cscheid

38 ms·

361.
by cscheid 16y ago
Not so fast. You can still get nondeterministic behavior with things like thermal effects (think temperature affecting clock speed affecting relative timing between threads running on different processors)
362.
by cscheid 16y ago
It is incredibly painful, and there are some egregious offenders which I find hard to believe went through any testing (I'm looking at you, New York Times iPad app).
363.
by cscheid 16y ago
The other way to think of emails is as datagrams. Sometimes you'll get a reply, sometimes you won't. Or, from the other side: sometimes you'll be able to answer, sometime you won't. Don't feel bad about the latter case. If you don't hear ba
364.
by cscheid 16y ago
Logo?
365.
by cscheid 16y ago
(disclaimer: I work there) Actually, the AT&T research labs are alive and well: http://www.research.att.com Also, Bell Labs went to Lucent when AT&T split, not because of SBC.
366.
by cscheid 16y ago
Swivel tried, and it died: http://news.ycombinator.com/item?id=1786982
367.
by cscheid 16y ago
How exactly does Haskell have dynamic scoping? An appropriate monad does not count.
368.
by cscheid 16y ago
Anything with dynamic scoping belongs squarely in the ancient category. We're about a decade overdue for replacing Elisp with Scheme.
369.
by cscheid 16y ago
You're right that it is a problem, but defining objective criteria before knowing exactly what it is that you're looking for is potentially just as bad. This is part of the reason folks like Tukey defended statistical graphics early on: exp
370.
by cscheid 16y ago
"Only for networks? No" That's because it looks like he reinvented Parallel Coordinates: http://peltiertech.com/Excel/Charts/ParallelCoord.html
371.
The Rise and Fall of Swivel.com(eagereyes.org)
22 points by cscheid 16y ago|4 comments