[00:21:01] 10Scoring-platform-team (Current), 10drafttopic-modeling: Rewrite draft topic scripts to fetch linked pages and prepare training data - https://phabricator.wikimedia.org/T193594#4173355 (10awight) [06:54:05] PROBLEM - puppet on ORES-web01.Experimental is CRITICAL: CRITICAL: Puppet has 20 failures. Last run 1 minute ago with 20 failures. Failed resources (up to 3 shown): Service[ssh],Service[atop],Package[ntp],Service[systemd-timesyncd] [07:23:19] RECOVERY - puppet on ORES-web01.Experimental is OK: OK: Puppet is currently enabled, last run 30 seconds ago with 0 failures [11:27:28] 10Scoring-platform-team (Current), 10articlequality-modeling, 10User-Ladsgroup, 10artificial-intelligence: Train and test wp10 model for fawiki - https://phabricator.wikimedia.org/T190050#4174281 (10Ladsgroup) Announced the new campaign. [12:13:23] 10Scoring-platform-team, 10ORES: hspell warnings during extraction are annoying - https://phabricator.wikimedia.org/T192937#4174410 (10Ladsgroup) p:05Triage>03Lowest [12:53:32] o/ [14:29:36] o/ [14:32:39] halfak: o/ i saw adam's pings [14:32:47] Hey! [14:33:07] So yeah, it looks like there might have been a big hiccup in training/testing :| [14:33:30] Hey Amir1 :) [14:33:45] halfak: no, i don't think there's anything to worry, it seems a bug that got introduced when I was cleaning up for production :D [14:34:47] halfak: if you remember i was using two scripts initially for experiments one to fetch text - https://github.com/wiki-ai/drafttopic/pull/14 and one to extract dependents - https://github.com/wiki-ai/drafttopic/pull/15 [14:35:23] halfak: the fetch text script used the "talk_page_title" key to get the article contents since that was the only thing linking talk pages to their content namespaces [14:35:49] Oh great! [14:35:55] That's very good news :) [14:36:08] halfak: but when I deprecated those scripts in favour of `revscoring extract` i missed the point of why i had those in the first place - because rev scoring extract cannot fetch using article title [14:37:04] halfak: just because `rev_id` was present as a key it seemed to work out of the box, but was fetching wrong information, so i think i need to restore those 2 scripts [14:37:05] Seems like the rev_id in the dataset you uploaded to figshare should be the content page rev_id. [14:37:27] Could we update the scripts you used to gather that data and update the dataset? [14:37:51] halfak: yes, but we don't have any way to get that without making another query everytime which seems inefficient, i can rename that to sth like "talk_rev_id" [14:38:34] halfak: yes I'll update the dataset and re-introduce the fetch_text and extract_from_text which I think are needed [14:38:59] halfak: thankfully, because the results were copied from ACL draft which in turn came from experiments, we're good :D [14:39:31] What about the original query you made to gather the sample? [14:40:41] halfak: original query was fetching article contents using talk page title - https://github.com/wiki-ai/drafttopic/pull/14/files#diff-06429832f70700145e96dce2bf92f69fR93 [14:40:59] 10Scoring-platform-team (Current), 10JADE: [discuss] JADE schema format (endorsements?) - https://phabricator.wikimedia.org/T193643#4174967 (10Halfak) [14:47:43] https://github.com/wiki-ai/wikilabels/pull/236 [14:47:44] :D [14:48:14] 10Scoring-platform-team, 10Wikilabels, 10Easy: pytests for wikilabels - https://phabricator.wikimedia.org/T171082#4175055 (10Ladsgroup) [14:48:16] 10Scoring-platform-team, 10Wikilabels, 10Google-Code-in-2017: Provide a pytest for database of wikilabels - https://phabricator.wikimedia.org/T179014#4175054 (10Ladsgroup) 05Open>03Resolved [14:48:52] 10Scoring-platform-team (Current), 10Wikilabels: The great test making of Wikilabels - https://phabricator.wikimedia.org/T171080#4175067 (10Ladsgroup) [14:48:53] 10Scoring-platform-team, 