00:00:25 bakkdoor_ [n=bakkdoor@81-163-26-99.visitor.congress.ccc.de] has joined #lisp 00:00:25 -!- bakkdoor [n=bakkdoor@81-163-26-99.visitor.congress.ccc.de] has quit [] 00:01:31 -!- bakkdoor_ [n=bakkdoor@81-163-26-99.visitor.congress.ccc.de] has quit [Client Quit] 00:01:31 bakkdoor__ [n=bakkdoor@81-163-26-99.visitor.congress.ccc.de] has joined #lisp 00:05:29 -!- netaust1n [n=austinsm@160.79.78.72] has quit [Connection timed out] 00:05:50 LiamH [n=nobody@pool-72-75-101-81.washdc.east.verizon.net] has joined #lisp 00:09:50 Is it possible to declare ignore in a loop portably? SBCL does the right thing when the variable is named "ignore" (loop for (key ignore) on *all-generated-tests* by #'cddr collect key) but CCL doesn't. 00:11:54 -!- xan [n=xan@62.57.40.18.dyn.user.ono.com] has quit [Read error: 110 (Connection timed out)] 00:12:04 LiamH: use NIL. 00:12:33 pkhuong, thanks 00:18:06 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 00:18:20 -!- Phoodus is now known as White_Flame 00:18:29 -!- White_Flame is now known as Phoodus 00:18:48 ... Why are you even destructuring for ignore in that case? 00:18:59 Hun [n=Hun@81.163.116.190] has joined #lisp 00:19:51 ahaas: my CS4 dvd came in the mail. i wasn't expecting this sort of shrinkwrap-hell. the bastard gave me a chore for the whole day. 8gigs unpacked and a trillion files. my 1gb of ram faces inflation and i might need a laptop mouse and a second monitor. WOW! 00:20:13 -!- avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has quit ["Leaving."] 00:20:16 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit ["Sto andando via"] 00:20:31 fusss: Eek. And I thought the GIMP was bad... 00:20:55 I thought CS was Counter-Strike. :-) 00:21:17 hfoo [n=h@p5B17DA3A.dip.t-dialin.net] has joined #lisp 00:21:32 I figured it for Creative Suite, but Counter-Strike is almost plausible... 00:21:45 fusss: Wow! We've always done the digital download. And for the last several version, we only do it on my partner's mac, which is hefty. 00:21:46 the GIMP? Gimpie is 40mbs of delicious goodness. the gimp can fit in one pixel in that OLED billboard of adobe products. 00:22:01 *nyef* finds eval-for-inspect in the SBCL source and despairs. 00:22:01 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 00:22:22 *tic* messes with (:emacs-rex ...), trying to figure out the flow of control. 00:22:53 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 00:23:21 ... That can't be right... It sets - to command -after- evaluating? 00:23:33 clhs - 00:23:34 http://www.lispworks.com/reference/HyperSpec/Body/a__.htm 00:24:46 -!- dash_ [n=dash@dslb-084-057-046-221.pools.arcor-ip.net] has quit [Read error: 60 (Operation timed out)] 00:25:08 Heh. And this is one of those pages where it refers to "/ (variable)", where / is a link to the function and variable is a link to the glossary entry. 00:26:31 -!- bakkdoor__ is now known as bakkdoor 00:29:11 dash_ [n=dash@dslb-084-057-027-128.pools.arcor-ip.net] has joined #lisp 00:31:45 Yeah, that's clearly broken. 00:33:12 netaust1n [n=austinsm@cpe-67-243-48-35.nyc.res.rr.com] has joined #lisp 00:34:19 ManateeLazyCat [n=user@222.212.132.169] has joined #lisp 00:34:19 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 00:35:57 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 00:36:03 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 00:36:11 -!- andrewy [n=irssi@cl-53.lax-01.us.sixxs.net] has quit [Operation timed out] 00:39:21 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 00:40:41 So.. Where do I find my dear friend SWANK:ARGLIST-FOR-ECHO-AREA? 00:40:48 -!- antoni [n=antoni@220.pool85-53-4.dynamic.orange.es] has quit [Client Quit] 00:41:33 rvirding [n=rvirding@h40n5c1o1034.bredband.skanova.com] has joined #lisp 00:42:00 -!- chris2 [n=chris@p5B16B2AA.dip0.t-ipconnect.de] has quit ["Leaving"] 00:42:28 karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has joined #lisp 00:42:59 I don't know, but that reminds me. 00:43:19 jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has joined #lisp 00:43:26 For some reason, one of my generic functions is showing up in slime with an arglist involving (... &key wait (wait nil)). 00:43:47 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 00:43:58 Now, I have to have done something to provoke just such an arglist, but... WTF? 00:44:53 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit [Client Quit] 00:45:14 Hrm... I'd better rethink what I'm trying to do before I end up with a tortuous LOOP or a TAGBODY... 