[15:34:21] halfak, I also like the robot :) [15:34:33] :D [15:34:47] Was just about to start working on that diagram again. [15:35:26] go for it. I'm working on other slides. [15:35:28] * guillom waves/ [15:35:34] o/ guillom [15:35:41] morning, guillom! [15:36:30] J-Mo, do you know what % of newcomers hostbot doesn't invite? [15:36:53] of all newcomers, or newcomers who meet the invite criteria? [15:37:01] All newcomers. [15:37:05] (the answer to both is no, but it would be easier to find all newcomers) [15:37:10] no :/ [15:37:27] but from the spot-checks I've done, 4000-6000 per day sounds right [15:37:40] stats.wikimedia.org? [15:38:36] I'll check on the new registrations between October 20, 2014 and January 12, 2015. [15:40:16] 632448 [15:40:19] http://quarry.wmflabs.org/query/5786 [15:41:18] So we would have sent invitations to 2.3% of newly registered users. [15:42:38] * halfak runs a super-long query to find out what proportion of those ever saved an edit. [15:42:44] Didn't take that long! [15:42:49] 155439 [15:43:17] 9.5% of newly registered users who save an edit some time in their first 6 months. [15:43:26] ... 10 months. [16:07:30] The lack of welcome bots on en.wiki leaves a lot of space for wild niche welcome bots it seems. :) [16:37:15] o/ J-Mo [16:37:24] hey halfak [16:37:56] I just saw your comment on slide 6. [16:38:35] yah [16:39:03] Slides 6-13 are intended to together, but after that, I was just building fodder for discussion. [16:40:34] that makes sense. I just don't know if we have to go into that much depth about the rise/decline. let's discuss at our meeting, halfak? [16:40:41] OK [16:41:08] When I do that sequence of slides, it usually takes about 1 minute. [17:49:03] J-Mo, I don't have an invite to the hangout for the showcase. Can you invite me? [17:49:12] yah [17:50:19] halfak, actually I only have the event invite that you RSVPd to. Maybe Leila knows? [17:50:21] Also https://docs.google.com/presentation/d/1cBRnw_AMF9VpyDCm_YdxyUFSldcxOKpWeRf9Dp3342k/edit#slide=id.gd03bfc9a3_0_0 [17:54:22] J-Mo, see slide 27 [17:54:26] Also, it' [17:54:32] s important that we present first [17:54:38] I need to leave early for another meeting. [17:55:02] sounds good. I think we're the only ones presenting [17:55:07] slide 27 is great! [17:55:13] Cool :) [18:06:00] * halfak looks for lzia [18:07:47] hello halfak, J-Mo. [18:07:56] o/ leila [18:08:11] are you guys ready for testing? [18:08:26] pretty much. need 5 minutes. [18:09:08] sent you the Hangout link, J-Mo, halfak. If one of you can join to test that'd be great so we can let IT go back to their work. :-) [18:22:33] Hm, the youtube livestream doesn't appear to be working. : \ [18:22:51] Starting in 8 min [18:22:54] Sorry for the confusion! [18:23:03] Ah, OK. [18:23:18] Thanks. : ) [18:28:13] hi i_jethrobot. [18:28:34] Hey leila! [18:30:43] Here we go! [18:31:01] * guillom frantically refreshes. [18:31:20] Just clicked "live" [18:31:35] Yep! [18:31:41] Loading [18:32:20] success : ) [18:32:27] :-) [18:32:32] * halfak waves at Siko from behind his screenshare [18:33:53] * i_jethrobot cheers for Teahouse and HostBot! [18:38:38] If you have questions for Aaron or Jonathan, please share them here, I'll keep track of them and ask them at the end of the presentation. [18:41:11] Question out of curiousity for aaron-- has the rate of incoming new accounts that do not contribute fluctuated much since the "firehose" period? [18:41:24] (i.e. with zero edits) [18:41:53] made a note of it, i_jethrobot. [18:47:16] Ironholds: you are famous. ;-) [18:48:09] leila, what did I do? :/ [18:48:28] Got named in a Teahouse example :) [18:48:34] Jonathan and Aaron are presenting on Teahouse: the example Jonathan showed just now showed you. :D [18:48:44] On Scotch whiskey, no less. : P [18:48:52] haha, i_jethrobot. [18:51:31] ahh [18:54:00] wow, that sample size! : ) [18:55:04] yeah, that's nice, i_jethrobot. My