:33] sjoerddebruin: Looking at the filter… what the hell [11:16:41] ("wbeditentity-create" in summary [11:16:41] & added_lines irlike "\b(https?://|www\.)\S{10}" [11:16:41] &! 'WE-Framework gadget' in summary [11:16:41] &! new_size < 500) [11:17:00] Yeah, it doesn't seem pretty good to me. [11:17:03] So basically item creations with URLs in them?! [11:17:23] It's only a warning, though… if you post the edit again, it will succeed [11:17:40] That's what I did in my bot sometimes to bypass bogus warning filters [11:17:44] https://www.wikidata.org/w/index.php?title=Special:AbuseLog&wpSearchFilter=34 [11:18:06] It seems to block pasting Wikipedia url's as descriptions. [11:18:46] oh, interesting page! [11:18:50] I also see some legit stuff, but maybe bots should be excluded. [11:20:53] Most bots are probably autoconfirmed [11:21:28] I'm not a experienced abusefilter reader. :P [11:21:45] It should have a display like WDQS, where it describes the filter in human language. [11:22:25] Using AbuseFilter on Wikidata content is inherently very scary [11:23:52] hoo: does your bot do the post twice (so, within the same login session)? bc rerunning my bot as a whole gives the same error... [11:24:22] egonw: Yes, it just retries the edit [11:24:32] exact same form data being sent [11:25:03] ok, will discuss with sebotic if his PBB bot lib can do that... [11:56:38] Hello. Let's say, I want to add a few statements the concept of loan ( https://www.wikidata.org/wiki/Q189539 ). Loans usually have ( Issue date, Listing date on some market, Loan percent rate, Loan term, Loan originator (company, individual), Loan collateral, Loan amount, etc.). These are quite general properties, available to all loans. How could I include such basic knowledge into the page about Q1895 [11:56:39] 39 ? [12:01:48] Mindey: we currently don't model schema infromation like this (that is, a class will not tell you what properties an instance and or should have). Wikidata's data model is not class-based, but poroperty based. [12:02:20] A property can have information about on instances of which class it would typically be used, or should even be expected. [12:02:32] DanielK_WMDE: Mindey some WikiProject do have "onotologies" pages [12:02:45] which explain what property a class should have [12:02:49] However, since I don't think we have individual loans as items, we probably also don't have properties for describing individual loans-. [12:03:10] harmonia: instances of the class, yea. do we have many instances of the class "loan"? [12:03:22] DanielK_WMDE: yah I don't think so [12:03:29] I don't find any actually [12:03:39] DanielK_WMDE, I see. But wouldn't "has property", or "often has property" properties would accomodate definition of schema information? [12:04:08] Mindey: as we said, we have help pages for that [12:04:19] but we don't have many instances of loans [12:04:38] so we have no need for this specific page [12:04:43] Well, you have instances of pictures, drawings of famouse painters. [12:04:54] yes [12:05:30] Mindey: an example: https://www.wikidata.org/wiki/Wikidata:WikiProject_Theatre/Properties [12:06:14] Wouldn't we benefit from having anothe category altogether, e.g., we right now have Q - concept, P - Property, and I - Instance? [12:06:27] Mindey: we have P1552 (has quality) which seems to be used like this [12:06:27] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:06:32] It doesn't look appropriate to add instances to WikiData, as Q. [12:06:39] Q is like a concept, which could have its instances. [12:06:54] Mindey: compare https://www.wikidata.org/wiki/Q4130#P1552 [12:06:55] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:06:55] Mindey: uh no [12:07:28] We could have a concept of "Famouse drawing", and instances would be a different thing.l [12:07:40] Mindey: all Q-items are instances (classes are also instances - of the class "class") [12:07:48] Hmmm... [12:08:09] Even