00:03:41 george [n=george@189.107.176.74] has joined #scheme 00:08:09 Granted, Riastradh, but I would be very surprised if eq? ever returned #t in such a case. 00:08:24 The definition in R5RS is non-operational. 00:12:39 -!- jcowan [n=jcowan@72.14.228.129] has quit ["Leaving"] 00:14:19 -!- bweaver [n=user@75-148-111-133-Chattanooga.hfc.comcastbusiness.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:17:59 -!- masm [n=masm@bl10-244-45.dsl.telepac.pt] has quit ["Leaving."] 00:19:30 -!- pavelludiq [n=quassel@83.222.175.184] has quit [Read error: 104 (Connection reset by peer)] 00:24:58 kilimanjaro [n=kilimanj@cpe-173-172-99-25.austin.res.rr.com] has joined #scheme 00:26:43 incubot: The definition in R5RS is non-operational! 00:26:47 well, i mean operational in the sense of the actual VM rather than operational semantics in general :) 00:31:05 metasyntax` [n=taylor@pool-71-127-125-129.aubnin.fios.verizon.net] has joined #scheme 00:31:44 -!- Nshag [i=user@Mix-Nantes-117-4-155.w193-250.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 00:34:59 schmir` [n=schmir@p54A93FAB.dip0.t-ipconnect.de] has joined #scheme 00:38:28 incubot: pizza 00:38:31 i would like to order a pizza 00:38:35 me too 00:39:57 I would like to eat a pizza, but I'm not in the mood for ordering one right now. 00:41:24 *mario-goulart* too 00:48:48 -!- Arelius [n=Indy@208.87.217.194] has quit [Read error: 104 (Connection reset by peer)] 00:48:59 Arelius [n=Indy@208.87.217.194] has joined #scheme 00:49:10 -!- dzhus [n=sphinx@95-24-113-2.broadband.corbina.ru] has quit ["gone asleep"] 00:50:35 CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has joined #scheme 00:54:15 incubot: go, lang, go! 00:54:18 this is the free lang! 00:54:42 evil is no go! 00:56:29 -!- Arelius [n=Indy@208.87.217.194] has quit [] 00:56:34 -!- mabes [n=mabes@66.236.74.194] has quit [Remote closed the connection] 00:57:08 Arelius [n=Indy@208.87.217.194] has joined #scheme 01:02:40 anyone know how fast the sort function in guile is? 01:03:21 -!- mario-goulart [n=user@67.205.85.241] has quit [Remote closed the connection] 01:05:43 what's a "less predicate"? 01:06:08 standard input:17:1: In procedure sort! in expression (sort (quote #) length): 01:06:09 standard input:17:1: Wrong type argument in position 2 (expecting less predicate): # 01:06:26 < 01:06:32 Presumably, a procedure of two arguments that returns true if the first one is considered less than the second, and false otherwise, under a total ordering of its domain. 01:07:37 hmm. 01:08:01 And I presume that the sorting procedure runs in O(n log n) average time, and possibly worst-case time, for a list of length n, and performing O(n log n) comparisons in the average (or worst) case. 01:10:08 yay :) I made a function that sorts lists by length 01:10:21 -!- schmir` [n=schmir@p54A93FAB.dip0.t-ipconnect.de] has quit [Read error: 113 (No route to host)] 01:11:26 -!- floodleboodle [n=nobody@Wabe8.w.pppool.de] has quit [Read error: 145 (Connection timed out)] 01:16:52 -!- mrsolo [n=mrsolo@nat/yahoo/x-blbhqdwycpcofiez] has quit ["Leaving"] 01:25:57 QinGW [n=wangqing@203.86.89.226] has joined #scheme 01:26:22 -!- npe [n=npe@94-224-251-223.access.telenet.be] has quit [No route to host] 01:26:53 mabes [n=mabes@bmabey.fttp.xmission.com] has joined #scheme 01:28:13 -!- Axioplase_ is now known as Axioplase 01:29:35 sladegen: are you in the "issue 9" camp? 01:30:47 i don't really care... 01:31:14 *sladegen* googles glang. 01:31:51 -!- rstandy [n=rastandy@net-93-144-228-61.t2.dsl.vodafone.it] has quit [Read error: 60 (Operation timed out)] 01:32:58 glang.sf.net... i guess in 10 years it will be just called the language :-} 01:37:26 -!- ve [n=a@cpat002.wlan.net.ed.ac.uk] has quit [Read error: 104 (Connection reset by peer)] 01:38:14 ve [n=a@vortis.xen.tardis.ed.ac.uk] has joined #scheme 02:00:07 -!- copumpkin [n=pumpkin@c-24-63-67-154.hsd1.nh.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 02:01:01 copumpkin [n=pumpkin@c-24-63-67-154.hsd1.nh.comcast.net] has joined #scheme 02:03:13 floodleboodle [n=nobody@Wb588.w.pppool.de] has joined #scheme 02:12:39 -!- kilimanjaro [n=kilimanj@cpe-173-172-99-25.austin.res.rr.com] has quit ["Leaving"] 02:18:47 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 02:21:53 -!- thermal_ [n=thermal@c-24-18-249-200.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 02:23:51 Riastradh: We need a test suite asap! 02:27:40 OK. Get right on it! 02:28:20 scary stuff! 02:30:39 foof [n=user@FLH1Ahu226.osk.mesh.ad.jp] has joined #scheme 02:31:46 -!- Arelius [n=Indy@208.87.217.194] has quit [] 02:34:11 rudybot_: eval (map round (list 2.5 3.5 4.5)) 02:34:11 mejja: ; Value: (2.0 4.0 4.0) 02:37:11 Hmm. 02:37:16 Riastradh: C back end also... 02:37:19 -!- george [n=george@189.107.176.74] has quit ["Konversation terminated!"] 02:37:34 george [n=george@189.107.176.74] has joined #scheme 02:37:35 -!- mabes [n=mabes@bmabey.fttp.xmission.com] has quit [Remote closed the connection] 02:38:25 double double_round (double x) { return (double_truncate ((x < 0) ? (x - 0.5) : (x + 0.5))); } 02:38:47 La la la 02:39:36 It's been that way for twenty years, too. 02:39:47 Gee. 02:45:12 (Cleverly, SSE doesn't include any rounding operations.) 02:47:38 kilimanjaro [n=kilimanj@cpe-173-172-99-25.austin.res.rr.com] has joined #scheme 02:54:34 Rounding is for sissies. Real mean truncate their floats. 02:56:04 It doesn't have any operations to truncate floats, either. It can convert them to integers, truncating them in the process, or convert double-precision floats to single-precision floats, losing precision in the process, but that's all. 02:56:49 Riastradh: eval (list (< 1.0 2.0) (> 1.0 2.0)) O_O 03:17:20 Oh, gee, thanks, C. The standard `round' function doesn't break ties to evens; it breaks ties away from zero. 03:18:08 mabes [n=mabes@bmabey.fttp.xmission.com] has joined #scheme 03:20:16 -!- mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 03:20:31 seamus_android [n=AW@host86-179-43-46.range86-179.btcentralplus.com] has joined #scheme 03:20:47 ada2358 [n=ada2358@login-linux.ccs.neu.edu] has joined #scheme 03:23:35 -!- MichaelRaskin [n=MichaelR@195.91.224.225] has quit [Remote closed the connection] 03:24:04 MichaelRaskin [n=MichaelR@195.91.224.225] has joined #scheme 03:28:43 -!- QinGW [n=wangqing@203.86.89.226] has quit ["Leaving."] 03:31:29 -!- ada2358_ [n=ada2358@login-linux.ccs.neu.edu] has quit [Read error: 110 (Connection timed out)] 03:39:10 -!- seamus_android [n=AW@host86-179-43-46.range86-179.btcentralplus.com] has quit [Read error: 54 (Connection reset by peer)] 03:44:12 saccade_ [n=saccade@209-6-54-113.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #scheme 03:51:25 -!- blackened` [n=blackene@ip-89-102-22-70.karneval.cz] has quit [] 03:57:58 -!- MichaelRaskin [n=MichaelR@195.91.224.225] has quit [Remote closed the connection] 03:58:27 MichaelRaskin [n=MichaelR@195.91.224.225] has joined #scheme 04:06:46 minion: memo for mejja: The rounding and comparison problems discussed on 2009-11-12 should be fixed now. 04:06:46 Remembered. I'll tell mejja when he/she/it next speaks. 04:09:15 sepult` [n=levgue@xdsl-87-78-131-89.netcologne.de] has joined #scheme 04:09:35 -!- sepult [n=levgue@xdsl-78-35-200-168.netcologne.de] has quit [Read error: 60 (Operation timed out)] 04:09:47 -!- tjafk [n=timj@e176195217.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 04:10:03 tjafk [n=timj@e176201109.adsl.alicedsl.de] has joined #scheme 04:19:09 -!- sepult` [n=levgue@xdsl-87-78-131-89.netcologne.de] has quit [Client Quit] 04:21:11 sepult [n=levgue@xdsl-87-78-131-89.netcologne.de] has joined #scheme 04:23:31 klutometis: Issue9 seems to be a good^Hf name, besides most obvious golang... 04:27:03 incubot: where are you going for the Christmas holidays? 04:27:05 Working on holidays? 04:27:17 incubot: not me. But perhaps you're forced to toil in the mines. 04:27:20 activity; toil; employment; sometimes, specifically, 04:27:26 sorry to hear that. 04:33:31 -!- jonrafkind [n=jon@crystalis.cs.utah.edu] has quit [Read error: 113 (No route to host)] 04:34:03 jeapostrophe [n=jay@69.169.141.110] has joined #scheme 04:35:37 -!- jeapostrophe [n=jay@69.169.141.110] has quit [Client Quit] 04:40:43 shyam_k [n=user@unaffiliated/shyam-k/x-8459115] has joined #scheme 04:41:34 just to make sure, there isn't any "identity element" for the cons operation? I mean, is there any A which makes (cons A B) evaluate to B? 04:42:17 That is correct. There is no such A. 04:42:27 There is, however, an A such that (APPEND A B) = B. 04:42:48 -!- flippo [n=frivol@unaffiliated/flippo] has quit ["Reality reasserts itself sooner or later."] 04:44:09 Oh, I know, it's (REVERSE B)! 04:44:35 foof wat 04:46:20 No, foof, that's the inverse of B, for which APPEND will yield the identity. After all, lists form a nice group. 04:49:39 shyam_k` [n=user@117.204.83.230] has joined #scheme 04:49:59 ops power-failure:( 04:51:32 Riastradh: ok.. but append can't be a primitive, right? 04:52:39 *can't be-> isn't 04:52:43 What's your definition of `primitive'? 04:53:11 dysinger [n=dysinger@166.129.153.79] has joined #scheme 04:53:20 hmm.. ok ok..:) 04:54:19 -!- MichaelRaskin [n=MichaelR@195.91.224.225] has quit [Read error: 113 (No route to host)] 04:54:38 -!- shyam_k` [n=user@117.204.83.230] has left #scheme 05:01:36 -!- Jacolyte [n=blaine@unaffiliated/jacolyte] has quit ["leaving"] 05:03:56 MichaelRaskin [n=MichaelR@213.171.48.239] has joined #scheme 05:04:25 -!- shyam_k [n=user@unaffiliated/shyam-k/x-8459115] has quit [Read error: 110 (Connection timed out)] 05:12:41 -!- kazzmir [n=kazzmir@98.202.86.149] has quit [Remote closed the connection] 05:13:04 kazzmir [n=kazzmir@98.202.86.149] has joined #scheme 05:18:48 kenjin [n=kenjin@163.152.180.51] has joined #scheme 05:19:14 -!- kenjin is now known as Guest79603 05:22:41 -!- annodomini [n=lambda@wikipedia/lambda] has quit [] 05:24:53 -!- samth [n=samth@c-76-24-220-74.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:28:51 -!- sepult [n=levgue@xdsl-87-78-131-89.netcologne.de] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 05:51:59 -!- Guest79603 is now known as kenjin2 05:55:26 jonrafkind [n=jon@98.202.86.149] has joined #scheme 06:04:20 -!- CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has quit [Remote closed the connection] 06:09:51 -!- ASau [n=user@83.69.227.32] has quit ["off"] 06:18:22 -!- splork [n=ben@dsl092-075-228.bos1.dsl.speakeasy.net] has quit ["Computer has gone to sleep"] 06:18:54 splork [n=ben@dsl092-075-228.bos1.dsl.speakeasy.net] has joined #scheme 06:22:04 -!- ski_ [n=md9slj@remote1.student.chalmers.se] has quit ["Lost terminal"] 06:26:42 ski_ [n=md9slj@remote1.student.chalmers.se] has joined #scheme 06:32:42 -!- dysinger [n=dysinger@166.129.153.79] has quit [] 07:14:23 ASau [n=user@host92-230-msk.microtest.ru] has joined #scheme 07:23:00 mmc [n=mima@cs27122078.pp.htv.fi] has joined #scheme 07:28:30 PongoTwistleton [n=user@dslb-094-219-072-159.pools.arcor-ip.net] has joined #scheme 07:29:14 -!- PongoTwistleton [n=user@dslb-094-219-072-159.pools.arcor-ip.net] has quit [Client Quit] 08:01:57 -!- mmc [n=mima@cs27122078.pp.htv.fi] has quit [Read error: 110 (Connection timed out)] 08:03:05 -!- kilimanjaro [n=kilimanj@cpe-173-172-99-25.austin.res.rr.com] has quit ["Leaving"] 08:07:09 -!- Fufie [n=innocent@80.203.225.86] has quit ["Leaving"] 08:13:01 pavelludiq [n=quassel@83.222.175.184] has joined #scheme 08:15:26 a-s [n=user@nat-240.ro.66.com] has joined #scheme 08:25:07 -!- proq [n=user@unaffiliated/proqesi] has quit [Read error: 110 (Connection timed out)] 08:30:13 Jafet [n=Jafet@unaffiliated/jafet] has joined #scheme 08:31:12 Nshag [i=user@193.250.85.206] has joined #scheme 08:39:33 ponzao___ [n=vesam@xdsl-83-150-86-25.nebulazone.fi] has joined #scheme 08:56:21 mmc [n=mima@esprx01x.nokia.com] has joined #scheme 08:58:32 -!- jonrafkind [n=jon@98.202.86.149] has quit [Read error: 110 (Connection timed out)] 08:58:32 Arelius [n=Indy@208.87.217.194] has joined #scheme 08:59:21 -!- xwl__ [n=user@esprx01x.nokia.com] has quit [Remote closed the connection] 09:01:45 Fufie [n=poff@Gatekeeper.vizrt.com] has joined #scheme 09:12:43 -!