Rust vs. Go post is up

[Eric S. Raymond]

Mark Atwood <fallenpegasus at gmail.com>:
> If you want a randomly different thing to relaxation hack on with Go,
> instead of an IRC server, consider a Matrix.ORG server.   A Matrix.org
> server that is binary and performant actually has a good chance of
> significant uptake.   But, your choice, your project.

That's not a bad idea and I may pursue it.  The code I've already
written could probably be repurposed for it.  But I promised Susan a
learning project in sysarching a while back and *she* wants an IRC
server - she says the existing implementations aren't very good. And
it's not like that'd even be a major effort really, not when I can
write protocol state machines practically in my sleep.

> Thank you for your work, Eric.  I had high hopes for Rust.  I think that
> extremely high learning curve for the languages is even more of a killer
> for us than swarm/crate/roadmap issue.

Arguably, yeah.  I mean, if somebody with my decades of experience in 23
different exotic languages has that much trouble after studying the
documentation *carefully*, what's it gonna do to J. Random Junior Dev?
Doesn't bear thinking about.

It's a damn shame.  The underlying theory of Rust is pretty
interesting, but something went wrong on the way to turning that
theory into a tool one can use.  Ah well, early days yet.  Maybe
they'll fix it.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>