Page MenuHomePhabricator

Halfak (Aaron Halfaker, EpochFail, halfak)
Principal Research Scientist

Projects (19)

Today

  • Clear sailing ahead.

Tomorrow

  • Clear sailing ahead.

Wednesday

  • Clear sailing ahead.

User Details

User Since
Oct 21 2014, 6:05 PM (267 w, 6 d)
Availability
Available
IRC Nick
halfak
LDAP User
Halfak
MediaWiki User
EpochFail [ Global Accounts ]

Hi! I'm a socio-technologist. I do science so that I can build new technologies for social systems.

You can find me as:

Recent Activity

Today

Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.

Bam: https://github.com/wikimedia/revscoring/pull/460

Mon, Dec 9, 8:58 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.

I wonder if there is something we can do from our end to minimize the issue in the short term.

Mon, Dec 9, 6:08 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T240243: Add accraze to analytics-privatedata-users.

I approve from my side. Andy will need this access to be able to use Hadoop-related resources for building and analyzing machine learning models.

Mon, Dec 9, 5:53 PM · Analytics, Operations, SRE-Access-Requests
Halfak created T240243: Add accraze to analytics-privatedata-users.
Mon, Dec 9, 5:52 PM · Analytics, Operations, SRE-Access-Requests
Halfak added a comment to T237605: Create kerberos principals for users.

@elukey, @ACraze not being in analytics-privatedata-users seems like an oversight. I'll check on that and come back to re-request when we've got it squared.

Mon, Dec 9, 5:49 PM · Analytics-Kanban, Analytics
Halfak added a parent task for T199834: Implement a MW API module for interacting with Jade entities: T229968: Write Jade extension documentation.
Mon, Dec 9, 5:37 PM · Core Platform Team Workboards (Clinic Duty Team), Patch-For-Review, Scoring-platform-team (Current), Jade, Documentation
Halfak added a subtask for T229968: Write Jade extension documentation: T199834: Implement a MW API module for interacting with Jade entities.
Mon, Dec 9, 5:37 PM · Scoring-platform-team (Current), Documentation, Jade
Halfak moved T223788: Implement hunspell dictionary for euwiki article quality model from Pending deployment to Done on the Scoring-platform-team (Current) board.
Mon, Dec 9, 5:23 PM · Patch-For-Review, Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak moved T223794: Use Spanish badwords/informals in the cawiki edit quality models from Pending deployment to Done on the Scoring-platform-team (Current) board.
Mon, Dec 9, 5:22 PM · Scoring-platform-team (Current), good first task, editquality-modeling, artificial-intelligence
Halfak moved T232005: Limit rev_ids in an ORES request to 50. from Pending deployment to Done on the Scoring-platform-team (Current) board.
Mon, Dec 9, 5:22 PM · Scoring-platform-team (Current), good first task, ORES
Halfak moved T238839: ORES deploy -- Late November, 2019 from Review to Done on the Scoring-platform-team (Current) board.
Mon, Dec 9, 5:22 PM · ORES, Scoring-platform-team (Current)
Halfak added a comment to T233448: Review prometheus ORES rules for completeness.

Nice work, Cole. You've reverse engineered correctly!

Mon, Dec 9, 4:59 PM · Patch-For-Review, ORES, Scoring-platform-team
Halfak added a comment to T236713: Improve drafttopic training data pipeline.

I responded to a bunch of notes and I moved the data to a machine-readable format. See https://github.com/halfak/wikitax

Mon, Dec 9, 4:54 PM · NewcomerTasks 1.1, Research
Halfak added a comment to T235184: Generate word vectors for ar, cs, en, and ko using FastText.

Copy-pasted from T235183:

Mon, Dec 9, 4:35 PM · artificial-intelligence, drafttopic-modeling, revscoring, Scoring-platform-team
Halfak added a comment to T235183: Experiment with different vector lengths for ar, cs, en, and kowiki topic models. .

I found https://fasttext.cc/docs/en/unsupervised-tutorial.html. It seems like a great tutorial for generating word vectors. I think we should start here with length 50 vectors and compare them to length 100 vectors.

Mon, Dec 9, 4:34 PM · artificial-intelligence, drafttopic-modeling, revscoring, Scoring-platform-team

Thu, Dec 5

Halfak updated subscribers of T239942: ORES articlequality for euwiki works differently in production.

