[00:10:26] DarTar, this is the list we've used for GS: https://meta.wikimedia.org/wiki/List_of_Countries_by_Regional_Classification [14:20:56] J-Mo: heya! would the webinar be youtube streamed or something?} [14:21:12] hi Yuvi. Yeah, it will be. [14:21:19] lemme send the link to the page… [14:21:23] cool [14:21:31] quiddity: ^ [14:22:04] https://plus.google.com/u/0/b/117126229120807344920/events/cngcjcmm1tcn1omqpalcm686044 [14:22:37] by the way, YuviPanda: Quarry is really slow for me right now? It may be because I keep running and then halting queries. [14:22:38] J-Mo: oh, no youtube live? [14:22:40] oh [14:22:44] * YuviPanda checks [14:23:02] J-Mo: as in, lots of queueing? [14:23:04] * YuviPanda checks [14:23:08] thanks [14:24:33] J-Mo: hmm, they seem to be hitting an odd bug and getting stuck :| hit run again? [14:25:08] will do [14:25:22] J-Mo: I think general workaround for now is 'if it is queued for more than 5s, hit submit again' [14:25:32] J-Mo: I'll monitor things throughout [14:28:48] J-Mo: hmm, all queries that return a datetime seem to be getting killed. fixing now [14:29:32] awesome. thanks! [14:37:24] J-Mo: can you re-run http://quarry.wmflabs.org/query/287 [14:38:27] yah [14:38:55] J-Mo, why in god's name are you awake and working [14:38:57] bah, unfixed. [14:39:23] Ironholds: I'm running a webinar, teaching people MySQL with Quarry [14:39:33] ahh [14:39:37] ...go back to bed [14:39:45] no!!! [14:40:41] J-Mo: run again? [14:40:47] k [14:41:07] Ironholds: I pushed a bunch of changes to DMZ [14:41:21] cool! [14:41:23] Ironholds: and have one more coming up that actually does something (measure edits per country for any wiki for any time period) [14:41:32] I actually almost certainly won't be able to come and work on it this week :/ [14:41:35] J-Mo: hmm, failed again. looking [14:41:37] things ar.e..tremendously hectic :( [14:41:37] Ironholds: yeah, that's fine [14:41:56] but hey, worst-case you have an idea now for an awesome research project :D [14:42:20] YuviPanda: does the problem only occur with certain kinds of queries? [14:42:43] J-Mo: with dates, yeah [14:44:34] J-Mo: aaand 'decimals'. am fixing now [14:44:46] cool [14:45:21] so much a.m. >.< [14:47:51] J-Mo: should work now, try submitting that query again? [14:48:03] k [14:48:18] yes!!! [14:48:20] worked [14:48:25] J-Mo: \o/ cool. [14:48:30] J-Mo: anything else that is stuck? [14:49:47] checking [14:54:09] YuviPanda: this query seems to be stuck: http://quarry.wmflabs.org/query/294 [14:54:31] J-Mo: hit run again? [14:55:23] did. still frozen [14:55:41] J-Mo: it might actually be just running... [14:55:57] since I see no errors [14:56:05] it took about 4 seconds when I ran it through my secure labs connection [14:56:14] just now [14:57:52] J-Mo: hmm, I hit run on a copy (http://quarry.wmflabs.org/query/295) and it *is* running [14:57:59] no errors, at least. [14:58:28] J-Mo: starts in 2mins, no? [14:58:46] yeah :( [14:58:59] J-Mo: others seem to be working fine. [14:59:00] hope that these errors/lags don't persist [14:59:23] yeah, me too [15:00:08] J-Mo: just that query seems to be stuck, others are going fine [15:00:46] okay, I'll wing it [15:00:52] starting in 2 min [15:03:35] quiddity: the webinar is at https://plus.google.com/u/0/b/117126229120807344920/events/cngcjcmm1tcn1omqpalcm686044 if you want [15:04:03] i'm in the hangout, to pass along any questions that you don't answer first :) [15:04:24] quiddity: is it live for you? [15:04:56] not yet [15:04:58] MatmaRex: hangout URL https://plus.google.com/u/0/b/117126229120807344920/events/cngcjcmm1tcn1omqpalcm686044 [15:06:32] ashley: https://plus.google.com/u/0/b/117126229120807344920/events/cngcjcmm1tcn1omqpalcm686044 is the link [15:09:04] I do now [15:11:11] hm, Flash needed, apparently :-( [15:30:23] YuviPanda: - how can can I authenticate quarry with my bot [15:30:36] * YuviPanda waves at OrenBochman [15:30:57] would the login token work ? [15:31:18] OrenBochman: so, you get an account at tools.wmflabs.org. Bots can run from there as well. We