00:02 (quit) mheld: Quit: mheld
00:03 (quit) stamourv: Read error: Connection reset by peer
00:03 (join) stamourv`
00:04 (quit) samth: Ping timeout: 250 seconds
00:05 (join) samth
00:29 (nick) offby1 -> ophphby1
01:00 (quit) mwolfe: Remote host closed the connection
01:11 (join) jonrafkind
01:26 Lajla: (call/cc jonrafkind)
01:27 Lajla: Ohhh
01:27 Lajla: (eqv? (compose) values) actually returns true
01:41 jonrafkind: rudybot: (eqv? (compose) values)
01:41 rudybot: jonrafkind: your sandbox is ready
01:41 rudybot: jonrafkind: ; Value: #t
01:41 jonrafkind: wtf
01:44 Lajla: jonrafkind, makes sense, no?
01:45 Lajla: values is the identity function
01:45 Lajla: From the perspective of a function that can take arbitrary arguments.
01:45 Lajla: It returns all those arbitrary arguments
01:46 jonrafkind: rudybot: (compose)
01:46 rudybot: jonrafkind: ; Value: #<procedure:values>
01:46 jonrafkind: i did not know that
01:48 Lajla: jonrafkind, why not? I mean, values is basically the identity function, and a composition of nothing is the identity function.
01:48 Lajla: the identity function itself is the identity of functional composition and all that.
01:58 (join) Sam__
01:58 Sam__: Is my function "summe" a higher order function ? http://codepad.org/2tihMehw
01:59 Lajla: Sam__, ah
01:59 Lajla: At last, you are back
01:59 askhader: Sam__ No.
01:59 Lajla: It is not.
01:59 Lajla: A higher order function is a function which takes a function as argument.
02:00 jonrafkind: Sam__, if you passed `f' into summe then `summe' would be higher order
02:01 Lajla: What you basically want is this: http://codepad.org/XDThKK1p
02:02 Lajla: jonrafkind, see, you can do it.
03:13 Sam__: i read in my book a higher order function is a function which takes a function as a argument AND/ OR has a Functions as output
03:13 Sam__: is this wrong ?
03:21 Lajla: No
03:21 Lajla: that's cool.
03:27 Sam__: Ok  is this (apply + (map abs '(1 2 3 -4)))   higher order func ?
03:28 Lajla: Both map and apply are higher order functions
03:28 Lajla: abs is not.
03:29 (quit) jonrafkind: Ping timeout: 250 seconds
03:39 (join) aLeSD
03:48 (quit) Sam__: Ping timeout: 265 seconds
04:26 (quit) aLeSD: Ping timeout: 246 seconds
05:22 (quit) unkanon-work: Ping timeout: 255 seconds
05:23 (join) unkanon-work
05:44 (quit) unkanon-work: Ping timeout: 240 seconds
05:46 (join) unkanon-work
06:21 (nick) ophphby1 -> offby1
06:37 (join) sam__
06:38 sam__: How i can romove a quote ? ''(0 0 1) ->  '(0 0 1)  ??
06:39 offby1: cdr
06:39 offby1: rudybot: ''(0 0 1)
06:39 rudybot: *offby1: your scheme sandbox is ready
06:39 rudybot: *offby1: ; Value: (quote (0 0 1))
06:39 offby1: rudybot: (cdr ''(0 0 1))
06:39 rudybot: *offby1: ; Value: ((0 0 1))
06:39 offby1: oops :-)
06:39 offby1: rudybot: (second ''(0 0 1))
06:39 rudybot: *offby1: ; Value: (0 0 1)
06:39 offby1: *ahem*
06:47 clklein: sam__: ''(0 0 1) is shorthand for (quote (quote (0 0 1)), which is a list where the first element is the symbol quote and the second element is the list containing 0, 0, 1
06:47 clklein: quote is surprising at first
06:47 sam__: Nad
06:48 sam__: And how i can easy remove the first ' ?
06:49 clklein: You have a list containing two elements: 'quote and '(0 0 1).
06:50 clklein: You want to extract the second element of that list.