@akosiaris, could you look at my work here and see what you think? The above commands give you a nice, minimal example of the problematic differences between some of the ORES hosts.

Thu, Dec 5, 10:59 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.
$ ssh ores1001.eqiad.wmnet "cat 'de' | strace -e open,openat enchant -l -d en" &> ores1001.strace
$ ssh ores1009.eqiad.wmnet "cat 'de' | strace -e open,openat enchant -l -d en" &> ores1009.strace
$ diff ores1001.strace ores1009.strace 
34c34
< open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_ispell.so", O_RDONLY|O_CLOEXEC) = 4
---
> open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_aspell.so", O_RDONLY|O_CLOEXEC) = 4
35a36
> open("/usr/lib/x86_64-linux-gnu/libaspell.so.15", O_RDONLY|O_CLOEXEC) = 4
39,41d39
< open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_hspell.so", O_RDONLY|O_CLOEXEC) = 4
< open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 4
< open("/lib/x86_64-linux-gnu/libz.so.1", O_RDONLY|O_CLOEXEC) = 4
45c43
< open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_aspell.so", O_RDONLY|O_CLOEXEC) = 4
---
> open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_hspell.so", O_RDONLY|O_CLOEXEC) = 4
47c45,46
< open("/usr/lib/x86_64-linux-gnu/libaspell.so.15", O_RDONLY|O_CLOEXEC) = 4
---
> open("/lib/x86_64-linux-gnu/libz.so.1", O_RDONLY|O_CLOEXEC) = 4
> open("/usr/lib/x86_64-linux-gnu/enchant/libenchant_ispell.so", O_RDONLY|O_CLOEXEC) = 4
57,59c56,58
< open("/usr/share/hunspell/en_AU.aff", O_RDONLY) = 3
< open("/usr/share/hunspell/en_AU.dic", O_RDONLY) = 3
< open("/usr/share/hunspell/en_AU.aff", O_RDONLY) = 3
---
> open("/usr/share/hunspell/en_US.aff", O_RDONLY) = 3
> open("/usr/share/hunspell/en_US.dic", O_RDONLY) = 3
> open("/usr/share/hunspell/en_US.aff", O_RDONLY) = 3
Thu, Dec 5, 9:49 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.
$ ssh ores1001.eqiad.wmnet "/srv/deployment/ores/deploy/venv/bin/python -c 'import enchant; print(enchant.Dict(\"en\").provider)'"
<Enchant: Myspell Provider>
$ ssh ores1009.eqiad.wmnet "/srv/deployment/ores/deploy/venv/bin/python -c 'import enchant; print(enchant.Dict(\"en\").provider)'"
<Enchant: Myspell Provider>
Thu, Dec 5, 9:39 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.

I made extra-double sure there was nothing different about what spelling packages were installed.

Thu, Dec 5, 9:17 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.

Alright! I figured out how to replicate the issue between machines without using any of our code. Instead, I'm using the CLI for enchant directly. When run as I have below, the utility prints out misspelled words. So when it prints nothing, it thinks the world is a correctly spelled english word. When it prints out the word, it thinks it is misspelled.

Thu, Dec 5, 9:08 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak added a comment to T239942: ORES articlequality for euwiki works differently in production.

I confirmed that all nodes have the same git hash.

Thu, Dec 5, 8:53 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak triaged T239942: ORES articlequality for euwiki works differently in production as High priority.
Thu, Dec 5, 6:54 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak updated subscribers of T239942: ORES articlequality for euwiki works differently in production.

@kevinbazira and I looked into this. We ran the following command to look up the installed versions of the enchant dictionaries we would like to use.

Thu, Dec 5, 6:53 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak created T239942: ORES articlequality for euwiki works differently in production.
Thu, Dec 5, 6:51 PM · Patch-For-Review, artificial-intelligence, articlequality-modeling, ORES, Scoring-platform-team
Halfak closed T234222: Onboarding Kevin Bazira -- Accounts and Access as Resolved.
Thu, Dec 5, 6:07 PM · Scoring-platform-team (Current)
Halfak updated the task description for T234222: Onboarding Kevin Bazira -- Accounts and Access.
Thu, Dec 5, 6:06 PM · Scoring-platform-team (Current)

Wed, Dec 4

Halfak updated subscribers of T238877: Write Huggle labels to Jade.

I've been talking to @Capt_Swing and @ACraze about the "suspicious" flag. We have a few good options we're thinking about and will extend the description of this task when we think we have something good.

Wed, Dec 4, 9:56 PM · Huggle, Jade, Scoring-platform-team
Halfak moved T238877: Write Huggle labels to Jade from Untriaged to Monitor on the Scoring-platform-team board.
Wed, Dec 4, 9:55 PM · Huggle, Jade, Scoring-platform-team
Halfak moved T229976: Design New Filters controls for Jade from Untriaged to New development on the Scoring-platform-team board.
Wed, Dec 4, 9:39 PM · Scoring-platform-team, Jade
Halfak triaged T229976: Design New Filters controls for Jade as Low priority.
Wed, Dec 4, 9:39 PM · Scoring-platform-team, Jade
Halfak moved T235003: Clean up naming conflicts around writing secondary schema data for Jade from Untriaged to Maintenance/cleanup on the Scoring-platform-team board.
Wed, Dec 4, 9:39 PM · Scoring-platform-team, Jade
Halfak removed a project from T235003: Clean up naming conflicts around writing secondary schema data for Jade: Epic.
Wed, Dec 4, 9:39 PM · Scoring-platform-team, Jade
Halfak removed a project from T229976: Design New Filters controls for Jade: Epic.
Wed, Dec 4, 9:39 PM · Scoring-platform-team, Jade
Halfak triaged T235003: Clean up naming conflicts around writing secondary schema data for Jade as Low priority.
Wed, Dec 4, 9:38 PM · Scoring-platform-team, Jade
Halfak updated subscribers of T237605: Create kerberos principals for users.

I'd like to request creds for the engineers on my team as we'll (hopefully) be using hadoop a lot more soon. Usernames: accraze and kevinbazira

Wed, Dec 4, 9:01 PM · Analytics-Kanban, Analytics
Halfak added a comment to T237605: Create kerberos principals for users.

I need creds for stat100*. My username is halfak

Wed, Dec 4, 9:00 PM · Analytics-Kanban, Analytics
Halfak claimed T236546: "wikilabels" Cloud VPS project jessie deprecation.
Wed, Dec 4, 7:33 PM · Scoring-platform-team (Current), Cloud-VPS (Debian Jessie Deprecation)
Halfak added a comment to T236546: "wikilabels" Cloud VPS project jessie deprecation.

Oof. I'm on it.

Wed, Dec 4, 7:20 PM · Scoring-platform-team (Current), Cloud-VPS (Debian Jessie Deprecation)
Halfak updated the task description for T234222: Onboarding Kevin Bazira -- Accounts and Access.
Wed, Dec 4, 5:31 PM · Scoring-platform-team (Current)
Halfak updated the task description for T234222: Onboarding Kevin Bazira -- Accounts and Access.
Wed, Dec 4, 5:30 PM · Scoring-platform-team (Current)
Halfak edited projects for T229976: Design New Filters controls for Jade, added: Scoring-platform-team; removed Scoring-platform-team (Current).
Wed, Dec 4, 5:30 PM · Scoring-platform-team, Jade
Halfak edited projects for T235003: Clean up naming conflicts around writing secondary schema data for Jade, added: Scoring-platform-team; removed Scoring-platform-team (Current).
Wed, Dec 4, 5:30 PM · Scoring-platform-team, Jade
Halfak moved T238839: ORES deploy -- Late November, 2019 from Active to Review on the Scoring-platform-team (Current) board.
Wed, Dec 4, 5:29 PM · ORES, Scoring-platform-team (Current)
Halfak claimed T226055: Write a cookbook for the workaround for getting LFS to gerrit.
Wed, Dec 4, 5:27 PM · Scoring-platform-team (Current), Gerrit, ORES
Halfak moved T188235: Add language support for nowiki from Pending deployment to Done on the Scoring-platform-team (Current) board.
Wed, Dec 4, 5:27 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence
Halfak moved T130291: Train/test edit quality models for nowiki from Review to Pending deployment on the Scoring-platform-team (Current) board.
Wed, Dec 4, 4:33 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling

Tue, Dec 3

Halfak committed rOEQ1907e4c1cce7: Train/test edit quality models for nowiki (#217) (authored by kevinbazira).
Train/test edit quality models for nowiki (#217)
Tue, Dec 3, 3:25 PM
Halfak added a comment to T236713: Improve drafttopic training data pipeline.

