[13:32:10] <halfak>	 o/
[13:32:51] <halfak>	 hey HaeB.  Are you around?  Looks like you're not away, but it's super early in PT. 
[13:33:07] <halfak>	 I was wondering if you knew when the next research newsletter might run. 
[14:11:19] <halfak>	 Looks like Ed17 is going to try to get the next issue out. 
[14:11:23] <halfak>	 https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/Newsroom
[15:13:36] <HaeB>	 halfak: yes, saw that
[15:15:27] <HaeB>	 ...i had been envisaging to publish it separately this week, but that would of course be preferabl
[15:24:00] <halfak>	 I'm thinking I'll bother Ed17 to find out how serious he is at some point today. 
[15:24:21] * halfak really wants to get his politically hazardous paper review out there. 
[16:02:49] <apergos>	 what paper is that, halfak?
[18:18:26] <halfak>	 Showcase will start in 12 minutes
[18:29:08] <halfak>	 Hey folks.  We're just about to kick things off. 
[18:31:36] <halfak>	 Stream should be live 
[18:31:42] <quiddity>	 working :)
[18:31:46] <halfak>	 great !
[18:31:47] <halfak>	 :) 
[18:31:48] * lzia claps for Research Showcase.
[18:31:55] <halfak>	 Ping me with questions
[18:32:28] <J-Mo>	 halfak is Shilad frozen for you?
[18:32:35] <halfak>	 Noppe
[18:32:42] <J-Mo>	 I can hear him, but his video is frozen. 
[18:32:49] <J-Mo>	 k. long as it's just me :)
[18:32:54] <halfak>	 Oh  Maybe he just hasn't started his slide deck
[18:33:01] <halfak>	 Hmmm yeah it is frozen
[18:33:04] <Nettrom>	 yeah, I was wondering the same thing
[18:33:19] <quiddity>	 (Reminder that you/we can adjust the default video quality with the cog icon, which sometimes auto-adjusts to 240)
[18:33:30] <quiddity>	 I saw mouse movement, video is fine. :)
[18:33:37] <halfak>	 good deal
[18:33:49] <lzia>	 halfak: it may be worth interrupting and asking Shilad to reshare his screen.
[18:33:53] <halfak>	 I messaged shilad to tell him we're not seeing things. 
[18:34:20] <Nettrom>	 thanks halfak!
[18:34:37] <halfak>	 I don't think he's talking through his first few slides. 
[18:34:50] <halfak>	 BTW: https://www.google.com/url?q=https%3A%2F%2Fcommons.wikimedia.org%2Fwiki%2FFile%3AShilad_Sen%2C_Wikimedia_Research_Showcase%2C_April_2017.pdf&sa=D&sntz=1&usg=AFQjCNHRrCnm0J-4x6MAtxhAplMEWXrbNQ
[18:34:54] <halfak>	 Darn you google
[18:35:01] <halfak>	 https://commons.wikimedia.org/wiki/File:Shilad_Sen,_Wikimedia_Research_Showcase,_April_2017.pdf
[18:35:02] <lzia>	 halfak: interrupt and ask for resharing.
[18:35:07] <lzia>	 it's unfortunate, it's hard to follow.
[18:36:41] <halfak>	 was workign in testing!
[18:36:51] <lzia>	 uhu. it happens. don't worry. 
[18:37:20] <lzia>	 thanks, halfak.
[18:38:06] <halfak>	 I had no idea bhecht was a macalester grad 
[18:38:13] <lzia>	 :P
[18:38:32] <halfak>	 http://atlasify.northwestern.edu/
[18:39:56] <halfak>	 http://matt.might.net/articles/crapl/
[18:40:40] <halfak>	 http://shilad.github.io/wikibrain/
[18:44:02] <halfak>	 lzia, some interesting stuff for recsys, right?