- schmir [n=schmir@mail.brainbot.com] has quit [Read error: 104 (Connection reset by peer)] 09:16:42 npe [n=npe@195.207.5.2] has joined #scheme 09:22:34 schmir [n=schmir@mail.brainbot.com] has joined #scheme 09:25:40 SvekloC [i=markush@unaffiliated/sveklo] has joined #scheme 09:32:33 -!- SvekloC [i=markush@unaffiliated/sveklo] has quit [Remote closed the connection] 09:32:39 -!- crink [n=crink@unaffiliated/crink] has quit ["Leaving."] 09:35:26 xwl__ [n=user@esprx01x.nokia.com] has joined #scheme 09:52:57 -!- MichaelRaskin [n=MichaelR@213.171.48.239] has quit [Remote closed the connection] 09:53:20 MichaelRaskin [n=MichaelR@213.171.48.239] has joined #scheme 09:58:52 QinGW1 [n=wangqing@203.86.89.226] has joined #scheme 09:59:32 -!- QinGW1 [n=wangqing@203.86.89.226] has quit [Read error: 104 (Connection reset by peer)] 10:01:27 QinGW [n=wangqing@203.86.89.226] has joined #scheme 10:04:49 -!- QinGW [n=wangqing@203.86.89.226] has quit [Client Quit] 10:05:10 QinGW [n=wangqing@203.86.89.226] has joined #scheme 10:16:01 -!- kenjin2 [n=kenjin@163.152.180.51] has quit [Read error: 60 (Operation timed out)] 10:28:46 -!- QinGW [n=wangqing@203.86.89.226] has quit [Read error: 104 (Connection reset by peer)] 10:41:59 rstandy [n=rastandy@net-93-144-180-123.t2.dsl.vodafone.it] has joined #scheme 10:55:12 mario-goulart [n=user@67.205.85.241] has joined #scheme 11:00:05 Edico [n=Edico@unaffiliated/edico] has joined #scheme 11:21:33 http://paste.lisp.org/display/90290 11:27:42 masm [n=masm@bl7-193-45.dsl.telepac.pt] has joined #scheme 11:29:36 -!- Axioplase is now known as Axioplase_ 11:40:26 edgard_ [n=edgard@189.60.22.76] has joined #scheme 11:40:31 -!- schmir [n=schmir@mail.brainbot.com] has quit [Remote closed the connection] 11:41:54 -!- edgard_ [n=edgard@189.60.22.76] has quit [] 11:43:51 -!- Jafet [n=Jafet@unaffiliated/jafet] has quit [Read error: 110 (Connection timed out)] 11:58:38 Jafet [n=Jafet@unaffiliated/jafet] has joined #scheme 12:07:07 hi! (define (f x) (+ x y)) if i enter this at repl, the interpreter would not shout at me that y is not bound. only after trying to eval (f 10) i get an error. even compilers, like chicken, give errors only at runtime. why? with lexical scoping, isn't it possible to determine that statically? 12:09:36 jeapostrophe [n=jay@69.169.141.110] has joined #scheme 12:15:41 Almost all scheme compilers do much less work than it is possible for them to do :-) 12:18:05 so it is always possible to statically determine whether a variable is bound, just noone does that? 12:19:06 In my ideal universe, a scheme compiler would *at least* be compelled (by standard) to check for arity correctness and pattently obvious type errors such as (cons (string-length "fubar")). 12:19:44 Which affect the latently typed nature of Scheme exactly 0%, if I'm not mistaken. 12:20:01 (define (cons ... 12:20:05 Lalala 12:20:32 Sorry, too early for lisp/scheme lesson - need more coffee. 12:20:39 if the answer to my question is "yes", we just check only those cases where cons binds to builtin 12:21:18 Summermute: Seeing how it is possible for them to do an infinite amount of work, that seems a tautology :) 12:22:07 In your case you could have a situation like this, though: file1.ss (define y 10) file2.ss (load file1.ss) (define (add x) (+ x y) (add 2) => 12. So y would be globally bound. 12:22:22 All syntax errors the total fault of my IRC client :-) 12:22:57 M-x pedantry 12:23:58 Summermute: well, but at the time the analisys is done, the interpreter has access to both files and can signal an error in pre-runtime 12:24:11 My hypothetical "do more work" scheme compiler would be compelled to give big fat warnings upon seeing unbound variables when not lexically apparent that they are bound in the global scope. So the above (load "file1.ss") would give an error. 12:24:51 So you seem to want a lint for scheme 12:25:05 There's one available for Gauche. 12:25:12 foof: true enough 12:25:13 #;1> (define (f x) (+ x y)) 12:25:13 Warning: the following toplevel variables are referenced but unbound: y (in f) 12:25:21 gribozavr: That's what Chicken tells me 12:25:25 Is that not enough? 12:26:12 sjamaan: i don't use chicken much, i just put that define into a file and called csc. it was silent 12:26:16 gribozavr: well, again, need more coffee, but that is likely true as well (just vaguely dreaming of IDE's of many years' past) 12:26:34 gribozavr, sjamaan: csc won't warn anything. 12:26:57 Doesnt PLT have a pretty cool static checker? I actually use PLT, but I've never tried it. 12:27:00 mario-goulart: I don't know; I always use modules and then you most certainly do get a warning (maybe even an error) 12:27:30 modules are allies in the war against load 12:28:18 ORLY? 12:28:58 Jafet: the redefine of language "primitives" is indeed an issue. No good simple answer at this level of caffeine intake (maybe need stronger drug) 12:29:20 "ORLY?" 12:29:37 wait - think i got it - i'm relatively new to irc lingo 12:29:48 (o-o) 12:29:57 gribozavr: Even chibi will give you that warning. 12:32:08 at the great risk of sounding very, very naive (too late), I think we need a new "reference scheme" that after the bitter infighting regarding its definition, is carefully put together for *pedagogical* purposes to show future implementers what is more-or-less easily accomplished. 12:33:06 Chicken 3 had a -G option for such warnings. 12:33:15 And, no not something like a toy Scheme from EOPL or the latest and greatest textbook featuring a circular interpreter :-) 12:33:33 luz [n=davids@189.122.90.116] has joined #scheme 12:33:51 Chicken 4 will warn by default inside modules, and also recommend what module to import from if it's a known name (e.g. you used `fold' without importing srfi-1). 12:33:58 blackened` [n=blackene@ip-89-102-22-70.karneval.cz] has joined #scheme 12:34:24 That is, indeed, a very nice feature! 12:34:40 I believe that even works for user-defined modules 12:35:27 After all (relevant to the work in front of me), 'tis alot easier for a computer to do a big tree search of directories and files than for a human to do so. 12:36:07 sjamaan: yes, you need to install them and run chicken-install -u though 12:37:18 Oh come on, can't it load up a CD or start combing the Net :-) 12:37:49 schmir [n=schmir@mail.brainbot.com] has joined #scheme 12:38:47 Wait a minute, in all seriousness, why wouldn't it start with the CWD and any/all project related directories/files (don't know chicken's user files org scheme) 12:40:06 then comes another question -- why is this just a warning? to allow (eval ...) or direct manipulation with environments? 12:40:18 gribozavr: It might be defined later 12:40:32 (think mutually recursive functions) 12:41:03 Well, the "load" trick described above relegates a multitude of sins to "warnings" 12:41:37 And then someone mentioned the redefinition of language defined "primitive" operations.... 12:44:37 I kind of like ML's "top-level" scheme, where you can redefine away, but no new value/name bindings are "retroactive." Applied to scheme primops, would make for a much safer (saner?) language, but would throw tons of "true believer" minimalists in to total fits requiring immediate medical attention. 12:45:15 Not really, it'd just be lexical scoping taken to extremes 12:45:38 binding to values instead of bindings to locations 12:45:56 Summermute: How do they create mutually recursive functions at toplevel? 12:46:59 Just about to comment on that - mutually recursive definitions are "vexing" and have many interesting options.. Being an "old man" by Net standards, I kind of like CL's (labels) forms. 12:48:05 Haskell does this whacky topological sort of all a files definitions, automagically making them all mutually recursive (it's a "lazy" language though) 12:48:21 fancy 12:48:22 thehcdreamer [n=thehcdre@94.198.78.26] has joined #scheme 12:49:49 It's just implicit letrec 12:49:56 Nothing horribly strange 12:51:32 And much more tractable with a lazy eval language 12:52:37 it's not a simple letrec: allows one to run code between evaluations of right-hand-sides. (define (a) ... b ...) (display ...) (define (b) ...) 12:52:40 Oh, and SML (don't know Ocaml) uses the "and" keyword between function definitions putting their names into their respective scopes. (It uses "andalso" for what we earthlings would typically use boolean "and" for) 12:55:09 Hmm, I'm no haskeller, but that side effecting "(display)" call smells like monad'ville. It's alot easier to imagine the topological sort strategy in the complete absense of side effects, which of course, is not strange in Haskell'ville. 12:56:59 The Monad of the Haskellville 12:58:51 I don't know about Haskell, but I do have a more-or-less uncritical conviction that quite a few SML features could add some nice things to Scheme without ruining the "schemeliness" 13:00:25 Just a silly example (not so silly really) is PLT's abandonment of mutable cons cells! That's a huge deal if you think about it. (Yeah, there's mcons, etc. - but they're outcasts at this point). 13:04:35 To me, the next step is a "datatype" (sort of mutually recursive define-structure) style construct for defining algebraic data types: (datatype List ((cons ?, List) Nil)) Where we use ? in place of type holders allowing for Scheme's latent typing. 13:06:28 The goal, to me, is to have *lower level primitives* than Lists that can define Lists, and then break up Scheme into nice little modules instead of one big monolithic collection of functions where cadddr lives side-by-side with string->symbol (which looks pretty bizarre if you look at it from afar). 13:09:50 (import [scheme symbol]) looks alot more "orderly" and "modern" as far as languages go. 13:11:25 Also make linked executables smaller depending on the organization of the runtime (even though we live in the era of "The Giant DLL/.SO Runtimes") 13:12:02 If you persevere and keep going, in a couple of years you just might reinvent Common Lisp. 13:13:32 Yeah, around 20 years ago I implemented a full CL minus CLOS. Packages - blech! But still a nice set of sequence functions, probably not outdone until STL finally came along. 13:15:14 As much as people (rightly) bang on the flaws of C++ templates, STL is truly, truly a work of both genius and very hard work. RELEVANCE: first STL research was done on Scheme, at SGI I believe. 13:16:50 I wonder whatever happened to that early Stepanov/Lee Scheme code base, btw? 13:17:40 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #scheme 13:19:31 CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has joined #scheme 13:22:23 Jafet: you comment on CL makes me reflect sadly - our "modern" languages are apples that never really fall very far from the tree(s): Pascal, Lisp, Fortran, Smalltalk. How much time we waste adding a complex number type here, exceptions over there, blah blah. 13:22:55 Erm, none of those are modern languages 13:23:15 In the end, it's all simplistic control flow, numbers, characters(more numbers), strings (vectors of chars/numbers), symbols(interned strings or numbers depending upon your interpretation) and trivial data structures (anything from cons cells to hash tables). Sort of a depressing thought for a very dreary USA East Coast morning. 13:24:05 Jafet: No, those are the "old trees" from which our "modern" language apples have fallen. Hence the phrase, "an apple never falls far from the tree." 13:24:07 There are healthy exceptions to all those, unfortunately they tend to appear in different languages 13:24:16 I see 13:24:25 kilimanjaro [n=kilimanj@cpe-173-172-99-25.austin.res.rr.com] has joined #scheme 13:24:55 Apparently, autocad's lisp does not have any form of let-binding. Or macros. 13:25:43 -!- Arelius [n=Indy@208.87.217.194] has quit [] 13:26:34 I mean, we had a few primitives for basic concurrency, threads/mutex/locks/etc. and we're falling all over ourselves trying to write correct programs. Will we ever get languages that can actually model interesting things - contracts, physical assemblies of parts, relationships between people, semantic relationships between documents (semantic web?). This morning, alas, I feel no closer. 