[03:54:41] Hi all [10:57:22] halfak_ https://phabricator.wikimedia.org/T117999 :D [10:57:35] I'll review the work on other languages too [10:57:42] often people report things in the wrong place :p [16:04:24] Cool! I [16:04:28] 'll pick that up today [16:04:44] * halfak_ is doing work work when he should be doing fun work :( [17:45:00] o/ kenrick95 [17:45:37] FYI, we just added support in ORES for a ton of wikis -- including idwiki. [17:45:55] https://meta.wikimedia.org/wiki/Objective_Revision_Evaluation_Service#Support_table [17:46:18] I'm working on announcement stuff today [18:04:40] * yuvipanda waves at halfak [18:05:03] * yuvipanda is helping do a mediawiki userscripts workshop today [18:05:11] Hey yuvipanda! [18:05:13] halfak: any new ideas for ORES related userscripts let me know [18:05:30] Yeah! You could show people ScoredRevisions [18:05:37] It's super simple and could be extended to do more. [18:05:44] right so ideas for extension [18:05:46] https://github.com/he7d3r/mw-gadget-ScoredRevisions [18:06:04] maybe file issues? [18:06:11] Hmm... Can't do that right now. [18:06:43] or just tell me :D [18:06:47] and I'll file them [18:06:50] not just now [18:08:43] The scoring strategy for Category pages is kind of lame. I honestly don't understand it. [18:09:15] The system also does not use the wp10 models at all. That might be very nice for category pages. [18:09:37] * halfak looks for an example [18:10:14] It would be great if pages like this had article quality assessments from ORES: https://en.wikipedia.org/wiki/Category:Stub-Class_medicine_articles [18:10:37] You'd need to figure out the associated article for each talk page and provide the most recent revision to ORES for assessment. [18:11:52] So far, we have been doing this manually. See https://meta.wikimedia.org/wiki/Research:Screening_WikiProject_Medicine_articles_for_quality [18:15:42] Hey kenrick95, I responded to your email. [18:15:50] Come talk to me :D [18:16:27] halfak: aah the enwp10 model sounds nice to use [18:16:51] halfak: do we even have a thing that just shows people the quality ratings per page? [18:16:53] :D [18:17:02] * yuvipanda adds to list [18:17:03] yuvipanda, good Q. I don't think so. :) [18:17:18] BTW, we have an article quality model deployed for frwiki too. :) [18:17:30] The AI hydra gets stronger. [18:17:47] nice [18:17:57] yeah so that sounds like a nice thing to build [18:20:30] +1 [18:20:46] Where are you giving the workshop? [18:20:50] yuvipanda, ^ [18:21:11] halfak: the office [18:21:17] this was planned weeks ago [18:21:19] so [18:21:28] halfak: we capped it at 20 people and have a lot more interest! [18:21:38] Gotcha. Remote participants? IRC channel? [18:21:42] halfak: and I really like doing workshops, and had stopped for a few years... [18:22:08] halfak: no, unfortunately. in my experience workshops for people getting started don't work well when 'mixed' - gotta be full remote or full in person [18:22:24] yuvipanda, hokay. [18:22:29] halfak: but I can try organize a full remote one in a few weeks [18:22:50] Makes sense. [18:22:54] this is the first one in the office, so [18:23:00] BTW, this is all up to date if you want to point people to it: https://meta.wikimedia.org/wiki/Objective_Revision_Evaluation_Service [18:23:02] halfak: are you interested in learning how to write userscripts / gadgets? [18:23:06] halfak: aesome yes [18:23:15] yuvipanda, rather, interested in helping out. :) [18:23:22] :D [18:23:24] * halfak writes lots of userscripts / gadgets [18:23:30] Or... enough I guess. [18:23:35] halfak: so one thing I wanted to do was to do a "AI in Wikimedia" workshop [18:24:01] Yeah. That's something I'd really like to do at the MW Dev summit, but we're supposed to have "discussions" instead. [18:24:17] halfak: I can organize this for right after the dev summit if you can stay a bit extra [18:24:58] usually the workshops I organize are targetted towards answering the question "I know X technology, how can I use it to help wikimedia?" and since most of them are super basic they end up being filled by people outsie the movement [18:25:07] I've been talking to people about doing that. We'll need to work it out soon so that switching my flights is cheap/free. [18:25:52] yeah, +1 [18:25:54] I'll push on DarTar a bit and get back to you. [18:26:00] do include me in those conversations! [18:26:03] oh ok [18:26:05] cool [18:28:20] How many hours do you devote to the workshop? [18:28:25] yuvipanda, ^ [18:28:45] halfak: to the entire organizing etc? [18:29:04] halfak: I have spent about 1.5-2h over the last couple weeks co-ordinating with faciliites to get the space ready. [18:29:08] Na. Just day-of [18:29:14] halfak: ah. so this one is just 3h [18:29:17] Cool [18:29:19] mostly because it's new [18:29:31] the ones I enjoyed the most when I helped run them were 3 *nights* long [18:29:39] started at 9PM and went on till 4-5AM [18:29:55] made some amazing friends during those. super intense and fun [18:30:01] don't think those'll happen here though [18:31:06] I wouldn't mind a serious hackathon event around the dev summit. [18:31:29] +1 [18:31:52] it's probably not going to be organized by the current organizers of the dev summit itself though [18:52:35] Yeah. It seems they really don't want to do that. [18:52:41] * halfak likes hackathons a lot [18:52:49] Workshop-athons :D [20:32:43] See https://meta.wikimedia.org/wiki/Talk:Objective_Revision_Evaluation_Service#More_Artificial_Intelligence_for_your_quality_control.2Fcuration_work. [20:33:31] And I'm off. Have a good one folks! [20:33:31] o/