[18:46:18] <lzia>	 paying very careful attention, halfak. :)
[18:46:44] <halfak>	 I've been trying hard to get Wikibrain on Wikimedia infra so we can start playing around with it too :D 
[18:47:02] <quiddity>	 :)
[18:47:28] <halfak>	 https://meta.wikimedia.org/wiki/Grants:IEG/WikiBrainTools
[18:47:50] <halfak>	 https://phabricator.wikimedia.org/T161554
[18:47:51] <lzia>	 halfak: question for when there is question time. How is the mapping between a Wikipedia article and categories done?
[18:48:01] * halfak adds all the relevant links to the backchannel. 
[18:48:05] <halfak>	 lzia, got it
[18:51:06] <halfak>	 One of my fav. psych articles: https://en.wikipedia.org/wiki/Overjustification_effect
[18:52:28] <J-Mo>	 paper Shilad is talking about now: https://pdfs.semanticscholar.org/be0a/e35c4cf47c4ce543481f888bde8a4cfe2cd1.pdf
[18:52:40] <J-Mo>	 http://dl.acm.org/citation.cfm?id=2675285
[18:53:39] <halfak>	 I nominate this for best visualization of the year. Look at Africa!
[18:55:31] <halfak>	 This one: https://commons.wikimedia.org/w/index.php?title=File:Shilad_Sen,_Wikimedia_Research_Showcase,_April_2017.pdf&page=38
[18:56:56] <halfak>	 https://en.wikipedia.org/wiki/Tobler%27s_first_law_of_geography
[18:57:11] <Nettrom>	 yay Tobler!
[18:58:18] <halfak>	 https://movielens.org/
[18:58:25] <halfak>	 https://grouplens.org/datasets/movielens/
[19:00:10] <Ziko>	 great spatialisation
[19:01:14] <halfak>	 We're gonna be giving that a hug of death, I'm sure.  
[19:01:14] <halfak>	 https://en.wikipedia.org/wiki/Slashdot_effect
[19:05:08] <lzia>	 halfak: you may want to share https://meta.wikimedia.org/wiki/Research:Wikipedia_Navigation_Vectors with Shilad afterwards, in the context of word2vec.
[19:05:23] <halfak>	 lzia, I think he's just about to talk about it :D
[19:05:28] <halfak>	 He was working with ellery on it
[19:05:32] <halfak>	 Yup.  here's we go
[19:05:32] <halfak>	 :D
[19:05:36] <lzia>	 ok. :)
[19:06:00] <lzia>	 (there is some delay.;)
[19:07:01] <J-Mo>	 question for Shilad: how did they compare word2vec vs navigation vectors? 
[19:07:27] <J-Mo>	 just in terms of accuracy, or also a user study?
[19:08:34] <lzia>	 halfak: I should switch to another meeting. I'll check the YouTube link for the answer to my question if you get a chance to ask it. Thank you!
[19:08:47] <halfak>	 OK!
[19:08:50] <halfak>	 J-Mo, got it
[19:08:57] <J-Mo>	 ty
[19:09:28] <Taha>	 A question for later: is there any fundamental difference between a cartograph and a network visualisation based on Force Atlas algorithm or similar network layouts?
[19:10:07] <halfak>	 Taha, got it
[19:12:34] <halfak>	 \o/
[19:12:43] <halfak>	 https://figshare.com/articles/Monthly_Wikipedia_article_quality_predictions/3859800
[19:13:19] <Nettrom>	 \o/
[19:14:51] <Nettrom>	 I guessed MILHIST would be related to what he was about to say… same thing we found in our 2015 paper
[19:15:24] <halfak>	 Warncke-Wang, M., Ranjan, V., Terveen, L. G., & Hecht, B. (2015, April). Misalignment Between Supply and Demand of Quality Content in Peer Production Communities. In ICWSM (pp. 493-502). http://www.brenthecht.com/publications/wikipedia_supplydemandquality_icwsm2015.pdf
[19:16:05] <Nettrom>	 also https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#September_2015
[19:16:39] <halfak>	 Wrapping up.  Questions!