10Wikilabels, 10Easy, 10User-Ladsgroup: pytests for wikilabels - https://phabricator.wikimedia.org/T171082#4175068 (10Ladsgroup) 05Open>03Resolved a:03Ladsgroup [14:49:21] 10Scoring-platform-team (Current), 10JADE: [discuss] JADE schema format (endorsements?) - https://phabricator.wikimedia.org/T193643#4175071 (10Halfak) It could look something like this: ``` { "entity": { "type": "diff", "rev_id": 376901 }, "schemas": { "damaging": [... [14:49:46] wiki-ai/wikilabels#358 (badges - f074a00 : Amir Sarabadani): The build passed. https://travis-ci.org/wiki-ai/wikilabels/builds/373968633 [14:50:59] halfak: i'm thinking of adding extract_from_text using offile `text` to revscoring, as revscoring currently can only extract using rev_id's ,that sounds ok? [14:51:24] *offline [14:51:25] (sorry got side-tracked) [14:51:26] looking now [14:51:59] If you think this is easier, let's do it that way. [14:55:53] both ways is same, i think having in revscoring wil allow other projects to use it in future [14:57:17] Not sure what you mean by that. [14:57:23] Fetch_text shouldn't be in revscoring [14:58:07] And I was thinking that fixing the original dataset we pull from figshare was one way, while adding a fetch_text utility that gets the first revision based on title to drafttopic would be the other way [14:58:10] codezee, ^ [14:59:08] AFK for a bit [15:01:18] halfak: okay, but regarding extracting the first revision, I think we discussed in one of the very earliest meetings that we can't know in which revision the article got tagges with a particular WikiProject tag so it seemed natural to take the latest revision [15:01:29] *got tagged [15:01:50] it need not be the first revision, when an article gets tagged [15:08:07] codezee, when it got tagged doesn't matter. [15:08:31] It was always in a content space whether it was tagged or not. [15:08:37] We're predicting future taggings. [15:08:42] Using the first revision text [15:15:58] I'm going to step away from the computer for a little while. I'll likely miss staff meeting. I gave Amir1 a few updates & questions to deliver for me. [15:17:34] o/ [16:44:15] Amir1: There is a thing I should have mentioned. [16:44:29] Are you... looking for things to do, by any chance? [16:45:01] codezee: hi! Did you see my question in the drafttopic repo? [16:45:49] It's possible that I'm wrong, but it looks like we trained on the latest revision of talk pages, when we should have been pulling the rev_id of the oldest revision of content pages. [16:48:03] Amir1: ^ this is what I was going to mention to you as well. There's some fun (not sarcastic) coding to do, to scrape things via MW API. [16:48:27] halAFK: Is there a deadline for the drafttopic paper, before which we might be able to revise the data and results? [17:10:28] amir1, awight: Sorry about this, I've hit a silly error I can't seem to fix. If either of you has time, I'd appreciate another pair of eyes on this. I'm trying to install/run that talk page cleaner that Aaron sent the other day, and I've cloned the git repository, but when I try to install it, I get the following error: [17:10:28] https://en.wikipedia.org/w/index.php?title=User:Ewitch51/sandbox&action=edit&redlink=1&preload=Template%3AUser+sandbox%2Fpreload [17:11:47] ewhit_: That link isn't working for me... [17:11:48] I've tried Googling, and it looks like I might need to re-install Python. Although I can't tell if that's the fix, or if it's something to do with PsychoPy (a program I have for another project). Can either of you tell more about what's happening here? Sorry, I know this is probably a simple fix, I just can't figure out what it is! [17:12:03] maybe the page isn't saved yet? [17:12:51] ah, yes, thank you! [17:12:53] should be fixed [17:13:02] k I see it [17:13:36] The problem is just that it's trying to install stuff globally-- use a virtualenv