I just shared a gdoc with @MMiller_WMF and @Isaac with a completed taxonomy. Once they have had a look at my work, I'll convert this to something machine readable and we can start experimenting with re-training the model.

Tue, Dec 3, 12:15 AM · NewcomerTasks 1.1, Research

Mon, Dec 2

Halfak added a comment to T232899: Newcomer tasks: dismiss suggestions.

Nothing I could use this for right now in Jade but I'd be interested in an analysis that could tell us how many newcomers don't find their interested topics represented and what king of topics seem to be missing.

Mon, Dec 2, 4:06 PM · NewcomerTasks 1.1, Design, Growth-Team, GrowthExperiments-Homepage
Halfak updated subscribers of T106869: Implement priority prediction functionality for WikiProject tasks module.

Hmm. I'm not sure that @Sumit was ever working on this.

Mon, Dec 2, 3:44 PM · Reports-bot, Scoring-platform-team
Halfak updated the task description for T188235: Add language support for nowiki.
Mon, Dec 2, 3:39 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence
Halfak added a comment to T188235: Add language support for nowiki.

Wow that page is fantastic. Eventually, I'd like to move these badwords lists to wiki pages. This is an interesting model for that.

Mon, Dec 2, 3:32 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence

Tue, Nov 26

Halfak moved T130291: Train/test edit quality models for nowiki from Active to Review on the Scoring-platform-team (Current) board.
Tue, Nov 26, 3:59 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak moved T188235: Add language support for nowiki from Active to Pending deployment on the Scoring-platform-team (Current) board.
Tue, Nov 26, 3:59 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence

Mon, Nov 25

Halfak added a comment to T188235: Add language support for nowiki.

Awesome! This is the type of work we're looking for in the language assets. It takes fluent local to help us make sense of what should and should not be matched. Generally, I think matching the animal "fox" is good if it is often used in an insulting way. In an article about Fox or Mammals generally, we'll get some signal that the word is more common there.

Mon, Nov 25, 8:09 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence
Halfak added a comment to T223383: Improve Czech Language assets.

Looks like I can't access the task. I've registered with GCI using my wikimedia email address (ahalfaker@wikimedia.org)

Mon, Nov 25, 7:21 PM · Google-Code-in-2019, Scoring-platform-team, good first task, ORES
Halfak updated the task description for T234222: Onboarding Kevin Bazira -- Accounts and Access.
Mon, Nov 25, 5:40 PM · Scoring-platform-team (Current)
Halfak moved T212370: Design Jade entity UI from Pending deployment to Done on the Scoring-platform-team (Current) board.
Mon, Nov 25, 5:37 PM · Scoring-platform-team (Current), Jade

Thu, Nov 21

Halfak added a comment to T223383: Improve Czech Language assets.

We're using these word lists to build artificial intelligences (AIs) for Wikipedia. These AIs help Wikipedia editors catch vandalism, measure the quality of articles, and sort articles by topic. By extending these word lists, you're helping the AIs get a better sense of what's happening in a given language.

Thu, Nov 21, 11:51 PM · Google-Code-in-2019, Scoring-platform-team, good first task, ORES
Halfak added a parent task for T183381: Deploy Jade extension MVP to production: T238877: Write Huggle labels to Jade.
Thu, Nov 21, 10:48 PM · Scoring-platform-team (Current), Core Platform Team Legacy (Watching / External), Epic, Goal, Services (watching), Operations, TechCom, Jade
Halfak added a subtask for T238877: Write Huggle labels to Jade: T183381: Deploy Jade extension MVP to production.
Thu, Nov 21, 10:48 PM · Huggle, Jade, Scoring-platform-team
Halfak created T238877: Write Huggle labels to Jade.
Thu, Nov 21, 10:45 PM · Huggle, Jade, Scoring-platform-team
Halfak added a comment to T223383: Improve Czech Language assets.

Also note that we'd be happy with anyone taking on this task in any language -- serbian, czech, french, etc.

Thu, Nov 21, 7:52 PM · Google-Code-in-2019, Scoring-platform-team, good first task, ORES
Halfak added a comment to T223383: Improve Czech Language assets.

Awesome! Thanks for picking this up!

Thu, Nov 21, 7:45 PM · Google-Code-in-2019, Scoring-platform-team, good first task, ORES
Halfak added a parent task for T223788: Implement hunspell dictionary for euwiki article quality model: T238839: ORES deploy -- Late November, 2019.
Thu, Nov 21, 5:26 PM · Patch-For-Review, Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak added a parent task for T223794: Use Spanish badwords/informals in the cawiki edit quality models: T238839: ORES deploy -- Late November, 2019.
Thu, Nov 21, 5:26 PM · Scoring-platform-team (Current), good first task, editquality-modeling, artificial-intelligence
Halfak added subtasks for T238839: ORES deploy -- Late November, 2019: T223794: Use Spanish badwords/informals in the cawiki edit quality models, T223788: Implement hunspell dictionary for euwiki article quality model.
Thu, Nov 21, 5:26 PM · ORES, Scoring-platform-team (Current)
Halfak created T238839: ORES deploy -- Late November, 2019.
Thu, Nov 21, 4:39 PM · ORES, Scoring-platform-team (Current)
Halfak moved T223794: Use Spanish badwords/informals in the cawiki edit quality models from Review to Pending deployment on the Scoring-platform-team (Current) board.
Thu, Nov 21, 4:38 PM · Scoring-platform-team (Current), good first task, editquality-modeling, artificial-intelligence
Halfak moved T223788: Implement hunspell dictionary for euwiki article quality model from Active to Pending deployment on the Scoring-platform-team (Current) board.
Thu, Nov 21, 4:38 PM · Patch-For-Review, Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak updated the task description for T226055: Write a cookbook for the workaround for getting LFS to gerrit.
Thu, Nov 21, 4:18 PM · Scoring-platform-team (Current), Gerrit, ORES
Halfak closed T232494: Require git-lfs in ORES hosts as Resolved.
Thu, Nov 21, 4:18 PM · Puppet, ORES, Scoring-platform-team (Current)
Halfak moved T226055: Write a cookbook for the workaround for getting LFS to gerrit from Active to Review on the Scoring-platform-team (Current) board.
Thu, Nov 21, 4:11 PM · Scoring-platform-team (Current), Gerrit, ORES
Halfak added a comment to T226055: Write a cookbook for the workaround for getting LFS to gerrit.

And now it contains the LFS commands you need to run.

Thu, Nov 21, 4:11 PM · Scoring-platform-team (Current), Gerrit, ORES

Wed, Nov 20

Halfak added a comment to T226055: Write a cookbook for the workaround for getting LFS to gerrit.

I started some work here for describing how to build the wheels. This is part of the same deployment prep process so it should be included.

Wed, Nov 20, 10:19 PM · Scoring-platform-team (Current), Gerrit, ORES
Halfak edited projects for T226055: Write a cookbook for the workaround for getting LFS to gerrit, added: Scoring-platform-team (Current); removed Scoring-platform-team.
Wed, Nov 20, 10:19 PM · Scoring-platform-team (Current), Gerrit, ORES
Halfak claimed T188235: Add language support for nowiki.
Wed, Nov 20, 10:18 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence
Halfak added a comment to T188235: Add language support for nowiki.

Aha! Turns out that these language assets exist. @jeblad, could you review what we have here? https://github.com/wikimedia/revscoring/blob/master/revscoring/languages/norwegian.py

Wed, Nov 20, 10:17 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence
Halfak moved T237807: gerrit: scoring/ores/editquality takes a long time to git gc from Untriaged to Monitor on the Scoring-platform-team board.
Wed, Nov 20, 10:16 PM · Release-Engineering-Team-TODO, Release-Engineering-Team (Development services), Scoring-platform-team, Gerrit
Halfak added a comment to T237807: gerrit: scoring/ores/editquality takes a long time to git gc.

This could be an old repo that used before we had git LFS. We should be able to find out and get it cleaned up if I'm right in the process of addressing T238746: Clean up ORES and related gerrit repos.

Wed, Nov 20, 10:15 PM · Release-Engineering-Team-TODO, Release-Engineering-Team (Development services), Scoring-platform-team, Gerrit
Halfak added a comment to T130291: Train/test edit quality models for nowiki.

Here's the example: https://github.com/wikimedia/editquality/pull/201

Wed, Nov 20, 10:13 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak renamed T130291: Train/test edit quality models for nowiki from Deploy edit quality models for nowiki to Train/test edit quality models for nowiki.
Wed, Nov 20, 10:13 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak claimed T130291: Train/test edit quality models for nowiki.

