00:00 (quit) dnolen: Ping timeout: 244 seconds
00:05 mithos28: Does anyone know of any use cases for dynamic-wind besides resource cleanup?
00:09 (quit) jao: Ping timeout: 244 seconds
00:14 (quit) mithos28: Remote host closed the connection
00:14 (join) mithos28
00:26 (quit) thesnowdog: Read error: Connection reset by peer
00:29 (join) thesnowdog
00:30 (join) ambrosebs
00:35 (quit) getpwnam: Ping timeout: 245 seconds
00:36 (quit) hash_table: Ping timeout: 245 seconds
00:42 (quit) thesnowdog: Read error: Connection reset by peer
00:43 (join) thesnowdog
00:47 (quit) jeapostrophe: Ping timeout: 246 seconds
00:54 asumu: mithos28: the paper "Constraining Control" mentions implementing fluid variables/parameters as one application.
00:55 asumu: Also the only different between control/prompt & shift/reset is whether the captured continuation contains a prompt or not.
00:55 mithos28: Do continuation marks supercede that implementation mechanism?
00:55 asumu: Probably.
00:55 mithos28: why are shift/reset called static versus control/prompt called dynamic delimited control?
00:57 asumu: Because if the captured continuation itself uses control, it can't jump past the continuation application.
00:57 asumu: So this statically delimits futher control.
00:58 asumu: IIUC, this is what enables the original shift/reset type & effect system to work.
00:59 mithos28: ok, so with control I could have a procedure that calls control multiple times to escape a prompt, where as I cannot do that with shift because each time I add the prompt back
01:02 (quit) ambrosebs: Ping timeout: 268 seconds
01:03 asumu: Well, it's more that if you have (reset (... (shift k (+ 1 (k 5))))), the application of k can't cause the (+ 1 []) to be ignored, since any further jumps go to (+ 1 []).
01:03 (nick) billyoc -> billyocZzz
01:03 asumu: So it doesn't let you clobber the expression you use in the shift.
01:04 asumu: Both control/shift retain the outer prompt.
01:04 mithos28: ah, ok
01:05 mithos28: I'm trying to understand your proposal to change call/cc and it seems like you are requiring it to invoke the abort handler, and then hope that that reinstates the old continuation
01:06 asumu: Essentially, yes.
01:07 asumu: Which means in some contexts, call/cc might not work. But it's a tension between letting call/cc work or breaking the invariants of prompts.
01:07 mithos28: call/cc would only work with abort handlers that work the default way
01:08 mithos28: the issue I see is with barriers
01:08 asumu: What's the issue?
01:09 mithos28: that I install a prompt, install a barrier, do some work, capture a continuation to the prompt, do some more work, invoke the continuation
01:10 mithos28: If I actually have to go down to the prompt and invoke the abort handler, then I cannot reinstate the continuation because it contains a barrier
01:10 mithos28: its similar to the dynamic-wind issue
01:10 asumu: Oh, you're saying now you can't re-enter with that continuation. Hmm, interesting.
01:11 (quit) mceier: Quit: leaving
01:11 mithos28: because the current implementation is smart and doesn't actually jump up that much
01:16 asumu: Yeah, you're right that it doesn't work as it is. At least, not if you actually implment call/cc with call-with-composable-continuation.
01:16 asumu: That's pretty unfortunate.
01:20 mithos28: The issue is the magic of call/cc being able to inspect the continuation frames
01:24 asumu: I wonder if we could get away with acting like normal call/cc when it's in the same prompt but otherwise act like an abort.
01:24 asumu: Implementation-wise, there's already a clear distinction in the two cases so it seems implementable.
01:25 mithos28: when what is the same prompt (as what)?
01:25 asumu: The continuation application is in the same prompt.
01:25 asumu: As the prompt for the captured continuation.
01:28 mithos28: I think you would also need the caveat; and the abort handler is the default handler.