13:27:51 samth [n=samth@c-76-24-220-74.hsd1.ma.comcast.net] has joined #scheme 13:28:04 As much as we site Autolisp, it's essentially an historical mistake. If not for the giant installed base, we'd have AutoBasic instead :-) 13:29:00 We do have VisualLisp... 13:29:01 In fact, it's been many years, but I recall seeing Autolisp code that looked pretty much *LIKE* basic and nothing much like typical Lisp code. 13:29:44 (set a foo) (set b bar) (if (= a b) (set c q)) - very "imperative" style code, IIRC. 13:30:46 Jafet - I didn't know that? I knew that in recent months Microsoft blessed essentially "Visual Ocaml" via giving F# official status. What is VisualLisp? 13:31:52 Some cut-rate AutoLisp with ActiveX bindings 13:32:00 I suppose one could call "Autolisp" a "VisualLisp" on total steroids! 13:32:22 No kidding? A product for $$$$? 13:33:57 -!- thehcdreamer [n=thehcdre@94.198.78.26] has quit [] 13:35:42 Maybe a little controversial, but if Visual F# worked more-or-less flawlessly (IDE and all) on the "holy trinity" of user-boxen (Mac, Win, Linux), I probably drop everything else and do 99% of my code in F#. 13:36:02 *sjamaan* runs in terror 13:36:35 That would ultimately depend on the portability and adoption of .NET 13:36:47 Which presents Microsoft a neat dilemma, if I may offtopically add 13:36:48 I mean, as for scheme, one could *write* a reasonable scheme interpreter in F# (pick your ML dialect) in a week or less :-) 13:37:49 xwl [n=user@123.115.124.13] has joined #scheme 13:37:55 Jafet: Yeah, I'd much prefer good old compiled binaries and put all this bytecode/jit stuff behind us, personally. (BTW, my CL of 20 years ago was a byte code lisp) 13:38:16 Bad old 13:38:26 ?? 13:38:47 annodomini [n=lambda@c-75-69-96-104.hsd1.nh.comcast.net] has joined #scheme 13:40:40 On the old front we have the famed p-pascal (??) and Smalltalk Bytecode and VM. We're talking LATE 1960's technology! Do we only have brain cells left over from a 2 decade baby-boomer drug induced binge to recycle 40 year old technology??? ;-) 13:41:12 You're trolling now. 13:41:19 Have fun 13:41:50 Jafet: Yes, just having fun. Guess the cafeine finally kicked in. Back to schemely matters. 13:43:37 So, anyone up on the latest new regarding the small version/large version split specs of our favorite language? 13:44:02 -!- splork [n=ben@dsl092-075-228.bos1.dsl.speakeasy.net] has quit ["Computer has gone to sleep"] 13:44:24 I mean post r6rs, bless its more-or-less DOA soul 13:45:19 gribozavr: in the REPL variables will always be assumed to exist, if you try run the code as a toplevel program ALL schemes MUST give you an error (else that scheme is not really scheme) 13:45:59 and that must be an error at compile time, the program should never be able to run like that 13:49:53 -!- annodomini [n=lambda@wikipedia/lambda] has quit [] 13:50:19 Unless one finds/uses an ancient tiny scheme interpreter that automagically binds all unbound variables to '() 13:50:48 But, I guess that wouldn't really count as a scheme anymore. 13:53:04 Summermute: I like F# too, just havent coded in it, but I like what I see 13:55:37 I think it's on YouTube (but might be on, what, "Channel 9", or the msoftie site), but there are some really good interviews with some of the F# dudes - very astute folk. In another interview, I seem to recall that all the LINQ code (modern monads from Peyton Jones point of view) is already millions of lines of code. 13:56:41 *zbrown* thinks Summermute's discourse isn't near as entertaining as Wintermute's ;) 13:56:58 Oh, and like other CLR languages, generics are monomorphized (sp?) for non-pointer sized arguments - so no boxing of doubles. 13:57:20 *Jafet* invades zbrown's dreams 13:57:44 -!- Nshag [i=user@193.250.85.206] has quit [Read error: 145 (Connection timed out)] 13:57:51 hmmm I should read that book again 14:00:10 I guess my point here (and I more or less loathe microsoft), is that building a total "environment" or "ecosystem" for any language that would compete with the F# class of (1) raw language processign technology and (2) sheer man hours on IDE, libraries, etc. - well, is possible only by a really small group of giant companies at this point: MS, Oracle/Sun, IBM, Cisco, Software AG Europe (???).... 14:00:40 *zbrown* doesn't loathe MS. 14:00:51 then again, they will be signing my paychecks in about 2 months 14:00:53 :-P 14:01:06 zbrown: Wintermute has to be the coolest AI character in pretty much any work of fiction :-) 14:01:16 davids [n=davids@189.122.90.116] has joined #scheme 14:01:35 Summermute: Neuromancer is the coolest techno-hackerish-book in all of fiction :) 14:01:39 -!- luz [n=davids@189.122.90.116] has quit [Read error: 110 (Connection timed out)] 14:01:54 That's pretty good, considering it doesn't even have it's own personality... 14:02:14 -!- npe [n=npe@195.207.5.2] has quit [Remote closed the connection] 14:02:44 npe [n=npe@195.207.5.2] has joined #scheme 14:03:51 zbrown: On that we are birds of a feather. BTW, almost went to work as a program manager at MS circa 1990 (already "too old" to sling code according to their recruiter). Really smart group of folks to work with in any case. Best of luck to you there. 14:03:53 Summermute: you chose wrong language. 14:04:21 I can type a little Urdu, but I don't have the right keyboard here. 14:04:22 Summermute: go SML, Caml, Clean & other real functional programming languages. 14:05:01 mmm I don't think I'd count F# as any less a functional language than those, its a funky mix of Haskell and OCaml 14:05:12 Well, F# is 99% Ocaml, with a rather nice simplified syntax option. 14:05:18 -!- npe [n=npe@195.207.5.2] has quit [Client Quit] 14:05:35 npe [n=npe@195.207.5.2] has joined #scheme 14:05:39 That's why I have F# envy and not "Visual Basic" envy :-) 14:06:16 As for me, Scheme needs more release engineering. 14:06:26 Any implementation. 14:08:04 You have enough code to reuse, but it is either buggy and unsupported 14:08:04 or supported to the extent that it changes in arbitrary way with the time. 14:08:20 edgard_ [n=edgard@189.60.22.76] has joined #scheme 14:10:12 OCaml is too convoluted for my casual use, so I end up keeping SML installed on my laptop. I did install Clean, and just didn't have the time, although it looks brilliant. I have an ongoing "when nothing is on TV" working through a Haskell tutorial with ghci on my box. And then, of course, PLT's DrScheme (better error messages) and Emacs eat up most of my CPU cycles (which is lame given how many of them there are these days - SETI at home?) 14:11:16 ASau: I'm pretty sure I agree totally. 14:11:26 *ASau* uses SML as a kind of ATP for prototype projects. 14:11:32 ATP? 14:11:44 Authomatic theorem prover. 14:11:51 ATP is fun :) 14:11:59 I'm in a grad level class for that right now. 14:12:02 Summermute: I wrote an implementation of LINQ in R6RS scheme, but not quite the whole thing, only operates on 'enumerables' and not 'Queryable' yet, but that is for version 3, allowing different macro generation for backends 14:12:19 Ah. Like using yacc to work out the kinks in a grammar, and then hand coding it for much better error reporting :-) 14:12:27 it's nice, as I use LINQ everyday at work 14:12:36 well C# 14:12:59 leppie: very cool! 14:13:11 but after 2 years of Scheme, I must say I wwrite much better C# code, very functional, minimal mutation 14:13:36 zbrown: are you using Coq and tools like that? 14:13:48 Summermute: nah, we work mostly in FOL's not HOL's 14:14:03 Ah 14:14:52 ASau: Back to your point - I think you've really hit on something (or made me hit on something) 14:15:06 I suspect if Scheme had a stronger release engineering cyle, I could be convinced to use Scheme nearly exclusively 14:15:31 We have N scheme implementations out there (N +1 every few hours :-) 14:15:31 I don't know it well but I know it well enough to like it :) 14:15:44 What we don't have are two "suites". 14:16:02 First: A compliance suite. Pass = scheme; Fail = not scheme 14:16:30 Second: a reference scheme implementation to validate libraries. Pass = scheme library; Fail = not scheme library 14:17:22 -!- edgard_ [n=edgard@189.60.22.76] has quit [] 14:17:26 All hail the cottage industry language? 14:17:38 Ok and then Third, some standard, reference command line "scmtool" utility to do some basic things like install a single module or a packaged up library (zip,tar,ar) 14:19:40 Within those parameters, we could still have a loosey goosey scheme world, but there would be a "core" of software technology that is basically "validated" for the more no-nonsense "I just want to reliably code scheme in some reasonably safe, perhaps compiled/optimized and cross platform way" crowd of coders. Too radical? Any obvious flies in the ointment? 14:20:02 Jafet: ??? 14:22:25 Jafet: If I understand you, just passing a reference suite doesn't necessarily make either a scheme compiler or a set of modules "commercial quality". It would just mean: it runs, produces good results; produces errors when it should; installs (for modules) via the reference utility. 14:23:14 Whether such an item would be worth $$$ is another matter altogether? In this tech environment, my bet would be on "mostly free." 14:23:31 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 14:24:00 I think you are proposing to solve a cultural issue via purely technical means 14:24:24 Summermute: the problem isn't in standards. 14:24:35 Jafet: Yes, in some sense, I probably am violating that dictum. 14:24:36 The problem lies in practical things. 14:25:07 You can use R4RS pretty successfully. 14:25:21 But compliance suites, or benchmarks and so forth are really A TOTAL NORMALITY in many other language cultures. 14:25:29 If there isn't a perceived urgent need to standardize certain aspects of scheme implementations, nobody is going to do it 14:25:48 -!- xwl [n=user@123.115.124.13] has quit [Remote closed the connection] 14:25:49 Jafet: that's what I'm talking about. 14:25:51 Other languages are used for tasks that lead to that sort of need. 14:26:12 Problems don't arise in standartization area. 14:26:33 bweaver [n=user@75-148-111-133-Chattanooga.hfc.comcastbusiness.net] has joined #scheme 14:26:35 C++ has one of the worst specification documents for any mainstream language that ever existed, and yet implementations interoperate very well, often well beyond that required by documented standards. 14:26:41 And 99% of scheme users are intro-to-programming students, so one could say a brilliant generation or two of language designers are essentially pissing into the wind.... Sorry dreary morning. 14:27:14 Knowledge gained from learning scheme does not become worthless because scheme is a cottage industry language. 14:27:19 99% of C or C++ users are students as well. 14:27:54 We don't consider that problem. 14:28:02 As much as I agree with your sentiment, Summermute, interoperability will not just appear from the wind. 14:29:02 Jafet: Excellent point on C++. I used to read (why, I can't fathom) C++ report pretty faithfully, and the excruciating level of inspection of various corner cases in the various implementations was really,really AMAZING - and we're talking mostly Bell Labs luminaries writing the articles back in those days. 14:30:02 zbrown, since Scheme is not a piece of software, which Scheme's release engineering are you not happy with? 14:31:10 -!- jeapostrophe [n=jay@69.169.141.110] has quit [] 14:31:22 ASau: That's true, but the pie is way bigger, and so the 1% left over are writing device drivers, or wicked fast parallel graph processing algorithms for super computer clusters in C/C++ :-) 14:31:24 samth: eh I am mostly in agreement with Summermute with regard to some sort of compliance suites for libs and scheme impls. At least for a particular release (ie: R5RS or w/e) 14:32:03 zbrown, that's not release engineering 14:32:11 what scheme do you use? 14:32:22 does it have problematic release engineering? 14:32:36 samth: well I just grabbed a word that Summermute used carelessly, my point is that I'd like to see some sort of reference impl 14:32:36 what's the C++ release engineering process? 