things like https://www.wikidata.org/wiki/Q318 ? [12:08:11] (galaxy) [12:08:14] Mindey: but a great many Q-items are "proper" instances. People and places. [12:08:56] Mindey: sure, Galaxy is a class, right? "is a" means "instance of". [12:09:05] So, maybe "proper" instances (which are the proper nouns), should be identified by something different, e.g., "I"? [12:09:21] Mindey: why? I see no advantage [12:09:41] Also, it's not always clear what is an instance or not. [12:09:53] DanielK_WMDE, this would allow for clearer separation of "proper" and non-"proper" instances. [12:09:54] One interesting debate was whether Ethanol is an instance or a class. [12:10:05] Ethanol is a class, obviously. [12:10:27] Mindey: but that separation isn't clear, and i sdee no use in enforcing it. [12:10:33] A molecule of ethanol, which I have under my microscope, and have asssigned an ID in my database, would be an instance :) [12:10:55] I do see use in associating typical or defining properties with a class. P1552 seems liek a good way to do that [12:10:55] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:11:01] we had the problem for the colour, too [12:11:06] instances or classes ? [12:11:15] Very clear use: let's say we want to index (or refer to) all instances in any databases in the world. [12:11:18] it's not so clear cut most of the time [12:11:29] Mindey: well, we use queries [12:11:44] Mindey: Ethanol being a class may be obvious to you, but not to many other people. There was great debate about it, and when you look at similar cases, they are often treated as instances. [12:12:18] Mindey: what does "all instances" mean, and why should it exclude Ethanol? [12:13:19] Mindey: is "New Yourk City" an instance? It changes over time, so isn't it really a class, made up of all the instances-in-time? [12:13:30] Ontology is a tricky thing :P [12:13:49] By instances, I would mean actual things of the world, which could be owned by people, or simply exists as physical things in outer world, outer space, etc., identifiable by WCS/CCS coordinates, things that can have serial numbers. [12:14:15] Mindey: oh ? and how do you treat fictional entities? [12:14:26] .oO(Stachbot should not link P-ids to Phabricator) [12:14:29] Even things like products (product brands) are not real things, but classes of things. Only specific items of them are real things. [12:15:03] I my experience, the more you think of the class/item distinction, the less useful it becomes [12:15:12] the useful distinction is ditctated by the use case [12:15:26] https://www.wikidata.org/wiki/Q4130#P1552it's better to model it based on properties, instead of a categorical distinction [12:15:43] (and even if I thought otherwise, it's weay to late to fix that now) [12:16:21] harmonia, fictional entities... I think they are hypothetical instances. So, not items. Very interesting. [12:16:40] oops, sorry for the random paste. i meant to say "it's better to model it based on properties, instead of a categorical distinction" [12:16:54] Mindey: urgh [12:17:30] Mindey: So, is "Merlin" a single hypothetical instance, or a class? There are many quite different Merlin characters... [12:17:33] Mindey: well, I don't see what we would gain by implementing your idea [12:17:42] as i said, the distinction becomes blurry and pointless when you look closely [12:18:39] DanielK_WMDE, I like the modelling based on properties. It's definitely very flexible. We can add new P-items as necessary, and ennumerate the "claims" about the Q items. [12:18:57] that's exatcly the idea [12:19:06] This way, it becomes quite universal to model anything -- both the concrete instances, and abstract instances. [12:19:08] DanielK_WMDE, you can do both. Class based modelling is also able to deal with a lot of usecases [12:19:30] TomT0m: sure, that's why we have P31 [12:19:30] P31 Fork of P29 (An Untitled Masterwork) - https://phabricator.wikimedia.org/P31 [12:19:37] oh shut up stashbot [12:19:40] :) [12:19:47] anyway, any query can define a class [12:20:08] yea, and that's useful, in the context of the query [12:20:19] the point is that a *categorical* distinction is not useful. [12:20:32] what's a categorical distinction ? [12:20:57] So, going back to the question about the "Loan" concept (Q189539), and the ability to include information about the database properties (e.g., P "databases of loans usually have the following fields to store information about this kind of items"). [12:21:03] Can we have such a P ? [12:21:11] TomT0m: having a separate namespace and prefix. like Q vs P. [12:21:15] TomT0m: Mindey wanted to distinguish "instances" and "classes" by Q and I identifiers [12:21:19] (not loans specifically,..) [12:21:27] Mindey: we do, it's P1552 [12:21:27] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:21:29] ah, ok [12:21:33] no, not that, stashbot! [12:21:39] Mindey: I don't understand the property you want? [12:21:56] I have stashbot ignored, makes it much less annoying :P [12:22:14] harmonia: "instances of this class are characterized by the following property(s)" [12:22:41] I love how stashbot interventions are nearly always surreal [12:22:41] harmonia, in the sentence about "databse properties", I meant, database table field names, when storing items defined by specific Q. [12:23:06] What's the stashbot? D) [12:23:07] well, our data model isn't based on tables or columns... [12:23:20] oh, that would be https://www.wikidata.org/wiki/Property:P1552 [12:23:20] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:23:27] TomT0m: indeed [12:23:30] XD [12:23:39] Mindey: see for example https://www.wikidata.org/wiki/Q4130#P1552 - is that what you mean? [12:23:39] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:23:42] Masterwork From Distant Lands! [12:23:49] orrrrr. [12:23:59] it looks like a fantasy novel title :D [12:24:09] I tried models like "human has quality ..." qualified with date of birth [12:24:40] Mindey: stashbot links to pharbicator tasks, like T44783. Sadly, it thinks that IDs starting with P are phabricator paste pages, not wikidata properties... [12:24:40] T44783: Do not rerender whole client page when only sitelinks are updated - https://phabricator.wikimedia.org/T44783 [12:24:41] I thought I removed stashbot here... [12:24:51] there is also a property "property for this type" [12:24:55] sjoerddebruin: it's useful for tickets, but really annoying for P-ids [12:24:57] Mindey: as i said, generally we list the relevant properties on the WikiProject page [12:25:07] DanielK_WMDE: can it be adjusted then? [12:25:12] sjoerddebruin: i have no idea [12:25:17] DanielK_WMDE, I understand that. But in the real world, there are many databases that do. If, let's say, I want to refer to a specific record in some specific database, a quite informative way, that I see, is to use two Qs, namely Instance(Q1="Loan", Q1="Bank of Lithuania", ID_IN_THE_BANK). [12:25:34] The same happens on Phabricator though, maybe pastes should have a different prefix [12:25:35] Instance(Q1="Loan", Q2="Bank of Lithuania", ID_IN_THE_BANK). [12:26:06] This way, we could index things existing in any database. [12:26:12] Mindey: such a database needs to be designed for the concrete use case. [12:26:31] mapping database fields to Q-ids or P-ids would be quite useful, I agree [12:26:33] E.g., Instance(Q1="Tweet", Q2="Twitter, Inc.", ID_IN_THE_INSTITUTION). [12:26:42] but that would have to be done by the designers of that database [12:27:40] Mindey: in fact, I would love to see DataHub use P-ids and Q-ids to define the meaning of records and fields. https://datahub.io/dataset [12:27:43] (or, maybe Instance(Q1="Web post", Q2="Twitter, Inc.", ID_IN_THE_INSTITUTION) [12:28:05] :) [12:28:07] Mindey: that's useful infor for a specific database. it's nothing we can record on wikidata. [12:28:16] DanielK_WMDE: hey, have