01:28 mithos28: But at first glance, it seems ok
01:29 mithos28: and that would work fine for scheme compatibility because it has no notion of abort-handler/prompt so the behavior would always be the same
01:30 asumu: Right. Although if you ignored the handler when it's not the default, that's still a problem for TR.
01:30 asumu: So it may still have to be incompatible there.
01:31 mithos28: I'm saying that if you implement an abort handler that does the same thing as the normal handler but with a logging statement, you would hit the barrier
01:37 (join) ambrosebs
01:40 mithos28: I think your point is that 'if there is a custom abort handler, call/cc needs to invoke it'
01:41 asumu: Hmm, yes, that should actually be good enough.
01:41 asumu: Although not sufficient if you want things like dynamic-wind compatibility.
01:42 mithos28: I think if you have a custom abort handler, you need to trigger the dynamic winds
01:44 asumu: Why's that? call/cc could just ignore handlers if it's safe to do so to trigger dynamic winds as it currently does.
01:44 asumu: (and it should be safe if the continuation being installed is the same as the current one)
01:46 asumu: Also, is this supposed to raise a barrier error? https://gist.github.com/3549450
01:46 asumu: Seems like as documented, this should only remove a barrier and thus be safe.
01:47 mithos28: I'm viewing it as call/cc is like call-with-composable and invoking =abort/invoke the composable one, as you sent in the email
01:48 mithos28: in that world a custom abort handler needs to be triggered, but if you are trying to maintain old behavior I see the argument for what you said
01:50 asumu: Yeah, it's unfortunate because I think proper use of prompts can make the call/cc+dynamic-wind behavior unnecessary.
01:50 mithos28: The example seems like it should work to me, but I also get the barrier
01:50 asumu: But we can't really break existing invariants just with that argument.
01:51 asumu: Okay, will submit it as a bug and see what Matthew thinks.
01:52 mithos28: I changed it and replaced it with a dynamic-wind printing 'before and 'after
01:53 mithos28: and got before, after, 5
01:54 mithos28: And the barrier should be eqivalent to the printing of the before, so it looks like an error to me
01:54 Haffe: What is the rationale behind the draw-ellipse taking the dimensions of the bounding box as arguments instead of a mid point, and half-axis lenghts?
01:56 (quit) ambrosebs: Ping timeout: 268 seconds
01:57 mithos28: No idea, maybe the underlying graphics libraries make it easier for that way to work
01:57 Haffe: wx does it this way as well.
01:58 asumu: Haffe: I imagine it's compatibility with Cairo or perhaps wx.
01:58 Haffe: I wrote a tiny wrapper to convert midpoint + half-axis to bounding box.
01:58 asumu: racket/gui was originally tied to wx.
01:58 mithos28: cairo looks like it does it the other way from a quick glance
01:58 (join) ambrosebs
01:58 Haffe: Is there a way to submit patches?
01:58 mithos28: but it could be from the original wx
01:59 mithos28: yeah github pull requests work pretty well
01:59 asumu: mithos28: also, thanks for pointing out that continuation barrier issue. Hadn't thought about barriers enough.
02:00 asumu: I also don't think it's in the Redex model, so its semantics might be weird.
02:00 mithos28: no problem
02:00 mithos28: Isn't it the same as a dynamic-wind pre-thunk that throws an exception
02:01 mithos28: but only on the second time through
02:02 asumu: That sounds right, but I can't say for sure.
02:07 mithos28: asumu: Can I join the interested parties on that bug, or am I stuck just polling it?
02:09 asumu: mithos28: bug replies usually get sent to -dev. I can probably add you directly if you want though.
02:10 mithos28: I don't see them in dev. They always end up in my inbox, where dev and users are filtered under a label
02:10 mithos28: Yeah, If you could add me that would be nice
02:12 (join) tfb
02:13 asumu: mithos28: you should be on the notify-list now
02:13 mithos28: thanks
02:14 asumu: Oh, you're right. Bug notifications don't go to dev but a separate list. It's commit replies that get to dev.
