[11:31:14] 10Jade, 10Scoring-platform-team (Current), 10DBA, 10Operations, and 2 others: Introduce a new namespace for collaborative judgments about wiki entities - https://phabricator.wikimedia.org/T200297 (10daniel) Quick meta-point: @Halfak wrote > This conversation is getting very Producty but if this is what Te... [14:24:15] (03PS1) 10Umherirrender: Improve param docs [extensions/JADE] - 10https://gerrit.wikimedia.org/r/482302 [14:51:04] (03CR) 10Hoo man: [C: 03+2] Improve param docs [extensions/JADE] - 10https://gerrit.wikimedia.org/r/482302 (owner: 10Umherirrender) [14:54:23] o/ [14:56:55] (03Merged) 10jenkins-bot: Improve param docs [extensions/JADE] - 10https://gerrit.wikimedia.org/r/482302 (owner: 10Umherirrender) [14:58:39] 10ORES, 10Scoring-platform-team (Current), 10Growth-Team, 10MediaWiki-Recent-changes, and 5 others: Wikidata recentchanges API request with ORESscores parameter returns 0 results - https://phabricator.wikimedia.org/T209459 (10Lydia_Pintscher) Can this be closed? [15:01:06] (03CR) 10jenkins-bot: Improve param docs [extensions/JADE] - 10https://gerrit.wikimedia.org/r/482302 (owner: 10Umherirrender) [15:03:04] 10ORES, 10Scoring-platform-team (Current), 10Growth-Team, 10MediaWiki-Recent-changes, and 5 others: Wikidata recentchanges API request with ORESscores parameter returns 0 results - https://phabricator.wikimedia.org/T209459 (10Addshore) 05Open→03Resolved Yup [15:36:40] (03PS6) 10Halfak: Updates for revscoring 2.3.0. [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/481210 (https://phabricator.wikimedia.org/T212530) [15:37:48] Looks like articlequality is fully updated. No longer a WIP. [15:40:12] Going to staging in wmflabs. [15:45:37] 10Scoring-platform-team (Current), 10articlequality-modeling, 10draftquality-modeling, 10drafttopic-modeling, and 3 others: Rebuild models for new revscoring (2.3.0) - https://phabricator.wikimedia.org/T212530 (10Halfak) I don't see an issue with updating articlequality from gerrit so I have removed the "(... [15:53:44] 10Scoring-platform-team (Current), 10articlequality-modeling, 10draftquality-modeling, 10drafttopic-modeling, and 3 others: Rebuild models for new revscoring (2.3.0) - https://phabricator.wikimedia.org/T212530 (10Halfak) I just added a `--force` to `git submodule update --init` in the fabfile.py and that l... [16:10:13] 10Scoring-platform-team (Current), 10articlequality-modeling, 10draftquality-modeling, 10drafttopic-modeling, and 3 others: Rebuild models for new revscoring (2.3.0) - https://phabricator.wikimedia.org/T212530 (10Halfak) I'm seeing a lot of this in the logs: ` [2019-01-04T06:25:36] statsd_send_metric()/sen... [16:27:39] wikimedia/editquality#438 (requirements_fix - 526f743 : halfak): The build passed. https://travis-ci.org/wikimedia/editquality/builds/475377286 [16:30:44] (03PS1) 10Halfak: Adds langdetect-1.0.7 [research/ores/wheels] - 10https://gerrit.wikimedia.org/r/482319 [16:30:58] (03PS2) 10Halfak: Adds langdetect-1.0.7 [research/ores/wheels] - 10https://gerrit.wikimedia.org/r/482319 (https://phabricator.wikimedia.org/T200297) [16:36:14] (03PS3) 10Halfak: Adds langdetect-1.0.7 [research/ores/wheels] - 10https://gerrit.wikimedia.org/r/482319 (https://phabricator.wikimedia.org/T212530) [16:36:56] 10Jade, 10Scoring-platform-team (Current), 10DBA, 10Operations, and 2 others: Introduce a new namespace for collaborative judgments about wiki entities - https://phabricator.wikimedia.org/T200297 (10Halfak) ^ Nevermind that. Was a copy-paste mistake [16:43:35] * halfak waits on a reviewer [16:52:19] * halfak contemplates self-merging. [16:52:30] I could just look at Jade for a bit. [16:52:32] I'll do that [16:56:41] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Diff view controls - https://phabricator.wikimedia.org/T212377 (10Halfak) [16:57:15] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10Halfak) [16:57:24] 10Jade, 10Scoring-platform-team, 10Design: Jade Wireframes: Entry view mode - https://phabricator.wikimedia.org/T212379 (10Halfak) [16:57:36] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Search integration - https://phabricator.wikimedia.org/T212380 (10Halfak) [17:42:10] o/ awight [17:48:46] heya [17:50:40] I have some quick reviews to help get me unblocked for the ORES deploy. [17:50:49] https://gerrit.wikimedia.org/r/482319 [17:51:12] https://github.com/wikimedia/editquality/pull/176 [17:51:25] The second one isn't really a blocker, but it's tightly related. [17:51:28] awight, ^ [17:51:30] sorry, won't be able to do it for a few hours [17:51:35] (away) [17:52:55] Gotcha. I might self-merge. It just adds the langdetect wheel needed by the new