[01:32:52] milimetric: how did the umapi stuff go? [01:33:04] i just pulled the latest revision [10:00:23] hey average, are you around? [10:39:50] hey average [11:01:46] hi drdee [13:23:21] ottomata: this is cool: http://blog.booking.com/dotfiles-distribution-at-booking.com.html [13:26:24] ah cool! [13:26:39] nice, that's awesome [13:26:50] only works when you use something like LDAP everywhere [13:26:51] but cool [13:27:37] but we are heading in that direction :) [13:27:48] also log stash is cool [13:28:00] i think we should pipe all our hadoop job log files into it [13:28:08] and do analytics about analytics [13:28:15] it's really neat [13:29:48] is this your day off? [13:55:31] yeah [13:55:34] i mean [13:55:38] i can work some, i'm about to head out now [13:55:42] we aren't biking so i'm online a bit [13:55:46] but i'm about to head out soon [13:57:51] just enjoy your day! [13:57:54] hey qchris! [13:58:09] Hi drdee :-) [13:58:17] back in austria? [13:58:48] Yes indeed. 12 hours sitting in a crammed night train :-) [13:58:56] that sounds wonderful! [13:59:06] It was indeed. [13:59:24] So I decided I migth as well strat idling around here as well. [13:59:34] totally! [13:59:45] what are you working on with gerrit? [14:00:21] qchris, are you on the ops mailing list? [14:00:27] Bringig a patch into shape that allows to replicate to github repositories under a different name [14:00:35] yeahhhHH!!! [14:00:39] i love that patch! [14:00:47] *gg* [14:00:52] hoooray i even just a few mins ago checked that bug report to see if there was an updated [14:00:59] No, I am not yet on the ops mailing list. [14:01:13] aye, s'ok, there's just a discussion going on there about packaging the kafka stuff [14:01:18] which is like java packaging stuff [14:01:47] Yes, you mentioned kafka in Amsterdam [14:02:02] I'll have a look threugh the mailing list archives. [14:02:55] hmm, not sure if they are public [14:03:00] yeah, no worries [14:03:08] thought you might be interested or potentially ahve advice [14:03:14] I'll CC you on the thread [14:03:45] qchris, what's your email addy? [14:03:53] ottomata: I just checked and cannot find it listed on https://lists.wikimedia.org/mailman/listinfo [14:04:38] So I guess that means no ops-archive reading for me :-/ [14:04:43] aye [14:34:29] ottomata: did you finish https://mingle.corp.wikimedia.org/projects/analytics/cards/385 [15:40:48] average: can you give me a quick update on what you are working on ? [15:41:58] drdee: nothing at the moment [15:42:11] why not? [15:42:30] busy with some other things, will resume working on the task in 1h [15:57:01] New review: Erik Zachte; "I'm aware there is such a thing as pod, but have never used it. So I can't comment on specific usage..." [analytics/wikistats] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/59847 [15:57:01] Change merged: Erik Zachte; [analytics/wikistats] (master) - https://gerrit.wikimedia.org/r/59847 [16:57:01] of interest to some here: https://github.com/dsc/limn-mediawiki-ext [16:57:16] ^^ robla, drdee, milimetric [16:57:31] it works :) [16:57:57] nice! [16:58:19] * robla is off to a meeting, but will look more later [17:03:14] nice, is there a testwiki where we can see it in action? [17:48:35] oops! missed your message dschoon, that's very cool [18:01:59] drdee: do we have an instance already? i've been testing locally. [18:03:10] drdee: what does webstatscollector actually power atm? [18:29:12] drdee: where can i find the official and unofficial wikipedia reader app reports ? [18:29:28] milimetric: ^^ [18:30:04] hi tfinc, the file is in the same spot, we haven't moved it: http://stats.wikimedia.org/kraken-public/webrequest/mobile/platform/ [18:30:14] http://stats.wikimedia.org/kraken-public/webrequest/mobile/platform/mobile_platform-daily.tsv [18:30:46] thanks [18:36:54] Sorry! I keep forgetting that the meeting is now 30min and not an hour [18:37:29] Hi J-Mo [18:37:36] J-Mo: https://www.mediawiki.org/wiki/Analytics/Dreams [18:37:36] hi! [18:38:11] * sumanah looks at https://www.mediawiki.org/wiki/Talk:Amsterdam_Hackathon_2013#What_I_did_in_the_hackathon_27993  [18:38:51] so J-Mo let me make sure I understand what is available to the public now [18:39:09] oh, I don't know myself exactly what's available yet [18:39:10] there's of course the general dashboards via Limn [18:39:12] lot of moving parts [18:39:21] I think drdee will correct me if I am wrong [18:39:47] so: 1) the existing Limn dashboards, such as reportcard [18:39:56] 2) the general MediaWiki API (no key needed!) [18:40:02] yah [18:40:13] 3) the UserMetrics API now nicely documented at https://www.mediawiki.org/wiki/UserMetrics/Guide [18:40:56] 4) http://stats.wikimedia.org/ [18:41:02] (more about that at https://www.mediawiki.org/wiki/Analytics/Wikistats ) [18:42:15] So, J-Mo, were you aware of all 4 of those items? [18:42:22] yah. [18:42:35] and don't forget Toolserver (and then Tool Labs) [18:42:44] so, what was your question exactly? [18:42:50] Right. 5) replicated DBs via Tool Labs [18:43:02] that's the basic categories of publicly accessible stuff; (3) sort of counts as publicly accessible because one has to have a key [18:43:06] sorry, I'm not sure I made myself clear in the hangout. [18:43:24] I made this list to follow up on my line " so J-Mo let me make sure I understand what is available to the public now" [18:43:24] (eom) [18:43:43] right! got it :) [18:43:46] in order for us to foster a community of users, we have to make sure we know what we can offer them [18:43:55] agree [18:44:18] should User Metrics be added to https://meta.wikimedia.org/wiki/Research:Data ? [18:44:46] * sumanah was checking what she could recall without going to https://meta.wikimedia.org/wiki/Research:Data , forgot a few things [18:44:54] I'm not sure, but it would make sense. [18:45:34] What I'm thinking about here is more than just presenting the list of curated resources to community members, though. [18:46:02] Go ahead [18:46:11] I think I know what you are gonna say, but go ahead and say it :) [18:46:13] I'm thinking particularly of encouraging people to use these tools to perform research related to the IdeaLab and the Individual Engagement Grants program... [18:46:44] and to provide a little human infrastructure to guide them, suggest things, foster peer support etc. [18:46:59] was that what you predicted? :P [18:47:28] I was generally predicting that you would say we should actively, conversationally encourage people to use the data sets [18:47:37] hehe. who would have thunk? [18:48:01] I see Teahouses everywhere I go. Forgive me. I'm working on it in therapy. [18:48:02] J-Mo: I understand you are heads-down prepping for your viva [18:48:04] hahahaha [18:48:17] J-Mo: I don't know whether you use GMail at all, but there is a "teahouse fox" theme [18:48:19] yeah, but I'm still 20 hours a week, same as always. [18:48:20] it is soothing [18:48:30] Oh, Ima get that. [18:48:47] over the course of the day, the fox goes around a little rural scene, drinking tea, napping, etc. [18:48:56] that is THE BEST. [18:49:07] so, I like these Analytics/Dreams questions. [18:49:16] Please do add more! [18:49:17] they could be good seeds, if we provide the right context for them. [18:49:25] like, come up with a few more to put on the page [18:49:30] re EE maybe [18:49:52] I think what's missing for these to be really engaging (or what I would want to see anyway) is a little more of the WHY these questions are interesting, and what we could do with the info. [18:50:01] yeah, word. [18:51:00] so I'm going to be putting together this "Learning & Evaluation Hub" on meta sometime in the hopefully-not-so-distant-future, which will be kind of a research arm of the IdeaLab. I'd love to foster some connections with the existing volunteer dev community when we go live. [18:51:24] sounds good [18:51:25] lot of details still vague and ownership/leadership are being tugged over, but it's GONNA happen... [18:51:28] how long