[22:55:49] DarTar, Ironholds: Are we ready to office hours? [22:56:25] hey - I'm here if you guys need me [22:57:54] Understood [23:00:12] halfak: yep :). 30 seconds. [23:00:35] right. We got anyone here FOR office hours? I note a lot of people have started work already :) [23:01:39] nobody? [23:01:49] Maybe they are late. [23:02:01] Is this is like academia, assume 15 minutes of lateness. [23:02:21] I'm not nobody [23:03:15] Hi wolfgang. Is this your first time using our tool to evaluate article feedback? [23:07:19] halfak: I've never used the article feedback tool [23:07:28] Mostly I'm just here because I'm curious [23:07:34] ahh [23:07:45] well, in the absence of anyone else: any questions you'd like answered? :) [23:07:52] No, not really [23:08:47] I mean, the answer to life, the universe, and everything would be nice, but I don't think this is the right place to ask. [23:08:54] 42 [23:09:08] Or rather, the question, not the answer [23:09:23] [[WP:42]] will do that [23:10:05] What's 6 x 9 (in base 13) [23:10:13] ^? [23:10:50] heh [23:10:58] hmn. It seems there is nobody :/ [23:11:06] I guess a lot of people have already started ;p [23:11:34] Indeed. We have 6 workers who have done a substantial amount, but also a lot in waiting. [23:15:45] It looks like cardsagainsthumanity.com just sent off $70k to the Wikimedia Foundation. [23:16:09] Donating to the WMF is becoming pretty popular. [23:16:16] cool [23:16:36] I was going to do it but I figured it'd basically be fiscal masturbation [23:16:41] (to be crude ;p) [23:16:41] Yeah lol. [23:16:51] mind you, most of us play CAH, so... [23:17:00] CAH? [23:17:06] first word "cards"... ;p [23:17:13] ahh [23:18:05] I didn't know it was a thing until they posted about donating. Good advertising. [23:19:26] inorite? But yeah, it's like "apples to apples" how you actually play it [23:19:48] most casual office hours ever [23:20:08] Missing Chzz? :) [23:20:10] Indeed. [23:20:46] I wonder if Chzz will ever come back [23:20:59] I have no comment at this juncture ;p [23:21:36] Well, this is kind of the opposite extreme though :) [23:21:53] you mean how nobody is swearing at us? ;p [23:22:04] $70000 is a pretty big donation [23:22:18] I cant see that being anything other than true generosity [23:22:37] it's not like they could actually get 5000 people to buy the game just because they donated to us [23:22:38] actually, to be frank [23:22:52] I suspect it's "we built this really cool thing everyone buys, we can afford to do this" in part. [23:23:01] I mean, if I was running them...that's the sort of thing I'd do. [23:23:15] I had never heard of them before [23:23:29] also you can apparently download the game for free, the paper version is just for convenience [23:23:39] so I think theyre not really in it for the money [23:23:46] not as much as some people anyway [23:24:02] it's a good party game though? [23:24:24] interesting name [23:25:26] it's a fantastic party game [23:25:41] assuming your friends are all disturbed (and mine are. Which creates an interesting chicken-and-egg problem) [23:28:00] Ironholds: you always miss me to ping :/ [23:29:02] Ironholds: and I can do CHzz's job ^^ [23:29:24] Ironholds: btw do you have any stats if somebody is actually READING the feedback? [23:29:37] moreover any study if that stuff changes the articles? [23:29:53] DarTar/ halfak are good people to comment on this - I know they've been looking into it :) [23:30:39] mabdul|busy: we are not logging visits to the FeedbackPage, but: [23:30:49] These are good questions that are surprisingly difficult to explore. Sadly, we don't have a very good means for editors to mark which feedback they used to fix an article. [23:31:23] 1) we have data on how many comments are moderated as a function of a source (article view, vs central page vs watchlist view) [23:32:00] We are also working to concentrate the usefulness of feedback so that people will get more out of the feedback page. We're building a machine learning architecture to help editors more efficiently find useful feedback. [23:32:20] Hopefully, once