bascule

41 ms·

421.
All the crypto code you've ever written is probably broken(tonyarcieri.com)
112 points by bascule 14y ago|128 comments
422.
by bascule 14y ago
For what it's worth, I've done two libev bindings, rev/Cool.io (originally in 2008) and nio4r in 2011
423.
by bascule 14y ago
> I'm not sure I understand how EventMachine "guts the Ruby internals" EventMachine does not use the Ruby IO primitives (e.g. TCPSocket, UDPSocket, etc) as the basis of its IO abstraction, and instead has reimplemented its own set of pr
424.
by bascule 14y ago
Hello, author of Celluloid here. I think DRb pretty cool and has long been underutilized. That said, there are a few fundamental design problems with the way it works that I think DCell solves: 1. Distributed systems really need to be built
425.
by bascule 14y ago
Then I assume your development workstation is an exact duplicate of a production server, and is serving a production load? Even getting a staging environment to realistically resemble production is an incredibly difficult undertaking. Your
426.
by bascule 14y ago
No, you're wrong, an extremely scant number of applications can completely avoid allocating memory in the old generation. Your best bet would probably be to use the Disruptor approach of preallocating everything then avoiding any subseque
427.
by bascule 14y ago
If you're confused about what InvokeDynamic is... Previously languages like Ruby suffered on the JVM as the set of JVM instructions available were designed around the Java programming language. This meant that for invoking Ruby methods and
428.
by bascule 14y ago
Yes, each Ruby method needs to get executed 50 times to JIT to JVM bytecode, and from there it takes an additional 10,000 calls for HotSpot to JIT that to machine code. Any hotspots therefore need to be hit at least 10,050 times in order to
429.
by bascule 14y ago
"rvm get head; rvm install jruby" also works ;)
430.
by bascule 14y ago
> Slow startup / test runs. At one point, loading large amounts of files (like during rails bootup) was much slower on jruby than MRI. You can develop on MRI and deploy on JRuby. Several people, including myself and Square, develop JRub
431.
by bascule 14y ago
Also note that for Fixnum dependent benchmarks, JRuby has to allocate memory due to boxed integers, whereas on the de facto interpreter it's able to use tagged pointers which allow Fixnums to be treated as simple numeric values. This may be
432.
by bascule 14y ago
> So with about the same CPU time, JRuby will use up to 100x memory in worst case? The JVM often trades increased memory usage for increased performance. That said "100x memory" is total hyperbole, even in the context of a microbenchmar
433.
by bascule 14y ago
> "Accumulo's cell-level security makes it possible to set access control for individual pieces of data using 'visibility tags.'" Are these so-called "visibility tags" really crypto-capabilities?
434.
by bascule 14y ago
"What I like about TypeScript - so far - is that TypeScript’s static typing could enable better tooling with warning squiggles, easy statement completion, plus smart refactoring. You also get easy navigation around code, as well as find ref
435.
by bascule 14y ago
I love this book but I'm not sure "Hacker's Delight" is really a proper name for it. More like "The Bit Twiddler's Guide To Fame And Fortune"
436.
by bascule 14y ago
Closed source is a definite dealbreaker: how do I know there aren't horrible flaws in their cryptographic protocols? Short of reverse engineering their binaries there's not you can do. The features AeroFS provides seem incredibly similar to
437.
by bascule 14y ago
Vitamin D supplements are generally D3 by design because that's the form most readily absorbed by the body. That said you're probably better off getting it from catfish or salmon (which also contain omega fatty acids, a "good" fat which you
438.
by bascule 14y ago
I thought it was funny that this blog post didn't mention getting outside as a potential solution. That said, it's not like it's a dichotomy: I take a daily vitamin D supplement, try to eat as much catfish and salmon as I can, and try to ge
439.
by bascule 14y ago
Extensive work has already gone into improving Rails startup time. If you want to further improve it, try removing railties you aren't using.
440.
by bascule 14y ago
The release notes for Django 1.4, for example, list support for both Time Zones and Selenium. Time Zones were added to Rails 2.1 in 2008. selenium-on-rails also debuted in 2008. Does it not seem strange to you that it took 4 years after Rai
441.
by bascule 14y ago
I'm friends with Jose and quit developing my language Reia because I thought Elixir was better. I've since gone on to develop Celluloid and haven't looked back ;)
442.
by bascule 14y ago
Last I checked out Django they had just added a bunch of features from Rails 1.2. (Edit: sorry, Rails 2.1) Python suffers from their web framework community being too fragmented, so none of the projects are as mature as Rails.
443.
by bascule 14y ago
Cramp is based on Rack, just like Rails... there is nothing stopping you from using the two in conjunction.
444.
by bascule 14y ago
> But that's old news, and doesn't need to be re-hashed now Well gee Jeremy, thanks for rehashing it then! ;) I work on a legacy Rails 2.x app at my day job. Wading through the alias_method_chain soup is not fun. I for one am glad the t
445.
by bascule 14y ago
They talk about servers with "128, 256 and 512 processor cores" but how effectively can that be leveraged given ARM's weakly consistent memory model? Wouldn't systems like the JVM which have the technology to utilize this many cores struggl
446.
by bascule 14y ago
The difference between the output of SHA512 and random data is negligible, provided that the master password is of non-negligible entropy. This is a necessary property of a secure hash function. HMAC or HKDF would probably be more appropria
447.
by bascule 14y ago
Does this have any applications to elliptic curve cryptography?
448.
by bascule 14y ago
Not sure what you mean by "parallelizes better". However Erjang and Kilim on the JVM have managed to outperform the Erlang VM (thanks to a shared heap they provide zero-copy messaging unlike the Erlang VM) and with InvokeDynamic Erjang shou
449.
by bascule 14y ago
The main difference between CSP in Go and the actor model in Erlang is that actor mailboxes are many-to-one whereas CSP channels are many-to-many. If you were to model CSP using actors, channels would be actors.
450.
by bascule 14y ago
It's pretty slow, among other things

More ›