00:46:34 -!- locklace [i=locklace@gateway/tor/x-edd5938d84384cae] has quit [Remote closed the connection] 00:46:49 -!- HET3 [n=diman@chello084114161225.3.15.vie.surfer.at] has quit ["This computer has gone to sleep"] 00:49:49 nyef: has semi-working (works when hot-patched in) support for conditional moves, by patching up IR2 (depends on previous work that splits testing and branching, which lets us call functions from :conditional VOPs) 00:51:41 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 00:53:14 locklace [i=locklace@gateway/tor/x-b67ae81a270bd5c6] has joined #lisp 00:53:45 Good Night! 00:53:48 pkhuong: Hunh. Neat. 00:53:49 -!- mrSpec [n=SomeOne@88.208.105.1] has quit ["c Ya!"] 00:54:11 Does it cross-build correctly as well, or only as a hot-patch? 00:54:47 nyef: haven't tested a full build yet. 00:55:07 Okay. Still neat, though. 00:55:47 I'm trying to get a better idea of the sort of patterns we might wanna have for IR2 rewrites. 00:56:37 *karlw* realizes how bad he is at political tactics. 00:57:17 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 104 (Connection reset by peer)] 00:57:43 I'm at the point of trying to figure out the structure of the main loop for a simple debugger. 00:59:33 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 00:59:33 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 01:00:03 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 01:02:00 JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has joined #lisp 01:02:33 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Client Quit] 01:03:11 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 01:03:35 Uh, does anyone know where I can find detailed criticism of Lisp? 01:03:45 -!- specbot [n=specbot@common-lisp.net] has quit [Excess Flood] 01:04:12 specbot [n=specbot@common-lisp.net] has joined #lisp 01:04:31 here, if you're very patient 01:04:42 -!- specbot [n=specbot@common-lisp.net] has quit [Excess Flood] 01:04:47 grr! 01:04:58 specbot [n=specbot@common-lisp.net] has joined #lisp 01:05:15 tomasf [n=tom@c-8dd9e555.024-204-6c6b7012.cust.bredbandsbolaget.se] has joined #lisp 01:05:47 -!- specbot [n=specbot@common-lisp.net] has quit [Excess Flood] 01:06:10 -!- rvirding [n=rvirding@h40n5c1o1034.bredband.skanova.com] has left #lisp 01:06:10 -!- Yuuhi [n=user@p5483FB09.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 01:06:12 Or is it true that my only obstacle is FUD? 01:06:20 tomasf: did you intentionally flood specbot off? 01:06:50 specbot [n=specbot@common-lisp.net] has joined #lisp 01:08:33 rottcodd [n=user@ppp59-167-42-106.lns2.cbr1.internode.on.net] has joined #lisp 01:09:14 disumu [n=disumu@p57A26D70.dip.t-dialin.net] has joined #lisp 01:09:18 Didn't gabriel write a critique, or am I misremembering the author? 01:10:40 -!- specbot [n=specbot@common-lisp.net] has quit [Excess Flood] 01:12:08 karlw: What would be your reason for wanting such detailed criticism? 01:13:08 The main issues seem to be that CL doesn't have a big enough standard library and can't interface to C. 01:13:12 FUD would only be a problem for you if you lack character. Its principal application is getting clients, employers, etc. to accept. 01:13:39 s/accept/reject/ 01:14:06 karlw: That doesn't answer my question. 01:14:07 character and/or an ability to make a mature judgement on your own 01:14:21 *inability 01:14:26 -!- tomasf [n=tom@c-8dd9e555.024-204-6c6b7012.cust.bredbandsbolaget.se] has left #lisp 01:14:36 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 01:14:46 karlw: I think the "can't interface to C" bit is definitely FUD. Where did you pick that up? 01:15:09 specbot [n=specbot@common-lisp.net] has joined #lisp 01:15:17 chandler: CS majors. 01:15:27 All of them?! 01:15:38 -!- specbot [n=specbot@common-lisp.net] has quit [Excess Flood] 01:15:38 karlw, nyef, http://groups.google.com/group/comp.lang.lisp/browse_thread/thread/18cc7a823112559c/e86335fb4cc6c367 01:15:48 Grr. 01:16:30 karlw: it would help if you told us why you want such detailed criticism. 01:17:04 so he can respond to people who make it? 01:17:49 that doesn't make a lot of sense 01:17:55 -!