translatewiki model. Sound OK to you? [17:53:23] version 1.0.7. I built it on stat1007. I actually rebuilt all of the wheels to see if there were any secondary dependencies (nothign we don't already have) [18:00:22] (03CR) 10Halfak: [C: 03+2] Adds langdetect-1.0.7 [research/ores/wheels] - 10https://gerrit.wikimedia.org/r/482319 (https://phabricator.wikimedia.org/T212530) (owner: 10Halfak) [18:09:08] (03CR) 10Awight: [C: 03+2] "countersigned" [research/ores/wheels] - 10https://gerrit.wikimedia.org/r/482319 (https://phabricator.wikimedia.org/T212530) (owner: 10Halfak) [18:13:23] \o/ thanks [18:13:32] Looks like we're working on ores-staging [18:15:54] 10Scoring-platform-team (Current), 10articlequality-modeling, 10draftquality-modeling, 10drafttopic-modeling, and 3 others: Rebuild models for new revscoring (2.3.0) - https://phabricator.wikimedia.org/T212530 (10Halfak) https://github.com/wikimedia/ores-wmflabs-deploy/pull/101 includes the new `--force` o... [18:28:15] (03PS7) 10Halfak: Updates for revscoring 2.3.0. [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/481210 (https://phabricator.wikimedia.org/T212530) [18:32:18] (03CR) 10Awight: [C: 04-1] "Still needs the articlequality update" [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/481210 (https://phabricator.wikimedia.org/T212530) (owner: 10Halfak) [18:34:40] awight, do we have a specific task about the articlequality mirroring issue? [18:34:45] I can't seem to find one. [18:35:21] naw I was rattling cage bars for all of the issues at once, but didn't manage to make a task for the one that remained [18:35:28] doing so now [18:35:58] OK great. I'm going to step away for lunch. I'll be rattling something when I get back :) [18:36:27] the LFS objects failing to mirror for other repos is... quite the dealbreaker for our work process though [18:41:31] 10Scoring-platform-team, 10Gerrit, 10GitHub-Mirrors, 10Release-Engineering-Team: articlequality repo mirroring is broken - https://phabricator.wikimedia.org/T212962 (10awight) [18:41:51] 10Scoring-platform-team, 10Gerrit, 10GitHub-Mirrors, 10Release-Engineering-Team: articlequality repo mirroring is broken - https://phabricator.wikimedia.org/T212962 (10awight) p:05Triage→03High [20:43:55] I'm changing locations. Back online in ~30 minutes. [20:54:59] HareJ: I'd say our "hearts and minds" propaganda campaign is off to a strong start ;-) [21:31:23] awight, did you work on the expense report for the Berlin trip yet? [21:33:00] HareJ: no, it seems quite minimal [21:33:07] want to share a gdoc? [21:33:23] this is the template: https://docs.google.com/spreadsheets/d/1XYpkJJ29t2U8cE-xiRIz7cAMElZo-3D21q1Cul_dL2g/edit#gid=1483315145 [21:34:00] for per diem i plan on requesting for lunch and dinner for each of the days we were there, except for one dinner on friday night because halfak picked up the bill on that one, and breakfast was included with the rooms so I'm not including that. Does that sound right? [21:34:31] That is exactly what I remember. [21:34:36] maybe some train tickets [21:40:03] This coffeeshop blocks IRC :| [21:41:07] Rude [21:49:38] Oh wow. It doesn't. It just takes 10 minutes to start a connection! [21:49:42] Not gonna trust that. [21:56:37] I was thinking, what kind of coffee shop blocks IRC specifically [21:58:28] "There's this technology from the 80s someone might try to use, better block it just in case." [22:03:36] ^ HTTP explodes. [22:08:06] Re. terminology around Jade. https://en.wikipedia.org/wiki/Wikipedia:Polling_is_not_a_substitute_for_discussion#Use_of_polls_when_discussing_Wikipedia_articles [22:08:31] There's a lot of cautionary language about premature polling and not employing democracy. But I think we can harvest terminology from this. [22:08:39] We get "proposal" and "endorsement" [22:10:33] See also https://en.wikipedia.org/wiki/Wikipedia:Requests_for_comment [22:12:49] 10Jade, 10Scoring-platform-team, 10Design: Jade Wireframes: Entity view mode - https://phabricator.wikimedia.org/T212379 (10Halfak) [22:13:17] ^ working on spec'ing out what belongs in the wireframe [22:17:54] What work product should I expect from this exercise? [22:19:29] 14 [22:19:34] Some wireframes :) [22:19:38] Exciting! [22:29:06] Yes. I'm working myself into the mood and getting my subconscious working. [22:29:17] Gotta get some Jade shower-thoughts. [22:29:27] That are not related to social-dynamics [22:40:16] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10Halfak) Looking at this now. I'm wondering if we're looking for anything beyond designing the filter-controls for the new changes filters. E.g. "show me changes... [22:41:35] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Diff view controls - https://phabricator.wikimedia.org/T212377 (10Halfak) [22:49:59] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10Harej) Right, we shouldn't need anything terribly novel in terms of design here. A wireframe would just show what the current watchlist feed would look like with... [22:54:51] 10Jade, 10Scoring-platform-team, 10Design: Jade Wireframes: Entry edit mode - https://phabricator.wikimedia.org/T212374 (10Halfak) Would this necessarily be separate from the view mode? I'm guessing not, but it might be good to spec out a set of patterns for making various types of judgment data editable.... [22:55:14] awight: just wondering. How clearly does releng know that we are blocked on them and gerrit LFS stuff? [22:55:45] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10awight) At one point we were looking at nesting the judgments in new RC layouts, either indenting under the target revision if it also appears in the page of resu... [22:56:05] 10Jade, 10Scoring-platform-team, 10Design: Jade Wireframes: Entry edit mode - https://phabricator.wikimedia.org/T212374 (10Harej) Even if they are the same page, a page might look different while you're in edit mode compared to if you're just reading it. (It might not be, though! Depends on what you come up... [22:57:17] halfak: No one has been alerted specifically, yet, I've left this as a "High" prio task in the normal places. [22:57:38] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10Halfak) We could still put the consensus judgment information inline for many judgement types. It would be very useful for editquality. It might be less useful... [22:57:41] It seems fine to push on that task a bit, since getting new models out will be so nice. [22:57:59] Is there anything else urgent in revscoring 2.3? [22:58:23] CIDR will be great, but FU-berlin is not kicking our doors in at the moment [22:59:12] Other than the models, the biggest change is fixing some schema stuff that -analytics wants for EventStream stuff. [22:59:32] It makes me uncomfortable to not be able to deploy. [22:59:35] also nice to have, for sure! [22:59:40] haha welcome to my LIFE [22:59:43] lol [23:00:18] I'm wrapping up for the day. Thanks for the discussion in the Jade tasks. I didn't get to building the wireframes yet, but I have a good sense for what I want to work on first. [23:00:19] There have been regressions in our dev pipeline, for sure. [23:00:32] It's probably time for another push on that generally [23:00:56] * HareJ wonders how the Trust and Safety people got involved in the Jade discussion [23:01:02] * HareJ is happy they are, for the record. [23:01:06] halfak: I was thinking after the fact, that we're getting hearts and minds propaganda results already. [23:01:21] We should have more of these "workshops" eh [23:01:39] Re. dev pipeline. Good idea. Let's use gum in the short term while we focus on Jade though. [23:01:40] every friday, video and irc office hours [23:01:59] I think it might be good if we have more meetings like these with certain stakeholders. [23:02:00] +1 for the workshop. Margeigh has been organizing a lot of multi-stakeholder meetings for product strategy things. [23:02:16] Maybe not everyone we can think of, but certain key people who we think could be useful for the project. [23:02:20] halfak: the gum was blown clear off the hull by this articlequality snafu [23:02:22] We could do something like that and I think she might even help. :) [23:02:36] s/articlequaliy/LFS [23:02:41] ^ [23:02:43] yeah [23:03:17] HareJ & awight: if we were to engage more purposefully in some workshops, I think we might "accidentally" socialize some of our teams needs and why we're super awesome as well. [23:03:27] :D [23:03:38] I think a little sunlight never hurts [23:03:43] Is there like a calendar of these workshops? I'm happy to go to meetings but I'm usually not aware of them. [23:04:13] Haven't been any I have been invited to in a while. I think they were the precursor to the 3-5 year strategy documents they have been working on. [23:04:20] E.g. augmentation, scale, etc. [23:04:49] sounds like a fun way to try out Jade ideas, either way. [23:05:20] sort of like a general council of self-managing Wikimedia projects would proceed in a parallel, hippie'd out universe. [23:20:57] 10Jade, 10Scoring-platform-team, 10User-Testing, 10Design: Jade Wireframes: Watchlist integration - https://phabricator.wikimedia.org/T212378 (10awight) >>! In T212378#4856318, @Halfak wrote: > We could still put the consensus judgment information inline for many judgement types. It would be very useful f...