provide machines + sql access to the same databases via mysql commandline / any programming language [15:31:57] my bot runs localy [15:32:16] OrenBochman: you can create an account there, and use a 'ssh tunnel' to access the sql interface while still running bots locally [15:32:21] that's why i like quarry [15:32:24] although of course, I"d reccomend running them from toollabs :) [15:32:59] yuvi I got a tool labs account [15:33:19] OrenBochman: ah, right. so, you can run: [15:33:20] ssh -C -N -L 3306:enwiki.labsdb:3306 tools-login.wmflabs.org [15:33:24] on your local machine [15:33:35] and that'll setup a ssh tunnel from your local machihne [15:33:36] *machine [15:33:43] so you can connect via that [15:33:52] username and password are in replica.my.cnf in your homedir on toollabs [15:34:03] but I [15:34:33] but I'd realy love to access the db using the gateway you made.... [15:34:54] logistics of labs is difficult [15:34:57] OrenBochman: heh, sadly Quarry is just a GUI. [15:35:16] OrenBochman: true, but with the ssh tunnel method, you don't have to deal with labs at all, other than copy the user account credentials from there [15:35:18] no api [15:35:37] no API, no. I could build one if there's demand :) [15:36:12] J-Mo: cmd and + will increase text size of your browser [15:36:20] quiddity: ^ [15:36:25] didn't work :( [15:36:36] my bot is a lib for hackers to access the api without too much hassle from R [15:36:47] OrenBochman: the MW Api? [15:36:57] yes [15:37:06] All example queries so far, are at https://meta.wikimedia.org/wiki/Grants:Evaluation/Library/Wikiresearch_webinars#Part_2:_Becoming_Wikimetrics [15:37:17] acessing the DB would be even a great bonous [15:38:07] but the lib is curl based [15:38:08] OrenBochman: ah, indeed. but unauthenticated access to the DB has been nop'd by both legal and ops in the past, so you'll need to have some form of labs authentication. You'll also need to agree to Labs ToU before that every time, and there's no easy way to do that with the API unfortunately. [15:38:32] I'll write a blog post about making that ssh tunnel super easy later today / tomorrow if you want. [15:39:13] my lib supports mw login via api [15:39:36] right, but this is different, since to have *any* access to the DB you need to agree to Labs ToU [15:40:57] > right, but this is different, since to have *any* access to the DB you need to agree to Labs ToU [15:40:59] OrenBochman1: ^ [15:41:30] but you wrote a gateway [15:41:50] OrenBochman: indeed, and you need to login with a wikimedia account before you can access it :) [15:41:59] and there's a 'I agree to labs ToU' there [15:42:07] cool! [15:42:30] OrenBochman1: I'll think of a way to get API access [15:43:20] cool [15:44:22] J-Mo: people want zoooom :) [15:44:29] quiddity: ^ [15:45:09] p.s. I've already coded sql queries for a bunch of the research demos - it would be a big effort to recode it via the api [15:45:47] OrenBochman1: true. you should publish them on Quarry! :) [15:46:25] OrenBochman1: I'll soon be adding functionality where you can download CSV or JSON or TSV output of a query [15:46:51] they take a while to run but I'll try them out [15:48:04] * OrenBochman1 missile defence rocks [15:48:29] I already support parsing of json! [15:48:58] and R loves csv [15:49:08] my code run in R [15:49:44] :D [15:49:53] TSV > CSV, and R handles TSVs beautifully [15:50:02] * Ironholds goes back to his cave, having responded to both of his stalkwords [15:51:36] hi Ironholds - nice lib you wrote! [15:51:54] wait, which one? [15:52:07] the r api [15:52:10] WikipediR? Thanks! I need to add unit tests and vignettes to it ;p [15:52:18] but I just roxygenised the documentation which was fun [15:52:28] I've been writng much the same but using curl [15:53:27] I'l be presenting it in about amonth at a user group meeting [15:54:01] I hope [15:54:03] YuviPanda, does SQL (quarry) care about linebreaks? [15:54:16] quiddity: nope, you can line break things [15:57:46] J-Mo: that query isn't going to complete, it got killed. Works