06:50 clklein: Do you know how to get 7 out of '(3 7)?
06:51 sam__: yea  cdr  lst
06:51 sam__: hmm i know hat you want to say me
06:51 sam__: ok  i try  something   with my code
06:53 clklein: (cdr '(3 7)) is '(7). I think you want (car (cdr '(3 7)) = 7
06:54 clklein: alternatively, (second '(3 7)) = 7
06:55 offby1: why does that sound so familiar
07:11 (quit) coldhead: Quit: Bothallchoractorschumminaroundgansumuminarumdrumstrumtruminahumptadumpwaultopoofoolooderamaunsturnup
07:38 (join) Yann3
07:38 (quit) Yann1: Ping timeout: 240 seconds
07:42 (join) masm
07:48 (join) Qvist_
07:56 (quit) sam__: Ping timeout: 265 seconds
08:05 (join) mheld
08:31 Lajla: clklein, I find that this is the part that silly way of saying that 'quote is a symbol goes wrong.i
08:33 (join) MayDaniel
08:35 clklein: offby1: because you jumped right to the answer without explaining why :)
08:36 clklein: Lajla: I don't understand. (symbol? 'quote) is true.
08:37 Lajla: I was under the impression that a function call evaluated each sub expression in the environment of the function call.
08:37 Lajla: What you mean to say is 'In most environments the datum 'quote _evaluates_ to the symbol quote.
08:38 Lajla: 'quote however is a list of two identifiers/symbols quote and quote.
08:38 Lajla: In things like ''quote or (let ('-) '3) it shows just how easily it can break, I really don't like it.
08:38 Lajla: Especially that defenders of seeing like that often don't do so consistently and get caught up in it.
08:39 Lajla: Like jonrafkind first saying that a symbol evaluates to itself because 'quote is a symbol and later having to admit that the idea is meaningless because everything evaluates to itself under this model.
08:40 clklein: I don't think it's accurate to say that 'quote is a list.
08:40 Lajla: Of course it is, it just _evaluates_ to a symbol, and only does so in specific environments.
08:40 clklein: In the same way that (+ 1 5) is not a list
08:41 Lajla: (ऋऋ+ 1 5) is a list, it however evaluates to a number
08:41 Lajla: You can at max say that considered an expression its _value_ is a number.
08:41 Lajla: But since racket is not purely functional, an expression is not identical to its value.
08:41 Lajla: And this still depends on the environment.
08:41 clklein: When you say it's a list, you're referring to one particular representation of the abstract syntax
08:42 clklein: and it's not even the one that Racket uses
08:42 Lajla: In the environment created by (let ('-) 'quote) is most certainly not a symbol.
08:42 Lajla: I'm referring to a definition that doesn't break up on itself.
08:43 Lajla: In any case saying 'quote is a symbol is incomplete. You can make it work, but you have to say 'in an environment which .... 'quote is a symbol'
08:43 (join) lucian
08:43 Lajla: Because what it "is" depends on the environment it is situated in.
08:43 Lajla: Which is very nonstandard use of the term "is" in programming languages, but not for the term "evaluates to"
08:44 Lajla: Also, how are you going to explain the difference between syntax and function calls in this idea?
08:44 Lajla: Clearly if you say that (+ 1 2) "Is" a list et cetera, you no longer say that a function call evaluates its sub forms.
08:44 Lajla: So what does syntax do, it devaluates them?
08:44 clklein: I have no idea what you're talking about
08:45 Lajla: clklein, like I said if 'quote "is" a list or not in your model depends on the environment.
08:45 clklein: No
08:45 Lajla: In the local environment created by (let ((quote +)) ...)
08:45 Lajla: 'quote isn't a symbol.
08:46 Lajla: The syntax becomes erronous.
08:46 clklein: I'm just objecting to you referring to every piece of abstract syntax as a list
08:46 clklein: That's not useful or correct.
08:46 Lajla: It doesn't break apart though.
08:46 (quit) MayDaniel: Read error: Connection reset by peer
08:46 Lajla: Under (let ((quote +)) ...)
08:47 Lajla: I'm fine with calling it syntax and not saying it's a list.
08:47 (quit) _danb_: Ping timeout: 255 seconds
08:47 Lajla: But there is still a destinction between "is" something and "evaluates to" something.
08:47 Lajla: "is" is irrespective of surrounding context and environment at the very least in standard terminology.
08:47 clklein: OK
08:47 Lajla: 'quote is an expression which evaluates to a symbol in most contexts.
08:48 clklein: that's fine
08:48 clklein: I don't disagree.
08:49 Lajla: clklein, well, I do disagree with you, sorry. =(
08:49 Lajla: But
08:49 (join) dnolen
08:49 Lajla: I would say that your view is also valid if you extend it and say" 'quote is a symbol in an environment that meeds certain criteria"
08:50 Lajla: Then you just use the word "is" different than I do, albeit noncontradictory and consistently
08:50 clklein: Lajla: You're welcome to
08:50 Lajla: Variety is the spice of life and diversity the source of creativity.
09:14 (join) mceier
09:15 (quit) lucian: Remote host closed the connection
09:16 (join) jao
09:16 (quit) jao: Changing host
09:16 (join) jao
09:16 (quit) Qvist_: Quit: leaving
09:18 (join) corruptmemory
09:32 (quit) mheld: Quit: mheld
09:34 (join) Yann1
09:35 (quit) Yann3: Ping timeout: 240 seconds
09:37 (join) Samy
09:41 Samy: I want generate all possible list of lists like this '('(1 0 0) '(1 0 0) ' (1 0 0)  each clip can be (1 0 0) or ( 0 1 0) or ( 0 0 1) .  Can pls  say to me how i can do that ?
09:45 (part) mattmight
09:45 Lajla: See, this is what you get, now you let him to believe that '('(1 0 0) '(1 0 0) '(1 0 0)) is a list of lists rather than a list of lists of lists.
09:45 Lajla: Samy, what you want is '((1 0 0) (1 0 0) (1 0 0))
09:48 Samy: yea
09:48 Samy: I want this
09:48 Samy: I must generate all variations of it
09:48 Samy: you know how i can do that ?
09:49 (quit) dnolen: Quit: dnolen
09:50 Samy: I need  lists like this  '((1 0 0) (1 0 0) (1 0 0)) or  '((0 1 0) (1 0 0) (1 0 0)) or '((0 0 1) (0 1 0) (1 0 0)).....
09:56 (join) mheld
10:08 Samy: Pls help :D
10:19 Samy: I need  lists like this  '((1 0 0) (1 0 0) (1 0 0)) or  '((0 1 0) (1 0 0) (1 0 0)) or '((0 0 1) (0 1 0) (1 0 0))..... generate all posible variations
10:19 Samy: can some one tell me how i can do that ?
10:21 (join) MayDaniel
10:23 unkanon-work: you probably need a permutations function
10:25 unkanon-work: http://www.dreamincode.net/forums/topic/19935-recursive-permutations-in-scheme/
10:25 rudybot: http://tinyurl.com/38hlh9k
10:29 Samy: yea thx  i did it i think
10:29 (quit) mheld: Quit: mheld
10:44 (join) lucian
11:09 (quit) lucian: Remote host closed the connection
11:18 (quit) stamourv`: Quit: ERC Version 5.2 (IRC client for Emacs)
11:25 (join) anRch
11:25 (quit) jeapostrophe: Quit: jeapostrophe
11:28 (join) mheld
11:35 (join) stamourv
11:44 (join) jonrafkind
11:50 (quit) MayDaniel: 
12:03 bremner: anyone know why "define-type" in #lang plai is so expensive? It takes like 10 seconds to compile a single form in DrRacket, with tons of garbage collection on a 3GHz Core II Duo.
12:03 samth: bremner, you could look at the expansion
12:05 (quit) Samy: Quit: Page closed
12:07 (join) carleastlund
12:08 bremner: do you mean look at the plai source, or something fancier?