14:32:58 forget that I said release engineering, its not much my point 14:33:00 Bad. 14:33:12 samth: yes "release engineering" probably isn't a good term for the literal problem 14:33:25 Jafet: whats bad? 14:33:27 zbrown, how would a reference implementation solve anyone's problems 14:33:40 I think Jafet was referring to C++ rel eng 14:33:42 Although I'm intrigued by who r6rs versioned libraries might end up being used in practice. 14:33:43 The C++ release engineering process. 14:33:54 there's a pretty extensive test suite for r6rs 14:33:58 whoop- "by how r6rs..." 14:34:11 samth, if people wanted one and didn't have one, then one would solve their problem 14:34:23 The devil's in the preconditions. 14:34:24 but why would people want one? 14:35:00 PongoTwistleton [n=user@dslb-094-219-072-159.pools.arcor-ip.net] has joined #scheme 14:35:18 sorry, want one what? Lost the thread.... 14:35:23 i don't know of any languages that have a reference implementation and a standard 14:35:30 Summermute: reference impl 14:35:33 why would someone want a reference implementation? 14:35:55 samth: and not really any other languages have quite as many implementations as scheme 14:36:07 there are lots of implementation of both C and C++ 14:36:11 samth: why would a community NOT want a reference implementation? 14:36:13 -!- PongoTwistleton [n=user@dslb-094-219-072-159.pools.arcor-ip.net] has left #scheme 14:36:25 What would be the point of a reference implementation? 14:36:32 First, it would set a standard for module/library compliance. 14:36:34 Summermute, b/c writing one takes someone's time, that they could spend doing something useful 14:36:41 like chatting on irc 14:36:45 or writing libraries 14:37:13 Summermute, what standard would it set that would not already have been set by the, uh, standard? 14:37:37 I still think Scheme should have a common standard library, like libc, that is mostly implementation independent 14:37:51 samth: but if we do a little informal arithmetic, the time spent on a reference implementation could save a BUTT LOAD of time down the road debugging the oodles of library code that makes up most language environments. 14:38:13 leppie, r6rs comes with a decent one, doesn't it 14:38:21 *zbrown* needs to think a little more deeply on this before he expresses further opinions. The idea is not yet solidified. 14:38:54 Summermute, why do you think that the C or C++ or Ada communities have not come up with one then? 14:39:14 Jafet: YEA for std lib!!! Booo for old style monolithic "interpreter legacy" scheme design :-) 14:39:30 Jafet: mostly odds and ends, I have not used any code in IronScheme from the 'reference' implementations from 5.91 - 5.97 14:39:45 leppie: I think the std lib would be nice 14:39:46 the only code I did use was the same defintions of some procedures 14:40:00 leppie, well I'm just going by what the PLT docs seems to say about r6rs 14:40:18 for example my enums library is 100% written in R6RS 14:40:27 Summermute, i don't see the connection between standard libraries and interpreter-based design 14:40:29 samth: That's a very, very good question. They seem to achieve a sort of "compliance" via the marketplace, for lack of a better term. 14:40:40 Summermute, "consensus". 14:40:51 not that i'm a big fan of only caring about interpreters 14:41:06 Although if you used pragma in gcc 1.x, it'd start nethack. 14:41:07 Summermute, they have a very precise standard 14:41:08 Jafet: Far beyond "consensus" is "user pressure." 14:41:54 languages defined by 'reference implementations' seem mostly to be those where the designers can't come up with a spec 14:41:55 I would love for Scheme to have a comprehesive spec like the MS CLR has 14:41:56 like python 14:41:56 samth, yes, the standards clearly and precisely define a massive class of common behaviours as undefined 14:42:19 I just realized the problem: It's totally heresy, but the problem is that the scheme designers and implementors are driving the whole process. If we were using their tools for 20 years to calculate ballistics trajectories or something, the scheme universe would be VERY VERY different. 14:42:22 Jafet, these days GCC has started taking more advantage of that 14:43:02 Summermute, i'm confused 14:43:12 what does ballistics have to do with anything? 14:43:21 Summermute: how would you like a typed scheme for the CLR in the spirit of F#? 14:43:41 leppie, how about in the spirit of Typed Scheme :) 14:44:17 samth: I am going to try keep the syntax similar to Typed Scheme 14:44:19 samth: not a necessary relationship, just a history of "sitting down to the REPL and Emacs and having the whole std lang at one's fingertips." It's not *bad* per se. I just prefer a language setup better from the beginning for a *huge* volume of forthcoming libraries. 14:44:49 but I am yet to figure out how I am gonna handle cases like map and fold-left as per your(?) paper on that 14:45:04 leppie, yeah that would probably be hard on the clr 14:45:17 you should look at one of the tech reports cited in our paper 14:45:33 which talks about doing similar things in the context of generics in a class-based lang 14:45:46 not too hard, i think, I have 2 options currently, and will probably settle for a combination 14:45:47 samth: python is an example of yet another odd but still "good thing" - the "one true implementation" approach. Gotta say it's really worked well for perl, python and probably some others (squeak?) 14:46:28 Summermute, pick a Scheme implementation, and treat that as the "one true implementation" 14:46:38 "The one true implementation that runs on the one true platform, x86-32" approach? 14:46:39 that's certainly how I work with Scheme 14:46:39 leppie: now you're teasing me! 14:47:17 Pretty much that's what I do - PLT (probably like many others just judging from IRC and the mailing list) 14:47:38 1st for 8 arguments or less, will be 'properly' typed in the sense of CLR byte code, beyond that, I will follow probably the approach used in Typed Scheme, by decoration the functions with type contract, and evaluate the contract at compiel time, and just tell the proc to accept it 14:48:17 leppie, sorry, it's not cited, but look at the paper here: http://www.ccs.neu.edu/scheme/pubs/NU-CCIS-08-01.pdf 14:48:20 im trying to avoid runtime casts and type checks 14:48:30 But I am bummed out that it's butt slow, and then something fast like Larceny is basically incapable of "adopting" all of PLT's libraries and goodies. 14:48:44 Summermute, it really isn't 'butt-slow' 14:49:16 ooo, thanks samth, now that looks familiar :) 14:49:59 annodomini [n=lambda@130.189.179.215] has joined #scheme 14:50:12 the idea is to have 9 such interfaces (well they are in IronScheme already, in both typed and untyped versions) 14:50:34 leppie: are you actually banging on this in a serious way???? 14:51:06 and beyond that, either use the approach described before, or use some kind of generative compile time procedure 14:51:15 Sorry, 9 interfaces to what? 14:51:17 so your procedure will be a template 14:51:30 Summermute: look at that pDF link 14:51:33 borism [n=boris@213.35.235.143] has joined #scheme 14:51:38 bottom of first page 14:51:41 I definitely wll 14:51:54 Mono? 14:51:55 its like Func etc 14:52:04 but using an interface 14:52:26 but I am going to use vurtual methods instead, like F# does 14:52:30 virtual 14:53:02 currently mine is a delegate wrapped inside such a class, so it is pretty slow 14:53:25 mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has joined #scheme 14:53:25 Summermute: yes, I have been 'thinking/designing' this in my head for about a year now 14:53:52 at one stage I was thinking of using the F# compiler internals for code generation 14:54:16 but at 9MB for that assembly, I cant... 14:54:32 Why? 14:54:33 mejja, memo from Riastradh: The rounding and comparison problems discussed on 2009-11-12 should be fixed now. 14:54:38 One of the things, actual 2 of the things, that really get my goat about PLT's typed-scheme is (1) it isn't integrated into the compiler/jit to elide no-longer type checks and (2) I seem to run into some bug with it just about every other day - bugs which people happen to actually know about but are usually undocumented. 14:54:38 so next best bet, write the whole damn thing in Scheme :) 14:55:01 So in a way, my "ideal language" is just a pain in the ass to work with in reality! 14:55:24 mejja: I pride myself with the fact that IronScheme's installer is only 1.2MB 14:55:44 -!- mabes [n=mabes@bmabey.fttp.xmission.com] has quit [Remote closed the connection] 14:55:52 Summermute, did you see the recent changes for vectors? 14:56:29 mabes [n=mabes@bmabey.fttp.xmission.com] has joined #scheme 14:56:36 leppie: That's ok -- it will pass. 14:56:39 (Or die.) 14:56:50 samth: you just brought a tear to my eye!!!!! Really, is there some doc page or a mailing list I missed (there's lots of them to miss)????? 14:57:01 are you on the plt-scheme list? 14:57:10 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 14:57:37 see my email entitled 'New vector library' 14:57:38 yes, but I haven't checked my mail in about 24 hours (after many years, I just find it too distracting to have mail up 100% of the time) 14:57:48 i sent it on the 8th 14:57:50 I'm on my way there 14:58:07 my code editor/IDE's (I wrote in 100% C#) setup is only 640KB, and the text rendering is on par with VS2008 and faster 14:59:29 leppie: you know that 640K is the 666 of the PC world. I'd add a few bytes just in case :-) 15:00:08 why? it's friday the 13th and my last release was in april, version 0.2.0.87 15:00:10 :p 15:00:40 -!- mabes [n=mabes@bmabey.fttp.xmission.com] has quit [Remote closed the connection] 15:00:49 i should be on target to win the lottery today 15:00:50 leppie: Um, why not just integrate your language into VS in the first place (I have it, but no, don't use it - used to be a Borland C++ man) 15:01:06 VS *is* a rich environment these days. 15:01:26 I have a syntax editor plugin for VS2008 15:01:41 And I imagine CLR integration/etc - well, the whole toolset integration shebang must be pretty good for getting work done. 15:01:43 hoping to expand it a bit more 15:02:07 yeah, I can even do 'some' step debugging directly in VS :) 15:02:33 leppie: I see, I don't know the VS lingo. 15:02:35 my web framework is 400 lines of SCheme code, rest just leeches off ASP.NET 15:03:16 Summermute: like any source debugger, I can step thru the code expression by expression 15:03:28 step in and out of functions 15:03:39 leppie: I'm starting to have naughty thoughts. 15:03:41 look at local variables 15:04:01 and one in the environment 15:04:19 not quite like F#, but enough to track down nasty bugs 15:04:47 for more than 18 months I wrote/developed IronScheme without a debugger for the Scheme code 15:04:48 OK. I think my boxed VS is2005. Get to some juicy screen shots and your word that it mostly works and i'll upgrade and give it a good working over :-) 15:04:55 I rarely use it now 15:05:24 Oh, I'm just making the connection. You're behind IronScheme 15:06:01 yeah 15:06:28 So, if I get you right, now you're (1) integrating with the VS tool set and (2) implementing something akin to type-scheme on top of it, but with much better CLR/"optimization" 15:07:59 Am I roughly on target? 15:08:00 (1) not so much (2) not yet at implementation, need to figure out many details still, but I'm in no rush, also work been keeping me too busy to really focus on it 15:08:07 brb 15:08:29 sorry: i'm still irc starved - "brb" 15:10:04 Summermute, if you could see if the new vector stuff works for you, that would be great 15:12:06 samth: Just read Vol 51 Issue 31 Message #7 - Thanks for your efforts! 15:12:12 dysinger [n=dysinger@166.187.193.173] has joined #scheme 15:13:00 I'm in the middle of (1) cobbling together an emacs mode for my "tnl" language and (2) working out the details of the module system (based on r6rs mostly). 15:13:10 tnl? 15:14:16 But then I'll get back to my library of typed-scheme vector utilities I plan to use to build the actual compiler and give it a good working over hopefully. Actually, it sounds like you will already have written most of my planned t-s vec utilities and then some! 15:14:46 -!