on s3.labsdb, but not on s4.labsdb. I'll follow up with springle later on [15:58:41] OrenBochman1: title your queries! :) [15:59:33] not sure what they do anymore :-) [16:01:29] OrenBochman1: heh [16:05:36] god thing I wrote a blog post about it [16:06:00] any how this last one takes about an hour [16:06:08] will it time out ? [16:06:13] OrenBochman1: yeah, 10min limit [16:08:12] J-Mo: it's in the works, btw :) waiting some upstream fixes, though :) [16:08:17] (re python on the web) [16:09:14] \o/ [16:09:49] Ironholds: I'd like to use an example of an api query using a generator - an idea for a interesting example [16:10:04] OrenBochman1, hmmn. [16:10:07] a generator.. [16:10:21] what about chunks of revision text? [16:10:33] desktop version versus mobile version and then controling for section=N in the latter case [16:10:37] chunks ? [16:10:42] sections [16:10:51] I'd suggest diffs but the API has sucky diff error messaging (ROAN.) [16:11:20] I'll be porting you diff code in a couple of days [16:11:38] to? [16:11:45] to work using curl [16:11:49] RCURL [16:12:37] ahh [16:12:38] isn't httr just a wrapper around RCurl? ;p [16:12:39] about revision - I tought a revision is one chunk [16:12:50] no idea [16:13:32] I think it isn't [16:15:04] huh [16:15:11] J-Mo: quiddity \o/ [16:15:20] yo! [16:15:21] (and whoever was the other person facilitating it, I have no idea) [16:15:49] thanks YuviPanda! It was Maria Cruz, our Communication Coordinator. [16:16:49] J-Mo - nice talk [16:17:02] much better than the last one I listened to [16:17:22] J-Mo: that query that was getting killed works now, I've switched database hosts [16:19:01] awesome. Thanks YuviPanda and OrenBochman1. [16:19:12] also, quiddity is awesome. [16:20:15] * quiddity was just quietly hanging out, learning, and sipping coffee. :) [16:20:31] oh, and adding links to the webpage. [16:20:41] 4 of my favourite things! [16:21:22] * YuviPanda was eating fake chicken [16:21:45] Ironholds: httr does not depend on RCurl [16:22:51] it uses httpuv  [16:22:51] huh [16:22:53] ahhh [16:33:01] is it possible to remotely connect to the mysql db via ssh [16:34:38] OrenBochman1: it is, that's the 'ssh tunnel' I was talking about :) [16:35:08] https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Help#Configuring_MySQL_Workbench has some help, and I"ll write more detailed help later on [16:35:10] gotta go now [16:35:21] thanks [16:45:58] * Ironholds is all excited for the presentation [16:47:58] :-) [16:52:32] btw is there a way to pretty print json output from the api ? [16:57:02] goood question [16:57:14] I think there is; legoktm is my API point of reference, so he may know if he's awake [17:03:46] lzia, ping :) [17:04:14] so, the presentations: could I go either first and be quick or second but definitely after 11:30? Some guy is coming round to drop off some keys at 12 and I don't want to have to peg it in the middle of the talk :p [17:10:40] Ironholds: I just just codded a getPageTemplates which uses the generator=templates [17:10:52] aha [17:13:05] next implementing a continuing query [17:14:46] send me a link to what you come up with, templates in R sound..interesting. [17:16:14] I wanted the to to list user box on user pages [17:16:46] ahh [17:16:53] lzia, around? :/ [17:35:58] DarTar, is leila alive? ;p [17:36:05] hey [17:36:10] she is, next to me [17:36:20] setting up a hangout on air [17:36:27] to see if we can share a link in advance [17:36:38] isn't it in an hour? [17:36:55] My message was "so, the presentations: could I go either first and be quick or second but definitely after 11:30? Some guy is coming round to drop off some keys at 12 and I don't want to have to peg it in the middle of the talk :p Sorry to be a hassle :/" [17:37:37] (okay, going out for 5 mins) [17:45:26] Ironholds: yes it’s starting at 11.30 PT but this time we’re sharing the streaming link well in advance [17:45:55] -> https://twitter.com/WikiResearch/status/502149251628294144 [17:46:05] hey Nettrom [17:50:17] Ironholds, yt? [17:50:22] not