00:42 (quit) jonrafkind: Ping timeout: 240 seconds
05:18 (join) masm
11:16 (join) bremner
11:54 (join) BW^-
11:54 BW^-: hi guys - questions
11:54 BW^-: what's racket's threading model, internally?
11:54 BW^-: i'm interested to know what former mzscheme really is - for instance, http://en.wikipedia.org/wiki/PLT_Scheme tells of a macro API, what about it?
11:55 BW^-: the "JIT" VM is completely made for racket only?
12:03 (join) jonrafkind
12:12 (join) winxordie
12:30 (part) djahandarie
12:49 (join) anRch
12:59 (quit) masm: Ping timeout: 245 seconds
13:02 (join) masm
13:11 chandler: BW^-: I'm not sure I understand what you mean by "former mzscheme".
13:11 BW^-: it's now racket
13:11 BW^-: anyhow, i got things from the manual -
13:11 BW^-: the threading model is green
13:12 BW^-: and the macro API is, I suppose, what's mentioned on http://docs.racket-lang.org/guide/languages.html = customizable reader and some other thing
13:12 BW^-: regarding the VM, is it an "interpreter" like java or does it do anything like native compilation?
13:12 BW^-: what is delimited and composable control is about? apparently racket has it (http://www.cs.utah.edu/plt/publications/icfp07-fyff.pdf)
13:14 (quit) anRch: Quit: anRch
13:14 chandler: The macro API is based on the macro system known as `syntax-case'. The reader is also customizable, yes.
13:15 chandler: I don't understand what ``"interpreter" like java'' means, given that Java is (almost always) JIT-compiled.
13:15 chandler: Racket has a JIT on some platforms, is executed through a bytecode interpreter on others, and can also be compiled to native code via C (but this is as I understand it usually slower than the JIT).
13:16 chandler: Delimited and composable continuations are a model of continuations that allow just part of a continuation to be reified and to be called in a composable fashion. It's a very flexible model for building control-flow operators.
13:25 BW^-: hm
13:25 BW^-: what's a possible application for it? (delimited..)
13:28 chandler: Generators and coroutines are natural applications. For instance, in a generator you would only want to capture the portion of the continuation between where the generator was called and where the next result is yielded.
13:30 chandler: Using a full continuation has several disadvantages. It can cause spurious calls to `dynamic-wind' guards, and it can result in an apparent memory leak as objects that would not otherwise be retained are held by irrelevant frames of captured continuations.
13:31 (quit) BW^-: Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org
13:38 chandler: derrida: Also look at Geiser, which provides a set of tools for working with an interactive Racket in Emacs: http://gitorious.org/geiser/mainline/trees/master
14:52 (join) mije
14:53 mije: hi there
14:55 jonrafkind: sup
14:56 mije: racket is up
14:56 jonrafkind: the floor is down
14:58 mije: roses are red
14:59 jonrafkind: pigs cant fly
15:00 mije: they can when they are on wings
15:00 mije: i'm looking for someting to do in order to learn racket
15:02 jonrafkind: http://htdp.org/
15:03 j3parker: ^ ++
15:04 mije: yeah i read it
15:04 mije: sicp as well
15:05 bremner: plai should work with racket
15:07 mije: my difficulties are about how object and functional programming can be merged
15:07 bremner: oh. well. You could read CTM, but it isn't scheme based
15:08 bremner: "Concepts, Techniques, and Models" is the start of the title.
15:09 mije: i'm ok with racket per se, but i'm a bit puzzled about how to use wisely all the possibilities it allows
15:09 jonrafkind: you shouldn't even try
15:09 jonrafkind: i mean, use what works at the moment
15:10 jonrafkind: if you realize that you can write something in an OO manner, then do that
15:10 mije: i'm from an oo background and i have this reflex to see things as a bunch of fragmented states at first
15:15 mije: i guess something like http://www.katacasts.com/ dealing with little puzzles solved in racket would help people to get a grasp on the language
15:15 mije: and it would be fun as well
16:31 (join) sizur
16:31 sizur: congrats on new name!
16:31 sizur: i think what you guys are doing is great
16:33 sizur: for some reason parenlog is not working...  just trying to run example
16:35 sizur: http://pastebin.com/UrmsKLdG
16:37 (quit) mije: Ping timeout: 265 seconds
16:42 eli: sizur: This looks like a result of a minor incompatible update, I've told Jay about it, so he'll probably fix it soon.
16:43 sizur: eli: thanks
16:43 (part) winxordie
20:07 (quit) masm: Quit: Leaving.