#haskell - Fri 27 Apr 2007 between 00:00 and 00:17

NY Lost Funds



cedricshockjcreigh: Yeah.
mgsloan: Trying to call things with tuples can get ugly. I think the apply bifs take lists. Not sure; there's a way to do it but it's been a while since I erlanged up.
sorearbifs?
cedricshocksorear: built in functions.
sorearare builtins magic in erlang?!
cedricshocksorear: The atoms are not strings, thought they are uniquely identified in code by one (and I believe it can be backwards looked up so symantically they probably are). The bifs aren't magic, but that's where you need to look in the erlang docs for almost everything interesting.
sorearcedricshock: (guessing) strings with fast equality, like Lisp atoms?
mgsloanare there any agent-based static typed functional languages?
not that there's much better than erlang - every language has it's downsides
kpreidjcreigh: pattern matching is not especially associated with static typing
cedricshocksorear: Yeah. I think theres a maximum of 2^32 atoms in an erlang cluster, so you can probably guess how they're implemented...
PseudonymUsing nanotech, obviously.
jcreighkpreid: sure; I can see how it could work in a dynamically typed language. But in practice it seems to be used more by statically typed languages.
cedricshockmgsloan: If you just want the pattern matching benefits of static typing that's very easy to do in erlang; make tuples which have the first value be an atom of their type.
sorearwhat? not md5? ;)
well sha1 now
kpreidjcreigh: I hadn't especially noticed... perhaps we've seen different languages
mgsloanright, it's actually more about the general benefits of static typing than anything else :)
sorearIf you have an atom collision, you're screwed. but you don't care because you have prize money!
jcreighkpreid: probably. I'm thinking of Haskell and ocaml specifically.
sorear: CRC-32, no doubt. :)
kpreidStatic, no patterns: C, Java, C#, ...; Static, patterns: Haskell, ML. Dynamic, patterns: Prolog, Erlang, E. Dynamic, no patterns: Perl, Python, ...
jcreighE?
cedricshockkpreid: What do you mean by patterns?
kpreidhttp://www.erights.org/
lambdabotTitle: Welcome to ERights.Org
kpreidcedricshock: the sort which bind variables to parts of a data structure
and can fail
cedricshockkpreid: D'oh, what we were just talking about.
mvanierj xmonad
cedricshockDynamic: ALL PATTERNS (only way to write): Q, stratego (might be static)
mgsloan: You can get quite a few of the statically typed benefits, since erlang is statically typed, there's just exactly one type.
mgsloan: But none of the interesting ones :(
actionkpreid thinks pattern matching is a Very Good Thing because otherwise you get to write horrific nested conditionals to test and destructure
kpreidthinks pattern matching is a Very Good Thing because otherwise you get to write horrific nested conditionals to test and destructure
...which probably means I've just gotten used to patterns and seek to emulate them when I don't have them
s/emulate/approximate/
dibblegopattern matching is "nested conditionals to test and destructure", only short and concise
cedricshockHmm, I should make a compressor and decompressor for my run-length encoding lists.
nohopeHello :)
sorearhello!

Page: 2 9 16 23 30 37 44 51 58 65 

IrcArchive

NY Lost Funds