till it goes up? [18:51:46] yeah…. not sure. Currently above my head :/ [18:51:57] hopefully midsummer? [18:52:47] https://www.mediawiki.org/wiki/User:KatieIreneC this might be of interest to you J-Mo [18:52:57] she got accepted to our internship program and is working on writing this doc [18:53:07] we figure it will be useful to people who do how-to-edit outreach [18:54:06] yeah, this is going to be great. [18:54:19] (btw, I've never looked at https://wikimediafoundation.org/wiki/User:Jtmorgan before. Hahahahah "Captain Swing". Also, what year did you graduate from St. John's?) [18:54:50] thanks for pointing out her work to me. I'm 2005. are you a Johnny too? [18:56:07] * sumanah added a few of the quotes to https://en.wikipedia.org/wiki/Captain_Swing  [18:56:19] No, I have some Johnny friends who graduated ~2002 [18:56:43] names? I started in 2001 (dropped out for a year) [18:56:54] I love this "Sir, This is to acquaint you that if your thrashing machines are not destroyed by you directly we shall commence our labours. Signed on behalf of the whole ... Swing" [18:57:00] Sarah Peters, Mirabai Knight, Moss Collum? [18:57:23] definitely remember Sarah and Mirabai, though they probably don't remember me. [18:59:21] * sumanah wants to use "we shall commence our labours" in a mega-ominous way at some point [18:59:28] * sumanah is ominous pretty rarely [19:00:02] Anyway! OK, so the Learning & Eval Hub. Will it be drafted on meta someplace? [19:00:34] yeah, that quote used to be my twitter byline. Of course, my handle is kind of confusing, since I'm not ACTUALLY a collective pseudonym for a Luddite/Labor resistance movement. [19:01:07] yeah, some rough notes and preliminary stuff here: http://meta.wikimedia.org/wiki/Grantmaking_and_Programs/Learning_&_Evaluation_portal [19:01:23] * sumanah laughs at the accusations-of-Socrates bit in your IdeaLab bio [19:01:35] but as I said, the dev on it has basically ceased until we can figure out some internal (political) stuff. [19:01:52] NICE!!!!! So glad someone got that easter egg. [19:02:00] I was a political science major [19:02:16] * J-Mo bump [19:02:21] * sumanah bump [19:02:26] (was that a fistbump?) [19:02:55] yeah, or just metaphorically bumping your thread to the top, out of agreement. [19:03:10] https://meta.wikimedia.org/wiki/Grantmaking_and_Programs/Learning_%26_Evaluation_portal/Tools looks totally very useful [19:03:32] so https://meta.wikimedia.org/wiki/Program_evaluation_basics:_the_program_impact_model and stuff like that will live in this portal? [19:04:45] yes, that's the idea. But there has been some… misunderstanding? conflicting understandings? of the relationship between Programs and Grants, and the role of evaluation, so we're all kind of fragmented right now. Which is bad. [19:04:59] ahhhhhhhhhhh [19:05:13] were you in the quarterly review last week? [19:05:15] so I'm letting others hash it out and I'm focusing on IdeaLab with Siko and Heatehr. [19:05:18] *heather [19:05:23] yep! [19:06:39] J-Mo: when was the last time someone advertised IdeaLab to the researchers' mailing list? [19:07:21] not recently, if at all. But now that I'm more involved, I'll make a point of doing so (probably after first phase of our current redesign, say 2 weeks out) [19:07:31] OK. Yeah, please do that [19:07:51] and when you do, please consider also sending it to Labs-l, the Wikimedia Labs mailing list, and/or the Toolserver and bots-l lists [19:08:04] I can fwd to some of those [19:08:11] we're setting it up to have more featured content, to make the idea pages themselves more findable/browsable, and to have some more obvious and engaging calls to action. [19:08:16] nod [19:08:38] btw, at the Amsterdam hackathon this past weekend, I spread the word about Grants to several of the participants. My experience is that some of them know about some of the grants, but none of them knew about all of them [19:08:59] for SURE. We want to keep IdeaLab going even when there isn't a current