instead, for sanity. [17:13:45] Also, is the project python2 or 3? [17:13:47] (or both?) [17:13:50] ah ok. Thank you! Python 3 [17:14:11] virtualenv -p python3 VENV_DIR_OF_YOUR_CHOOSING [17:14:24] ooh, perfect, thank you! [17:14:32] I usually do ~/venv to be lazy and reuse across projects, or PROJECT_DIR/.venv if I want isolation [17:14:57] u familiar with the venv dance? [17:15:17] Thank you, I will save this info! I am currently not, but will look up the documentation for it [17:15:29] once you have the venv dir, you'll need to "source venv/bin/activate" to make the virtualenv python the default for that shell session. [17:15:37] Has to be re-run in each shell fwiw [17:15:48] blergh gross, thanks for letting me know! [17:15:51] hehe [17:16:14] once you have the venv activated, I recommend "pip install -r requirements.txt" and "pip install -e ." for this kind of project [17:17:04] the "-e" option is like "python setup.py develop" instead of "python setup.py install", it means that the repo is linked into your python libraries rather than copied, so any changes you make will be reflected immediately [17:17:26] ahhh ok [17:37:56] awight: i've explained to Aaron, it isn't any problem with the paper or repo, just that during cleanup for deployment I did away with the `fetch_text` and dependent extraction scripts in favour of revscoring extract [17:38:42] awight: what i failed to realize at the time was I had introduced those scripts for the purpose of extracting article contents using "talk_page_title" which is the only link that connects talk pages with their content pages [17:39:26] codezee: Is the figshare data what was used to train the checked-in model? [17:39:37] awight: see this - https://github.com/wiki-ai/drafttopic/pull/14/files#diff-06429832f70700145e96dce2bf92f69fR93 this is the original script that was used in experiments [17:40:00] cool, thanks for the pointer [17:40:03] awight: yes the figshare dataset was used but the "rev_id" is useless there, "talk_page_title" is of interest [17:40:29] awight: i'll revert back the two scripts which I had not merged and declined [17:40:43] hmm, this is hard to parse through... [17:41:02] awight: you mean the script or this whole thing? [17:41:50] i'll also cleanup figshare and remove rev_id which I don't think is useful [17:41:54] Would you mind revising https://meta.wikimedia.org/wiki/Research:Automatic_new_article_topics_suggestion#Results to explain what was done to arrive at the model used in the paper? [17:44:33] IMO it would be best to check in the scripts actually used, and maybe put "DEPRECATED" or something in there. [17:44:58] or uh if it's in git history but deleted, that's fine too. Just as long as it's in master and not an abandoned branch. [17:47:05] i'll see where all i have to make things clearer [17:47:51] codezee: I'd be happy to go through with you, I tried to repeat the experiment yesterday... [17:49:54] codezee: ah also, you might be interested in the Makefile changes in my PR [17:51:13] I tried to automate as much as possible, and you can see where my attempts broke down [17:52:07] Pretty sure that revscoring extract will use the rev_id and nothing else, which is why I'm poking at what was actually used for the paper's data [17:53:13] awight: this should fix things - https://github.com/wiki-ai/drafttopic/pull/22 i'll test once on ores-misc before its merge ready [17:57:10] awight: can we push the meeting to one hour from now? [17:57:18] I need to be somewhere before it closes [17:57:28] (on-time Germans :) [18:06:36] Amir1: lol, certainly. I'll do a rough pass myself [18:08:37] cool [18:29:41] awight: i'll add the makefile rules in some time, that'll make things much easier in the next PR [18:31:02] codezee: btw I think I have a sane plan for how to replace PAWS with remote API requests... [18:32:01] awight: that