we are done, more editors will have good reason to check the feedback pages. [23:32:24] 2) we collected raw pageview data for a sample of AFT5-enabled articles to measure how much traffic they get: short answer, a ridiculously small amount of traffic [23:33:04] mmmh, ok. and do you have data about the person who gave the feedback if they have edited any page or created an account? [23:33:31] DarTar: did you exspected something different actually? [23:33:50] mabdul|busy: no, that confirms what we were expecting [23:34:03] DarTar: and let me guess: the pages with very high page views like the actual season of [Place any TV series in here] [23:34:09] mabdul: Yes (on editor conversions). Let me pull that up. [23:34:28] http://www-users.cs.umn.edu/~halfak/publications/Making_Peripheral_Participation_Legitimate/halfaker13making-preprint.pdf [23:34:48] no wmf server? XD [23:34:48] ^^^ This is the best summary we have on feedback quality and converted new editors at the moment. [23:35:00] mabdul|busy: even pages with a lot of traffic on the main article get very few visits to the feedback page [23:35:04] We have a writeup on meta. It is in less good shape. [23:36:16] halfak: i will read that in metro tomorrow. looks interesting actually [23:36:27] :) I'd be interested to have your feedback. [23:37:20] halfak: is that the most recent version of the preprint? I should update my copy [23:37:24] did you also created also a study related to the reviewers of the feedback? (I think that is something very interesting actually: some wikipedia-inhouse-sphere-study) [23:37:34] Yes. That is the best pre-print we have. [23:37:44] halfak: k [23:38:30] mabdul|busy: we have usage data on feedback reviewers, what questions are you interested in? [23:39:01] some of this data is available here: http://toolserver.org/~dartar/fp/ [23:39:13] heh. good question. *click* [23:40:27] mabdul|busy: drop us a line if you have any question that is not answered there --> User_talk:DarTar [23:40:59] will do, maybe christmas. I'm very productive working on my "to do quee" these days XD [23:41:04] halfak: should we add the acknowledgments to the preprint [23:41:06] ? [23:41:24] mabdul|busy: sounds good [23:41:30] what was autoflagged btw? [23:41:35] and autohide [23:41:47] DarTar: Hmmm... It looks like this is an old version after all. Let me fix it. [23:41:58] mabdul|busy: that's when the abuse filter goes "this is bad" or sufficient non-monitors mark it as bad for it to be automatically hidden [23:42:24] ah... makes sense [23:42:31] * DarTar waves at Ironholds [23:42:40] DarTar: huh? [23:43:03] I didm;t say hi and was not really lurking here until you pinged me [23:43:49] DarTar: That link is fixed now. [23:43:55] Complete with acks [23:44:00] great [23:44:25] ahh [23:44:35] This is the final submission, so if you find a typo, keep it to yourself (unless it is a problem). [23:44:42] DarTar: where's my SQL access? I've got nothing to do all christmas break ;p [23:44:49] Play Skyrim! [23:45:07] Ironholds: stat1 access still waiting on ops approval [23:45:39] andrew is ready to set you up as soon as we get the green light [23:45:59] DarTar: Who turns these lights green? [23:49:04] halfak: pm [23:51:30] well, that's ominous [23:51:41] Yeah. It's bad. [23:51:46] :P [23:52:39] OK Folks. Last call for the AFT feedback evaluation volunteering. Anyone out there? [23:53:23] tuuumbleweed. [23:54:03] I've never seen the stuff in my life and apparently it grows in IRC chatrooms. [23:55:13] heh! [23:55:43] I think I came a bit late. What's feedback evaluation? [23:57:32] We're working on building a set of classified feedback submissions from AFTv5. [23:57:38] legoktm: lol [23:58:16] legoktm: to translate it into english: we're asking people to rate a completely randomised set of feedback posts according to the normal criteria (oversight, useful, etc) [23:58:39] I built a tool to make it easy. Want to see? [23:58:48] http://toolserver.org/~grphack/aft/test.html [23:59:39] Ironholds: Thanks. Yeah I don't mind. How many do I need to do? and when?