[19:17:00] <apergos>	 grar
[19:17:10] <apergos>	 I missed the showcase, was busy merging/deploying
[19:18:08] <halfak>	 switchover has been rough today :S
[19:18:40] <halfak>	 J-Mo, feel free to ask your own question
[19:18:41] <halfak>	 :) 
[19:18:44] <halfak>	 You're up next
[19:19:01] <halfak>	 I can ask too if you like :D
[19:19:03] <J-Mo>	 I will. Lot of background noise here, but I can ask quick and then re-mute
[19:19:16] <halfak>	 cool
[19:19:54] <ragesoss>	 really cool! that visualization of quality vs popularity for a topic cluster would be a great basis for an article finder designed to help people (like professors running Wikipedia assignments) to find high-impact low-quality articles to work on.
[19:19:56] <halfak>	 J-Mo, audio was very rough. 
[19:20:06] <J-Mo>	 yeah, I was afraid of that
[19:20:08] <apergos>	 I was actually in dumpsland but it was rought there too
[19:20:19] <apergos>	 anyways: slide deck? video? since some of us suck
[19:20:19] <halfak>	 ragesoss, right :DDDD
[19:20:29] <quiddity>	 No question, but oodles of appreciation. New ways to explore topics, and to add layers of metadata, is near the top of my wishlist.
[19:20:31] <halfak>	 Video is in topic
[19:20:36] <apergos>	 sweet
[19:20:37] <apergos>	 thx
[19:20:39] <quiddity>	 apergos, https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#April_2017
[19:20:41] <halfak>	 https://commons.wikimedia.org/wiki/File:Shilad_Sen,_Wikimedia_Research_Showcase,_April_2017.pdf
[19:20:50] * halfak keeps the wiki up to date 
[19:21:02] <J-Mo>	 I'm going to follow up with Shilad re: https://meta.wikimedia.org/wiki/Research:Evaluating_RelatedArticles_recommendations
[19:22:56] <halfak>	 Any more Qs?  I'm gonna start asking my own >:) 
[19:23:35] <isaacj>	 I'd be curious to hear him talk about the possibility of explaining the distances between points in Cartograph to users
[19:24:09] <isaacj>	 i.e. why are two points close by
[19:24:24] <halfak>	 isaacj, got it
[19:25:21] <halfak>	 https://en.wikipedia.org/wiki/Page_rank
[19:27:20] <isaacj>	 thanks!
[19:27:22] <halfak>	 Relevant: Vig, J., Sen, S., & Riedl, J. (2009, February). Tagsplanations: explaining recommendations using tags. In Proceedings of the 14th international conference on Intelligent user interfaces (pp. 47-56). ACM.  https://pdfs.semanticscholar.org/bcb8/1d41eabebf44a98b37673be66bf5b5bae13e.pdf
[19:28:30] <Nettrom>	 the project that Aaron mentioned that we’re working on is here: https://meta.wikimedia.org/wiki/Research:Automated_classification_of_article_importance and we have a draft lit. review of related works on https://meta.wikimedia.org/wiki/Research:Studies_of_Importance
[19:30:47] <basvb>	 thanks, interesting.
[19:30:51] <quiddity>	 Thanks Shilad, and thanks halfak for organizing!
[19:32:07] <halfak>	 \o/  Thanks for attending!  
[19:33:21] * halfak wipes brow.  Good show. 
[20:56:18] <matanya>	 halfak: planning any progress on https://phabricator.wikimedia.org/T162297 ?
[20:56:54] <halfak>	 matanya, I honestly haven't spent time on it since we last talked. :( 
[20:56:59] <halfak>	 I'm really sorry to say that. 
[20:57:20] <matanya>	 fair enough, you are a busy person
[21:00:35] <halfak>	 matanya, I think a good next step would be to write up a proposal to fund a research project into steward workload and the risk of not managing steward tasks well. 
[21:00:51] <halfak>	 I'd like to find the right grad student to advise in taking measurements. 
[21:01:17] <matanya>	 ok, what length kind of proposal ? FDC like ?
[21:02:49] <halfak>	 Project grant
[21:03:02] <matanya>	 ok
[21:14:23] <halfak>	 (Sorry jumped into a meeting)