02:15 mithos28: yeah, that matches with my observations
02:22 (join) mceier
02:40 (quit) thesnowdog: Read error: Connection reset by peer
02:55 (join) thesnowdog
03:00 (quit) mithos28: Quit: mithos28
03:02 (join) mithos28
03:03 (join) danking
03:07 (quit) jonrafkind: Ping timeout: 246 seconds
03:08 (quit) Kaylin: Read error: Connection reset by peer
03:09 (join) billyocZ`
03:11 (join) djcoin
03:11 mithos28: asumu: how do you determine if the continuation is the same as the current one? the model uses the existence of dynamic-wind frames and there might not be those there
03:12 (join) lewis1711
03:14 (quit) billyocZ`: Remote host closed the connection
03:14 mithos28: I think you can do it if you add comparing the prompts to the calculation sameDWs
03:15 (quit) ambrosebs: Ping timeout: 260 seconds
03:16 (join) ambrosebs
03:21 asumu: mithos28: I don't know if that's enough, since prompts need to be uniquely labeled for that.
03:21 asumu: IIRC, the model doesn't label prompts in a way to distinguish them except textually.
03:22 mithos28: yes, you would need to add unique labels to prompts in addition to the tag
03:22 asumu: In the implementation, a continuation mark is installed with the prompt tag as the key and the pointer to the prompt object as the value.
03:22 asumu: (so it's doable there)
03:23 mithos28: the [dw] step alread adds a fresh label to the dw frame
03:23 (quit) tfb: Ping timeout: 264 seconds
03:24 mithos28: I am suggesting a similar step, and seperating out prompt from % like dynamic-wind from dw
03:35 (quit) ambrosebs: Ping timeout: 264 seconds
03:38 (quit) mithos28: Quit: mithos28
03:41 (join) ambrosebs
04:04 (join) Jeanne-Kamikaze
04:06 (quit) em: Ping timeout: 272 seconds
04:08 (quit) bjz: Quit: Leaving...
04:08 Haffe: Hmmm. In the drawing toolkit. Is there a convinient way to scale a bitmap?
04:09 Haffe: Ah, yes there is.
04:09 Haffe: scale
04:18 (quit) ambrosebs: Read error: Connection reset by peer
04:19 (join) ambrosebs
04:19 (join) jesyspa
04:20 (join) em
04:42 (join) add^_
04:43 (quit) ambrosebs: Ping timeout: 276 seconds
04:47 (join) ambrosebs
05:04 (quit) add^_: Quit: add^_
05:09 (quit) jyc: Remote host closed the connection
05:15 (join) add^_
05:17 (quit) thesnowdog: Ping timeout: 245 seconds
05:18 (quit) Nisstyre: Ping timeout: 246 seconds
05:19 (join) ivan\
05:20 (join) Nisstyre
05:26 (join) thesnowdog
05:46 (quit) jesyspa: Ping timeout: 264 seconds
05:48 (join) jesyspa
05:49 (join) masm
05:57 (quit) em: Ping timeout: 240 seconds
06:11 (join) cdidd
06:25 (join) didi
06:29 (join) MayDaniel
06:30 (quit) jesyspa: Quit: leaving
06:37 (quit) bitonic: Remote host closed the connection
06:37 (join) bitonic
06:37 (quit) bitonic: Remote host closed the connection
06:38 (join) bitonic
06:38 (quit) vu3rdd: Remote host closed the connection
06:50 (quit) ambrosebs: Ping timeout: 264 seconds
06:55 (join) em
07:00 (part) dca: "ERC Version 5.3 (IRC client for Emacs)"
07:20 (join) ambrosebs
07:34 (quit) add^_: Quit: add^_
07:36 (quit) ambrosebs: *.net *.split
07:43 (join) add^_
07:55 (join) jeapostrophe
07:55 (quit) jeapostrophe: Changing host
07:55 (join) jeapostrophe
08:13 (quit) acarrico: Ping timeout: 252 seconds
08:21 (quit) add^_: Quit: add^_
08:29 (join) acarrico
08:41 (join) jao
08:41 (nick) jao -> Guest911
08:42 (join) jao`
08:43 (nick) jao` -> jao
08:54 (quit) didi: Remote host closed the connection
08:56 (join) getpwnam
08:56 (quit) jeapostrophe: Read error: Operation timed out
08:56 (join) hash_table
09:07 (join) newby
09:07 newby: hi all
09:08 newby: quick question: can the executables cresated in racket be sold commercially?