you seen this? https://gerrit.wikimedia.org/r/#/c/306750/ [12:28:17] we can record typical (characteristic) properties [12:28:42] DanielK_WMDE, "we can record typical (characteristic) properties" --- Yes!! [12:28:50] That would be great :) [12:29:01] Typical characteristics. [12:29:14] Amir1: yes, but i haven't reviewed it yet. Last week and this week was filled with meetings and offsites and stuff, and now I'm a bit ill and working from home. [12:29:20] I'm in the process of going through my review backlog. [12:29:30] well, right now, i'm procrastinating on IRC :P [12:29:45] :D [12:29:51] ^__^ [12:29:52] I hope you feel better soon [12:30:01] Mindey: i told you the appropriate property, and pointed you to an example. does that work for you? [12:30:08] Amir1: thanks! [12:30:52] DanielK_WMDE, "has_quality" ( https://www.wikidata.org/wiki/Q4130#P1552 ?). So, I could use this one? [12:30:52] P1552 Masterwork From Distant Lands - https://phabricator.wikimedia.org/P1552 [12:31:01] I guess so :) Alright. [12:31:06] Mindey: yep [12:31:17] DanielK_WMDE, Great! [12:31:43] Thanks :) [12:31:54] Mindey: thanks for making me find that out. I dodn't know about thart property :) [12:32:06] XD [12:32:11] * DanielK_WMDE is not an active wikidata editor, he mostly works on the software [12:32:26] WikidataFacts you trollbait :p [12:32:32] when using PetScan to create items, how does one add a default description? [12:33:00] The create function of petscan is very limited, you can use QuickStatements though [12:33:02] WikidataFacts btw, same guy who wrote the description for that "ethnicity" {{dubious}} wrote this in itwiki https://it.wikipedia.org/w/index.php?title=Discussioni_utente:Kopiersperre&diff=prev&oldid=79565363 [12:33:02] 10[1] 04https://www.wikidata.org/wiki/Template:dubious [12:33:04] * Mindey actually, is just a guest, but we are using WikiData API, to give suggestions for https://infty.xyz users... And would like to use the Qs and Ps, to formalize definitions of goals for mankind :) [12:34:18] Mindey: awesome, that's exactly what wikidata is for! [12:34:46] DanielK_WMDE, Wow. So, I think, I'll stay here for longer. Actually, I find it quite much fun, in conceptualizing. [12:35:15] it is indeed :D [12:35:58] I have my own little theory of "everything" = ( substance ) X ( essence ) = ( Agents, Things, Places, Events, ... ) X ( Needs, Goals, Ideas, Plans, ... ) ( https://www.mindey.com/index.html ). [12:36:19] Always looking for ways to make sense of the world. [12:36:38] +1 [12:36:52] that could be Wikidata motto actually [12:36:59] LeaAuregann_WMDE: what do you think? [12:38:21] Alphos: it’s weird, that name sounds familiar, and I kind of had it mentally filed under “legit”, I think [12:38:25] but that looks terrible [12:38:41] it does, doesn't it ? [12:39:49] Amir1: I added a comment. Looks good, nearly there! [12:40:14] awesome [12:40:15] thanks [12:42:54] WikidataFacts_ : https://www.wikidata.org/w/index.php?title=Q22686&diff=314533991&oldid=314533454 [12:47:28] Alphos: nothing beats https://www.wikidata.org/w/index.php?title=Q21621117&diff=prev&oldid=357387609 though [12:48:11] sjoerddebruin: LOL [12:48:34] sjoerddebruin : seems legit ! [12:48:48] sjoerddebruin: O_O [12:48:55] that's [12:48:58] ... [12:49:17] * harmonia doesn't know what it is exactly, actually [12:49:40] we really should have a page with the more improbable edits [12:52:13] wtf? Q7048763 is "the real Whites from Europe"? Where do hispanics come from again?... [12:52:51] https://www.wikidata.org/w/index.php?title=Q7048763&type=revision&diff=349273480&oldid=331516192 [12:54:44] WikidataFacts_: you beat me to it! [12:54:58] DanielK_WMDE: feel free to improve it :) [12:55:03] DanielK_WMDE shush, the iberic peninsula is NOT part of europe ! #itisknown [12:55:17] WikidataFacts: naw, i was going write "ethnicity defined by the US Census Bureau". same thing. [12:55:28] Alphos: heh [12:55:29]