sure what you mean by be quick [17:50:41] It starts at 11:30, we need few minutes to cover the survey results. [17:50:52] You want to start at 11:35? [17:51:08] hey Nettrom. [17:51:22] back [17:51:23] so, just to figure out logistics, how much time do you need for your presentation? [17:51:25] DarTar, cool! [17:51:39] leila, ideally? but it looks like I may be able to hnadle the poblem [17:51:43] will know in the next 10 mins [17:51:57] that's always easier, Ironholds. ;-) [17:52:25] hey DarTar [17:52:32] hey [17:52:43] leila, yup :) [17:52:46] also, guess what I did! [17:52:52] I finished my presentation more than an hour before I gave it! [17:52:53] so we’re going to test the a/v in a moment [17:52:56] Ironholds: w00t [17:53:01] and it contains really interesting demographic factoids I think are paper-worthy! [17:53:03] that makes me a bit nervous [17:53:08] about the behavioural differences between old and new contributors [17:53:13] and how readers behave in varying circumstances [17:53:22] can’t wait to seeing it [17:53:32] well you'll have to, it's not for another 40 mins [17:56:02] can someone send me the links so we can test the audio setup etc? [17:56:06] I mean, if that'd be helpful ;p [17:56:09] otherwise I'd be wasting your time [17:56:29] We're sending you in 4 min [17:56:35] the testing starts only then [17:56:36] ;-) [17:56:41] breathe deeply Ironholds. [17:57:40] okay! [17:59:45] leila, please communicate to dartar that not only did I complete this ealy [17:59:50] it is not white-background-black-text [17:59:55] * Ironholds waves hands, makes scary noises [18:09:25] OrenBochman1: do you want &format=jsonfm? [18:16:03] Ironholds: hundreds of people watching you [18:16:13] no, nobody showed up yet [18:16:43] haha [18:16:49] hello [18:17:52] Nettrom: are you in MN at the moment? [18:18:00] DarTar: yes [18:18:08] kk [18:19:29] Nettrom, can you do me a favour? [18:19:30] Ironholds: +1 https://gerrit.wikimedia.org/r/#/c/155274/ ? [18:19:34] Go to U.S. bank and shout at them [18:19:53] they don't have any branches in boston. Ever tried getting a deposit and rent together with a frickin' ATM? :p [18:20:03] ouch, sorry to hear that [18:20:09] legoktm, done [18:20:27] Nettrom, it'll work out. Just ask em to open one in Boston. They're midwesterners, they'll do it out of politeness! [18:20:28] thanks, +2'd [18:22:11] so guys, I understand we’ll go in this order: dartar (intro - 30 sec); leila (quick update on survey - max 5 minutes); morten (25 + 5); Oliver (25+5) followed by general questions, time allowing [18:22:30] the slots don’t add up to 60 mins but that’s ok [18:22:44] how does that sound? [18:22:51] DarTar: sounds good to me! [18:23:46] BTW, I've been aiming for about 25 mins in total w/questions, so I'll probably try to be done around the 20 mark, we'll see [18:25:58] Nettrom: perfect [18:28:20] we’re going to take a few more minutes for people to show up [18:28:40] Ironholds: bring back the BBC screen! [18:29:00] O [18:29:02] I'll brb [18:29:15] DarTar, I didn't do it, Nettrom did! [18:29:42] ahhh [18:36:29] can we request speakers now? [18:36:36] I want Neil Gaiman. He's great. [18:36:51] I'm running AV for this presentation, any comments please ping me. :-) [18:36:51] I want to hear what he thinks about....ooh, ANYTHING. [18:36:54] Ironholds, we didn't accept submissions from R&D [18:37:09] okay, consider that made under my volunteer account ;p [18:37:10] you can bring it up in the next R&D staff meeting. ;-) [18:49:21] Hey folks. Can someone post the youtube link? [18:49:49] https://www.youtube.com/watch?v=wgnnVG7sLQ0 [18:49:53] halfak, ^ [18:50:23] ty [18:50:38] np [18:50:49] Flat bike tire. Just walked 5 miles :( [18:51:12] oh oh. :-( [18:53:52] President! [18:55:27] The quality re-assessment that Morten is talking about: https://meta.wikimedia.org/wiki/Research:Screening_WikiProject_Medicine_articles_for_quality [18:58:17] If you have questions for Morten, this is a good time to bring it up. I'll relay them for you. [18:59:07] *claps* [18:59:40] What's