«We cannot have a society in which, if two people wish to communicate, the only way that can happen is if it's financed by a third person who wishes to manipulate them.»

— Jaron Lanier

Now that I’m trying to work with the API, this seems like a dumb question. Checked exceptions are too burdensome.

Show thread

@qbit now that I’m actually trying to use it, I think you’d want to avoid LockInterruptably in pretty much any business logic. It throws a checked exception. That’s also kind of weird (seems like a user interrupt is a clear runtime exception to me).

Java’s Lock interface provides the methods lock and lockInterruptably, the former not responding even to user interrupts.

In what world would I ever want to lock non-interruptably?

Incidentally, I had been reading nobs(1) as "knobs," but I understand now that it should be pronounced "no-b-s" (bs for backspace, but one could think of more creative expansions).

Show thread

if(~ $termprog 9term || ~ $termprog win){
fn git {
`{which git} $* | nobs
}
}

Adding a second DoT provider to my forward zone seems to have resolved things:

thread0.recursion.time.avg=0.674038
thread0.recursion.time.median=0.382061

Show thread

Internet is slow. It can't be DNS. There's no way it's DNS.

total.recursion.time.avg=9.375286
total.recursion.time.median=7.4386

It's DNS.

Reading a blog post from 2009 where someone mentions a 787 Dreamliner and realized that after more then 10 years I’ve still never seen one in the wild

I'm relieving stress my cultivating my fortunes file. Send me your best!

@solene Fantastic. I think I’ve found my tribe with the crowd.

OpenBSD hackathon #k2k20 started. deraadt committed the associated picture

(I am not at k2k20)

I could be happy with X11, if only it didn’t use that architecture and weren’t written that way

Semver is a meta-API, and developers who are cavalier about violating it can’t be trusted to created stable contracts

Just timed a full build-and-test in Go at .95s. I’ll need to re-implement this in Java anyway, so I’ll be curious to see what the time run returns there.

Show thread

I might be speaking too soon, because my reference is ~30s to run a full build and test with Maven, whereas 5ms in Go is just the test. So not an apples-to-apples comparison.

Show thread
Show more
Mastodon

A private Mastodon instance