Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bascule
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
64 ms
·
571.
▲
by
bascule
15y ago
You don't have a problem with one of the most trafficked sites for programming/web startup-related news implementing HTTP incorrectly? Do you ignore whether your HTML is valid just because the browser rendered it correctly?
572.
▲
by
bascule
15y ago
My favorite part of HN's headers: the lines are separated by naked LFs instead of CRLF, in violation of the HTTP spec
573.
▲
by
bascule
15y ago
Desktop Linux users are flocking to a distro that looks like an uglier version of Windows 2000? "Tonight we're gonna GUI like it's 1999!" I guess if they had any aesthetic sense, they'd probably be switching to OS X...
574.
▲
by
bascule
15y ago
Google acquired Aperture Science?!?!
575.
▲
by
bascule
15y ago
Looks like it uses a kernel interaction of some sort to implement memory barriers in software (whereas Vega provided hardware memory barriers): http://www.azulsystems.com/products/zing/c4-java-garbage-col...
576.
▲
by
bascule
15y ago
Cliff Click rules everything around me
577.
▲
by
bascule
15y ago
Why does everything else on the page say RailwayJS except the headline, which says "RightwayJS"?
578.
▲
by
bascule
15y ago
Even if all your databases are running at LUDICROUS SPEED thanks to SSDs, his argument is still irrelevant. That same LUDICROUS SPEED can be applied to technologies which can be used to mitigate the computational times of slower languages (
579.
▲
by
bascule
15y ago
Please don't take what Cringely has to say as an accurate barometer of the Ruby community. While I'll freely admit that Rails (and web applications in general) is an enormous part of what people are using Ruby for, it's used for many non-we
580.
▲
by
bascule
15y ago
This article conflates "dynamic typing" with "weak typing" all over the place. First he whines that type declarations aren't mandatory (at least within the scope of what a type inferencer can resolve), claims the runtime won't be able to de
581.
▲
by
bascule
15y ago
Zed had two parsers: a server parser as part of Mongrel and a client parser as part of RFuzz. I actually used the client parser from RFuzz in my Rev/Cool.io libraries. I talked to Zed about his plans for these, he wanted to do a unified par
582.
▲
by
bascule
15y ago
I plan on eliminating Net::HTTP entirely and writing my own client using Carl Lerche's parser for his Picard library which is written in Ragel: https://github.com/tarcieri/http/tree/master/parser
583.
▲
by
bascule
15y ago
Here's the new one I'm working on: https://github.com/tarcieri/http httparty does a few of the same tricks, but I still find its API a bit obtrusive. At least it automatically parses JSON! (my library does this too, if an appropriate libr
584.
▲
by
bascule
15y ago
I just started working on a similar library for Ruby: https://github.com/tarcieri/http
585.
▲
by
bascule
15y ago
At your request I'm answering this question... I never used the word "fanboy" in my post. That said, there were two people who really thought the nextTick solution was a legitimately good idea. Contrary to the idea that they were just count
586.
▲
Node.js has jumped the shark
(unlimitednovelty.com)
321 points
by
bascule
15y ago
|
126 comments
587.
▲
by
bascule
15y ago
If I try to ask for a large number (say 1 million) I run out of RAM: $ node app.js FATAL ERROR: CALL_AND_RETRY_2 Allocation failed - process out of memory Somehow you're using O(n) RAM to do the calculation. Seems bad bro. This co
588.
▲
by
bascule
15y ago
Fibonacci isn't a kind of problem that should be bounded on RAM though. Something is fundamentally broken with the way this is implemented.
589.
▲
by
bascule
15y ago
Hmmm... $ time curl localhost:3000/1000000 curl: (52) Empty reply from server real 1m0.392s user 0m0.006s sys 0m0.004s This actually crashes the server because it runs out of memory: $ node app.js FATAL
590.
▲
Object Oriented Concurrency: A Deep Dive into Celluloid
(unlimitednovelty.com)
6 points
by
bascule
15y ago
|
1 comments
591.
▲
by
bascule
15y ago
For speed
592.
▲
by
bascule
15y ago
In what manner is HTTP "unreliable" that IIOP isn't?
593.
▲
by
bascule
15y ago
I always liked the term PORC (Plain Old Ruby Classes) for the Ruby analogue of a POJO
594.
▲
by
bascule
15y ago
This looks cool, but I wish it were http://socket.io compatible
595.
▲
The Trouble with Erlang (or Erlang is a ghetto)
(unlimitednovelty.com)
50 points
by
bascule
15y ago
|
1 comments
596.
▲
by
bascule
15y ago
Apple's "leaner install base" for Lion seems to ship with Oracle's VisualVM tool, which won't work until you install Java
597.
▲
Funargs: Ruby-like blocks for Erlang
(groups.google.com)
1 points
by
bascule
15y ago
|
0 comments
598.
▲
by
bascule
15y ago
It's τ/2. The area of a circle is τr²/2. You may be familiar with the idea of x²/2 from calculus: it's an integral, which can be used to compute areas.
599.
▲
by
bascule
15y ago
The section of the Tau Manifesto on the area of the circle (which the author describes as pi's coup de grace) was the one that convinced me tau makes more sense.
600.
▲
by
bascule
15y ago
"Barriers are slow" in what way? Azul was hitting 10-20ms pauses while successfully garbage collecting 500GB+ heaps using hardware memory barriers on their Vega architecture, which had up to 768 CPUs. Cliff Click, Azul's principal JVM archi
More ›