- Jafet [n=Jafet@unaffiliated/jafet] has quit ["Leaving."] 15:15:46 I'm writing a compiler for a language provisionally called "tnl" in PLT Scheme, that's all. Think of it as ML meets Scala meets CL's family of control structures - but totally different :-) 15:16:02 if i've left out any useful vector functions, let me know 15:16:33 ok i am back 15:17:14 Summermute: so another weird bastard of a language :) 15:17:15 Sure, or I'll take the one's I've written /will write myself and post them to that cool "lisp code board" thingy and you can judge for inclusion or not. 15:19:15 zbrown: yes and no. There are "corner cases" that are hard - I do NOT like Scala's "case class" solution to integrating classes and algebraic data types. Just too much damn typing for a language designed from scratch (typed-scheme gets a pass here). 15:19:16 -!- leppie [n=lolcow@196-210-146-247-tvwt-esr-2.dynamic.isadsl.co.za] has quit [Read error: 54 (Connection reset by peer)] 15:19:37 leppie [n=lolcow@196-210-146-247-tvwt-esr-2.dynamic.isadsl.co.za] has joined #scheme 15:21:13 I'm adding ML's infixr style operator definitions. It compiles to object code supportingn C, standard (windoze), Pascal calling conventions, so the "FFI" really just has to deal with nastiness of the GC memory management scheme. 15:24:28 Promise - last off topic post. Instead of Scheme/Dylan style macros, I have "syntax-modules" which are compiled in advance and use a completely built in lex/yacc to tranlate pretty much any language into native "tnl" (or the first level intermediate representation). So I can do stuff like (import (prefix "sql/" (all [db sql]))) and then have delimiters around SQL code - fully type checked, etc. 15:24:33 FufieToo [n=poff@Gatekeeper.vizrt.com] has joined #scheme 15:25:20 Paraselene__ [n=None@79-68-173-7.dynamic.dsl.as9105.com] has joined #scheme 15:25:30 There's a standard escape syntax for embedding tnl code in "syntax-module" defined languages: ${tnl-expr} does the trick. It's pretty cool. 15:27:39 Whew! Sorry for all that. I just get excited about my baby :-) Hope you understand. 15:27:59 mabes [n=mabes@66.236.74.194] has joined #scheme 15:31:23 Nshag [i=user@Mix-Nantes-117-4-149.w193-250.abo.wanadoo.fr] has joined #scheme 15:32:36 any letrec experts on the channel? 15:33:16 gribozavr, probably Chicken only warns so that it doesn't halt the compilation process, during which you can get multiple warnings for all the undefined variables. In some Scheme systems, though, such as Scheme48, you can later define the name (in the appropriate context, using its module system), and the definition will automatically be reflected in the place of reference, enabling quicker interactive development. 15:33:57 Summermute: I use it really well. 15:34:06 Summermute: http://ironscheme.codeplex.com/wikipage?title=IronScheme%20Visual%20Studio%202008%20Integration 15:34:06 -rudybot_:#scheme- http://tinyurl.com/yj3rwqw 15:34:17 gribozavr, `Oops, while compiling and loading this huge pile of code, I see that I forgot to define FOO in this one utility module. C-x b utility.scm RET C-M-x -- all better now, and no need to reload that huge pile of code.' 15:34:35 leppie:can you repoint me to the PDF, I've scrolled too far 15:34:48 http://www.ccs.neu.edu/scheme/pubs/NU-CCIS-08-01.pdf 15:35:58 -!- sstrickl [n=sstrickl@pool-129-44-182-54.bos.east.verizon.net] has quit [] 15:36:15 Summermute: what do you want to know about letrec, it has given me enough headaches 15:37:22 jimrees [n=jimrees@ita4fw1.itasoftware.com] has joined #scheme 15:38:50 leppie: real quick perusal of a few pages - how does this relate/not relate to the "typed-scheme" paper on typing things like (map (a..g -> r) (Listof a) ... (Listof g) -> (Listof r)) 15:38:58 Sorry for the ad hoc notation 15:39:03 -!- Fufie [n=poff@Gatekeeper.vizrt.com] has quit [Read error: 110 (Connection timed out)] 15:39:45 i think this explains more from the point of view of the underlying 'language/runtime' 15:40:16 more implementation specific, which is nice for me :) 15:41:42 gotcha - Just about once per day I seriously consider redesigning "tnl" to take the Haskell and ML approach of all functions being (A -> R) with lists and tuples and records making up the "mess" of an otherwise one arg/one return function application scheme. 15:41:48 unfortunately I dont read denotational semtics definitions so well :*( 15:42:15 looks like pure greek to me 15:42:37 the paper or my stupid example? 15:42:55 the paper, your example is perfectly understandable :) 15:42:56 jlongster [n=user@75-148-111-133-Chattanooga.hfc.comcastbusiness.net] has joined #scheme 15:43:11 thanks :) 15:44:07 does anyone have a nice 'cheat sheet' for denotational semantics? 15:44:33 holy batshit - now I see what you mean - the 2nd half of this paper is a week long effort with a magnifying glass and google near at hand for me. 15:45:42 i guess it cant be too had to read if you use it frequently 15:45:49 s/had/hard/ 15:45:56 One nice thing about scheme is that despite it's minor, if vexing, incompatibilities, it's a pretty darn understandable nice notation. 15:46:22 and to think the parenthesis bothers some, bleh 15:46:27 leppie: there is no royal road to denotational semantics. 15:48:02 maybe I should try playing with PLT Redex again, but that just confused me like hell too 15:48:12 I read an article by Stephen Wolfram (sp?) of Mathematica fame, and he points out that despite popular mythology, mathematical notation is really a COMPLETE AD HOC MESS. And that this was a big issue in coming up with mathematica's supported syntax. I studied quite a bit of mathematical logic in my college days, and I can concur from that point of view. 15:48:37 leppie, there's now a good book on Redex 15:49:00 and it still wont gaurentee me understanding it :) 15:49:19 I just have the necessary background for all this theory :( 15:49:24 just dont have 15:49:58 masm: I'm not doubting you, but one would think you could go to Amazon and buy a book in two volumes: Volume 1: Denotational Semantics, Volume 2: Operational Semantics. And get over the constant notational (for me) headaches. 15:50:39 -!- a-s [n=user@nat-240.ro.66.com] has quit [Remote closed the connection] 15:52:10 Instead, I have aging but still terrifying memories of my professor Goldfarb devising a notation for formalized Turing machines on the black board more or less "on the fly" from his notes!!! 15:52:44 -!- drewfer [n=drewfer@67.58.77.138] has quit [Read error: 131 (Connection reset by peer)] 15:52:45 lol 15:52:54 BTW, I find even the Wikipaedia entries pretty weak on these topics. 15:52:59 drewfer [n=drewfer@67.58.77.138] has joined #scheme 15:53:44 perhaps it IS only meant for really smart people 15:54:20 -!- schmir [n=schmir@mail.brainbot.com] has quit [Remote closed the connection] 15:54:26 What? Wikipedia? 15:54:47 Weird thing is I'm pretty much a "literary type" who has always banged away on "left brain" problems because I find them pleasantly hard compared to writing haiku :-) 15:55:12 What are "left-brain" problems? 15:55:22 ASau: I can't spell to save my sould.Sorry. 15:55:52 dzhus [n=sphinx@95-25-177-162.broadband.corbina.ru] has joined #scheme 15:56:22 Mathematics and languages are the same thing. 15:56:53 with all this talk, hopefully it will inspire me to start documenting my ideas for IronScheme 15:57:19 I know this because I have hard core chronic fatigue. While there is compensation in damaged brains, etc. in general, the "left brain" is the ordered, reasoning, logical brain, while the "right brain"is the "creative", pattern matching, intuitive brain. Really, we need both. For 3 years I could not even read a book, because chronic fatigue clobbers the left brain so hard. 15:57:56 Summermute: so you fall asleep as soon as you start reading? 15:58:08 On an even more personal note, I'm writing my "tnl" compiler as a kind of "celebration" for somewhat regaining my old left brain capabilities. Yippie for me!!!! 15:58:34 Summermute: Damn... that must suck. 15:59:43 Oddly, not far from the truth. A kind of "brain fog" that is also somewhat like a kind of tiredness. it's weird. I still can't be around my own 3 year old son for more than about 2-3 hours, because the stimulus makes me literally nearly fall asleep on my feet. 15:59:51 Yes - it does suck. 16:00:22 -!- ASau [n=user@host92-230-msk.microtest.ru] has quit ["off"] 16:01:47 But now i can read, and kinda get get through tech papers, and do more with computers than just surf finance news or whatever (I literally watched Buffy reruns on DVD for nearly 1 full year - that was my level of function). So life, well, sucks less. And Scheme is helping is suck even less :-) 16:03:53 -!- FufieToo [n=poff@Gatekeeper.vizrt.com] has quit ["Leaving"] 16:03:55 sounds like post traumatic depression almost 16:04:25 from a functional level 16:06:17 -!- drewfer [n=drewfer@67.58.77.138] has quit [Read error: 104 (Connection reset by peer)] 16:06:29 drewfer [n=drewfer@67.58.77.138] has joined #scheme 16:06:52 -!- samth [n=samth@c-76-24-220-74.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 16:06:56 ahh, CFS == yuppy flu 16:07:07 my cousin got that currently :( 16:07:54 Shouldn't get into it, but it's an immune/neurological disorder that creams the HPA access (A as in Adrenaline), so yes, it's very related to PTSD and stimuls response. Oddly, due to the immune problems, I've been to the hospital so many times, I've already been diagnosed with 'REAL' PTSD, where I can have vivid images of being hit by cars, cut by knives, dropping my baby,etc. 16:08:48 Really classic PTSD stuff at this point. Sounds weirder than it is. I'm mostly an educated, middle class guy in a nice urban neighborhood who dresses preppie and drives a Mercedes and what not. I just can't work, and so I'm mostly a "shut in." 16:09:05 -!- drewfer [n=drewfer@67.58.77.138] has quit [Read error: 104 (Connection reset by peer)] 16:09:17 drewfer [n=drewfer@67.58.77.138] has joined #scheme 16:09:40 leppie= exactly, yuppie flu. Starts with this bizarre horrendous flu that lasts for weeks - feel like you're gonna die. very strange. 16:09:43 so there is no real treatment? no drugs to help? 16:09:53 ejs [n=eugen@137-9-178-94.pool.ukrtel.net] has joined #scheme 16:11:23 Anyway, last week in Science they *FINALLY* tied CFS to an obscure retroviruus called XMRV which in the medical community is only of interest to date because it shows up in men with prostate cancer. But there's an 80% correlation between CFS and XMRV - which is a big correlation in biology. So maybe a cure is on the way???? A man can dream. 16:12:57 Clonapin helps even out the nervous system, but it's addictive and can be somewhat sedating. Other than that - NEVER have you seen so many snake oil salesmen than at the book store section on Chronic Fatigue. It's more random and stupider than the dieting scheme books!!!! 16:13:14 eish 16:13:51 -!- npe [n=npe@195.207.5.2] has quit [No route to host] 16:14:11 wont some adrenalin stimulants like cipralex or ritalin help (i guess ritalin will probably make it alot worse) 16:15:10 Oddly, it's usually the "type-A" folk who get it - I was a CEO of a medium sized tech company. Keith Jarret was a famous jazz pianist. The woman who wrote Sea Biscuit has it. Pretty bizarre. Seems to be the XMRV *PLUS* some genetic/environmental/behavioral proclivity to get the disease. 16:16:02 -!- ejs [n=eugen@137-9-178-94.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 16:17:04 Yea, like i said it's weird. One get's tired because the nervous system is OVER stimulated, so it sort of konks out after after a few hours (I can sleep on and off for 2-4 hours all day/nigh for months at a time). So stuff like Ritalin or Aderol are no good - I pity the CFS folks who need Aderol for their ADHD :-) 16:18:50 Back to letrec. I'm sure there's some lame #cfs irc channel somewhere (most are ladies bitching about how they feel tired after work and their neck hurts - online CFS community is LAME) 16:19:55 lol 16:20:00 LETREC. So let's start with CL LABELS and Scheme's LET* 16:20:01 lets stick to letrec 16:20:20 there are some good papers on 'fixing letrec' 16:20:42 LABELS is functions only - all defined as lambdas and "bound" simultaneously, if I read that right. 