Twitter | Search | |
Jordan ⚛️
13,658
Tweets
505
Following
31,193
Followers
Tweets
Jordan ⚛️ 57m
Replying to @hisophiabrandt
A great testament, because TypeScript is a really good experience when you consider all the developer tools and integrations.
Reply Retweet Like
Jordan ⚛️ retweeted
Eduardo Rafael 7h
Ok that's nice, ReasonML Native + esy. Running on a Android 9 device, with ARM64 A HTTP server talking to a Redis, fully statically linked. Guess what? Now it's running with MUSL, so it works even outside of the chroot. Code:
Reply Retweet Like
Jordan ⚛️ 1h
Replying to @TimSweeneyEpic
Another language feature which affects the trade offs is if this effectfullness is completely inferred and checked by the type system.
Reply Retweet Like
Jordan ⚛️ 1h
Replying to @TimSweeneyEpic
They also allow us to “scope” the globalness of that mutable state so that certain regions of our code can have imperative implementations/dependencies (without plumbing) but from the outside are as good as pure.
Reply Retweet Like
Jordan ⚛️ 1h
Replying to @TimSweeneyEpic
Agree with thesis and the sweet spot for language design. We’re also seeing new lang features that change the trade offs - Effects (Eff/OCaml Algebraic Effects) which allow func languages to perform mutation without plumbing down the stack. Better than OO and FP? Perhaps!
Reply Retweet Like
Jordan ⚛️ retweeted
Tim Sweeney 10h
There are really only two types of computations: calculating stuff about an unchanging thing, and calculating a new version of a thing that is changing over time. Recognizing this simplified debate about functional versus imperative programming.
Reply Retweet Like
Jordan ⚛️ 22h
Replying to @Sander_Spies
I learned to appreciate that each shift reduce error is telling you something.
Reply Retweet Like
Jordan ⚛️ 22h
Replying to @ken_wheeler @Rklawton
“I want fewer opportunities!”
Reply Retweet Like
Jordan ⚛️ retweeted
CRON BON JOVI 24h
Replying to @ken_wheeler
BUT I WANT TO COMMUTE 45 MINUTES EACH WAY TO SIT AT A DESK WITH HEADPHONES ON SO THAT IM NOT LONELY
Reply Retweet Like
Jordan ⚛️ Nov 15
Replying to @kylemathews
💯
Reply Retweet Like
Jordan ⚛️ retweeted
Bryan Phelps Nov 14
Amazed by 's PPX for components / hooks in and ! Still have the benefit of compile-time, type-safe hooks... w/ much less boilerplate! 🚀🚀 Thanks for converting components and for bringing it to !
Reply Retweet Like
Jordan ⚛️ Nov 14
Replying to @ken_wheeler
“My area you say? Hey, that’s exactly where I live!”
Reply Retweet Like
Jordan ⚛️ Nov 14
Replying to @mvanier42 @growtofill
They realized they need to boil the frog.
Reply Retweet Like
Jordan ⚛️ Nov 14
Replying to @growtofill
Maybe they realized they were pushing it too fast, and/or they decided to listen to users?
Reply Retweet Like
Jordan ⚛️ Nov 13
I get the sense Apple is trying to slowly get us comfortable with the idea of not having physical keys. In theory there could be some really cool text input paradigm out there but we aren’t there yet and we are biologically wired/conditioned for tactile feedback.
Reply Retweet Like
Jordan ⚛️ Nov 13
Now, all Apple has to do is replace all of the keys with good keys!
Reply Retweet Like
Jordan ⚛️ retweeted
Dorsa Amir Nov 9
Some of the most interesting mimicry I’ve ever seen — the wings of this moth (Macrocilix maia) mimic two flies feasting on bird droppings.
Reply Retweet Like
Jordan ⚛️ Nov 10
Replying to @tekknolagi
Your function can be deoptimized just because someone calls it with weird data where keys are in the wrong order etc. I understand it’s the only way to get perf for dynamic languages (or partially dynamic) in practice. But people are moving towards static anyways.
Reply Retweet Like
Jordan ⚛️ Nov 10
Replying to @tekknolagi
JITs are such a crutch for languages. They aren’t predictable. And then programmers game the heuristics for better perf, so JITs are locked into old heuristics. They create larger warm up times. I would rather use a language designed from the ground up to be compiled/optimized.
Reply Retweet Like
Jordan ⚛️ retweeted
António N. Monteiro Nov 3
If you've been missing OCaml's "monadic let" operators in Reason, I opened a PR that adds support for them. Would appreciate any and all feedback!
Reply Retweet Like