@kevinbazira could you take on this task? I asked @ACraze and he agreed to help out. Here's an example of a pull request that adds a damaging and goodfaith model for another wiki.

Wed, Nov 20, 10:12 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak closed T130268: Complete nowiki edit quality campaign as Resolved.

Looks like this is done. Thank you @jeblad

Wed, Nov 20, 10:09 PM · artificial-intelligence, Scoring-platform-team, editquality-modeling, Wikilabels
Halfak closed T130268: Complete nowiki edit quality campaign, a subtask of T130291: Train/test edit quality models for nowiki, as Resolved.
Wed, Nov 20, 10:09 PM · Scoring-platform-team (Current), artificial-intelligence, editquality-modeling
Halfak merged task T238645: Wikilabels at nowiki into T130268: Complete nowiki edit quality campaign.
Wed, Nov 20, 10:08 PM · Scoring-platform-team, Wikilabels
Halfak merged T238645: Wikilabels at nowiki into T130268: Complete nowiki edit quality campaign.
Wed, Nov 20, 10:08 PM · artificial-intelligence, Scoring-platform-team, editquality-modeling, Wikilabels
Halfak added a comment to T238645: Wikilabels at nowiki.

Amazing! We'll get to work on the models. Thank you!

Wed, Nov 20, 10:07 PM · Scoring-platform-team, Wikilabels
Halfak added a comment to T238746: Clean up ORES and related gerrit repos.

All the old repos should be archived once we have this cleaned up.

Wed, Nov 20, 10:07 PM · Scoring-platform-team
Halfak triaged T238746: Clean up ORES and related gerrit repos as Medium priority.
Wed, Nov 20, 10:04 PM · Scoring-platform-team
Halfak added a comment to T238746: Clean up ORES and related gerrit repos.

https://gerrit.wikimedia.org/r/admin/projects/q/filter:ores

Wed, Nov 20, 10:03 PM · Scoring-platform-team
Halfak committed rORESWHEELS160c10834c80: Updates wheels for revscoring 2.5.1 (authored by Halfak).
Updates wheels for revscoring 2.5.1
Wed, Nov 20, 8:33 PM
Halfak closed T122680: Operation times out when processing 282788592 in Wikidata as Resolved.

Looks like it's working now.

Wed, Nov 20, 8:24 PM · Performance Issue, wb_vandalism
Halfak updated the task description for T238746: Clean up ORES and related gerrit repos.
Wed, Nov 20, 2:50 PM · Scoring-platform-team
Halfak created T238746: Clean up ORES and related gerrit repos.
Wed, Nov 20, 2:49 PM · Scoring-platform-team

Tue, Nov 19

Halfak reopened T188235: Add language support for nowiki as "Open".

Agreed that this task shouldn't be closed.

Tue, Nov 19, 2:45 PM · Scoring-platform-team (Current), Bad-Words-Detection-System, revscoring, artificial-intelligence

Mon, Nov 18

Halfak added a comment to T236713: Improve drafttopic training data pipeline.

I think having multiple levels makes a lot of sense and I'm interested in what a proposal for a better top-level grouping would be.

Mon, Nov 18, 8:48 PM · NewcomerTasks 1.1, Research

Thu, Nov 14

Marsupium awarded T150094: Add PAWS to interwiki links a Like token.
Thu, Nov 14, 4:53 PM · User-DannyS712, MediaWiki-Interwiki, PAWS

Wed, Nov 13

Halfak added a comment to T229401: Key-value extraction misses on Wikipedia:WikiProject Council/Directory/WikiProject template invocations.

Hi @dr0ptp4kt! Thanks for your patchset. Sorry for the confusion. Would you be willing to submit your changes as a pull request on github? https://github.com/wikimedia/drafttopic

Wed, Nov 13, 5:31 PM · Patch-For-Review, drafttopic-modeling, Scoring-platform-team

Nov 7 2019

Halfak added a comment to T231214: Refactor revscoring to handle session-orientation.

New changes! See https://github.com/wikimedia/revscoring/compare/9ff5ac176d6fb71f4ccc88bd43a1e36439cb4968...04bb9dfd6acc7da1c908e02d77ff2729a4cb3875 for what is new.

Nov 7 2019, 10:07 PM · artificial-intelligence, revscoring, Scoring-platform-team (Current)