- wormphlegm [n=george@c-98-234-189-214.hsd1.ca.comcast.net] has quit [] 01:18:16 if he has "cs majors" giving him critique, he doesn't need us to provide more of it 01:18:59 <_3b> well, he apparently doesn't have good 'cs majors' :) 01:19:01 his locally available "cs majors" may not be the best authority on lisp. 01:19:12 indeed. 01:19:16 I like how we're all using scare quotes for his cs majors :) 01:19:56 -!- lisppaste [n=lisppast@common-lisp.net] has quit ["Want lisppaste in your channel? Email lisppaste-requests AT common-lisp.net."] 01:19:56 -!- minion [n=minion@common-lisp.net] has quit ["Common Lisp IRC library - http://common-lisp.net/project/cl-irc"] 01:20:10 Gah! 01:20:12 <_3b> well, they worry about interfacing to C instead of CS stuff, so they sound more like 'applied programming' or whatever majors than CS 01:20:39 CS majors aren't allowed to write useful software? :( 01:21:05 if anything, they should be complaining about hygienic macros, pure functionality, monads, orthogonality, etc 01:21:41 and lack of automatic memory management 01:21:51 -!- benny` [n=benny@i577A2105.versanet.de] has quit [Read error: 113 (No route to host)] 01:21:57 haha 01:22:03 single dispatch, single inheritance 01:22:13 lack of method combinations 01:22:27 lack of introspection and a MOP 01:22:34 I doubt you'll find 1% of CS majors who would complain about the lack of method combinations 01:22:47 -!- Dextor [n=NoorDext@unaffiliated/noordextor] has quit ["Go Canada!"] 01:22:54 jsnell: right, but they *should* complain about that. 01:23:02 I've never needed a method combination 01:23:13 I never expect my choice of language to depend on their presence 01:23:13 jsnell: really? 01:23:36 I use append and progn often enough that I would miss them if they didn't exist. 01:23:50 I use the standard method combination quite a lot. Doesn't that count? 01:23:51 really 01:23:57 Uh, I think the main problem people have is with syntax. 01:23:58 and by `method combinations' I also meant the standar one with :before, :after, :around. 01:24:25 <_3b> karlw: are you trying to use it or sell it? 01:24:27 karlw: the main problem people have is that they are too narrow minded. 01:24:41 karlw: but you still haven't answered my question. 01:24:54 Nowadays, I bet a worse problem is not having a portable access to Java, more than C. 01:24:59 karlw, anyway. his post makes pretty standard criticisms. Clojure addresses at least the libraries issues, fwiw. 01:25:05 S11001001: aroundp 01:25:52 lisppaste [n=lisppast@common-lisp.net] has joined #lisp 01:25:54 We're pretty much shackled to ACL because of need for the jlinker (jfli seems to be LispWorks only). 01:26:43 Every lisp has a way to get at C. 01:26:51 benny [n=benny@i577A2105.versanet.de] has joined #lisp 01:26:51 rpg: o rly? 01:27:00 rpg: "we"? 01:27:05 my company. 01:27:12 rpg: don't think ABCL does :) 01:27:16 rpg: Please show me the C interface to ABCL. 01:27:17 (directly) 01:27:25 (not belonging to me, but the one I belong to). 01:27:26 minion [n=minion@common-lisp.net] has joined #lisp 01:27:32 specbot [n=specbot@common-lisp.net] has joined #lisp 01:27:32 OK, every cl but ABCL. 01:27:54 JNI 01:27:59 ABCL would solve your Java interface problem, though! 01:28:02 beach: I don't really understand why networking projects I'm involved in always use other scripting languages. 01:28:02 Anyway, I think difficulties of getting at Java are much more of an issue than getting at C... 01:28:19 Phoodus: JNI is incredibly primitive. 01:28:20 Phoodus: I think that counts even less than running a machine emulator to run the C :-) 01:28:34 karlw: and so having detailed criticisms of Lisp would improve the situation? 01:28:34 Phoodus: Oh, sorry. Misunderstood. 01:28:42 karlw: Perhaps because Lisp isn't a "scripting" language? 01:29:13 *rpg* always mentions the java interface issue, hoping someone will say "haven't you heard about this cool new CL-Java interface....?" 01:29:34 rpg: possibly not 01:29:43 rpg: didn't the Clojure guy write one for CL? 01:29:57 Well, what distinguishes scripting languages from other languages? 01:29:59 luis: maybe this is mhy moment.... 01:30:14 you'd want to get to Java if you wanted java libraries which you did not wish to reproduce and for which no analogue existed in CL 01:30:29 or to integrate into an existing java thing 01:30:31 karlw: they are slow, and usually designed by someone who doesn't know anything about language design. 