call for Individual Engagement Grant proposals, so a PR push will be really helpful. [19:09:13] it's never too early to start Labbing [19:09:17] * sumanah makes up a terrible verb [19:10:00] yeah, that's another thing in the works: getting better at advertising Grant programs as a whole, and directing people to the right ones. A little out of my remit generally, but definitely within scope when it comes to IdeaLab and L&E [19:10:09] * J-Mo labbing it up [19:10:27] so a possible pipeline might go: idea for research -> learning to use publicly available data sources -> quick prototyping via User Metrics, replicated databases & Labs, stats.wikimedia.org, limn -> idea for a bigger project with research & editor engagement implications -> IdeaLab -> grant proposal ? [19:10:45] WHAT IS YOUR THEORY OF CHANGE ON THIS, MR MORGAN [19:10:47] that's a beautiful workflow [19:10:54] :P [19:11:27] so yes, that's the kind of thing I'd like to see. I'd say that the missing piece there is to bring in collaborators along the way. [19:11:45] particularly people with useful skillsets to contribute to the project at the right time. [19:11:46] will you be at WikiSym? [19:11:52] yep! two papers accepted. [19:12:09] rock. Well, if you can run a speed dating session there, that would help [19:12:22] s/you/someone/ [19:12:34] yeah, that's another good idea. I can pick their big fat researcher brains for seed ideas. [19:12:41] they can all pick each others' brains [19:12:49] ewwww. [19:12:56] zombie orgy [19:13:03] I was thinking more like primates grooming each other [19:13:19] glad to see your mind went to a better place than mine. much more colleagial. [19:13:22] just move up Maslow's hierarchy of needs a few levels and that's what a conference is anyway [19:13:25] collegial? [19:13:30] collegial [19:14:05] anywho, I need to change my laundry and stop clogging this chan with my misspelled and inappropriate nattering. [19:14:09] ha! no prob. [19:14:21] I have added 2 or 3 things to you TODO list - success [19:14:42] one more thing I'd ask of you J-Mo [19:14:44] let's talk about this again? I'll ping you to set up a meeting if I can think of any reason to consume more of your time. [19:14:47] yah? [19:14:48] :) sure [19:15:17] you know that list of ~5 types of data researchers can get at, that's partly overlapping with [[Research:Data]]? Check on that & maybe update that data hub page? [19:15:49] that way we can run an informal user test or 2 on our researcher/tool developer community, ask them how easy it is to start working with that data, find the unexpected roadblocks & fix them [19:16:09] yeah, I can do that within the next week. [19:16:13] missing docs, wait-you-need-that-account, privilege gatekeepers, flaky servers, etc. [19:16:14] Cool [19:16:22] Great! Thanks [19:16:29] Happy laundering [19:16:38] thanks! ttyl [19:53:17] milimetric: how easy would it be to regenerate http://stats.wikimedia.org/kraken-public/webrequest/mobile/platform/mobile_platform-daily.tsv as platform, page views day1, page views day2, etc [19:53:47] so basically pivot on day right tfinc? [19:54:12] last person to try that was Andrew and he said he hit a snag, but should be pretty quick [19:54:26] priority wise, i'll have to defer to drdee [19:54:26] yeah, i'm making some bar charts in libre and it'll make my life easier if the data set is formatted like that coming in [19:54:37] yep, makes sense [19:56:43] hm, actually tfinc! this is quite easy with Hive. Because we can just write a typical pivot in SQL on top of the file that's there [19:57:18] milimetric: nice. is that a quick fix or should i add a story ? [19:57:58] it'd have to be a story otherwise i get in trouble i think (well, I don't know, drdee might be asleep). But we're kind of under crazy pressure these days with Visual Editor [19:58:04] thats fine [19:58:22] i'll script around it for now and add a story [19:58:57] k [22:06:00] nite everyone, gonna go figure out why i've been nauseous since I came back