Skip to content

Conversation

@schlawg
Copy link
Contributor

@schlawg schlawg commented Apr 11, 2025

  • don't recommended posts with less than 4 common likers
  • we could filter these in bin/mongodb/*.js scripts but with < 50k posts total, let's err on the side of storing more data in ublog_post.similar rather than less
  • boost default (LOW) blog tier for a fighting chance at the lower half of community listing page.

requires mongosh bin/mongodb/ublog-similar-full.js

lila can expect a few mongo exceptions while that's running

@ornicar
Copy link
Collaborator

ornicar commented Apr 12, 2025

I don't understand why the storage structure change. What's the benefit over filtering out count < 4 in the cron script?

@schlawg
Copy link
Contributor Author

schlawg commented Apr 12, 2025

because we want the data.

@schlawg
Copy link
Contributor Author

schlawg commented Apr 12, 2025

what if we want to more heavily favor larger intersections in the random sampling?

isn't it better not to have to run migrations every time we tweak the algorithm?

@ornicar ornicar merged commit 52f28f0 into lichess-org:master Apr 13, 2025
4 checks passed
@schlawg schlawg deleted the ublog-rec-filter branch April 24, 2025 17:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants