[14:10:42] Happy Monday, Science people [14:59:44] 'morning [16:02:34] morning scientists :) [16:12:54] o/ Ironholds [16:13:04] yo halfak! How goes? [16:13:19] Not bad. Yourself? [16:17:13] o/ everyone [16:18:20] hey Nettrom :) [16:19:15] hey Nettrom [16:40:56] * guillom pokes Ironholds. [16:44:36] guillom, poked [16:44:41] wb DarTar ! [16:44:52] hi there [16:46:45] Hey well-rested DarTar [16:46:59] bonjour [16:47:21] Ironholds: I'm planning to edit the Meta page soon-ish; just wanted to check you weren't planning to do the same. [16:47:50] (mostly the table) [16:47:50] guillom, nope, spending the morning finishing off pageviews :) [16:47:59] Ironholds: Alright :) [16:48:00] (and probably dealing with endless "have you finished the pageviews?" questions) [16:48:06] guillom – I came across a thread on Wikipedia and SEO you’ll very much enjoy, sending it once I get through my backlog [16:48:22] DarTar: Thanks! [16:55:39] DarTar: BTW we created https://phabricator.wikimedia.org/project/profile/1163/ that you might want to join. You're also encouraged to add issues once you're all caught up. [16:56:20] guillom: done, thx [16:58:20] DarTar! Hey dude! [16:58:27] Welcome back! [16:58:36] halfak! thanks, sir [16:59:24] DarTar, so, quick update on what happened while you were away [16:59:28] it feels like I’ve been away an awfully long amount of time, glad to see my desk is still there tho [16:59:46] halfak and I decided that reid's proposal was taking too long and we should just release the geodata, so that's set up [16:59:53] DarTar, is my cardboard doppleganger still watching over it? [16:59:58] legal decided we're now not allowed to delete the pageviews logs [17:00:13] halfak: still there looking at me from across the room [17:00:20] :D [17:00:20] and ottomata eloped with one of the new blade servers [17:00:29] so everything is pretty much normal for a monday. [17:00:40] (one of the things I just said is actually true. Guess which one!) [17:00:42] Ironholds: leila just told me that reid responded the day after I left, great timing :) [17:00:49] hmph ;p [17:01:08] Ironholds, in the call in 2 minutes. Must release new version of mwcites [17:01:13] np! [17:01:18] I'm writing queries [17:01:22] halfak: oh cool [17:03:54] but seriously we probably can't delete the sampled logs [17:04:02] Now I support, isbn, arxiv, doi & pubmed. [17:04:03] https://github.com/halfak/Extract-scholarly-article-citations-from-Wikipedia/releases/tag/v0.2.0 [17:04:23] Ironholds: is there a thread where I can read more about this? [17:04:46] DarTar, I'm working it through with Michelle et al and have asked them to send out a thing when they reach a firm decision [17:04:47] (I’ll get there eventually, my inbox excluding mailing lists is only 1.6K long) [17:05:08] halfak: fantastic [17:05:36] Ironholds: I imagine kevin is looped in? I’m syncing up with him at 11 [17:06:03] DarTar, nope! Hence the need for a thing when they reach a firm decision [17:06:44] hm ok, let’s touch base on this maybe later today so you can give me a bit of context? [17:07:06] sure [17:15:52] Fun fact: On Wikidata, "The Hobbit (film series)" both "follows" and is "followed by" "The Lord of the Rings trilogy". [17:16:20] Yay loops. [17:17:37] guillom, look at the categories in Category:Hidden categories on enwiki [17:19:09] Aaaaah. [17:19:43] Thas is so wrong. [17:19:47] That* [17:19:53] hahaha [19:54:34] halfak: Do you know if there is a production table with Interlanguage links? [19:55:15] ewulczyn, I think you can gather then from the pagelinks table. [19:55:32] The whole outgoing pagetitle is captured and that include language prefixes in the case of external links. [19:55:51] There's also a langlinks table I have never used. [19:56:52] It looks like there are 20.2 million rows in the langlinks table. [19:57:16] halfak, the NORMAL DISTRIBUTION IS SO COOL [19:57:19] ...this is unrelated [19:57:39] the CLT means I may have to replace log-normal as my favourite. Sorry, log-normal. [19:58:14] ewulczyn, I take it back. It looks like inter-language links do *not* appear in the pagelinks table and that langlinks is what you are looking for. [19:58:49] Ironholds, is that because you find log-normals everywhere and it's trivial to normalize? [19:59:13] halfak: Is langlinks a producton table or a one-off? [19:59:20] ewulczyn, production [19:59:28] did you look at the schema? [19:59:35] "SHOW TABLES" :P [20:00:20] halfak, no, it's because the Central Limit Theorem is mind-blowingly cool and makes me believe there might well be a god [20:00:23] and that he likes statistics [20:01:24] you know what is mindblowingly cool? relative simultaneity! still blowing my mind man. [20:01:25] Ironholds, :P [20:01:43] Or that "normal noise" is a simple mathematical property of randomness [20:02:03] I think you blew otto's mind [20:02:05] oh, he's back [20:02:14] haha cmd-w on wrong app :) [20:02:16] ottomata, yeah! Relativity is a mindblow [20:03:34] ja, i mean, most of it is, and I have settled from having my mind blown by it for a long time, but the consequences for 'now' is a mindblow that I am still realizing [20:07:06] halfak: So everythink in the enwiki db is a production table? There are some tables in there that looked to me like they are from single experiemnts. I checked langlinks before asking. It has some records with empty ll_titles, which made me suspect that it is not the canonical table. [20:07:44] ewulczyn, what version of "production" are you using. [20:07:57] All of those tables are in a "production" database on a "production server". [20:08:11] Should you trust the data in any of the tables ever? Never. [20:08:38] halfak: Thanks for the advice :) [23:07:20] DarTar, you wanted to chat today? [23:07:43] hey, yeah, if it’s not too late – in 10 mins ok? [23:07:52] sure, just gimme a hangout and I'll be there :) [23:07:52] tomorrow otherwise? [23:07:55] cool [23:09:49] Ironholds: sent, I’ll be there at 20 past at the latest [23:09:56] kk