16:21:11 yes, like a limited form of letrec that can bind only to procedures 16:21:22 So, not feeling too clever, LABELS semantics are clear. 16:21:40 that form is called 'fix' 16:22:06 -!- ski_ [n=md9slj@remote1.student.chalmers.se] has quit ["Lost terminal"] 16:22:20 Oh, I've seen "fix" in "Compiling with Continuations" (great book, IMHO) - where does that term, "fix" come from? 16:22:39 the fixpoint, from the Y combinator 16:22:53 (not that I really know what that mean, but I just know it) 16:23:40 dysinger_ [n=dysinger@32.177.36.147] has joined #scheme 16:23:51 Summermute: read this www.cs.indiana.edu/~dyb/pubs/fixing-letrec.pdf 16:24:01 Duh! I get it. Maybe I'll reterm tnl's "labels" to "fix", since excepting the module language, I'm generally seeking the most terse keywords :-) 16:24:15 you will probably understand it much better than I can 16:24:28 antoszka [n=antoszka@unaffiliated/antoszka] has joined #scheme 16:24:39 Ok, but I'm trying to follow a certain logic here first. 16:24:47 ok 16:24:49 Next is Scheme/Lisp's LET* 16:25:52 Riastradh: bork bork bork 16:26:33 That successively evaluates a rhs, binds it to a lhs, makes the lhs binding available to the next rhs and so on. Classically let* a = b; c = a * 2 is (let ((a b)) (let ((* a 2))) - if I got my parens right. 16:26:50 Just a series nested regular LET's, in other words 16:26:51 yes 16:27:28 the scope is important 16:28:03 xwl [n=user@123.115.124.13] has joined #scheme 16:28:07 sstrickl [n=sstrickl@129.10.110.45] has joined #scheme 16:28:12 so it's like putting in explicit { } in C 16:28:12 now we get to LETREC - is it just some bastard child of LABELS and LET* ???? I usually have a "machine model" of computations, but I don't for LETREC. 16:28:37 mejja, test case, please! 16:29:00 that';s what that paper deals with, deconstructing letrec into let and fix 16:29:19 and letrec* 16:29:24 Ah, wheel reinventing on my part 16:29:48 Summermute, there is an expansion for LETREC at the end of the R5RS. There is a simpler one at . The expansion for LETREC* is simpler still: (define-syntax letrec* (syntax-rules () ((LETREC* ((variable expression) ...) body0 body1+ ...) (LET ((variable ) ...) (SET! variable expression) (LET () body0 body1+ ...] 16:29:49 Riastradh: Do a clean build... 16:29:52 In IronScheme I still deal with it seperately :( 16:30:29 See, so far in "tnl" i just got rid of letrec. LABELS and LET (or named let) with LET always acting like LET*. Very, very simple and easy to follow what is going on. 16:31:35 i dont think let should always act like let* 16:31:54 would be better to express let* in terms of let 16:32:09 then you still have the best of both worlds 16:33:00 OK - I have more reading to do. end of r5rs and "Fixing Letrec". I'll give it 30 minutes, and then I'll otherwise go back to punting on a letrec style construct in my language and stick with "FIX" and "LET" :-) 16:33:19 Excuse me: I missed an ellipsis in my definition of LETREC* above. 16:33:27 I leave it as an exercise to the reader to figure out where. 16:33:28 Summermute: reading the base of R6RS is also very informative 16:34:01 Riastradh: I can find it online. 16:34:16 ? 16:34:27 R5RS 16:34:54 That's where there's an expansion of letrec toward the end, yes? 16:35:01 -!- drewfer [n=drewfer@67.58.77.138] has quit [Read error: 104 (Connection reset by peer)] 16:35:11 Yes. 16:35:21 drewfer [n=drewfer@67.58.77.138] has joined #scheme 16:36:48 Riastradh: IIRC, you're banging on MIT Scheme. If this recollection is right, what is the MIT Scheme community like (not personally, but in their uses/interest in MIT Scheme/professional background if relevant/etc.) ??? 16:39:02 -!- rudybot_ is now known as rudybot 16:41:40 -!- dysinger [n=dysinger@166.187.193.173] has quit [Connection timed out] 16:41:40 -!- dysinger_ is now known as dysinger 16:42:56 ski_ [n=md9slj@remote1.student.chalmers.se] has joined #scheme 16:47:35 -!- samth_away is now known as samth 16:48:10 langmartin [n=user@exeuntcha.tva.gov] has joined #scheme 16:51:29 sv21 [n=sv21@94.41.37.214.dynamic.ufanet.ru] has joined #scheme 16:53:19 -!- pavelludiq [n=quassel@83.222.175.184] has quit [Read error: 104 (Connection reset by peer)] 16:53:56 -!- saccade_ [n=saccade@209-6-54-113.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [Read error: 110 (Connection timed out)] 16:53:56 leppie: I note in the IronScheme screen shots that there are a ton of calls to clr-this and clr-that. Might there be some way to have a file/file(s) of declarations in a library/module instead and call these more like normal scheme functions, even if importing names with (prefix "clr/" ...) style???? 16:55:10 sure, you can just make a library, i just like showing them so people can see how easy it is to use the CLR 16:55:45 -!- rstandy [n=rastandy@net-93-144-180-123.t2.dsl.vodafone.it] has quit [Connection timed out] 16:55:49 with R6RS you can just rename them on import 16:56:30 Yes,but I'd rather me or someone get the job done in bulk :-) 16:56:46 I use the VS intellisense, so it's autotyped for me 16:56:51 Adamant [n=Adamant@unaffiliated/adamant] has joined #scheme 16:58:02 I have had ideas to reflect certain dll's or classes/types, but it would just end up making a huge mess of stuff you probably do not need 16:58:30 in terms of defined and exported variables 16:59:00 i dont like soilage from unused procs 16:59:26 -!- sv21 [n=sv21@94.41.37.214.dynamic.ufanet.ru] has quit [Client Quit] 16:59:40 not sure if understand what I mean 16:59:54 Those are very good points. 16:59:55 if you understand 17:00:49 But there's a balance somewhere. So,for example if this were C and I wantedto call fopen() and there was a cooler CLR way to do the same thing, I think I would want fopen() instead. See what I mean? 17:00:53 the clr-XXX is direct compiler syntax, iow it compiles to that exact method call with no extra frills, just like you would in any other .NET main stream language 17:01:13 Yet, that is a very good thing! 17:01:29 svb [n=svb@94.41.37.214.dynamic.ufanet.ru] has joined #scheme 17:01:45 Can those calls be used "higher order", like in (map) or (fold) for example? 17:02:02 nope, you would have to wrap it 17:02:50 or convert them to an IronSCheme callable object from some .NET Delegate 17:02:57 same thing really 17:04:39 the best scenario for higher order procedures on .NET is the approach F# uses, it's typesafe, and pretty fast 17:05:29 One idea that might be useful I've had to devise for tnl is "auto-lambda". tNL supports fancy lambda lists like common lisp - def (List 'a).filter(test:('a -> Bool), start = 0, end = -1, #test-not False):(List 'a). Get the rough idea - optional, keywords, rest args (not shown) 17:06:32 to me that's just all syntax sugar :) 17:06:44 So there's a certain compatibility between calls fn incr(x:Int, value = 1):Int and say fn negate(x:Int):Int 17:08:22 i still have this whole type inference thing to figure out for my typed scheme 17:08:51 The problem is, the compiler optimizes each combination of lambda list options when called "straight" - not "higher order" So here's the punchline - I have to have the compiler auto-wrap some function calls in a fn with a different lambda list in order to be call-compatible with what the higher order function expects. Make sense? 17:09:50 in scheme we just use apply :p 17:10:29 but yeah, i understand 17:10:56 jeapostrophe [n=jay@lallab.cs.byu.edu] has joined #scheme 17:11:03 that's similar to what I mean by compile time generative procedures 17:11:40 Well, I *COULD* have a single calling convention,and have implemented same years back. But let's look at keyword arguments - these can be reduced to a simple bitmask argument (in my case allowing for 30 bits thus 30 keywords). This is *STUPIDLY FAST* compared to scanning some Lisp &rest list for keyword value pairs, yes? 17:11:58 But it changes the calling convention for functions "featuring" keyword arguments. 17:12:45 Same goes for functions requiring multi-value return. I can make dedicated function entry points for optional arguments (I think many lisps have done this in the past). 17:13:58 But is all falls apart on simple stuff like (map string->list .....) because map doesn't know what function call convention it's getting in advance unless I wrap up string->list in the "calling convention package" that (map) expects. 17:14:35 that's why you should have something similar to APPLY, that is still fast 17:14:41 It's one of the basic language engineering principles: make shit butt fast, and make users pay for features they elect to use (higher order procedures in this case) 17:15:33 npe [n=npe@94-224-251-223.access.telenet.be] has joined #scheme 17:17:28 Well, I do plan to have a weird apply (tuples, structs, lists) because it's a manifestly typed langauge - but you can imagine there will be some ugly assembly code to make this "general case" work at all and then either (1) unwind itself from the call or (2) execute itself as a tail call. 17:17:32 -!- npe [n=npe@94-224-251-223.access.telenet.be] has quit [Client Quit] 17:17:54 npe [n=npe@94-224-251-223.access.telenet.be] has joined #scheme 17:19:21 I believe SPJ of Haskell fame has a paper on the options for dealing with higher order functions (I think currying is his issue in the paper) that has applicability to these same sorts of issues. 17:21:02 IIRC F# handle currying quite elegantly 17:21:46 BTW - can Iron Scheme run on the freebie Studio 2008 like F# can so I can avoid upgrading my 2005 to 2010 (that's the latest version,yes - do you know the cost here in the states?) 17:22:17 im still in doubt if I should go that extreme way 17:22:52 nope it cant, sorry you need at least standard version (I got MSDN subscription from work) 17:23:16 I would love for it to work in the express editions, but they forbid that 17:23:20 Among other luminaries now all in the MS camp, with SPJ doing research for MS as well, F# will soon become the "smartest" functional language out there, like it or not, I fear. 17:23:43 you need to use the whole 'shell' framework, and that's a lot of glue work 17:24:22 I just hope MS will concentrate on make it faster, as tail calls are rather expensive on the CLR 17:24:39 7 times slower than a normal method call 17:24:40 What was the "special" 2008 studio that supports f# then? Was the totally F# specific, because it wasn't packaged like Express C++, etc. 17:25:15 yes, you can download and extend the whole VS2008 Shell to your liking, no fees 17:25:46 it's a lot of work... 17:26:10 probably more code than I have written in my entire life 17:26:45 Tail calls are a bitch. If one heap allocates arguments (SML), then it's like one register assignment to the new args. If one stacks up one's arguments, there are a number of problems. 17:27:05 download the VS2008 SDK samples, etc, it's insane the amount of code that is needed for the simplest things 17:28:19 that's really whats holding me back from doing any more VS integration work 17:29:15 I am yet to find a simple way to intercept keypresses for IronScheme's editor window, so I can do some nice stuff 17:29:35 Yea, I have a nice x60s lenovo laptop, but the HD is "out of date" with "only" 80Mb. I'm planning to get a new netbook, and they tend to have 160Mb hard disks! In other words, one thing that holds me back (I do have masm and the express c++) from installing full VS2010 is just the raw disk space!!! 17:30:07 yeah, it's pretty big and slow monster 17:30:17 Summermute: You have what? 17:30:28 ? machine? 17:30:39 no, VS2010 :) 17:31:15 at least beta 2 is much better than the previous one 17:31:55 I use the new SQL Schema and Data Compare tools in it at work 17:32:08 but else I wont really use it 17:32:16 no, I have boxed 2005 I used to have installed on another box, but I'm putting off latest and greatest just because I don't want to spare the damn HD space. (And I'm pretty sure I'm putting Ubuntu on my new netbook too :-) 17:34:11 Also, I'm in a mode now where I'm using simple tools to write tools. I really don't want to start drowning in "tool heaven" and piles of documentation, if you know what I mean. PLT, Emacs, Open Office - more or less my world these days, with a few other languages installed for exploration. 