09:09 (join) add^_
09:11 newby: ?
09:12 newby: helloooo....
09:13 Araq: newby: I think so
09:14 (join) dnolen
09:21 newby: anyone know for sure? i understand that the licence requires modifications to be attibuted and commited back, but if nothing is done to 'change' the system: can (and have) any product been produced commercially with Racket (or its predecessor)?
09:23 asumu: newby: Racket is LGPL, so linking should be fine. However, if you really need to know for sure you need legal advice from a lawyer probably, not us.
09:23 asumu: (this is not legal advice)
09:24 (quit) hash_table: Ping timeout: 244 seconds
09:24 (quit) getpwnam: Ping timeout: 244 seconds
09:24 bremner: well, at least the intent of the copyright holders is to allow that, to the best of my knowledge
09:25 (join) bjz
09:26 samth: newby: yes, that's the intent
09:27 samth: the license page has some clarification on the intent of the copyright holders wrt licensing
09:27 asumu: newby: Also, Racket has been used in commercial software.
09:27 samth: and certainly "sold commercially" is fine -- the questions are all about what obligations you have to the people who you sell to
09:28 newby: ok, thanks for you input!
09:28 newby: have a good evening/morning
09:29 (quit) newby: 
09:33 (join) RacketCommitBot
09:33 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/j8C8LA
09:33 RacketCommitBot: [racket/master] ffi/unsafe/atomic: add `in-atomic-mode?' - Matthew Flatt
09:33 RacketCommitBot: [racket/master] include scribble/scribble-prefix.tex packages to scribble/sigplan - fbanados
09:33 RacketCommitBot: [racket/master] propagate prefix repair to to `scribble/jfp' and `scribble/lncs' - Matthew Flatt
09:33 (part) RacketCommitBot
09:44 (join) uselpa
10:19 (quit) mceier: Quit: leaving
10:21 (join) gridaphobe
10:30 (join) anRch
10:31 (join) getpwnam
10:31 (join) hash_table
10:33 (join) random_malice
10:33 (join) ambrosebs
10:33 (join) bro_grammer
10:36 (quit) hash_table: Ping timeout: 248 seconds
10:36 (quit) getpwnam: Ping timeout: 248 seconds
10:42 (join) RacketCommitBot
10:42 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/bx4YKw
10:42 RacketCommitBot: [racket/master] Simplifying case-> to union is only safe for 1-argument functions. - Sam Tobin-Hochstadt
10:42 (part) RacketCommitBot
10:46 (quit) ambrosebs: Ping timeout: 244 seconds
10:51 (join) ambrosebs
10:52 (quit) uselpa: Remote host closed the connection
10:56 (quit) dnolen: Ping timeout: 240 seconds
11:01 (quit) ambrosebs: Ping timeout: 244 seconds
11:02 (join) jeapostrophe
11:02 (quit) jeapostrophe: Changing host
11:02 (join) jeapostrophe
11:05 (join) uselpa
11:15 (quit) anRch: Quit: anRch
11:19 (quit) uselpa: Remote host closed the connection
11:20 (quit) acarrico: Ping timeout: 246 seconds
11:23 (quit) jrslepak: Ping timeout: 244 seconds
11:35 (join) Kaylin
11:36 (join) acarrico
11:37 (join) mceier
11:44 (part) lewis1711
11:47 (join) anRch
11:48 (quit) jeapostrophe: Ping timeout: 252 seconds
11:55 (join) vu3rdd
11:55 (quit) vu3rdd: Changing host
11:55 (join) vu3rdd
11:58 (join) jeapostrophe
11:58 (quit) jeapostrophe: Changing host
11:58 (join) jeapostrophe
11:59 (join) jrslepak
12:03 (quit) add^_: Quit: add^_
12:03 (join) getpwnam
12:03 (join) hash_table
12:04 (join) add^_
12:05 (quit) vu3rdd: Ping timeout: 240 seconds
12:06 (quit) bro_grammer: Ping timeout: 264 seconds
12:06 (quit) random_malice: Ping timeout: 264 seconds
12:08 (quit) thesnowdog: Quit: WeeChat 0.3.8
12:09 (join) close-paren
12:27 (quit) close-paren: Quit: close-paren
12:32 (quit) anRch: Quit: anRch
12:32 (quit) masm: Ping timeout: 268 seconds
12:33 (quit) djcoin: Quit: WeeChat 0.3.2
12:41 (quit) bjz: Quit: Leaving...