question is, why didn't they have the two groups the same size? [18:55:15] (keeping note of my question ;-) [18:57:50] J-Mo forgot that the next slide is cute [18:57:59] (full of plusses) [18:58:14] no speaker handle please. :P [18:58:31] Oh yeah! [18:58:32] Bah [18:58:39] I'm teasing you, halfak, mostly. [18:58:42] ;-) [18:58:45] Speaker handle! [18:58:55] hahaha! [18:59:09] I enjoyed the whole two seconds of all those plusses. : ) [18:59:52] The invite condition was better than the control across the board -- just significant in only two places. [19:01:06] n = 14766 [19:02:38] everyone, send your questions here for Aaron and/or Jonathan. I have one so far. :-) [19:05:49] Another interaction effect to consider: Other help space usage. : ) [19:06:17] made a note, i_jethrobot, as a comment. [19:06:40] do you have an example of other help spaces, i_jethrobot? [19:07:39] Sure-- the Help Desk and the Wikipedia Adventure are two active spaces. [19:07:48] https://meta.wikimedia.org/wiki/Research:New_editor [19:07:53] thanks, i_jethrobot. [19:08:26] Thanks halfak. : ) [19:09:34] agree that +16% is an awesome result! but (regarding jonathan's last remark) didnt some of the experiments by the E3/Growth team also show significant (if perhaps smaller) increases, at least in short term retention? [19:10:20] HaeB, is this a question for Jonathan I should ask? [19:10:54] leila: yes, but aaron might remember too [19:11:13] Articles for Creation is another help space-- though I'll note a nontrivial propotion of Teahouse guests are there because they need help regarding AfC. : P [19:11:20] ooki, I'll give room a chance to ask questions, if not, we will come back to this, HaeB. [19:13:35] hah. : ) [19:16:56] via the google hangout : Jeph Paul asks the question: "Are the datasets publicly available? Are the teahouse invites tagged ?" [19:17:54] made a note, abbey. [19:25:04] any other questions? :-) [19:25:09] halfak: thanks for the response, makes sense (i guess there also weren't much resources devoted to study longer term retention from each of these many experiments) [19:25:12] Question for either aaron or jonathan-- the Teahouse (and other help spaces) have been known to attract sockpuppet accounts from time to time, but I'm unclear on how frequent this happens, exactly. Any idea on how big of a problem this is? [19:25:26] (Or if it is not really a big problem.) [19:26:42] thanks everyone. :-) [19:26:50] OK, thanks. : ) [19:26:52] Hey folks. See https://meta.wikimedia.org/wiki/Research:Teahouse_long_term_new_editor_retention [19:26:59] We're still writing up results and extending the analysis. [19:27:03] * i_jethrobot applause [19:27:05] : ) [19:27:07] But you can read our work log there. [19:27:21] Thanks halfak and J-Mo. Great presentation. [19:27:24] I've got to run now, but I'll be back in this channel in 1 hour if you want to chat next steps,. [19:28:03] thanks, halfak. buy. [19:28:07] halfak - I'll be around. : ) [19:28:29] thanks i_jethrobot! [19:45:23] leila: I'm getting the slides ready for Commons this afternoon. halfak: do you have attribution for the pictures on slides 13 and 28? [19:46:50] Yup. Must have deleted the slides with the links [19:51:43] J-Mo, now labeled in the slide deck [19:51:50] thanks! [20:15:58] http://www.bbc.com/news/blogs-trending-34572462 [20:16:03] 'I can haz PDF.' [21:59:06] ^ brilliant [22:17:17] oh no, did I miss one of the thingies? [22:28:23] Showcase? [22:28:33] Or guillom post. http://www.bbc.com/news/blogs-trending-34572462 [22:28:34] :) [22:28:34] that's the thing [22:28:45] No, I could never miss one of those :) [22:30:41] J-Mo: the tool is down. [22:30:50] J-Mo: I think we cannot discuss it until it comes back up [22:31:04] boo!!!!!! [22:31:29] that's fine :) schedule me as soon as it's up and you wanna talk. [22:31:52] thanks, J-Mo. [22:31:56] it's up [22:31:56] :D [22:32:02] okay, let's get to the room. [22:32:09] kk