1,925 karma · joined August 20, 2025
Using libraries like this provide none of those assurances. Sure, you can improve performance with fine tuning , but then we're going back to doing what a model like jev was created to eliminate.
I don’t get why these need to be separate. One skill with progressive disclosure already handles this. That’s kind of the point.
the sites could implement their own search to make it all work if they wanted to. they just dont. but blaming virtualization is not tit.
"Taste" is one of these words that have been bandied about recently, in addition to "moat". I've never seen a productive conversation stem from either of those mentiosn.
Meta said it was considering doing the same because they saw how valuable excess capacity was.
You're misinterpreting what actually happened.