17:34:51 oh, and another thing, this whole free VS2008 shell is a whopping 384.4 MB in addition to my 1mb IronScheme package 17:35:12 at that point I say, screw that 17:37:49 Have you ever installed squeak or it's new slimmed down spinoff Pharo (really feature rich smallalk implementations from some of the pioneers still banging away on it last I knew). If not, RUN, don't walk, download and install - takes less than 3 minutes, really. Relatively *tiny* installs with embarrassing amounts of functionality. 17:38:20 So yeah, doing this little exercise will make you say "screw that" even louder :-) 17:38:21 i like that 17:39:49 We're talking experimental morphic GUI (arbitrary shapes, etc.), code and doc browsers and on and on and on - and next to no disk hit. Really impressive stuff all around. 17:44:09 I know you are doing the Lord's work "the CLR way," but a *GREAT* book is "Smalltalk: Bits of History, Words of Advice" All 100% language implementation with a nice bit of history thrown in to illustrate their points, when and why they made decisions, etc. Fantastic. I've probably read most of it at least twice. But one point is that at least back in the day, they had a *really* tight byte code design to shrink code size. Still may one of the tricks u 17:46:10 george_ [n=george@189.107.176.74] has joined #scheme 17:47:37 Summermute: Hum... are you sure about the book? Wasn't it the blue book that had the implementation? 17:48:58 There is a pdf version of "Smalltalk 80" (the book) that acm had given away a while back 17:49:00 Yes, I don't think it's been blue for a while, but I believe that's Adele Goldberg's "Smalltalk and its Implementation" or something like that. I have that one too. Fantastic stuff. 17:49:12 no kidding! 17:49:43 -!- george [n=george@189.107.176.74] has quit [Read error: 145 (Connection timed out)] 17:50:28 That's probably the Goldberg book - documents the vm, byte code, std library and even has a sample chapter on running simulations. In some parallel universe, we're all happily coding in SmallTalk :-) 17:51:31 "Bits of History, Words of Advice" was a separate book of papers. 17:52:19 BTW, while I'm reading about LETREC, has anyone made a relatively portable version of the Gabriel Benchmarks for Scheme? Or anything else more up to date that's might already be published on the Web? 17:52:27 Not me. I can't code in that, and I tried many times. When the code base reaches some significant size, I can't find my way around the system. 17:52:49 Newspeak seems to solve some problems with a decent module system. 17:53:54 Summermute: http://www.ccs.neu.edu/home/will/Larceny/benchmarks.html 17:54:21 Yeah, since long ago when i last touched smalltalk seriously, i noticed they went from a 3 pane code browser to a 4 pane code browser :-) As they say, "In comp sci, there is no problem that cannot be solved with another level of indirection." 17:55:11 george__ [n=george@189.107.176.74] has joined #scheme 17:56:00 But really, I think there's both coding style and types of programs that lend themselves to different languages and environments. I don't know if I would want to take one of dad's old 100's pages of COBOL program and try to cram it into my Smalltalk V/286 environment at the time. 17:56:11 -!- george_ [n=george@189.107.176.74] has quit [Read error: 60 (Operation timed out)] 17:57:39 But if one has got the "Smalltalk/Forth" style "lots of small routines" style, then Smalltalk is hard to beat for that kind of language (GC, latent typing, runtime type checking, lambdas, etc.) 17:59:47 mejja: I'm just guessing that the 'fake r6rs' set of benchmarks will give the most comprehensive overview for scheme implementations? 18:00:50 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #scheme 18:01:50 jonrafkind [n=jon@crystalis.cs.utah.edu] has joined #scheme 18:04:52 -!- mabes [n=mabes@66.236.74.194] has quit ["Leaving..."] 18:05:08 -!- george__ [n=george@189.107.176.74] has quit [Read error: 104 (Connection reset by peer)] 18:06:46 george__ [n=george@189.107.176.74] has joined #scheme 18:06:51 -!- pantsd_home [n=holden@174.21.219.195] has quit [Read error: 145 (Connection timed out)] 18:09:31 -!- foof [n=user@FLH1Ahu226.osk.mesh.ad.jp] has quit [Read error: 110 (Connection timed out)] 18:09:48 Larceny, Ikarus, Gambit, Bigloo (sometimes) - These seem to be the speediest - are they also well represented by users here on the #irc channel? 18:09:56 -!- borism [n=boris@213.35.235.143] has quit [Connection timed out] 18:10:43 whoops - #scheme channel? 18:11:14 i think bigloo and gambit has their own channels, for ikarus it's mailing list, dunno about larceny 18:11:29 Summermute: http://dynamo.iro.umontreal.ca/~gambit/wiki/index.php/Gambit_benchmarks 18:12:13 IIRC, Bigloo is it's own big system, but it's not 100% "standard scheme" 18:13:20 leppie: So #scheme isn't necessarily the forum of choice for the speedier implementations? 18:13:33 i think most schemes are like that 18:13:55 Summermute, that I can not say 18:16:48 -!- bombshelter13__ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has quit [Connection timed out] 18:17:17 ASau [n=user@83.69.227.32] has joined #scheme 18:20:01 Summermute: I'm an Ikarus (and, to a lesser extent, Ypsilon) user, FWIW 18:21:40 prince` [n=user@218.232.175.55] has joined #scheme 18:22:14 rotty_ If you mind, might you describe your Ikarus experience - emacs? debugging? libraries? extent of srfi support? ffi? pre-packaged ffi packages? regex or other handy built in types? Many thanks in advance. 18:23:49 Summermute: OK, here we go; devel environment first: 18:24:30 ok, I'm using Ikarus from inside Emacs, with Derick Eddington's enhanced scheme mode. Ikarus has a "debugging mode", which allows you to get a backtrace. 18:25:31 In my version of Derick's scheme emacs repo, you can jump to source locations from the stack trace 18:25:33 Is that "Quack" or another scheme mode? 18:25:36 theres another another user here that uses ikarus frequently, but not here now 18:25:47 (still need to push that) 18:25:51 weinholt 18:26:29 rotty_: ikarus pretty quiet lately hey 18:26:33 stack trace to code jump is pretty fancy 18:26:41 Summermute: https://code.launchpad.net/~derick-eddington/scheme-mode/derick-.emacs.d 18:27:05 leppie: yeah 18:27:18 How about the libraries, srfi's, ffi and general "richness" 18:27:23 ?? 18:27:38 Summermute: wrt. librararies: 18:27:42 Summermute: all the srfi's are there 18:27:55 impressive 18:28:18 First of all there's the R6RS SRFI collection: https://code.launchpad.net/~scheme-libraries-team/scheme-libraries/srfi 18:28:44 I'm a comprehension junkie (typed-scheme, hint, hint ....) 18:28:47 then there's various "personal" library collections 18:28:49 (sorry, a lot of them) 18:29:12 nowherman [n=pierre@lec67-4-82-235-57-28.fbx.proxad.net] has joined #scheme 18:29:12 -!- nowhereman [n=pierre@lec67-4-82-235-57-28.fbx.proxad.net] has quit [Read error: 54 (Connection reset by peer)] 18:29:18 by "personal" do you mean user contributed? 18:29:19 proq [n=user@unaffiliated/proqesi] has joined #scheme 18:29:30 Ikarus does have an FFI, which is not as fancy as say PLTs, but its something you can build upon 18:29:36 rstandy [n=rastandy@151.53.240.131] has joined #scheme 18:29:49 Summermute: yes 18:30:02 Summermute: generally cross platform general purpose libraries 18:30:14 but only by a single author in most cases 18:30:19 Is there some library/repository/install utillity? 18:30:25 there's ports of foof's matcher `irregex' and `fmt' 18:30:37 they are working on it, Summermute :) 18:30:49 of Riastradh's parscheme 18:30:57 not sure if IronScheme will every support that, but maybe some day 18:31:30 installer: yeah, I'm pretty close (for some value of close ;)) to a first release of dorodango 18:31:59 Summermute: the best place is to look on the Ikarus mailing list, gets the most R6RS traffic I think 18:32:04 Summermute: there's `sbank', which provides GTK+ bindings 18:32:18 and most stuff runs on quite a lot of implementations, but not all 18:32:20 18:32:25 yep 18:32:29 rotty_ :) 18:33:04 *rotty_* will be AFK for a few minutes 18:33:07 to be a bit plebian, on the ffi front, is there anything for the common databases, GUI (wxWindow or GTK2 seem popular candidates), tcp/ip, common net protocols http/smtp/pop3/ssh, etc. ??? 18:33:36 what impress me is the size of these libraries, they are all quite big/comprehensive 18:33:40 -!- ASau [n=user@83.69.227.32] has quit [Remote closed the connection] 18:33:51 ASau [n=user@83.69.227.32] has joined #scheme 18:34:31 Summermute: weinholt has done some work on the SSL/encryption/compression front, all pure scheme : 18:35:52 -!- prince` [n=user@218.232.175.55] has quit [Remote closed the connection] 18:36:44 So, in general it sounds like up-and-coming but good momentum. Not to be impolite, but does some university/grant keep it going? 18:36:46 tcp is normally provide by the implementation 18:37:05 keep what going? 18:37:48 work - financing - organization - etc. - the filthy lucre :-) 18:38:12 these projects? i think most are just exploration in people's free time 18:38:36 Summermute: regarding GTK+: http://rotty.yi.org/software/sbank/ 18:38:47 that's true at least for my stuff 18:39:13 at least me, I get no check for scheme :( 18:39:15 So it's like apache (or used to be) or something - open source at its core? pure volunteerism on the part of the central organizers? 18:39:38 Arelius [n=Indy@208.87.217.194] has joined #scheme 18:39:50 there are no central organizers -- pure anarchism, as it's meant to be ;-) 18:40:08 The reason I ask is that so many schemes seem (I'm exagerating, totally) to be mere vehicles for generating academic papers ;-) 18:40:28 plt has a handful of unis behind it 18:40:44 perhaps scheme is just a good vehicle for expressing innovation 18:41:04 yeah, I can see that from what I've seen of the paper trail. 18:41:55 leppie: I totally agree 18:42:22 mrsolo [n=mrsolo@nat/yahoo/x-talyzlhkoxjraojw] has joined #scheme 18:42:22 -!- floodleboodle [n=nobody@Wb588.w.pppool.de] has quit [Remote closed the connection] 18:42:38 *mbishop* wishes someone would buy his used copy of PAIP on Amazon 18:43:01 OTOH, sometimes I stare at the screen and think that my programming environment (support, etc.) might disappear with a single grant denial or job shift or whatever. Kind of spooky. 18:44:05 -!- mmc [n=mima@esprx01x.nokia.com] has quit [Remote closed the connection] 18:44:49 Just one example from the cool LTU blog is Cat, a very cool typed functional concatenative language (like Forth), which, if I'm not mistaken, just shut down as the principal switched interests or jobs or something. 18:44:59 -!- kniu [n=kniu@ELMUNDO.RES.CMU.EDU] has quit ["Leaving"] 18:46:16 those are random chances, I would not let them bother me 18:46:28 just like that guy that murderer his wife 18:46:33 -!- Arelius [n=Indy@208.87.217.194] has quit [] 18:46:40 ada2358_ [n=ada2358@login-linux.ccs.neu.edu] has joined #scheme 18:47:01 It does point out a highly valuable aspect of standardization and portability 18:47:14 murder tends to do that 18:47:39 to be honest I could code just as well from a prison cell probably :) 18:47:49 And also the generation of a genuine, contributing community, including opening up the language core to qualified contributors 18:48:08 languages die due to lack of users 18:48:16 Just to exagerate, I don't think gcc or nasm are going away anytime soon :-) 18:48:21 since 10,000 BC 18:49:50 In a prison cell, bootstrapping a forth using the keypad on an old cell phone :-) 18:51:03 16K and 2.5Mhz :-) 18:51:26 or as punishment, WIndows 3.1 and Visual Basic 2.0 18:52:49 Well, at least one could uninstall most of it and use GWBasic (did that come with 3.1?) to poke and peek one's way to a real language. But I admit, I'd need and x86 manual :-) 18:53:20 and remember the lovely debug command :) 18:53:54 dysinger_ [n=dysinger@32.177.83.159] has joined #scheme 18:54:07 holyshit, that is still on WIndows 7 18:54:33 those were the days. funny thing is that ~$80US products quickly took over the development market anyway. Don't know how many folks were really using those DOS tools for very long, anyway. 