Marყ̈n Haverbeke

"who has directed his entire effort to misleading me." https://en.wikipedia.org/wiki/Evil_demon 

At this point I'm treating the `data` attribute on composition events as Descartes' malicious demon

“Pony has no locks! So they definitely don't deadlock, because they don't exist.” http://tutorial.ponylang.org/capabilities/index.html …

So you get this kind of cargo-cult-programming-for-the-JIT

Benchmarking every micro code-shape decision isn't something I want to do. Just ignoring what I know about compilation also seems a shame

Often find myself writing code in a peculiar way because the way I understand JIT compilation leads me to believe that it might be faster

It's Saturday night and I'm chasing a bug in a collaborative undo history. Say what you want about my life choices. I found the bug & I'm OK

Parties need an introvert area where there is no music and people can have thoughtful 1-1 discussions.

Landed a rewrite of ProseMirror's history module today. We should be about ready to release a 0.6.0 next week. (Testing very welcome.)

One thing about the 'small module philosophy' is that it's a dangerous invitation to go crazy with noxious abstraction and indirection

The @ponylang tutorial Chapter 1: Fully memory safe & a program that compiles should never crash. Chapter 9: C FFI. (Still, )

I now want a JS dialect with Rust-style ownership semantics. (In general, is any work trying to apply those concepts to a GC-ed language?)

At this point the worst thing about JS is how hard it is to keep track of who owns what and who may mutate it

Wait, still no Array.splice-with-sane-parameters in ES6?

I am very happy to announce that I will be a resident at @recursecenter in June! https://www.recurse.com/blog/106-steve-klabnik-paul-fenwick-and-frank-wang-are-recurse-center-residents …

Can any (JSON/XML/etc) schema language express 'each of this element's children must have the same amount of children'?