@sjvipin we're unstalled now, but it's been a minute since we talked about this. Could you clarify exactly what is being asked for here (eg. a list of: link event records, links to edits where the citations were added, urls to cited articles)?

Aggregate backfill complete and looking good:

okay, so I distilled this down into the simplest terms I could without tossing out anything interesting.
This data was collected for the period from 2025-08-27 - 2025-10-01.
General awareness/usage of ORES-provided filtering is pretty low, however, the language agnostic revert risk filter was used more than 2x the next most popular ORES-provided filter during the above period.

This seems like we would have a very large denormalized table with multple entries per user and thread to track the many-to-many relationship. I could see this getting out of hand very quickly in terms of performance and scalability.

@GTrang I can see how this has impacted your workflow, but I'm not sure your proposed outcome is correct:

@Novem_Linguae thanks for filing this; we recently addressed a CVE and this is emergent behavior related to that change. We'll get on it!

We noticed this on an iPhone SE 2020

You could always try checking the resources made available the docker desktop vm that underpins your setup. Maybe make sure that it has enough ram + disk? I'm just throwing things at the wall at this point.

@Scardenasmolinar usually when I encounter issues like this, I just start destroying my local environment, eg.

docker system prune -a; docker volume prune -a

When that doesn't cut it, I've occasionally purged all docker packages and config and started completely over.

@Kgraessle can you give this a try as well?

@Scardenasmolinar This does look to be a local overlay fs driver issue to me; I did quite a bit of troubleshooting locally, and so far switching to a bind mount is the only thing that's fixed it. You and I were seeing somewhat different errors, so I was wondering if you could try running a restore in my test branch?
https://github.com/WikipediaLibrary/externallinks/tree/Jsn.sherman/T405925

just reusing this task for my reporting based on this code; currently rerunning

See T403495: [SPIKE] How could we present users of the moderator dashboard with their Impact?[4H] for initial investigation

note that we should probably write the code in such a way that tips on personaldashboard itself can be displayed via this mechanism.

see T402822: [SPIKE] How can we use DiscussionTools to retrieve the most active discussions on certain noticeboards? [8H] for initial investigation

I let this sit for long enough that I decided to rerun the script before writing up the findings. While waiting on that to run, I thought I'd move the data to a paste, but found an interesting bug/limitation: I can't seem to get phabricator to render a remarkup table in a paste, even though the same table renders in a comment.
See: P83475

for whatever reason, a remarkup table won't render in the paste, even if the exact same table renders in a comment:

It's still not clear to me that the benefits would outweigh potential costs on this. We don't currently have plans to make this change, so I'm marking as declined, but of course someone may reopen in the future.

verified on enwiki

verified on enwiki

just noting that we should add a ci step in the prs that validates the json file

@Samwalton9-WMF @OTichonova remind me: Are these meant on the dashboard or on other pages? From the description, it looks like they go directly in the dashboard.

Repository now exists at https://gerrit.wikimedia.org/r/admin/repos/mediawiki%2Fextensions%2FPersonalDashboard

request started https://www.mediawiki.org/wiki/Gerrit/New_repositories/Requests#mediawiki/extensions/PersonalDashboard

I've started a draft of the extension page:
https://www.mediawiki.org/wiki/Extension:PersonalDashboard