18:54:45 Wow! 18:54:59 I broke a lot of stuff fiddling 18:55:16 like testing out low level formats of friends hard drives 18:55:36 Hopefully still back in the days when one could do a full reinstall from about 17 floppy disks :-) 18:56:21 Arelius [n=Indy@208.87.217.194] has joined #scheme 18:56:49 no, by lucky chance he had a backup of the something of the something, that could be fixed by some norton utility 18:57:22 ndformat? 18:57:27 Is Norton still alive or did they go the way of dBase and Borland (rest their soul)? 18:57:28 that rings a bell 18:57:36 i have no idea 18:57:36 -!- ada2358 [n=ada2358@login-linux.ccs.neu.edu] has quit [Connection timed out] 18:58:30 You know my primary file manager on Windows is Total Comamnder, which is a clone of Norton Commander 18:58:38 ndformat. Hmmm. All I can recall is the utility where you see the erased files but you have to type in the first letter of the file name to "undelete" it. Remember that? 18:58:45 -!- Arelius [n=Indy@208.87.217.194] has quit [Client Quit] 18:58:51 incubot: Norton commander! 18:58:55 i suspect this channel has a very small minority of windows users, much less norton 18:58:55 flippo [n=frivol@unaffiliated/flippo] has joined #scheme 18:58:56 Good for you! 18:59:08 Is midnight commander still alive on Linux? 18:59:12 lol mejja :p 19:00:00 Summermute: Norton is now that antivirus people 19:00:32 its hard to believe... 19:00:33 Folks have so many boxen these days, I wouldn't be surprised if most folks didn't have at least 2 of the big 3 - Linux,Mac,Windoze. 19:02:02 i have a laptop with a broken screen that I wanna put linux on 19:02:03 mejja, a clean x86-64 build? 19:02:57 yes 19:03:09 -!- ponzao___ [n=vesam@xdsl-83-150-86-25.nebulazone.fi] has left #scheme 19:03:34 I have a xp laptop, with external keyboard,mouse,screen and that's ssh to an Ubuntu box where Emacs is essentially the complete environment. Then another windoze box which basically hosts a 1.5 TB disk and does nothing but serve files and run bittorrent. And a couple of other boxes that haven't been turned on in a few years, including a big honking Fedora box full of disks. 19:04:25 Norton antivirus - that's gotta be #3 behind symantec and mcafee (sp?) 19:04:34 personally I prefer one big monolithic machine with some VM's 19:04:39 -!- dysinger [n=dysinger@32.177.36.147] has quit [Connection timed out] 19:04:40 -!- dysinger_ is now known as dysinger 19:05:04 i wanna let the laptop serve as a NAS and host external USB drives 19:05:36 florgleborgle [n=nobody@Wb588.w.pppool.de] has joined #scheme 19:05:49 Good idea these days - harder a few years back. The reason I don't crank up my Fedorah behemoth is just because all the damn fans make TOO MUCH NOISE. 19:05:50 perhaps serve as transparent proxy as well, and maybe even do my broadband routing if I get time to set that up 19:06:41 leppie: does it have firewire? nice for getting a little speed out of the external disks 19:07:03 -!- svb [n=svb@94.41.37.214.dynamic.ufanet.ru] has quit [Read error: 104 (Connection reset by peer)] 19:07:27 what's a "transparent proxy" 19:08:03 -!- florgleborgle [n=nobody@Wb588.w.pppool.de] has quit [Remote closed the connection] 19:08:17 florgleborgle [n=nobody@Wb588.w.pppool.de] has joined #scheme 19:08:31 it just a proxy you go thru regardless, based on the router 19:08:51 just for http 19:09:07 caches grafics etcs 19:09:37 Ah, for getting a little extra speed surfing? 19:09:50 we still pay for bandwidth in my country :( 19:10:23 That's a shame - on what part of the world do you park yourself? 19:10:30 south africa 19:10:45 Here there's a very nice climate down there. 19:10:56 actually we pay for usage, not bandwidth 19:11:50 wow. you mean just raw time? that sucks. sounds like good old state owned/sanctioned communications network. 19:13:20 US$35 / 5gb after US$60 line rental per month 19:13:33 I have 2 '5gb packages' 19:14:20 so US$130 per month for 10gb @ 4mbit 19:14:29 Ow. 19:14:40 you know how fast that 10gb can go? 19:14:43 10hours 19:14:45 Damn! Is TCP/IP radio still alive :-) Do you have cable net access? 19:14:50 one way traffic 19:16:10 Riastradh: Correction: LIARC 19:16:24 i see some places are starting to offer uncapped offpeak access, at like US175/per month (excludes line rental) 19:16:53 leppie: you are a truly dedicated Schemer :-) 19:17:19 perhaps after the 2010 World Cup Football it will get better 19:19:37 mejja, I see a problem with x86-64 too. 19:20:10 I'm looking at the r6rs spec, and I can generally get behind fx vs. flonum arithmetic, but there's a routine (real->flonum x) which I can't figure out what it does? Isn't a real already a flonum? 19:20:44 No, Summermute. 19:20:55 The set of objects representing real numbers includes the set of flonums. 19:21:30 could be any number that real? returns #t 19:21:46 Ah. So might a flonum be a machine double, but an implementation's real support some arbitrary precision floats? Something like that? 19:21:58 complex numbers 19:22:18 Oh, are complex nums considered reals? 19:22:19 integers 19:22:21 An integer is a real number, Summermute. 19:22:47 No, complex numbers are not real numbers. 19:22:50 -!- synx [i=synx@gateway/gpg-tor/key-0xA71B0C6A] has quit [Remote closed the connection] 19:22:57 OK, so I have a basic terminology problem I have to fix in my head. Thanks. 19:23:05 a complex number with an exact 0 as the imaginary part is a real 19:23:49 not sure how much that can possibly happen in practice 19:24:14 synx [i=synx@gateway/gpg-tor/key-0xA71B0C6A] has joined #scheme 19:24:34 That sounds logical. If there's no imaginary thing involved, it has to be real. :-) 19:24:47 but some scheme may chose to represent all numbers as complex numbers 19:24:57 lol mario-goulart :) 19:25:08 very true 19:25:14 mejja, hmm: (integer? .6931471805599453) ;Value: #t 19:25:38 wow 64-bit is awesome :] 19:25:40 All numbers in standard Scheme (for just about any meaning of `standard Scheme') are complex numbers. 19:26:48 (No document that any reasonable person describes as a `Scheme standard' of which I am aware includes, for example, quaternions or octonions.) 19:35:56 -!- MichaelRaskin [n=MichaelR@213.171.48.239] has quit [Read error: 110 (Connection timed out)] 19:36:14 nefreat [n=nefreat@neustargw.va.neustar.com] has joined #scheme 19:40:04 Neat, NASA is sure now that they've found the cavern on the moon that Tintin explored. 19:52:44 Riastradh: The object 35251593985042150653049327/9007199254740992, passed as the first argument to integer-remainder, is not the correct type. 19:53:04 Yes, I've just fixed that. 19:54:10 -!- gnomon [n=gnomon@CPE0022158a8221-CM000f9f776f96.cpe.net.cable.rogers.com] has quit [Read error: 110 (Connection timed out)] 20:05:08 -!- mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 20:06:22 -!- proq [n=user@unaffiliated/proqesi] has quit [Remote closed the connection] 20:06:31 proq [n=user@unaffiliated/proqesi] has joined #scheme 20:13:41 -!- sladegen [n=nemo@unaffiliated/sladegen] has quit [Nick collision from services.] 20:13:51 sladegen [n=nemo@unaffiliated/sladegen] has joined #scheme 20:14:50 mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has joined #scheme 20:16:16 -!- mrsolo [n=mrsolo@nat/yahoo/x-talyzlhkoxjraojw] has quit [Read error: 60 (Operation timed out)] 20:21:12 Arelius [n=Indy@208.87.217.194] has joined #scheme 20:25:44 -!- Arelius [n=Indy@208.87.217.194] has quit [Client Quit] 20:38:00 MichaelRaskin [n=MichaelR@195.91.224.225] has joined #scheme 20:40:07 -!- dysinger [n=dysinger@32.177.83.159] has quit [] 20:40:28 mrsolo [n=mrsolo@nat/yahoo/x-ouydnrpdpcssqarg] has joined #scheme 20:42:01 bgs100 [n=ian@unaffiliated/bgs100] has joined #scheme 20:44:27 -!- Nshag [i=user@Mix-Nantes-117-4-149.w193-250.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 20:46:07 dysinger [n=dysinger@32.177.18.186] has joined #scheme 20:54:00 -!- rstandy [n=rastandy@151.53.240.131] has quit [Read error: 110 (Connection timed out)] 21:03:12 -!- CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 21:06:26 CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has joined #scheme 21:12:38 -!- metasyntax [n=taylor@75-149-208-121-Illinois.hfc.comcastbusiness.net] has quit ["Nichts mehr."] 21:15:54 mrsolo_ [n=mrsolo@nat/yahoo/x-mqnobqzopoytkzih] has joined #scheme 21:19:47 -!- annodomini [n=lambda@wikipedia/lambda] has quit [] 21:22:21 -!- jimrees [n=jimrees@ita4fw1.itasoftware.com] has quit ["Ex-Chat"] 21:22:45 -!- mrsolo [n=mrsolo@nat/yahoo/x-ouydnrpdpcssqarg] has quit [Read error: 110 (Connection timed out)] 21:28:10 -!- jeapostrophe [n=jay@lallab.cs.byu.edu] has quit [] 21:35:08 schmir [n=schmir@p54A9049C.dip0.t-ipconnect.de] has joined #scheme 21:40:05 msw_ [n=msw@adsl-67-113-140-106.dsl.sntc01.pacbell.net] has joined #scheme 21:51:10 bohanlon [n=bohanlon@pool-173-48-104-141.bstnma.fios.verizon.net] has joined #scheme 21:51:28 -!- bohanlon is now known as bohanlon_ 21:56:13 -!- mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 21:58:16 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 22:04:57 -!- Edico [n=Edico@unaffiliated/edico] has quit ["Ex-Chat"] 22:05:55 -!- schmir [n=schmir@p54A9049C.dip0.t-ipconnect.de] has quit [Read error: 113 (No route to host)] 22:13:01 -!- langmartin [n=user@exeuntcha.tva.gov] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 22:18:25 -!- mario-goulart [n=user@67.205.85.241] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 22:21:35 -!- sstrickl [n=sstrickl@129.10.110.45] has quit [] 22:23:22 mejja [n=user@c-49b6e555.023-82-73746f38.cust.bredbandsbolaget.se] has joined #scheme 22:32:35 -!- nefreat [n=nefreat@neustargw.va.neustar.com] has quit ["leaving"] 22:34:08 foof [n=user@FLH1Ahu226.osk.mesh.ad.jp] has joined #scheme 22:36:44 karlw [n=user@adsl-99-157-202-134.dsl.pltn13.sbcglobal.net] has joined #scheme 22:40:15 -!- bohanlon_ is now known as bohanlon 22:43:52 mmc [n=mima@cs27122078.pp.htv.fi] has joined #scheme 22:45:10 -!- CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 22:46:20 -!- davids [n=davids@189.122.90.116] has quit ["Client exiting"] 22:47:17 -!- karlw [n=user@adsl-99-157-202-134.dsl.pltn13.sbcglobal.net] has left #scheme 22:48:44 -!- samth is now known as samth_away 22:53:00 Fare [n=Fare@204.8.153.37] has joined #scheme 22:53:00 annodomini [n=lambda@c-75-69-96-104.hsd1.nh.comcast.net] has joined #scheme 22:56:22 sepult [n=user@xdsl-87-78-30-154.netcologne.de] has joined #scheme 22:56:39 Arelius [n=Indy@208.87.217.194] has joined #scheme 23:06:59 -!- sepult [n=user@xdsl-87-78-30-154.netcologne.de] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 23:09:02 -!- jlongster [n=user@75-148-111-133-Chattanooga.hfc.comcastbusiness.net] has quit [Read error: 113 (No route to host)] 23:10:21 -!- Arelius [n=Indy@208.87.217.194] has quit [Remote closed the connection] 23:10:42 Arelius [n=Indy@208.87.217.194] has joined #scheme 23:18:26 -!- msw_ [n=msw@adsl-67-113-140-106.dsl.sntc01.pacbell.net] has quit [Read error: 110 (Connection timed out)] 23:24:46 CaptainMorgan [n=CaptainM@c-24-62-183-102.hsd1.ma.comcast.net] has joined #scheme 23:27:46 -!- clog_ [n=nef@bespin.org] has quit [Read error: 60 (Operation timed out)] 23:28:18 clog [n=nef@bespin.org] has joined #scheme 23:29:16 sstrickl [n=sstrickl@nomad.ccs.neu.edu] has joined #scheme 23:42:55 -!- dzhus [n=sphinx@95-25-177-162.broadband.corbina.ru] has quit [Remote closed the connection] 23:45:39 -!- Fare [n=Fare@204.8.153.37] has quit [Read error: 110 (Connection timed out)] 23:45:56 -!- copumpkin [n=pumpkin@c-24-63-67-154.hsd1.nh.comcast.net] has quit [Remote closed the connection] 23:46:43 copumpkin [n=pumpkin@c-24-63-67-154.hsd1.nh.comcast.net] has joined #scheme 23:51:14 msw_ [n=msw@75-101-102-192.dsl.static.sonic.net] has joined #scheme