12:09 samth: i mean hit the 'macro stepper' button and see what your short program expands to
12:09 bremner: yeah, I thought you might mean that, but the button is greyed out :(
12:09 samth: hmm
12:09 samth: what language is drracket set to?
12:10 bremner: plai (declared in source)
12:11 samth: when i do that, the macro stepper button is available
12:14 bremner: oh wait, I was looking at the wrong thing, sorry.
12:14 (quit) anRch: Quit: anRch
12:21 (join) anRch
12:21 (quit) mheld: Quit: mheld
12:22 Yann1: Hi, how can I include a file that isn't a module?  I just want to split up a big file into logical sections.
12:24 samth: Yann1, I recommend splitting the file up into multiple modules
12:24 samth: is there a reason you can't do that?
12:24 Yann1: samth: but modules have another scope then
12:25 samth: why is that a problem?
12:25 carleastlund: Yann1, you can always put "(provide (all-defined-out))" in the modules, so that everything in them is available to their clients.
12:25 Yann1: samth: I would like to access 'global' variables
12:25 Yann1: carleastlund: yeah but not in the other direction
12:25 samth: put those 'global' variables in their own module, then all of them can be accessed from any module
12:26 carleastlund: Yann1, perhaps you want units?  They allow recursive linking.  Or what samth said.
12:27 bremner: samth: the expansion is about 340 lines. http://paste.debian.net/103896/  Is that as fast as can be expected then?
12:28 Yann1: carleastlund: but can't I just tell the compiler to include the code of a file without all this stuff?
12:29 carleastlund: Yann1, only in really hackish ways.  Modules are the feature designed for splitting up code into logical units.  Reinventing the wheel will almost certainly not work as nicely.
12:30 carleastlund: Look up "include", though.  I don't recommend it in general, but it sounds like what you want.
12:30 samth: bremner, what was the original code?
12:31 samth: most of that is the expansion of contracts, and is unsurprising
12:31 samth: nothing there should take 10 seconds
12:31 carleastlund: The problem with "include" is that if you include one file from multiple different places, they might each compile the code to something completely different.  Whereas a module is guaranteed to produce the same code for all clients.
12:32 Yann1: carleastlund: include testcase: file1: http://pastebin.com/aHsHwAxk file2: http://pastebin.com/NNj98NCg I always get this error:  http://pastebin.com/G9YMrkg8
12:33 carleastlund: Yann1, that's "load", not "include".  And by putting in "#lang scheme", you made both files into modules.
12:34 Yann1: carleastlund: sorry wrong testcase, but removing #lang worked! thanks!!
12:34 (quit) anRch: Read error: Connection reset by peer
12:35 carleastlund: Yann1, I'm not sure that "load" is going to work how you expect -- any "global variables" you import will be bound in the namespace used by "eval", but not in the actual module that imports them.
12:35 Yann1: carleastlund: im using include and now it works
12:36 carleastlund: Yann1, Okay, just giving you a heads up on a potential bug, so if you run into it down the road you'll know what's up.  But more power to you if you avoid it!
12:36 (join) anRch_
12:38 Yann1: carleastlund: okay. thanks
12:38 (quit) anRch_: Read error: Connection reset by peer
12:40 (join) anRch
12:44 (quit) anRch: Client Quit
12:46 (join) MayDaniel
13:16 bremner: samth: the original code is http://paste.debian.net/103901/
13:28 samth: bremner, that's basically instant for me
13:29 bremner: samth: what version of racket?
13:29 samth: 5.0.99.6
13:29 stamourv: bremner: are all your collects compiled?
13:29 samth: ie git from yesterday
13:31 bremner: stamourv: hmm. maybe not.  should the compiled stuff live in the same hierarchy under collects?
13:31 samth: yes
13:31 samth: bremner, try running 'raco setup plai'
13:32 bremner: yeah, there is no collects/plai/compiled directory. OK, so this is must likely the problem.
13:37 bremner: OK, I tried it on a different install with compiled collects and that is much faster.  So I have just to figure out my install problem.
13:37 bremner: thanks for hints.