next for suggestions in Wikipedia? Where are we missing out on opportunities for task routing? [18:59:48] leila, ^ [19:00:12] yup, halfak. will ask next [19:00:18] ty [19:02:12] Ooh! In collab. [19:02:16] <3 morten [19:02:31] yay Nettrom! [19:03:32] halfak, you may find mine interesting [19:03:35] I was going ask about suggestions within a wikiproject, thanks :-) [19:03:38] I found something weird about editors last night. [19:03:44] so, pay attention to that bit [19:03:57] :) [19:03:59] Cool [19:04:04] Looking forward to it. [19:07:40] Technical difficulties? [19:08:42] halfak, yes. The same issue you always have. If don't mute SF, people on Hangout will hear sounds that we don't hear [19:08:54] so mute/un-mute causes trouble [19:08:57] Gotcha. Seems to be good now [19:09:01] and then there are the mics. :-) [19:09:08] yeah, switches are hard. [19:10:33] " [19:10:40] "Bob's yer uncle" [19:11:27] Looks good to me [19:11:36] Situation is that Google hangout is producing noise from our account that *doesn't exist locally* [19:11:49] even if we mute the entire board (the only input) the account still produces noise. [19:11:53] I think that noise is called "Oliver" [19:11:58] Although, I can't tell what it sounds like. Is it an echo? [19:12:06] halfak: :-) [19:12:15] cndiv, I think it sounds pretty good on youtube right now. [19:12:28] halfak: Yeah, it's fine as long as I keep our local account muted [19:12:38] I see. [19:12:44] halfak: If I unmute, even though our mics and all inputs are muted, we get noise. [19:12:49] Although I can't tell what it sounds like. [19:13:08] * halfak looks for a good way to take a recording. [19:13:30] halfak: well it'll be on youtube afterwards, but with the flaws. [19:13:42] Oh yeah. That'll do. [19:14:42] halfak: the presentations themselves will be fine, the Q&A a little choppy. [19:14:48] Looks like mobile is solidly higher during the wee hours of the morning. [19:18:36] We also send our emails in the morning :) [19:22:28] Q for Oliver: It looked to me like the editing patterns had mobile consistently higher in the low hours. What's up with that? Isn't that a "difference"? [19:23:09] can someone tell me if they're hearing noise when O speaks? [19:23:48] halfak, there is one more question from the room, and then I'll bring up yours [19:23:53] ty [19:24:01] cndiv, sounds good to me so far [19:24:06] weird [19:24:19] halfak: did you mute us manually? sounds like a normal conversation between O and H? [19:24:21] I heard Howie just fine. [19:24:31] I'm listening to Youtube [19:24:43] Normal convo between O and H. [19:25:01] Just heard H cut in and that sounded good too. [19:25:12] grr. annoying. [19:27:03] actually local morning time is highest for donating [19:27:07] email and banner [19:28:19] No real studies where we aimed to convert readers. There's pushback against the general idea (what if readers suck at editing), but there's been talk about calling more attention to the edit links. It would be a good study to run IMO. [19:28:38] dang, just remembered a thing I missed when answering halfak's question... recommendation opportunity missing: supporting more languages [19:29:41] :D There were some questions about that at Wikimania. I pointed them to suggestbot and told them to demand it in their language if it didn't exist already. [19:29:47] Just join the talk sorry if it's covered already [19:29:55] but did you look at different languages? [19:30:05] leila, ^ [19:30:08] you will hear it soon, Taha [19:30:11] I just asked it [19:30:12] :) [19:30:16] Our previus researc showed daily and weekly patterns could differ from language to language [19:30:16] halfak: yeah, that is the way to get it :) [19:30:18] there is a bit of a delay [19:30:44] I pulled up the suggestbot page and told people to go use it on one of my talks. :) [19:31:03] all right! [19:31:05] okay, so [19:31:11] I couldn't see IRC or the audience [19:31:27] did that (a) go okay, (b) go really well and fascinate people, (c) piss off legal or (d) blow