[00:20:42] halfak: I was looking up your R:I redesign and realized that Reid’s proposal showed up as “approved by the RCom”, not your fault [00:20:59] those support categories probably need an overhaul more urgently than anything else [00:21:13] Yikes. Good catch. [00:21:19] I marked it at pending for now, but it still says “pending RCom approval” [00:21:22] Did you make the change already? [00:21:33] I did [00:21:43] for his project, not for the general support categories [01:24:38] DarTar, bandaid applied. This is a deeper issue. It'll be good motivation to clean up the support categories. [01:25:05] thank you [16:01:38] g'morning lzia & nuria [16:01:49] hello! [16:02:08] mornin' halfak. hopping on the bike. see you in an hour or so. [16:02:36] Have a good ride! [16:02:38] * halfak is jelly [18:13:05] hey lzia, are you in the office today? [18:13:15] I might use some help with the showcase if you’re around [18:13:51] I’m heading off to Montgomery in a sec [19:31:58] Where is the live stream please? [19:32:17] Adding to topic now see. http://youtu.be/yaj9dfHjkOA [19:33:26] Stream is not up yet [19:33:36] halfak: yes, starting at .35 [19:33:59] @halkfak: thanks! [19:34:10] halfak: are you taking question from IRC or shall I do it? [19:34:25] I'm happy to do it. :) [19:34:31] okay. great! thanks! [19:34:34] we're live [19:34:48] We just started. Stream should be moving in a few seconds. [19:34:59] Stream is up. [19:35:11] And Haitham is kicking us off. [19:35:19] Ping me if you'd like top ask a question. [19:35:35] * halfak raises hand [19:36:40] Good question. [19:37:00] * Nettrom thinks it would be great if the presentation was a bit less WMF-internal [19:37:08] 1 min. delay here on youtube. i now understand why halfak raised his hand .. i thought he was raising his hand to ask a qn :) [19:37:16] me too [19:37:29] heh. [19:37:33] But I'm sure halfak asks questions of himself on a regular basis. [19:37:40] hey Nettrom [19:37:44] :) [19:37:46] Maybe I just like feeling tall. [19:37:47] hey DarTar [19:37:56] what do you mean by less internal? [19:38:22] DarTar: Haitham makes a lot of references to previous meetings and addresses the crowd in the room, ignoring the reset of the audience [19:38:29] Good morning rdaiccherlb [19:38:33] got it, good point [19:38:35] "made a lot of", or something to that effect [19:38:47] Nettrom: the meeting Haitham referred to is public. ;-) [19:38:55] Nettrom, metrics meeting are public too [19:39:06] leila: cool! :) [19:39:23] but I don't think that should mean I'm assumed to have attended them [19:39:28] Still it's important to recognize that the remote audience is large. [19:39:33] but enough about that, carry on [19:40:03] ...including GS remote participants in the audience, hopefully. [19:42:22] Congrats on the large # of completed surveys, that suggests excellent outreach, especially if the diversity of the respondents was wide. [19:43:32] halfak: question when it's convenient: how was participant recruiting done? [19:43:46] Pine, got it. [19:43:50] Pine: I just asked it myself :) [19:43:59] don’t know if it came across [19:44:21] halfak: further question: was recruiting done on mobile as well as desktop? [19:44:25] they recruited via CentralNotice banners, selecting languages and countries [19:44:41] not stratifying participants [19:44:53] good q about mobile [19:45:13] I didn't really understand the last comment about the gender gap [19:45:25] Pine, I'll ask the whole thing since others in the room might have the same question. [19:45:35] ok [19:46:14] Background: I have had issues myself using Qualtrics on mobile, and someone complained to me about the same thing. [19:46:41] Haitham showed the chart of male vs. female users, and in some countries the bar for femaies was higher, but he said it didn't mean that there were more females, but rather than the gap was smaller. Can anyone explain? [19:46:41] Did he mean that women were more likely to identify their country in Argentina, Brazil, Phillipines but that doesn't mean that there are more women editing from those countries? [19:46:57] hmm [19:47:41] but then how does that relate to the gender gap question? [19:48:16] I'm curious about that too [19:48:54] guillom & Krystle, do you guys want me to raise a question? [19:48:55] I think the last question about the scale was about the same thing [19:48:58] I'm confused [19:49:20] halfak: I think Ellery just asked the same question, and