sounds nice... [18:32:31] although i'm not sure if api can handle queries containing list of hundreds of wikiprojects [18:32:40] which barred me in the first place from taking that approach [18:38:44] codezee: I was thinking we could iterate over projects, using these calls, https://en.wikipedia.org/w/api.php?action=query&generator=embeddedin&geititle=Template:WikiProject%20Accessibility&geinamespace=1&prop=info [18:39:13] that gives the list of all talk pages with the project template, then we sample 2,000 randomly. [18:45:12] 10Scoring-platform-team (Current), 10Research, 10Research-outreach, 10Epic, 10Paper: [Epic] Write paper about ORES as a socio-technical probe - https://phabricator.wikimedia.org/T121719#4176019 (10awight) 05Open>03Resolved [18:45:14] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Train and test damaging/goodfaith model for arwiki - https://phabricator.wikimedia.org/T189710#4176022 (10awight) [18:45:16] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Complete edit quality campaign for Arabic Wikipedia - https://phabricator.wikimedia.org/T131669#4176020 (10awight) 05Open>03Resolved a:03awight [18:45:18] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Train/test damaging & goodfaith models for Latvian Wikipedia - https://phabricator.wikimedia.org/T163006#4176025 (10awight) [18:45:20] 10Scoring-platform-team (Current), 10Wikilabels, 10editquality-modeling, 10artificial-intelligence: Complete Latvian Wikipedia editquality campaign - https://phabricator.wikimedia.org/T163005#4176023 (10awight) 05Open>03Resolved a:03awight [18:45:21] that seems reasonable, we can take first X results of each template, mix them up and sample 2000 [18:45:22] 10Scoring-platform-team (Current), 10ORES, 10Collaboration-Team-Triage (Collab-Team-This-Quarter): Deploy ORES advanced editquality models to lvwiki - https://phabricator.wikimedia.org/T192499#4176027 (10awight) [18:45:24] 10Scoring-platform-team (Current), 10editquality-modeling, 10Epic, 10artificial-intelligence: [Epic] Edit quality models (damaging/goodfaith) - https://phabricator.wikimedia.org/T130213#4176029 (10awight) [18:45:26] 10Scoring-platform-team, 10MediaWiki-extensions-ORES: Deploy ORES Review Tool for Latvian Wikipedia - https://phabricator.wikimedia.org/T163007#4176028 (10awight) [18:45:29] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Train/test damaging & goodfaith models for Latvian Wikipedia - https://phabricator.wikimedia.org/T163006#3183157 (10awight) 05Open>03Resolved [18:45:31] 10Scoring-platform-team, 10MediaWiki-extensions-FlaggedRevs, 10ORES: Decrease FlaggedRevs backlog by using ORES predictions models - https://phabricator.wikimedia.org/T165848#4176031 (10awight) [18:45:34] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Flagged revs approve model to fiwiki - https://phabricator.wikimedia.org/T166235#4176030 (10awight) 05Open>03Resolved [18:45:36] 10Scoring-platform-team (Current), 10articlequality-modeling, 10User-Ladsgroup, 10artificial-intelligence: Train and test wp10 model for fawiki - https://phabricator.wikimedia.org/T190050#4176034 (10awight) [18:45:38] 10Scoring-platform-team (Current), 10editquality-modeling, 10Epic, 10User-Ladsgroup, 10artificial-intelligence: [Epic] Implement code generation for model makefile maintenance - https://phabricator.wikimedia.org/T168455#4176032 (10awight) 05Open>03Resolved [18:45:40] 10Scoring-platform-team (Current), 10articlequality-modeling, 10User-Ladsgroup, 10artificial-intelligence: Article quality campaign for Persian Wikipedia - https://phabricator.wikimedia.org/T174684#4176033 (10awight) 05Open>03Resolved [18:45:42] 10Scoring-platform-team, 10editquality-modeling, 10artificial-intelligence: Complete edit quality campaign for Bengali Wikipedia - https://phabricator.wikimedia.org/T187339#4176036 (10awight) [18:45:46] 10Scoring-platform-team (Current), 10Wikilabels, 10editquality-modeling, 10Bengali-Sites, 10artificial-intelligence: Edit quality campaign for Bengali Wikipedia - https://phabricator.wikimedia.org/T174878#4176035 (10awight) 05Open>03Resolved [18:46:20] 10Scoring-platform-team (Current), 10MediaWiki-extensions-ORES, 10User-Ladsgroup: Scream properly when threshold stats fetch is failing - https://phabricator.wikimedia.org/T180587#4176038 (10awight) 05Open>03Resolved [18:46:23] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Complete edit quality campaign for Arabic Wikipedia - https://phabricator.wikimedia.org/T131669#4176037 (10alanajjar) Thanks all [18:46:26] 10Scoring-platform-team (Current), 10Release-Engineering-Team (Watching / External), 10Wikimedia-Incident: [Spike] Write reports about why Ext:ORES is helping cause server 500s and write tasks to fix - https://phabricator.wikimedia.org/T181010#4176040 (10awight) [18:46:29] 10Scoring-platform-team (Current), 10Operations, 10Patch-For-Review, 10Release-Engineering-Team (Watching / External), 10Wikimedia-Incident: Cache ORES virtualenv within versioned source - https://phabricator.wikimedia.org/T181071#4176039 (10awight) 05Open>03Resolved [18:46:34] 10Scoring-platform-team (Current), 10MediaWiki-extensions-ORES, 10MW-1.31-release-notes (WMF-deploy-2018-03-20 (1.31.0-wmf.26)), 10Patch-For-Review: Migrate ORES extension threshold config from old to new syntax - https://phabricator.wikimedia.org/T181159#4176041 (10awight) 05Open>03Resolved [18:46:39] 10Scoring-platform-team (Current), 10MediaWiki-extensions-ORES, 10MW-1.31-release-notes (WMF-deploy-2018-02-27 (1.31.0-wmf.23)), 10Patch-For-Review, 10User-Ladsgroup: Redesign onRecentChange_save hook handler for ORES - https://phabricator.wikimedia.org/T181335#4176042 (10awight) 05Open>03Resolved [18:46:48] 10Scoring-platform-team (Current), 10drafttopic-modeling: Fix response processing logic in drafttopic.fetch_page_wikiprojects - https://phabricator.wikimedia.org/T181522#4176043 (10awight) 05Open>03Resolved [18:46:58] 10Scoring-platform-team (Current), 10ORES, 10articlequality-modeling, 10User-Ladsgroup, 10artificial-intelligence: Rename "wikiclass" to "articlequality" - https://phabricator.wikimedia.org/T181820#4176044 (10awight) 05Open>03Resolved [18:47:08] 10Scoring-platform-team (Current), 10JADE, 10MediaWiki-Vagrant, 10Patch-For-Review: Vagrant role for JADE - https://phabricator.wikimedia.org/T182055#4176045 (10awight) 05Open>03Resolved [18:47:16] 10Scoring-platform-team (Current), 10ORES, 10Patch-For-Review: Make sure ORES is compatible with stretch - https://phabricator.wikimedia.org/T182799#4176047 (10awight) [18:47:19] 10Scoring-platform-team (Current), 10ORES, 10Patch-For-Review: Rebuild ORES wheels on Stretch - https://phabricator.wikimedia.org/T184135#4176046 (10awight) 05Open>03Resolved [18:48:00] 10Scoring-platform-team (Current): Host Google-News-word2vec.bin publicly - https://phabricator.wikimedia.org/T185147#4176050 (10awight) 05Open>03Resolved [18:48:03] 10Scoring-platform-team (Current), 10ORES, 10Collaboration-Team-Triage (Collab-Team-This-Quarter): Deploy ORES advanced editquality models to huwiki - https://phabricator.wikimedia.org/T192496#4176052 (10awight) [18:48:05] 10Scoring-platform-team (Current), 10editquality-modeling, 10User-Ladsgroup, 10User-Tgr, 10artificial-intelligence: Train/test damaging and goodfaith model for Hungarian Wikipedia - https://phabricator.wikimedia.org/T185903#4176051 (10awight) 05Open>03Resolved [18:48:15] 10Scoring-platform-team (Current), 10User-Ladsgroup: Fix or decide on edge cases of Makefile - https://phabricator.wikimedia.org/T186453#4176053 (10awight) 05Open>03Resolved [18:48:18] 10Scoring-platform-team (Current), 10Wikilabels: WikiLabels OAuth handshake doesn't work with HTTPS - https://phabricator.wikimedia.org/T186557#4176054 (10awight) 05Open>03Resolved [18:48:20] 10Scoring-platform-team (Current), 10Wikilabels, 10editquality-modeling, 10Bengali-Sites, 10artificial-intelligence: Edit quality campaign for Bengali Wikisource - https://phabricator.wikimedia.org/T186711#4176055 (10awight) 05Open>03Resolved [18:48:30] 10Scoring-platform-team (Current), 10JADE: Deploy JADE prototype in Beta Cluster - https://phabricator.wikimedia.org/T176333#4176057 (10awight) [18:48:32] 10Scoring-platform-team (Current), 10JADE: Build prototype JADE extension - https://phabricator.wikimedia.org/T187216#4176056 (10awight) 05Open>03Resolved [18:48:34] 10Scoring-platform-team (Current), 10ORES, 10Collaboration-Team-Triage (Collab-Team-This-Quarter): Deploy ORES advanced editquality models to cawiki - https://phabricator.wikimedia.org/T192501#4176059 (10awight) [18:48:36] 10Scoring-platform-team (Current), 10User-Ladsgroup: Train/test damaging/goodfaith model for Catalan Wikipedia - https://phabricator.wikimedia.org/T187732#4176058 (10awight) 05Open>03Resolved [18:48:38] 10Scoring-platform-team (Current), 10User-Ladsgroup: Edit quality campaign for Catalan Wikipedia - https://phabricator.wikimedia.org/T187771#4176060 (10awight) 05Open>03Resolved [18:48:40] 10Scoring-platform-team (Current), 10Global-Collaboration, 10MediaWiki-extensions-ORES: [Epic] Audit of pending ORES GUI deployments - https://phabricator.wikimedia.org/T187836#4176061 (10awight) 05Open>03Resolved a:03awight [18:48:51] 10Scoring-platform-team (Current), 10ORES: New precache endpoint isn't reporting its metrics correctly - https://phabricator.wikimedia.org/T187914#4176063 (10awight) 05Open>03Resolved [18:48:54] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES, 10Services (done): Change ORES rules to send all events to new "/precache" endpoint - https://phabricator.wikimedia.org/T158437#4176064 (10awight) [18:48:56] 10Scoring-platform-team (Current), 10Research-outreach, 10drafttopic-modeling: Present about draft topic model at Wikimedia Research Showcase. - https://phabricator.wikimedia.org/T188123#4176065 (10awight) 05Open>03Resolved [18:48:59] 10Scoring-platform-team (Current), 10artificial-intelligence: Build slide deck about AI at Wikimedia for Policy People - https://phabricator.wikimedia.org/T188124#4176066 (10awight) 05Open>03Resolved [18:49:01] 10Scoring-platform-team (Current), 10User-Ladsgroup: Fix or decide on edge cases of Makefile - https://phabricator.wikimedia.org/T186453#3944484 (10awight) [18:49:03] 10Scoring-platform-team (Current), 10editquality-modeling, 10User-Ladsgroup, 10artificial-intelligence: Simplify hewiki, cswiki, plwiki, svwiki - https://phabricator.wikimedia.org/T188270#4176067 (10awight) 05Open>03Resolved [18:49:05] 10Scoring-platform-team (Current), 10Packaging, 10Epic: [Epic] Support word2vec for production ORES models - https://phabricator.wikimedia.org/T187217#4176082 (10awight) [18:49:09] 10Scoring-platform-team (Current), 10revscoring, 10artificial-intelligence: Implement word2vec featurevector in revscoring - https://phabricator.wikimedia.org/T188445#4176081 (10awight) 05Open>03Resolved [18:49:17] 10Scoring-platform-team (Current), 10editquality-modeling, 10User-Ladsgroup, 10artificial-intelligence: JSON error in building models - https://phabricator.wikimedia.org/T188535#4176083 (10awight) 05Open>03Resolved [18:49:20] 10Scoring-platform-team (Current), 10Packaging, 10Epic: [Epic] Support word2vec for production ORES models - https://phabricator.wikimedia.org/T187217#3968196 (10awight) [18:49:22] 10Scoring-platform-team (Current), 10Packaging: Update ORES requirements to support revscoring 2.2.0 - https://phabricator.wikimedia.org/T188755#4176084 (10awight) 05Open>03Resolved [18:49:27] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: makefile generate makes impossible rules when no model is available - https://phabricator.wikimedia.org/T188777#4176086 (10awight) 05Open>03Resolved [18:49:38] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Simplify azwiki's labeling campaign - https://phabricator.wikimedia.org/T188781#4176087 (10awight) 05Open>03Resolved a:03awight [18:49:45] 10Scoring-platform-team (Current): Investigate word2vec memory issues with multiprocessing - https://phabricator.wikimedia.org/T189364#4176089 (10awight) 05Open>03Resolved [18:49:59] 10Scoring-platform-team (Current), 10ORES, 10User-Ladsgroup: Beta cluster ORES is emitting statsd errors - https://phabricator.wikimedia.org/T189605#4176090 (10awight) 05Open>03Resolved [18:50:05] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Train and test damaging/goodfaith model for arwiki - https://phabricator.wikimedia.org/T189710#4176092 (10awight) 05Open>03Resolved [18:50:36] 10Scoring-platform-team (Current), 10JADE, 10Documentation: Create mediawiki.org article for Extension:JADE - https://phabricator.wikimedia.org/T189938#4176097 (10awight) 05Open>03Resolved [18:50:53] 10Scoring-platform-team (Current), 10MediaWiki-extensions-ORES, 10MW-1.31-release-notes (WMF-deploy-2018-03-27 (1.31.0-wmf.27)), 10Patch-For-Review, 10User-Ladsgroup: Clean up old backward compatibility settings of $wgOresModels - https://phabricator.wikimedia.org/T189948#4176098 (10awight) 05Open>03R... [18:50:55] 10Scoring-platform-team (Current), 10ORES-Support-Checklist, 10User-Ladsgroup: Build ORES support checklist - https://phabricator.wikimedia.org/T189954#4176099 (10awight) 05Open>03Resolved [18:50:57] 10Scoring-platform-team (Current), 10articlequality-modeling, 10draftquality-modeling, 10drafttopic-modeling, 10artificial-intelligence: [Epic] Store ORES predictions for article quality, draft quality, and draft topic in MediaWiki - https://phabricator.wikimedia.org/T190471#4176102 (10awight) [18:50:59] 10Scoring-platform-team (Current), 10ORES, 10articlequality-modeling, 10draftquality-modeling, and 2 others: Discuss surfacing ORES for AFC/NPP - https://phabricator.wikimedia.org/T190464#4176101 (10awight) 05Open>03Resolved [18:51:01] 10Scoring-platform-team (Current), 10Operations, 10Patch-For-Review: Remove deprecated hosts from ORES scap config - https://phabricator.wikimedia.org/T191321#4176103 (10awight) 05Open>03Resolved [18:51:04] 10Scoring-platform-team (Current), 10MediaWiki-JobQueue, 10MediaWiki-extensions-ORES, 10User-Ladsgroup: Fix the ORES-extension grafana dashboard - https://phabricator.wikimedia.org/T191808#4176105 (10awight) 05Open>03Resolved [18:51:07] 10Scoring-platform-team (Current), 10ORES-Support-Checklist, 10User-Ladsgroup: Highlight wikis that has their campaign almost completed - https://phabricator.wikimedia.org/T191435#4176104 (10awight) 05Open>03Resolved [18:51:09] 10Scoring-platform-team, 10ORES: mwparserfromhell ParserError on Premier League - https://phabricator.wikimedia.org/T191861#4176107 (10awight) [18:51:11] 10Scoring-platform-team (Current), 10ORES, 10revscoring, 10artificial-intelligence: Upgrade ORES mwparserfromhell dependency to 0.5.1 - https://phabricator.wikimedia.org/T192038#4176106 (10awight) 05Open>03Resolved [18:51:21] 10Scoring-platform-team (Current), 10editquality-modeling, 10artificial-intelligence: Nulls appear in labeled data (merge_labels issue) - https://phabricator.wikimedia.org/T192362#4176108 (10awight) 05Open>03Resolved [18:51:27] 10Scoring-platform-team (Current), 10ORES: Rebuild all models for revscoring 2.2.2 - https://phabricator.wikimedia.org/T192917#4176109 (10awight) 05Open>03Resolved [18:54:29] flash flood warning. [19:11:41] I just got here [19:11:43] sup [19:11:49] nuthin' [19:11:56] I'm organizing tasks completed >.