13:44 stamourv: make install should compile all the collects
13:44 stamourv: did you interrupt it?
13:46 bremner: stamourv: I'm just investigating.  I'm alpha-testing debian packages, so it is likely a packaging error.
13:51 stamourv: ok
13:51 (join) anRch
14:02 bremner: hmm. In my build log, I see no "raco setup: made plai"   but rather "raco setup: made: <user>/plai"
14:03 (quit) MayDaniel: Read error: Connection reset by peer
14:03 (nick) unkanon-work -> rien
14:12 askhader: 5
14:32 bremner: hmm. It looks like I need to convince raco setup to ignore $(HOME)/.racket/$version/collects. Is there a flag for that?
14:43 samth: raco setup --no-user
14:44 bremner: perfect, thanks
14:45 bremner: I'm afraid I didn't completely understand the output of "racko help setup"
14:45 (join) mheld
15:21 (quit) mheld: Quit: mheld
15:40 (quit) anRch: Quit: anRch
15:51 (join) coldhead
15:51 (join) Yann3
15:51 (quit) Yann1: Ping timeout: 240 seconds
15:54 (join) mheld
16:06 (join) mwolfe
16:26 (join) MayDaniel
16:29 (quit) MayDaniel: Client Quit
16:41 (join) MayDaniel
17:03 (quit) tv|z: Ping timeout: 240 seconds
17:05 (join) tv|z
17:16 samth: eli, ping
17:29 (quit) MayDaniel: Read error: Connection reset by peer
17:49 (quit) Yann3: Ping timeout: 240 seconds
17:49 (join) Yann1
17:59 (nick) samth -> samth_away
18:10 _p4bl0: hi all
18:11 (quit) mheld: Quit: mheld
18:11 (nick) rien -> rien_home
18:12 _p4bl0: is there a way when drawing in a canvas to have a temp buffer (in order to draw in the buffer and then paste the whole buffer on the actual canvas, so I can avoid blinking when redrawing everything) ?
18:20 (join) jeapostrophe
18:51 (quit) parcs: Ping timeout: 240 seconds
18:52 (join) parcs
18:54 (nick) unkanon -> rien
19:28 (join) vivitron
19:32 (quit) masm: Quit: Leaving.
19:42 (quit) tv|z: Quit: leaving
19:46 (quit) mceier: Quit: leaving
20:20 (quit) mwolfe: Remote host closed the connection
20:29 (join) mheld
20:32 bremner: oops. If I pass --no-user to raco setup it still seems to pick up plai package in my home dir during testing. http://paste.debian.net/103949
20:33 (quit) jonrafkind: Ping timeout: 264 seconds
20:35 rien: _p4bl0: I can't help but just so you know, that's called "double buffering", it may help you google for an anser
20:35 rien: answer*
20:39 rien: _p4bl0: http://lists.racket-lang.org/users/archive/2009-April/032035.html
20:49 (quit) carleastlund: Quit: carleastlund
20:57 (join) Yann3
20:58 (quit) Yann1: Ping timeout: 240 seconds
21:49 (join) tv|z
21:53 (nick) cky -> Zba2Y7Pr3
21:56 (quit) tv|z: Remote host closed the connection
22:00 (join) tv|z
22:27 (join) rmitt
22:29 (quit) rmitt: Client Quit
22:30 (join) rmitt
22:30 rmitt: Why the "m" series for mutable pairs?  Why not stick to the Scheme standard on this one?
22:39 rmitt: Hello, I just asked a brilliant question.  Isn't everyone going to drop what they're doing and admire my brilliant question, offer to feed me and massage me, and then answer it?
22:39 Lajla: rmitt, because they it in reverse.
22:39 Lajla: THey want all the old code which used cons to use immutable pairs.
22:40 Lajla: And basically say that if you used pairs mutably you sucked anyway and your code deserves to be broken.
22:40 (quit) corruptmemory: Ping timeout: 240 seconds
22:40 Lajla: What I would rather have though is a list-cons.
22:40 Lajla: And a designated list type.