Haitham answered, but I still don't get it :) [19:49:35] No worries [19:49:39] kk [19:51:51] halfak: Not at this point, thanks - I think I'll want to revisit that part before framing a question [19:54:19] I personally am fond of Wiktionary, it would be interesting to see if that could become a growth driver, especially for language-specific content [19:54:33] DarTar: i think there was also some recruiting via email https://meta.wikimedia.org/wiki/Research:Global_South_User_Survey_2014#Countries_and_languages [19:54:51] HaeB: thanks [19:55:41] Intersting data on the native-language content demand, can you drill into that a little? what was the native-country-language being compated to? English? other? [19:56:11] SVentura, add to question list? [19:56:24] yes please :) [19:56:32] ok to wait the end [19:57:25] Note: "Mobile" probably only means the mobile site, not the mobile apps. IIRC the mobile apps can't display CentralNotice banners. [19:58:18] yes [19:58:59] ok thanks [20:00:09] can you talk more about internet connectiveity? is slow internet a real issue for wikipedia in GS? can we do something in terms of local servers in different geographies? [20:00:39] Question: [20:01:00] what would be the bext next research/step to do in the Global Soun\th? [20:01:01] dalek2point3, will ask [20:01:20] Ok Asaf is asnwering. [20:01:23] question from Leila first [20:01:24] halfak: thanks [20:01:32] SVentura, still ask? [20:01:47] I think Asak is answering [20:01:51] thanks [20:02:00] halfak: thanks! [20:02:12] kk [20:03:32] Leila: I had the same reaction. was expecting mobile to be way higher [20:03:48] yeah, SVentura. this is strange. [20:03:48] leila: that high desktop ratio might be survivor bias though ;) [20:04:38] HaeB: good point [20:04:45] Isn't more than half are using desktop [20:04:59] Looks like 75% if you add Desktop and Desktop & Mobile [20:05:12] HaeB: good point. if the sample is representative, this result is counter-intuitive. [20:05:27] but it may well be not representative as you mentioned HaeB [20:05:41] yesh, halfak. more than half. [20:07:20] halfak: we want to move on to the next presentation. let's keep all the other questions for the end of the talk. [20:07:23] if there are any [20:07:27] OK. [20:07:30] thanks! [20:07:36] dalek2point3, ^ [20:07:39] Sorry about that. [20:07:46] I've still got it in the queue though if we have time [20:08:05] halfak: we will have time at the end of the presentation per Dario [20:08:47] halfak: no issues :) [20:10:24] Alan gets points for not calling us "Wikipedia" [20:11:02] Nettrom and others who might be interested in that previous metrics presentation on Global South: https://commons.wikimedia.org/wiki/File:Key_theme_-_Global_South,_WMF_Metrics_Meeting_February_2015.pdf https://commons.wikimedia.org/wiki/File:WMF_Monthly_Metrics_Meeting_February_5,_2015.webm [20:11:27] :HaeB: thanks! [20:12:44] HaeB: yeah, thanks for the links! [20:30:17] interesting comment on "focus should be on the collaborative nature of the project - the people - rather than the data" [20:31:59] https://commons.wikimedia.org/w/index.php?title=File:Key_theme_-_Global_South,_WMF_Metrics_Meeting_February_2015.pdf&page=24 looks like bot-like welcomes cloud [20:35:12] foursquare, runkeeper, duckduckgo among sites using OSM data [20:43:00] i have one [20:43:48] dalek2point3, question for alan or for Haitham? [20:43:50] alan -- thanks for your talk. just a comment, in some research im doing i show that there is "good TIGER" and bad TIGER" -- comparing these, Im finding that Bad TIGER region had better community growth in the long run. that is something that you could perhaps use. [20:45:01] dalek2point3, do you want to describe what "good" and "bad" mean here so that I an relay that to Alan? [20:45:45] sure -- "corrected TIGER" that did not suffer from all the problems that he mentioned. they had been through the Census correction program while other counties were yet to be processed [20:46:03] dalek2point3: is that evidence for https://meta.wikimedia.org/wiki/Cunningham%27s_Law ? ;) [20:46:29] ironically, having the higher quality import hurt the community, was certainly surprsing to me [20:47:18] HaeB: thanks, didnt know about this! [20:48:13] what is a spacial regression exactly? [20:51:23] * notconfusing_ claps [20:51:50] any idea when the march event is happening? i might be in SF [20:52:15] dalek2point3: we're trying to figure it out. March 11 or