< [19:13:30] :P [19:25:33] 10Scoring-platform-team: Write JADE talk presentation for Wikimania - https://phabricator.wikimedia.org/T190692#4176164 (10awight) 05Open>03declined Rejected by committee. [19:25:42] awight: https://github.com/wiki-ai/ores/pull/250 [19:26:09] Amir1: Want to take a look? https://phabricator.wikimedia.org/phame/post/view/104/status_update_may_2_2018/ [19:26:40] I do [19:33:54] codezee: Did you say you're working on Makefile changes to run fetch_text utils? Just making sure I'm not stepping on your toes... [19:35:43] awight: it's great! [19:35:52] It doesn't have thinking emoji :P [19:36:00] aww :-{ [19:36:06] cos I'm boring [19:37:02] Noooo [19:37:12] hmm what [19:38:18] awight: yes, but before that i'm testing out fetch text, it seems to trip over redirects [19:42:01] I'm also rebuilding everything just to be sure! [19:42:20] codezee: fyi, my interests are: * understand exactly what was done to produce the original data, * rewrite the PAWS stuff to be more easily repeatable [19:42:46] I don't need the deprecated toolchain to be put back together perfectly, just want to analyze it... [19:42:49] 10Scoring-platform-team, 10Beta-Cluster-Infrastructure, 10ORES, 10User-Ladsgroup: On beta cluster, ORESFetchScoreJob got a HTTP 400 bad request from ores-beta - https://phabricator.wikimedia.org/T157790#4176208 (10Ladsgroup) 05Open>03Resolved a:03Ladsgroup This has been for a very long time. [19:43:08] awight: i think you can rewrite paws stuff independently, for understanding the pipeline you'll need the makefile rules which i am yet to add [19:47:30] kk, thanks. Make sure to base on top of my changes in https://github.com/wiki-ai/drafttopic/pull/21 , I think that'll save you some time! [19:51:10] okay, i've started fetch text on ores-misc will take one-two hours, will attend the remaining tomorrow [19:52:41] Amir1: I'll just add a "highlights" summary and send it out, then? [19:53:32] awight: yeah, that'd be great [20:22:28] 10Scoring-platform-team, 10Design: ORES logo should be replaced - https://phabricator.wikimedia.org/T193670#4176286 (10awight) [20:54:59] (03PS1) 10Krinkle: build: Remove left-over .jscsrc file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430490 [21:03:16] (03CR) 10Ladsgroup: [C: 032] build: Remove left-over .jscsrc file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430490 (owner: 10Krinkle) [21:05:38] (03PS1) 10Krinkle: tests: Remove unused selenium/LocalSettings.php file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430492 [21:07:03] (03Merged) 10jenkins-bot: build: Remove left-over .jscsrc file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430490 (owner: 10Krinkle) [21:12:01] (03CR) 10jenkins-bot: build: Remove left-over .jscsrc file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430490 (owner: 10Krinkle) [21:14:32] (03CR) 10jerkins-bot: [V: 04-1] tests: Remove unused selenium/LocalSettings.php file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430492 (owner: 10Krinkle) [21:23:32] (03Abandoned) 10Krinkle: tests: Remove unused selenium/LocalSettings.php file [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430492 (owner: 10Krinkle) [23:46:07] (03PS1) 10Krinkle: selenium: Use wdio-mediawiki and add standalone runner [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430515 (https://phabricator.wikimedia.org/T193088) [23:48:25] (03CR) 10jerkins-bot: [V: 04-1] selenium: Use wdio-mediawiki and add standalone runner [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430515 (https://phabricator.wikimedia.org/T193088) (owner: 10Krinkle) [23:51:38] (03CR) 10Krinkle: "This needs the job to run 'npm install' before running the tests. This is already done in other jobs (such as mwgate-npm-node-6-docker), b" [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430515 (https://phabricator.wikimedia.org/T193088) (owner: 10Krinkle) [23:56:12] (03PS2) 10Krinkle: selenium: Use wdio-mediawiki and add standalone runner [extensions/ORES] - 10https://gerrit.wikimedia.org/r/430515 (https://phabricator.wikimedia.org/T193088)