chunks? [19:31:46] Ironholds, can you re-join the Hangout? [19:31:52] sure! [19:31:56] cndiv wants to test few things [19:31:57] Ironholds: struck me that you've made some nice tools to have in the quiver [19:32:00] ..I've lost the link. PM it to me? [19:32:29] Thanks folks! [19:32:30] k [19:32:42] * halfak claps [19:32:55] halfak, find the editor variation between groups by editcount interesting? [19:32:59] or, lack thereof, rather [19:33:01] *time variation [19:33:06] you're being added Ironholds. [19:33:10] ta! [19:33:22] I can't see any notification [19:33:23] Ironholds, somehow I missed that. [19:33:28] aha, got it [19:33:29] * halfak rewinds the youtubes [19:33:30] k, you have it Ironholds. [19:33:52] great presos Nettrom and Ironholds [19:34:03] Ironholds: all your plots are belong to us [19:34:11] halfak: wait, i did not realize you could rewind these streams! :) [19:34:22] yo mako [19:34:29] that's useful, i managed to missed most of it this time [19:34:29] thanks DarTar [19:34:49] Ironholds: unless you have a standalone report on Meta, we should add the highlights here: https://meta.wikimedia.org/wiki/Research:Mobile_trends [19:34:55] halfak, TL;DR, edit patterns over 24-hour periods really similar for people w/ <=10, <=100, >=100 edits [19:35:04] DarTar, I'll write up a standalone, it seems interesting! [19:35:13] Did I shock you with a fascinating area of research, or something? [19:35:19] woo twitter traffic, that's new [19:35:24] chrismcmahon, what tools? [19:35:26] :) [19:35:46] Ironholds, that's pretty surprising [19:35:53] halfak, yeah, I want to quadruple-check it [19:36:02] and make sure I'm not failing to exclude bots or include bots or..something [19:36:20] also, leila’s suggestion on the dominance of enwiki and North America is an excellent point [19:36:22] Still seems plausible [19:36:24] agreed [19:36:48] just removing enwiki and NA will reveal a lot of interesting patterns [19:36:51] (e DarTar) [19:36:53] NA? [19:36:58] so seriously [19:36:59] NA? [19:37:00] North America [19:37:02] Ok [19:37:04] h [19:37:04] can someone tell me if it was fascinating or boring or what already [19:37:07] I still don't know [19:37:12] I only know twitter liked one of my jokes [19:37:12] I really liked it. [19:37:16] I want to see more [19:37:21] Ironholds: reader-vs-editor-plus-mobile-vs-desktop-by-timezone reports seems like tools to answer other questions [19:37:25] that is...awesome praise. [19:37:28] chrismcmahon, ah, gotcha. [19:37:29] same here, plus Oliver’s Law of Geodata is all over the twitterwebz [19:37:34] halfak, that's in fact perfect praise [19:37:38] DarTar, more people should follow it. [19:37:39] :) [19:37:40] like, all people [19:37:45] except people who I want datasets from [19:37:49] We'll add an exception for them [19:38:27] let’s figure out what we can share, I’d like aggregate datasets both from this work and the July 31 mobile trends review to be available on the DataHub [19:38:52] alright, lunchtime - ttyl all [19:38:55] okay! [19:38:59] I'm gonna go SIGN MY LEASE WOO [19:39:47] ooh [19:39:49] * Ironholds thinks [19:40:51] meganhernandez, yeah, sorry! I meant to specify mobile, I think, got evening traction for donations [19:40:52] doy [19:40:58] like, commute/bar time [19:41:56] I feel like this is...from the reaction, something I should write up and blog [19:42:01] but first, smoke and unwind and go get key. [19:48:29] anyway, i'm caught up now. that was great thanks everybody [19:49:50] o/ mako [19:50:08] Sorry to miss you and fail to have our discussion re cluster processing before Wikimania ended. [19:50:18] I'd like to pick up where we left off re. diff algorithms though. [19:51:46] halfak: oh yes! well, i'm coming to opensym this year [19:52:06] halfak: you will be there, right? if so, we can follow up then [19:52:15] Yup. Sweet! :) [19:52:29] I'll be in Berlin tomorrow night. I'm going to visit WMDE first [19:54:55] halfak: killer, i'm in like a day or so before [19:56:23] Cool. If you have time, I could try to organize some shenanigans with the