[14:33:36] bmansurov ciao :) [14:34:15] bmansurov: I have a question, do you have any idea why some section_id in the citation usage schema might be Null outside the infobox? [14:36:08] bmansurov: I can't seem to understand a pattern there, sometimes even different clicks to the same reference have have either "null" or "References" as section_id interchangeably [15:49:02] miriam_: sorry, in the middle of something else, I'll ping you once I'm freer. ;) [15:49:50] miriam_: I need to check this, but the lead section may have the null ID. [15:50:10] (that's the first section before the table of contents) [16:01:35] miriam_: that's correct, see the documentation for section_id: https://meta.wikimedia.org/wiki/Schema:CitationUsage [16:02:10] (I've also checked the code.) [16:03:28] bmansurov [16:03:30] thanks :) [16:06:06] bmansurov: sometimes references with NULL section_id are actually in sections other than main, could you get NULL if you click a reference from the hovering window ? [16:11:01] miriam_: let's debug this later today. Can you send an invite? [16:11:11] bmansurov [16:11:24] bmansurov: ehm I am busy till 7PM [16:11:43] miriam_: ok, let's do it async [16:11:54] i'll reply here betwee meetings [16:12:21] miriam_: can you give example data so that I can check the page? [16:12:34] bmansurov: let me see [16:12:42] bmansurov: can we do after the research showcase [16:14:07] miriam_: ok [16:20:48] bmansurov cool! [16:57:40] bmansurov: for the new editor interests study, are we doing any filtering of the sample users? [16:58:12] you know what I should just go read the meta page before I bother you. question withdrawn for now [16:59:04] question answered :) [17:00:09] okay new question lzia, bmansurov: in the 26k sample, we're filtering out blocked newcomers, right? [17:10:43] j-mo: I think so, the script that sends out emails checks various things like whether the user allows receiving emails, etc. [17:10:55] j-mo: I suppose if the user is blocked, they won't be able to receive emails. [17:11:02] bmansurov: so with all your recent changes in the recommendation-api, do you have more changes in mind or we should deploy? [17:11:32] Pchelolo: hi, can I get back to you on this. I need to check a couple of things. [17:11:41] ok, no rush [17:11:50] Pchelolo: thanks, I'll ping you later today. [17:11:56] kk [18:08:40] * leila reads [18:09:55] * leila will go heads down on finalizing her research showcase slides. [18:50:40] * leila welcomes Ramtin as she continues to work on research showcase presentation. [19:19:08] j-mo: thanks for updating the title and all [19:22:19] no problem leila. everyone: we're going to start the research showcase in about 10 minutes! see the chan topic for relevant links. If you have questions during the presentation, make sure to mention me and I'll capture your question and ask it to the speaker during the Q&A. [19:33:25] o/ [19:33:32] o/ halfak [19:34:15] the research showcase has begun! YouTube stream should start rolling any second now [19:37:15] hey all [19:37:29] \o [19:38:04] courtesy YouTube stream link: https://www.youtube.com/watch?v=RKMFvi_CCB0 [19:39:03] o/ [19:52:06] Nettrom: o/ [19:55:47] except for the Dutch :) [20:07:21] Were article lengths normalized by the article length distribution for each wiki? [20:07:37] There are more long articles about popular topics in bigger wikis than in smaller ones. [20:07:54] Pchelolo: let's not deploy today. I have to fix some things and I won't be able to fix them today. I'll ping you tomorrow when I'm done. [20:08:06] OMG geometric means, forgot about that. Someone’s ears must be burning [20:08:14] cool, no rush from my side [20:08:22] Pchelolo: ok, thanks [20:11:47] j-mo: if you’re holding the queue across channels (there’s a Q from halfak above), can I add myself to the queue? [20:13:40] DarTar, lol [20:13:59] Also yeah, I think I'd like to ask that question :) [20:15:11] DarTar, halfak: got it [20:16:15] I also have some questions: is this survey now a yearly event? if so, will the languages stay the same, or