01:30:39 luis: that's jfli, already rejected by rpg 01:30:40 whereas you'd want C for most system-level stuff 01:30:44 ABCL java api is based off jLinker 01:30:53 jsnell: I think it's http://foil.sourceforge.net/ 01:31:03 beach: hard to draw the line, though 01:31:07 karlw: But I take it you either cannot or do not want to tell us why you think having detailed criticism of Lisp would help you in any way. 01:31:09 what about Dylan, for instance? 01:31:10 karlw: In my observation, a scripting language is a language which favors syntactic brevity and kitchen-sink semantics, and usually includes a large number of incomplete and probably incorrect libraries. 01:31:34 They're also usually single-implementation, and that implementation is usually slow and poorly written. 01:31:43 kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has joined #lisp 01:31:46 jsnell: jfli seems to be lisp-works only. Hoping that someday I'll have enough downtime to port it. 01:32:08 isnty JFLI based on ussing jLinker? 01:32:09 sounds like the sort of thing that might be rather difficult to port 01:32:13 rpg: what about foil? 01:32:25 dos Allegro have good java bindings, then? 01:32:30 luis: hadn't heard of it before. Looking at it now. 01:32:37 rsynnott: Yes, quite nice really. 01:32:54 rpg: apparently not ported to non-Lispworks either, nhe. 01:32:56 rsynnott: I'm hoping not --- seems like there's some low-level plumbing only. 01:33:04 drat. 01:33:20 karlw: To put a fine point on "poorly written": Perl requires users to break circular references so that objects can be reclaimed by its reference counter. 01:33:38 Python originally had the same problem, but someone (other than GvR) bolted a mark & sweep collector on to the reference counter 01:33:39 grrr..... There seem to be a boatload of dead pointers to the Common Public License 1.0 on the web... 01:33:41 rajesh [n=rajesh@nylug/member/rajesh] has joined #lisp 01:33:46 chandler: wow! 01:33:52 Allegro used jLinker 01:34:06 rpg: it's a fairly liberal one 01:34:18 karlw: i do use common lisp for networking projects and i also use it to interface with C a lot 01:34:27 Python's scoping rules are so poorly designed that captured lexical variables can't be assigned to, because the assignment statement is interpreted as binding. 01:34:37 don't see what your issue is 01:34:54 rsynnott: I think I've read it before. but this is the third broken link today. I wish the hosting company had thought about what it was doing before it killed the location from which it was serving the license... 01:35:08 jLinker was reimplemented for ABCL since it was well documented enough spec for calling java.. that ABCL could adopt it.. some everyone doing java from lisp may as well adopt jLinkler 01:35:34 foil claims portability; seems appealing... 01:35:39 Ruby's implementation is improving to some degree, and will eventually feature a bytecode compiler and real Unicode support. For now, it is a slow interpreter with lousy support for full Unicode. 01:35:48 dmiles_afk: That would be fine with me! 01:35:57 rpg: http://opensource.org/licenses/cpl1.0.php 01:36:00 When last I checked, it also still captured the *entire* environment in closures, which prevents garbage collection of otherwise dead objects. 01:36:17 (yep, the fools changed the location and didn't forward it... 01:36:23 chandler: That's pretty devestating stuff. 01:36:38 rpg: seems to work over sockets, is that an issue? 01:36:44 I wonder how much work it would take to standardize language x (where x is Perl, Python, etc.) 01:37:08 luis: Primary issue for me would be ACL and SBCL, and I've already papered over the socket differences once for those two. 01:37:09 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 01:37:09 karlw: enormous amounts 01:37:10 luis: t 01:37:15 luis: thank you very much for the pointer! 01:37:16 How much effort would it take to convince the maintainers that a standardization process provided benefits that outweighed the ability to change the language rapidly by fiat? 01:37:47 (badly) broken unicode is a surprisingly common problem 01:38:03 rsynnott: Looks like they didn't consider the possibility of having multiple revs of the license... 01:38:12 chandler: it has to close over everything, there's no way to statically determine which variables are closed over 01:38:17 drwhen [n=d@209-112-181-104.static.acsalaska.net] has joined #lisp 01:38:19 On the other hand, there are some things that are not all that great for doing "scripting" work in Common Lisp. Subprocess control can be surprisingly frustrating, for example. 