Is this the algo? (from github)
SELECT id, row_number() OVER (ORDER BY ((GREATEST("weightedVotes", POWER("weightedVotes", 1.2)) + "weightedComments"/2) / POWER(GREATEST(3, EXTRACT(EPOCH FROM (now_utc() - created_at))/3600), 1.3) + (boost/5000::float) / POWER(EXTRACT(EPOCH FROM (now_utc() - created_at))/3600+2, 2.6)) DESC NULLS LAST, id DESC) as rank FROM "Item" WHERE "parentId" IS NULL AND NOT bio AND "pinId" IS NULL AND "deletedAt" IS NULL AND "weightedVotes" > 0 ORDER BY ((GREATEST("weightedVotes", POWER("weightedVotes", 1.2)) + "weightedComments"/2) / POWER(GREATEST(3, EXTRACT(EPOCH FROM (now_utc() - created_at))/3600), 1.3) + (boost/5000::float) / POWER(EXTRACT(EPOCH FROM (now_utc() - created_at))/3600+2, 2.6)) DESC NULLS LAST, id DESC LIMIT 2100;
Thank you for sharing the actual code. This is the promise of FOSS SN. We don’t need to trust @k00b or risk misunderstanding an explanation from the FAQ. We can verify for ourselves! Which file had this snippet?
reply
Maybe it was this one? stacker.news/prisma/migrations/20240228011144_user_values_view /migration.sql
I can't remember what I searched on right now. If you search for "weightedvotes" you get lots of results.
Honestly, this gives a clue to what might be going on, but to really say positively I'm not sure this is good enough.
reply
Hey “weightedvotes” is already a clue! I appreciate ya, cowboy! 🤠
reply