22:40 Lajla: Basically list cons demands its second argument to be of the list type.
22:40 Lajla: And only has a mutable car field.
22:40 rmitt: yes yes, a proper list
22:40 Zba2Y7Pr3: This is getting into Typed Racket territory.
22:40 Lajla: Yeah, it makes list? also run in constant time.
22:40 Lajla: Not at all.
22:41 rmitt: I don't see why it's a sin to use set-car! etc.  How else would you implement, say, a mutable stack or queue?
22:41 Lajla: Just a list-cons, which is like cons, except that what ever it produces has a type list, and it requires its second argument to be a type list.
22:41 Lajla: rmitt, it's not a sin.
22:42 Lajla: IT's perhaps just historically a bit uncomfortable that lisp never had a proper list type and ad hoced it.
22:42 Lajla: So that set-cdr! could make an object that once was a list not a list any more.
22:43 Zba2Y7Pr3: rmitt: What's the use case for a mutable stack?
22:43 rmitt: dunno
22:44 Lajla: rmitt, you can actually also use a vector for that.
22:44 (quit) tv|z: Quit: leaving
22:44 Lajla: A vector of two elements.
22:45 Zba2Y7Pr3: rmitt: If there is no use case for a mutable stack, why would you even mention it? O_o
22:45 Lajla: Freedom of speech, baby.
22:45 Zba2Y7Pr3: Heh.
22:46 rmitt: Forget I mentioned it.  I'm just in a tizzy because the standard has been abandoned.
22:46 (join) tv|z
22:47 Lajla: rmitt, #lang racket != scheme
22:47 Lajla: That is why they renamed it racket from PLT scheme.
22:47 Lajla: If you use #lang r5rs you get your old stuff back
22:47 rmitt: since proper lists are not a type in Scheme/Lisp, one could argue that set-car! & set-cdr! sit in the language more naturally than do set-mcar! & set-mcdr!
22:47 rmitt: Lajla: yes, I'm aware of that.
22:48 rmitt: Well the problem is that I want r6rs.
22:48 rmitt: ...which keeps set-car! etc..right?
22:48 Lajla: rmitt, yap.
22:48 Lajla: but it banishes them to their own library.
22:48 Lajla: rmitt, in Clojure, lists are actually a proper type.
22:48 Lajla: And there are no cons cells.
22:48 Lajla: I'd rather have both.
22:49 Lajla: Lists as a proper type AND cons cells.
22:49 Lajla: What I do really like about clojure though is the fact that nil, false, and the empty list are destinct values.
22:49 Zba2Y7Pr3: Supporting cons cells on the JVM sounds very icky. I guess that's why Clojure ditched it.
22:53 Lajla: supporting sanity on the JVM is very icky.
22:53 rmitt: r6rs is icky
22:53 Lajla: He had to make a lot of compromises in what he wanted to get it to run on the JVM.
22:53 Lajla: Most prominently no TCO, which he likes.
22:53 Lajla: rmitt, why?
22:53 rmitt: can't do anything unless you import, import, import
22:54 Lajla: Is that bad
22:54 Lajla: ?
22:54 Zba2Y7Pr3: rmitt: That's usually seen to be a feature, not a bug.
22:54 rmitt: fine
22:54 rmitt: more work for a programmer
22:54 Lajla: Well, it's also a feature that C can do whatever it wants if you access out of bounds
22:54 Lajla: still not desirable.
22:54 Lajla: rmitt, you come pretty far with (import (rnrs (6))) though
22:54 Zba2Y7Pr3: Lajla: Nasal demons are a feature? :-O
22:57 rmitt: these diverging standards are a mess
22:58 rmitt: ok, good night
22:58 (quit) rmitt: Quit: Page closed
23:16 (quit) tv|z: Quit: leaving
23:19 (join) tv|z
23:44 (quit) coldhead: Quit: Lukkedoerendunandurraskewdylooshoofermoyportertooryzooysphalnabortansporthaokansakroidverjkapakkapuk
23:54 (nick) Zba2Y7Pr3 -> cky