WMDE folks. [19:56:36] Germans do shenanigans, right? [19:58:32] that'd be awesome! lets follow up on email [20:00:10] sounds good [20:59:41] hey halfak, is our 1:1 on? [21:00:05] I got kicked out of the room we usually have for this meeting [21:00:16] let me know if you’re around and want to chat [21:37:15] DarTar, [21:37:19] Sorry I missed you [21:37:34] hey, no worries [21:37:47] I forgot that we had a meeting scheduled so late. [21:37:53] I’m available any time or can reschedule if you prefer [21:38:02] heh EU timezones... [21:38:03] That'd probably be good. [21:38:06] Looking at Friday. [21:38:11] sure [21:38:17] How early is too early? [21:40:23] Hmm... I'm guessing 8AM is too early. [21:40:39] I just grabbed a slot that you look available. Can you confirm and grab a room for it? [21:40:42] DarTar, ^ [21:40:53] lemme see [21:41:48] it works but we’ll have to keep it short, I have to walk to PLOS to give a lunch talk and I need to be there by 11.30 [21:42:02] I could stay in the office until 11 [21:42:12] so 30 mins, does that sound ok? [21:42:15] halfak ^ [21:42:35] Sure. [21:42:40] Sounds good [21:42:44] cool [21:45:50] * halfak --> bed [21:56:52] yay back! [21:57:05] who has a lease? I have a lease [21:57:09] also, man, this channel is exploding. [21:57:11] (In a good way) [22:07:42] Ironholds: congrats [22:07:49] pictures or it didn’t happen [22:08:35] DarTar, when I move in ;p [22:08:47] so, can people just verify whether my presentation was interesting and revealing please, just once, and then I will drop it. [22:08:49] fair [22:09:14] verify as in? Everybody seemed very interested and engaged [22:09:25] and we had Taha stalking on IRC [22:09:40] yeah, although he didn't say anything [22:09:41] I was not really following IRC tbh [22:09:52] undercover sociophysics [22:09:58] verify as in: for something I did in my spare time because what-if, did you find it interesting ;p [22:10:09] and also, if I can verify the editor patterns thing, would it maybe be worth writing up. [22:10:16] in the journal sense, not the blog sense. [22:10:20] totally [22:10:55] actually, both: I want an open version of that report before you transfer all your IP to some proprietary publisher [22:14:29] DarTar, hahaha [22:14:42] I was thinking of blogging it and then trying to put something together in LaTeX, as painful as I find LaTeX [22:14:52] I'll probably grab the raw AFT5 paper off bitbucket and replace the content ;p [22:15:00] Ironholds: interesting, but I think you jumped to some conclusions about implications of the findings. I thought the most interesting thing was the radically different patterns between readers and editors. [22:15:32] the amount of pain you can inflict yourself to learn new syntax is remarkable, but yes: learning LaTeX is definitely a good investment [22:15:40] chrismcmahon, yeah, as my disclaimer said at the beginning, this was largely supposition [22:15:47] the patterns were the interesting bit [22:17:11] Ironholds: differences suggest(!!) that attempting to convert readers to editors might be a waste of resources; follows that recruiting editors from some editor-pool would be a better use. [22:17:30] chrismcmahon, that was my main take, but I don't know. [22:17:40] I mean, it could just be the other way around [22:17:52] that the act of becoming an editor causes behavioural variations because it causes contextual variations [22:17:56] Ironholds: yeah, you'd need finer hypotheses, but you started with some interesting questions [22:18:00] and, of course, we don't know if editor read patterns are any difference. [22:18:13] ...although I could find that out [22:18:17] but that would be creepy as all hell [22:18:26] or rather, you went and explored some interesting areas and came back with stuff to follow up on [22:18:58] yup [22:19:01] Ironholds: seems like you can't avoid the creepy on this path :-) [22:19:08] no, but I mean really creepy [22:19:17] like unified read and edit logs in a dataset [22:19:23] I'd like to avoid *THAT* [23:20:18] YuviPanda, yt?