@thdxr I wondered the same thing, I thought I read a post that said Anthropic received a “letter”; was thinking that can’t be right, but also the least of my concerns right now… I miss fable
@steipete@morew4rd I want exactly this, but I want to also expose the running services back out so I can quickly QA its work. Is that possible? Cmux browser actually does this over ssh some how
Hey @mattpocockuk, was wondering about Evalite. I remember seeing you post about it a while back.
Did you end up picking something else up? Or do you think Evalite still has legs? Thanks for all that you do!
@thdxr people are making so many incorrect assumptions
that anyone at Anthropic asked for this (this 100% was all me)
that this is motivated by zig team’s no AI PR policy (nope)
I’m just tired of dealing with crashes and memory leaks & want language features to help prevent things
@lydiahallie I like Anthropic generally, but agreed with the poor taste here. At an absolute minimum Anthropic needs to stop playing games and attempting to spin pricing models to look as if they are doing the users a service. It makes it hard to tell good from bad. Be real with us, it’s okay
@RhysSullivan I like the idea generally, but even looking at Effect’s homepage, the section compare/contrasting with Effect vs without Effect. In every example I feel the Effect sample has a much higher cognitive complexity, and for what reason?
@mattpocockuk It’s like, as soon as I figure something out, +2/3 days go by and then there’s an article describing that exact thing. Feels like everyone is arriving at the same conclusions at the same time. Kinda cool
@MichelleBakels@g2i_co Leaving the best company I’ve ever worked with has been the hardest thing I’ve ever done ( professionally ) and the most rewarding. I’ve done that process 3 times now, same result each time.
Would recommend; congrats and very excited for you!
@kepano I just tried it this morning on the 245-page Mythos pdf and it failed badly and the outputs were all mangled. Converting pdfs is really hard, I think it has to probably be a Skill not a program, for a SOTA LLM for it to work properly.
@steipete@orangie I’m going to miss opus, it was the only model I ever really trusted. Hard for me to imagine anything else. Disappointed in Anthropic for sure.
Anthropic shutting down OpenClaw may turn out to be a strategic blunder, or strategic genius. The OpenClaw community will be the determiner of whether it is A or B. It's an interesting moment in history.
Personally I never bet against open source.
My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept):
Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow