00:02 (join) jesyspa
00:07 (quit) random_malice: Ping timeout: 244 seconds
00:07 (quit) bro_grammer__: Ping timeout: 244 seconds
00:14 (join) RacketCommitBot
00:14 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/oaVzjw
00:14 RacketCommitBot: [racket/master] When `plot-new-window?' is #t, `plot' and `plot3d' now create a frame in a new eventspace. This allows plot frames to be displayed programmatically. (Without this change, the frames are unresponsive until the REPL regains control.) - Neil Toronto
00:14 (part) RacketCommitBot
00:21 (quit) zyoung: Remote host closed the connection
00:25 (quit) antithesis: Quit: yes leaving
00:26 (join) antithesis
00:35 (join) nilyaK
00:55 (quit) gridaphobe: Remote host closed the connection
01:00 (join) drumond19
01:17 (join) RacketCommitBot
01:17 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/O1gmUQ
01:17 RacketCommitBot: [racket/master] Changed macros into functions + begin-encourage-inline; drops a few seconds from `images' compile, no measured performance penalty - Neil Toronto
01:17 (part) RacketCommitBot
01:44 (quit) drumond19: Remote host closed the connection
01:59 Cryovat: I was wondering about one thing in Typed Racket
02:00 Cryovat: If you specify a type as Vectorof Float, will that use a flvector, or a regular vector?
02:01 Cryovat: Once all the optimization magic is done
02:09 Cryovat: Hmm, looks like it's a regular one
02:12 Cryovat: Which of course makes sense, since vector-ref doesn't work on a flvector
02:12 Cryovat slaps self
02:16 (quit) antithesis: Quit: yes leaving
02:16 Cryovat scribbles mental note to investigate better before asking stupid questions
02:17 (join) antithesis
02:45 (join) kvda
03:03 (join) bluezenix
03:05 (quit) jonrafkind: Ping timeout: 248 seconds
03:08 (join) netrino
03:11 (join) neilv
03:14 (quit) nilyaK: Read error: Connection reset by peer
03:23 (quit) netrino: Quit: Ave!
03:39 (join) Kaylin
04:18 (join) bitonic
04:31 (join) cdidd
04:33 neilv: (UNSYNTAX (literal "#lang racket/base"))
04:41 (quit) bitonic: Quit: WeeChat 0.3.7
04:42 (join) snearch
04:52 (join) mceier
04:55 (quit) jyc__: Read error: Connection reset by peer
05:04 (join) papyrus
05:09 (quit) kvda: Quit: x__x
05:11 (join) masm
05:12 (quit) snearch: Quit: Verlassend
05:25 (quit) bluezenix: Quit: Leaving.
05:26 (join) bluezenix
05:28 (join) noelw
05:41 (quit) neilv: Ping timeout: 245 seconds
05:41 (quit) Kaylin: Quit: Leaving.
05:45 (quit) noelw: Quit: noelw
05:51 (join) noelw
05:55 (quit) noelw: Client Quit
06:05 (join) bitonic
06:06 (join) tfb
06:09 (quit) tfb: Read error: Connection reset by peer
06:10 (join) tfb
06:11 (quit) antithesis: Quit: yes leaving
06:11 (quit) bluezenix: Quit: Leaving.
06:11 (join) neilv
06:11 (quit) neilv: Changing host
06:11 (join) neilv
06:11 (quit) neilv: Client Quit
06:12 (join) mye__
06:14 (join) bluezenix
06:15 (quit) mye_: Ping timeout: 248 seconds
06:16 (quit) bluezenix: Client Quit
06:17 (join) bluezenix
06:30 (quit) bluezenix: Quit: Leaving.
06:31 (join) bluezenix
06:32 (quit) bluezenix: Client Quit
06:36 (join) bluezenix
06:47 (quit) bluezenix: Quit: Leaving.
07:05 (join) jeapostrophe
07:11 (join) bluezenix
07:11 (join) antithesis
07:24 (join) dnolen
07:34 (quit) jeapostrophe: Read error: Operation timed out
07:34 (quit) peeeep: Read error: Operation timed out
07:35 (join) peeeep
07:48 (quit) bluezenix: Ping timeout: 244 seconds
07:50 (join) bluezenix
07:52 (quit) bluezenix: Client Quit
07:55 (quit) Shvillr: Ping timeout: 248 seconds
07:56 (join) Shvillr
07:56 (quit) mceier: Quit: leaving
08:09 (join) bluezenix
08:15 (quit) bluezenix: Quit: Leaving.
08:27 (join) kanak
08:34 anonus: is it possible somehow to make an place-channel-get return after timeout ?
08:34 anonus: or there is no way except to make other place that send messages to first place after specified time ?
08:37 (join) random_malice
08:38 (join) hash_table
08:47 (join) bluezenix
09:26 (quit) hash_table: Ping timeout: 252 seconds
09:27 (quit) random_malice: Ping timeout: 252 seconds
09:37 (join) mithos28
09:42 (join) tyson1
09:45 (part) tyson1
09:57 (quit) mithos28: Quit: mithos28
10:14 (join) RacketCommitBot
10:14 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/_q0bPQ
10:14 RacketCommitBot: [racket/master] fix the contract context messages for the case of ->*'s #:rest arguments - Robby Findler
10:14 (part) RacketCommitBot
10:33 (join) getpwnam
10:34 (join) hash_table
10:35 (quit) bluezenix: Read error: Connection reset by peer
10:36 stamourv: Cryovat: That's not a stupid question at all! In fact, we thought about it.
10:39 stamourv: If the TR optimizer could prove that the vector you're working with doesn't escape (that is, it's only used within the module), that all functions that operate on it also don't escape, and that they only work on vectors that also don't escape (and that functions that work on them, etc.), then it could change all these vectors to flvectors, and change all these functions to use `flvector-ref' instead of `vector-ref'.
10:39 (join) bluezenix
10:39 stamourv: But that would require more analysis than we currently do.
10:41 stamourv: eli: BTW, I never found the struct property for promise forcing that you mentioned yesterday.
10:48 asumu: stamourv: collects/racket/private/promise.rkt
10:49 asumu: It doesn't appear to be exported publicly.
10:51 stamourv: Oh, I see.
10:51 stamourv: I was looking in the docs.
10:51 stamourv: If it's not exported, I'm not sure it's worth porting.
10:53 stamourv: asumu: You had a look at the hand-rolled generics lib in the parallel build stuff, right?
10:54 asumu: Not yet really, I just found it in a grep I was doing.
10:54 stamourv: Ok, I'll have a look then.
10:54 stamourv: Do you remember where it was?
10:55 asumu: collects/setup/parallel-do.rkt
10:56 asumu: Oh, it's using the generics thing from racket/class.
10:56 asumu: Which is not actually generic dispatch, AFAIK.
10:57 stamourv: Ok, so there's nothing for us to do, then?
11:01 asumu: Yeah, I dont't think so. These macros just bind generics for use with send-generic, which is faster than send.
11:01 stamourv: Ah.
11:01 stamourv: Ok.
11:01 (quit) jrslepak: Quit: This computer has gone to sleep
11:09 (quit) dnolen: Ping timeout: 260 seconds
11:25 bartbes: porting?
11:25 bartbes: what are you guys porting, and what are you porting it to?
11:26 stamourv: We just pushed support for generic functions for structs yesterday, and we've been porting parts of the code base to it.
11:28 bartbes: the question arises whether that improves it, though (since it obviously worked fine without), does it?
11:29 stamourv: It did, yes.
11:29 stamourv: But (IMO) generic functions can make code cleaner.
11:30 stamourv: That, and some parts of the code base were basically rolling their own.
11:31 Cryovat: stamourv: Thank you for your reply :)
11:33 Cryovat: Would there be a significant gain in switching my code to use flvector where relevant?
11:33 stamourv: Maybe?
11:33 stamourv: It really depends on your code.
11:34 stamourv: It should be a quick change, though, so you could try it out and measure.
11:34 stamourv: If you don't use any other vectors: `(require (rename-in racket/flonum [flvector-ref vector-ref]))'
11:35 stamourv is not sure whether he got the order right.
11:35 stamourv: rudybot: (require (renami-in racket/flonum [flvector-ref vector-ref]))
11:35 rudybot: stamourv: your sandbox is ready
11:35 rudybot: stamourv: error: #<undefined>:1:9: renami-in: not a require sub-form in: (renami-in racket/flonum (flvector-ref vector-ref))
11:35 stamourv: rudybot: (require (rename-in racket/flonum [flvector-ref vector-ref]))
11:35 rudybot: stamourv: Done.
11:35 Cryovat: Most of my vectors now actually are f32vector instances
11:35 stamourv: rudybot: (vector-ref (flvector 1.0 2.0) 0)
11:35 rudybot: stamourv: error: #<undefined>:1:1: compile: access disallowed by code inspector to protected variable from module: '#%flfxnum at: vector-ref in: flvector-ref
11:35 stamourv: ?
11:36 stamourv: Not sure what that was about.
11:36 stamourv: Cryovat: From the FFI?
11:36 Cryovat: Trying to piece together a simple GL game framework, haven't gotten beyond the level where I don't have to think about ffi yet
11:36 Cryovat: Err, poor phrasing, but I hope you understood my intent ;)
11:37 stamourv: So right now, you're taking f32vectors and converting them to regular vectors?
11:38 (join) mceier
11:38 Cryovat: This came more into play when I started thinking about how to do things when I've gotten out of the "C level"
11:41 Cryovat: I try to do as little involving the f32vectors as possible; mostly use them to move things into video memory
11:41 Cryovat: And then have shaders work on that data
11:42 Cryovat: I have made the unfounded assumption that working with ffi things would incur more overhead than native Racket data types
12:01 (quit) spanner_: Ping timeout: 248 seconds
12:03 (quit) tfb: Quit: sleeping
12:05 (join) netrino
12:05 (quit) bluezenix: Quit: Leaving.
12:11 (quit) antithesis: Quit: yes leaving
12:12 (quit) bro_down: Read error: Connection reset by peer
12:12 (join) antithesis
12:18 (join) spanner
12:20 (join) bro_down
12:23 (quit) karswell: Remote host closed the connection
12:27 (quit) bitonic: Quit: WeeChat 0.3.7
12:33 (join) karswell
12:34 (join) gridaphobe
12:39 (join) aalix
12:40 (quit) antithesis: Quit: yes leaving
12:45 (quit) karswell: Remote host closed the connection
12:52 (quit) mceier: Quit: leaving
12:53 (quit) peeeep: Ping timeout: 245 seconds
12:54 (join) sizz_
12:54 (join) peeeep
12:54 (join) bluezenix
12:54 (quit) sizz: Ping timeout: 248 seconds
12:56 (join) karswell
13:10 (join) antithesis
13:13 (quit) mye__: Quit: Leaving
13:38 (quit) hash_table: Ping timeout: 244 seconds
13:38 (quit) getpwnam: Ping timeout: 244 seconds
13:43 (join) __rahul__
13:57 (join) tfb
14:04 ozzloy: http://pastebin.com/yHv9bRVZ i'm under the impression that line 28 should fail since inexact numbers are represented with double precision floating point numbers.  this is from sicp exercise 1.7 http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-10.html#%_sec_1.1.7
14:05 (quit) netrino: Ping timeout: 244 seconds
14:06 (quit) tfb: Quit: gone
14:16 danking: ozzloy: What is your definition of fail? Do you expect it to diverge/run forever?
14:18 (join) netrino
14:19 (join) jacius
14:24 (join) bitonic
14:28 (join) mceier
14:32 ozzloy: danking, yes
14:32 ozzloy: run forever
14:33 ozzloy: i think i'll get that behavior if i reimplement with this http://docs.racket-lang.org/reference/flonums.html?q=ieee
14:37 (join) getpwnam
14:37 stamourv: ozzloy: Pure speculation: if this code does all its work inside the FP registers, they would use the FPU's increased precision. (was it 80 bits? maybe more.)
14:38 (join) hash_table
14:38 ozzloy: stamourv, wow, wouldn't that blatantly violate IEEE standards?
14:38 stamourv: To defeat that, store each guess in a box, and unbox it when you use it again. That should force it out.
14:39 stamourv: ozzloy: Maybe. I'm no numerical analyst, but IIRC that's how x86 does it these days.
14:39 ozzloy: also if that is the case then i could just increase the exponent to ... whatever the fractional size is in this 80 bit representation
14:40 ozzloy: oh
14:40 stamourv: ozzloy: http://gcc.gnu.org/ml/gcc-patches/2001-03/msg00466.html
14:41 ozzloy: i just realized i can't store (+ 1 (expt (expt 2.0 53.0) 2.0))
14:41 ozzloy: it's being truncated to 2^106.0 because the granularity up there is greater than 1.0
14:41 ozzloy: ooops
14:43 (join) jonrafkind
14:43 (join) anRch
14:52 (quit) anRch: Quit: anRch
14:56 bremner: anyone know an easy way to plot vectors from the origin in 3D (with PLoT)?
14:58 (quit) bluezenix: Quit: Leaving.
15:26 (join) Kaylin
15:31 (join) snearch
15:34 (join) bluezenix
15:36 (join) jeapostrophe
15:36 (quit) jeapostrophe: Changing host
15:36 (join) jeapostrophe
15:52 ozzloy: stamourv, well wow
15:52 (quit) karswell: Read error: Operation timed out
15:55 (quit) jonrafkind: Ping timeout: 252 seconds
15:57 (join) jonrafkind
15:57 (quit) jacius: Remote host closed the connection
15:57 (join) karswell
15:59 (quit) antithesis: Quit: yes leaving
16:12 stamourv: ozzloy: ?
16:13 (quit) bluezenix: Ping timeout: 244 seconds
16:14 (join) bluezenix
16:16 (quit) jeapostrophe: Ping timeout: 245 seconds
16:28 (quit) jesyspa: Ping timeout: 265 seconds
16:29 (join) jesyspa
16:36 (join) MayDaniel
16:43 (quit) netrino: Quit: Ave!
16:51 (join) kanak1
16:51 (quit) kanak: Ping timeout: 252 seconds
16:54 (quit) stamourv: Quit: ERC Version 5.3 (IRC client for Emacs)
16:54 (join) stamourv
17:01 (quit) jesyspa: Quit: leaving
17:09 (join) RacketCommitBot
17:09 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/1ixvuQ
17:09 RacketCommitBot: [racket/master] slideshow/pict: fix clipping problem with `standard-fish' - Matthew Flatt
17:09 RacketCommitBot: [racket/master] scribble/eval: fix error line wrapping - Matthew Flatt
17:09 RacketCommitBot: [racket/master] new error message convention - Matthew Flatt
17:09 (part) RacketCommitBot
17:10 asumu: Oh neat, new error messages.
17:14 jonrafkind: now we can look like haskell
17:15 stamourv: jonrafkind: How so?
17:17 asumu: The new errors also make type errors contract violations.
17:22 (quit) __rahul__: Ping timeout: 240 seconds
17:26 (quit) gridaphobe: Remote host closed the connection
17:34 (quit) snearch: Quit: Verlassend
17:35 jonrafkind: haskell error messages are multiline
17:35 (join) __rahul__
17:44 (quit) __rahul__: Quit: Leaving
17:45 (quit) papyrus: Ping timeout: 260 seconds
17:47 (join) RacketCommitBot
17:47 RacketCommitBot: [racket] plt pushed 5 new commits to master: http://git.io/5qolyQ
17:47 RacketCommitBot: [racket/master] Add reminder that unsafe-flexpt is currently the same as flexpt. - Vincent St-Amour
17:47 RacketCommitBot: [racket/master] Fix handling of NaN and infinities. - Vincent St-Amour
17:47 RacketCommitBot: [racket/master] Fix NaN and inf handling in flonum specific op types. - Vincent St-Amour
17:47 (part) RacketCommitBot
17:47 (quit) bluezenix: Read error: Connection reset by peer
17:47 (join) bluezenix
17:52 asumu: I bet a lot of languages other than Haskell have multi-line errors too.
17:53 jonrafkind: like c++ template errors
17:55 ozzloy: stamourv, that thread you linked to about extended precision
17:55 ozzloy: stamourv, btw, it turns out the problem was the granularity if the bigger number.  i couldn't just add 1 to it to get a new number whose square root was less than 1 away from the representable root
17:56 ozzloy: i had to add 2^54
17:56 stamourv: ozzloy: Ah, makes sense.
17:56 ozzloy: which is big, but not nearly as big as 2^106
17:56 ozzloy: so the result still has a root near, but not near enough, to the other root
17:57 ozzloy: er.. to the other ... the initial root
18:08 (quit) SHODAN: Quit: No Ping reply in 180 seconds.
18:08 (join) SHODAN
18:12 (quit) kanak1: Ping timeout: 260 seconds
18:13 (quit) MayDaniel: Read error: Connection reset by peer
18:18 (join) dented42
18:19 (quit) bluezenix: Quit: Leaving.
18:20 (join) jyc
18:25 (join) untrusted
18:26 Cryovat: When is 5.2.2 due? ;)
18:29 jonrafkind: august I think
18:30 Cryovat: The nightlies feel tempting then
18:31 (quit) karswell: Remote host closed the connection
18:39 (quit) mceier: Quit: leaving
18:40 (join) jacius
18:41 (join) karswell
18:48 (quit) karswell: Remote host closed the connection
18:50 (join) zyoung
18:51 (quit) Kaylin: Quit: Leaving.
18:53 (join) karswell
19:00 (join) dnolen
19:01 (quit) zyoung: Remote host closed the connection
19:15 (join) nilyaK
19:16 (quit) karswell: Remote host closed the connection
19:31 (join) jrslepak
19:32 (join) karswell
19:38 (quit) jonrafkind: Ping timeout: 265 seconds
19:46 (quit) aalix: Ping timeout: 260 seconds
20:25 (quit) cdidd: Remote host closed the connection
20:42 (quit) untrusted: Remote host closed the connection
20:56 (join) RacketCommitBot
20:56 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/Dh75dA
20:56 RacketCommitBot: [racket/master] racket/sandbox: add sandbox-propagate-exceptions - Matthew Flatt
20:56 RacketCommitBot: [racket/master] adjust `raco setup' replay of exception messages to handle multiple lines - Matthew Flatt
20:56 RacketCommitBot: [racket/master] correct a `raise-argument-error' conversion mistake - Matthew Flatt
20:56 (part) RacketCommitBot
21:09 (join) rick__
21:10 rick__: i was wondering if anyone knew of functions that were the opposite of above, and beside. (for instance if there was something  like below and to the left of)
21:23 rick__: ?
21:35 (quit) dnolen: Ping timeout: 248 seconds
21:43 (quit) sizz_: Quit: quitted.
21:44 (quit) rick__: Ping timeout: 245 seconds
21:44 (join) sizz
21:48 ozzloy: http://www.meetup.com/hac-dc/events/64947482/ going over sicp in dc area.  you're all invited
21:55 (quit) nilyaK: Quit: Leaving.
22:02 (join) nilyaK
22:03 (quit) nilyaK: Client Quit
22:05 (join) zyoung
22:05 (quit) ivan\: Quit: ERC Version 5.3 (IRC client for Emacs)
22:06 (join) ivan\
22:10 (join) jonrafkind
22:18 (quit) bitonic: Quit: WeeChat 0.3.7
22:21 (quit) masm: Quit: Leaving.
22:28 (join) nilyaK
22:37 (quit) jonrafkind: Ping timeout: 265 seconds
22:40 (quit) dented42: Ping timeout: 244 seconds
23:13 (join) dented42
23:48 (quit) nilyaK: Quit: Leaving.