March 25 [20:52:25] notconfusing_, https://en.wikipedia.org/wiki/Spatial_analysis#Spatial_regression [20:52:26] We will know it in 1-2 days [20:52:36] cool -- what is a good way to propose a talk or are you booked up? [20:52:40] tl;dr, a regression that accounts for the laws of geography. [20:52:51] https://en.wikipedia.org/wiki/Tobler%27s_first_law_of_geography [20:53:10] dalek2point3: please propose. we have a semi plan for the March showcase, but it's not finalized [20:53:11] leila: this is the paper: http://web.mit.edu/nagaraj/files/copyright_nagaraj.pdf [20:53:18] dalek2point3, I wasn't able to ask haitham your other question, but I'll PM his email address. [20:53:21] and if it doesn't work for March, we can always consider it for April [20:53:42] alright, thanks [20:54:10] dalek2point3: I'll pass it by the team and we will update, how about that? [20:55:00] great -- email would be best, thanks [20:55:13] im in town on mar 11 -- so would be a good excuse to visit [21:31:23] The http://dx.doi.org links seem to have stabilised around 20k [21:32:44] vs. 1733100 protocol-relative (on en.wiki) [21:55:47] Nemo_bis, what? [21:56:52] halfak: 99 % of dx.doi.org links on en.wiki are now protocol relative [21:57:04] Oh! I see [21:57:14] So they are prefixed with "//"? [21:57:22] Yes [21:57:32] Cool :) [21:57:38] Bot job? [21:58:05] No, just two templates, one module and the interwiki map IIRC [21:58:16] Even too easy, makes me too lazy to tackle the rest [22:02:51] halfak: and the funny example diff from #wikimedia-tech is https://en.wikipedia.org/w/index.php?title=Heat_shock_protein_90kDa_alpha_%28cytosolic%29,_member_A1&diff=647766236&oldid=647766064 [22:03:53] I must be missing something. What is this diff an example of? [22:06:04] Of how we get to such absurdly huge numbers of links to papers :) [22:06:17] Ahh. heh. [22:10:17] * halfak goes to gather the rev that adds the most cites. [22:13:21] :D [22:15:17] This one adds 375 cites that still exist in the article: https://en.wikipedia.org/w/index.php?diff=prev&oldid=208356562 [22:15:24] That's the biggest one I found. [22:15:27] Nemo_bis, ^ [22:15:54] "bibliographicmetadatabot" [22:16:23] moo [22:16:31] baa [22:18:46] uamaol: your example wasn't a record for total, halfak found https://en.wikipedia.org/w/index.php?diff=prev&oldid=208356562 ; dunno for per-line count [22:19:31] Note, that I am only looking for unique IDs and I only count 'em if they persisted to the current revision of the article at the time of the database dump [22:19:36] 2015-01-12 in this case. [22:19:57] Nemo_bis: I have absolutely no idea what you are talking about [22:20:00] Are you a bot? [22:20:07] lol [22:20:27] I met a human claiming to be Nemo_bis once. [22:21:39] See the top 10 revisions here: https://gist.github.com/halfak/bbaeb8e75ba268d04755 [22:24:44] But yeah, no offence Nemo_bis, but what are you actually talking about? [22:25:12] uamaol, https://en.wikipedia.org/w/index.php?title=Heat_shock_protein_90kDa_alpha_%28cytosolic%29,_member_A1&diff=647766236&oldid=647766064 [22:25:27] Apparently you guys were discussing revisions that added a ton of DOIs [22:25:48] I don't understand what this has to do with IRC logs [22:26:28] Did someone say something about IRC logs? [22:26:31] Nemo_bot [22:26:33] !logs [22:27:33] aww [22:27:48] I asked a question about IRC logs in #wikimedia-tech, and Nemo_bis got back to me and said something about them, then told me to go here about some search thing [22:27:58] Now I'm lost as hell as to why I'm here [22:28:12] uamaol: to get nice facts about the diff you linked [22:28:25] You are free to leave the channel, I'm not going to chain you to the ground :) [22:28:31] Oooooooh ok, fair enough [22:28:56] I posted it due to the amazment of a one liner having 15 citations. [22:29:17] I'm not saying you are, I'm just wondering what the hell you were referring to, lol [22:32:19] Think I might go to bed, good night! [22:35:41] DarTar, I am working on the revision hand-coding stuff and I'd like to run some ideas past someone. Got a few minutes to get on a call? [22:39:40] halfak: I’d love to, but I’m in meeting hell until later today :( [22:40:06] No worries. [22:40:30] lzia, :) Got some time for me to bounce some ideas off of you? [22:41:25] need to finish something for Hovercards, halfak. in 2 hours if you will still be around then? [22:41:41] Hmm.. Not likely. No worries. [22:41:48] sorry. :-| [22:42:25] No problem. :)