are there plans to change it and could you discuss why it would change or not? [20:18:34] j-mo: there is a question on the youtube channel :) [20:18:58] going there miriam_ [20:21:00] leila: cool, thanks for answering. that :) [20:23:22] Nettrom: sure. :) [20:25:51] this conversation reminds me of the "why is there no content about country music outside of enwiki?" question that B. Hecht brought up, and subsequently whether multilingual readers read different topics in different languages [20:26:04] *brings up in some of his talks [20:26:44] IIRC reading traffic from Norway is about 50/50 split between nowiki and enwiki, but we don't know if users read different things [20:27:30] and the competitor encyclopedia (SNL) focuses on content in Norwegian that isn't what Wikipedia has (e.g. they won't focus on pop culture) [20:28:36] a followup question to leila's remarks on the product/interface implications of these results: does the team happen to have concrete suggestions/ideas for products or changes that could be informed by what we learned from this work? [20:28:50] My prior is that the reading patterns ARE different between no and en, Nettrom. [20:29:19] what % of Norwegians have English as a second language? What are the demographics associated with ESL? [20:30:10] j-mo: captured HaeB’s Q above? [20:30:19] * DarTar checks YT [20:31:13] HaeB thanks for asking the "implications for design" question ;) [20:34:06] thanks leila ! [20:34:12] sure, HaeB [20:34:37] j-mo: I would suspect the ESL % in Norway is high (in 2012 it was around 85–86% in Sweden and Denmark), but English proficiency is also high (top 10 on EF's proficiency test) [20:37:21] Nettrom yeah, then I think we must assume that a whole lot of people are going to nowiki vs. enwiki for particular reasons, because they have a choice. I bet content coverage plays a role. [20:39:05] j-mo: agreed! I also remember the quick analysis done for the 2012 paper on similarity and differences in wikis, were we found that nowiki's unique articles are largely about culture [20:39:22] those are articles that aren't notable enough to exist outside nowiki [20:40:09] thanks for hosting and for the great presentation! :) [20:40:40] thanks everyone! great talk IMO [22:25:13] lzia: the result that 20% of sessions involve language switching is interesting. i just noticed that the language team appears to have arrived at much lower numbers previously: https://www.mediawiki.org/wiki/Universal_Language_Selector/Compact_Language_Links/metrics/data [22:25:23] ... "In June 2016 0.2125% of Wikipedia visitors in all languages clicked on interlanguage links. In June 2017 it was 0.4042%." [22:25:54] it's not quite clear to me though what definition of "visitors" was used there, would need a closer look [22:41:33] HaeB: that's super low compared to the results diego saw. [22:41:40] * lzia looks up the documentation place [22:42:12] HaeB: I wonder if people switch in other ways than clicking on interlanguage links (which would be fascinating to learn about) [22:43:11] HaeB: I have to wait for Diego to come back and ask him. made a note to talk in our next 1:1 [22:43:28] perhaps the results on the ULS page are about desktop only (still, wouldn't account for the difference) [22:43:45] HaeB: yeah. the difference is huge. [22:43:52] HaeB: we will let you know once we know more. [22:46:42] lzia: cool, thanks ! (btw, perhaps it's worth pinging amir directly about the interpretation of these numbers.. don't see him on IRC right now though) [22:47:54] HaeB: makes sense. thanks. [22:48:59] bmansurov, Ramtin: thank you for around the clock coverage of https://meta.wikimedia.org/wiki/Research:Voice_and_exit_in_a_voluntary_work_environment/Elicit_new_editor_interests#Second_Online_Experiment . I'm excitedly waiting to hear more about the responses. [22:50:20] bmansurov: re https://meta.wikimedia.org/wiki/Research:Voice_and_exit_in_a_voluntary_work_environment/Elicit_new_editor_interests#Stats : are those who didn't receive emails those who have Email_this_user feature off?