13:11 (quit) acarrico: Ping timeout: 244 seconds
13:27 (join) acarrico
13:34 (join) jonrafkind
13:34 (quit) jonrafkind: Changing host
13:34 (join) jonrafkind
13:52 (quit) jeapostrophe: Ping timeout: 245 seconds
13:57 (join) anRch
13:58 (quit) acarrico: Ping timeout: 244 seconds
13:59 (quit) bitonic: Remote host closed the connection
14:09 (quit) gridaphobe: Read error: Connection reset by peer
14:09 (join) gridapho_
14:12 (join) acarrico
14:19 (nick) gridapho_ -> gridaphobe
14:21 (join) masm
14:50 (join) bitonic
14:52 (quit) Kaylin: Read error: Connection reset by peer
14:54 (join) uselpa
14:58 (join) josdeha_rt
14:59 (quit) anRch: Quit: anRch
15:24 (quit) josdeha_rt: Remote host closed the connection
15:25 (quit) rapacity: Read error: Operation timed out
15:26 (quit) tauntaun: Ping timeout: 272 seconds
15:26 (quit) elliottc1ble: Ping timeout: 265 seconds
15:49 (join) Shvillr_
15:50 (nick) Shvillr_ -> Shviller
16:02 (join) ivan`
16:11 (quit) Jeanne-Kamikaze: Ping timeout: 246 seconds
16:24 (join) Jeanne-Kamikaze
16:34 (join) Kaylin
16:35 (quit) jonrafkind: Ping timeout: 245 seconds
16:41 (join) ambrosebs
16:48 (join) RacketCommitBot
16:48 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/LldZeA
16:48 RacketCommitBot: [racket/master] Comment and clarify top-level evaluation & barriers - Asumu Takikawa
16:48 (part) RacketCommitBot
16:48 (quit) scott_: Quit: Leaving
16:53 (quit) kanak: Ping timeout: 248 seconds
16:54 (join) jonrafkind
16:54 (quit) jonrafkind: Changing host
16:54 (join) jonrafkind
17:07 (quit) hash_table: Ping timeout: 276 seconds
17:08 (quit) getpwnam: Ping timeout: 276 seconds
17:10 (quit) ambrosebs: Ping timeout: 246 seconds
17:17 (join) jesyspa
17:19 (join) yoklov
17:26 (join) dnolen
17:26 (quit) dsantiago: Quit: Computer has gone to sleep.
17:39 (join) dsantiago
17:47 (quit) MayDaniel: Read error: Connection reset by peer
17:51 (quit) Jeanne-Kamikaze: Quit: Did you hear that ?
17:52 (join) tester
17:52 (nick) tester -> tester234
17:52 tester234: hi
17:53 tester234: i have a problem with ddracket 5.3 on win 64
17:53 (join) RacketCommitBot
17:53 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/wUqEOw
17:53 RacketCommitBot: [racket/master] Track mutable variables across modules. - Sam Tobin-Hochstadt
17:53 (part) RacketCommitBot
17:53 tester234: i get a contract violation n launch
17:54 jonrafkind: what is it
17:59 tester234: what i what ?
17:59 tester234: the contract violatin ?
18:00 tester234: violation*?
18:00 jonrafkind: yes
18:01 (join) scott_
18:01 (quit) scott_: Changing host
18:01 (join) scott_
18:03 (join) getpwnam
18:03 (quit) uselpa: Remote host closed the connection
18:03 (join) hash_table
18:07 (join) RacketCommitBot
18:07 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/3pFXaA
18:07 RacketCommitBot: [racket/master] fix `in-directory' to properly propagate filesystem errors - Matthew Flatt
18:07 RacketCommitBot: [racket/master] scribble: for Latex output, use `\Ssection', etc. for section - Matthew Flatt
18:07 (part) RacketCommitBot
18:14 (quit) dnolen: Remote host closed the connection
18:24 (join) dnolen
18:42 yoklov: why do the docs for sets say 'this operation runs in constant time' with a sidenote that says 'these operations actually run in O(log N)'?
18:42 yoklov: shouldn't it just say, uh, logarithmic time in the first place?
18:42 jonrafkind: haha wheres that
18:43 samth: yoklov: it says pretty clearly why it says that
18:43 yoklov: http://docs.racket-lang.org/reference/sets.html
18:43 yoklov: err
18:43 yoklov: why?
18:43 yoklov: or where?
18:44 jonrafkind: i dont think its clear at all
18:44 scott_: I remember seeing something like that in hashtables
18:44 scott_: http://docs.racket-lang.org/reference/hashtables.html
18:44 scott_: it actually is made clear there
18:45 jonrafkind: oh reasonably constant, i see
18:47 (quit) jesyspa: Quit: leaving
18:57 (join) jrslepak_
18:58 (nick) jrslepak_ -> jrslepak_neu
19:13 (quit) add^_: Quit: add^_
19:24 (quit) dnolen: Ping timeout: 244 seconds
19:24 (quit) jrslepak: Quit: What happened to Systems A through E?
19:24 (quit) acarrico: Ping timeout: 240 seconds
19:33 (quit) Nisstyre: Read error: Connection reset by peer
19:40 (join) RacketCommitBot
19:40 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/leXOhg
19:40 RacketCommitBot: [racket/master] Add comment about the current state of the latex character table. - Vincent St-Amour
19:40 (part) RacketCommitBot
19:40 (join) acarrico
19:42 (join) bjz
19:52 (quit) Kaylin: Quit: Leaving.
19:54 (quit) jonrafkind: Ping timeout: 272 seconds
19:54 (join) jrslepak
20:19 (quit) Nisstyre_: Read error: Connection reset by peer
20:20 (join) Nisstyre_
20:59 (quit) masm: Quit: Leaving.
21:09 (join) ambrosebs
21:53 (join) RacketCommitBot
21:53 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/4_bmJA
21:53 RacketCommitBot: [racket/master] fix check-error so it compares the rewritten error messages, - Robby Findler
21:53 RacketCommitBot: [racket/master] speed up the test-engine drracket tests some - Robby Findler
21:53 (part) RacketCommitBot
21:56 (quit) scott_: Quit: Leaving
21:57 (join) jyc
21:59 (join) jonrafkind
21:59 (quit) jonrafkind: Changing host
21:59 (join) jonrafkind
22:00 (quit) bitonic: Remote host closed the connection
22:05 (join) scott_
22:05 (quit) scott_: Changing host
22:05 (join) scott_
22:11 (join) billyoc
22:22 (join) jeapostrophe
22:22 (quit) jeapostrophe: Changing host
22:22 (join) jeapostrophe
22:37 (join) gridaphobe
22:41 (join) mithos28
22:55 (join) Kaylin
23:19 (quit) jeapostrophe: Ping timeout: 240 seconds
23:31 (quit) cdidd: Remote host closed the connection
23:36 (join) close-paren
23:38 (quit) gridaphobe: Remote host closed the connection