01:38:22 S11001001: I inserted page breaks before each CLHS-style node in . Does that look like an improvement to you? 01:38:49 could probably be fixed with a library for the purpose 01:38:51 jsnell: Why is that? (I don't actually know Ruby.) 01:38:56 (actually, isn't there one?) 01:39:16 chandler: I just did some subprocess control in perl (to start multiple cl jobs and get all the right sockets hooked up), and it was surprisingly painful. 01:39:42 Apparently, GVR didn't include good functional programming support because he thought it would make the syntax ugly. 01:39:59 luis: yes 01:40:21 Hey, someone's talking about subprocess control in lisp? 01:40:24 chandler: their version of "eval" can access the lexical bindings of the calling scope. so it's a fundamental language design problem, not just implementation sloppines 01:40:27 *nyef* perks up. 01:40:45 I need to do a fork/ptrace/exec thing. 01:41:05 How do I make it not suck? 01:41:05 nyef: Please look at Factor's interface for this. (I forget the name atm.) 01:41:10 and yes, it is quite amazing that the same trivial language design mistakes are done over and over again 01:41:13 nyef: they always suck :) 01:41:21 slava has done a very comprehensive job on the design. 01:41:22 echo-area [n=user@nat/yahoo/x-4f22e3f673adaaab] has joined #lisp 01:41:39 jsnell: How widely is that used? Couldn't Matz just declare by fiat that this no longer worked, in the name of not leaking objects everywhere? 01:42:04 probably a great risk of annoying users there 01:42:20 luis: incidentally, page 62 is labeled inc-pointer instead of incf-pointer 01:42:24 What are they going to do - learn Lisp / Scheme / Factor / etc? :-) 01:42:24 -!- ManateeLazyCat [n=user@222.212.132.169] has quit [Remote closed the connection] 01:42:43 I think people are finally becoming pissed off at the frequent massively incompatible changes to php, for instance, even though they make sense from a making the language less horrible point of view 01:43:02 rsynnott: no they won't, as long as it's horrible suckage 01:43:07 people are attracted to that I think 01:43:19 I think it's widely used, but I believe some implementations forbid aliasing the magical eval function into other names, thus making it statically detectable 01:43:20 Hey, I forgot to even mention PHP. I've forgotten it so well that I don't even know what parts of it to throw mud at. 01:44:09 jsnell: I seem to remember asking this once before, and the answer is that some of the ORM / CRUD skeleton systems use it due to the lack of macros. 01:44:50 dmiles pasted "pretty stack trace" at http://paste.lisp.org/display/72810 01:44:52 S11001001: ah, fixed now, thanks. 01:45:00 nyef: http://docs.factorcode.org/content/vocab-io.launcher.html 01:45:24 how about the exciting type punning in the comparison operators? 01:45:43 chandler: Thanks. 01:46:31 ManateeLazyCat [n=user@222.212.132.169] has joined #lisp 01:46:55 Speaking of type punning, I'm quite fond of Perl's duck typing. That's duck as in "get down or be killed by incoming fire", not "quack". 01:47:10 Heh. 01:47:10 <_3b> throw mud at php's big standard lib, which you can't update in pieces without updating the entire implementation :p 01:47:33 At a former job, our boss had implemented a TWiki plugin for tracking time. If you entered 0 in any field, it couldn't distinguish it from an empty field, so you couldn't say that a task was finished by entering 0 hours to do. 01:47:42 You had to switch to % done and enter 100. 01:48:14 That's... stunning. 01:48:53 win 01:49:36 -!- brianj_otter is now known as brianj_otterZZZ 01:53:18 Speaking of scripting languages, in my sotware-engineering course, I tell the students about a standard scenario where C++ is choses early on for speed, and then a scripting langauge is choses because of the lack of flexibility of C++, with a debugging nightmare and slow code written in the scripting language as a result. 01:53:44 beach, are you a professor? or TA? 01:53:57 I am a professor. 01:54:01 the software engineering class I took this past semester was.. special. 01:54:33 beach, what do you teach in your course? 01:55:18 jli: lots of stuff: standard SE activities and development models, chosing a programming language, interaction design, project management, etc. 01:55:48 *nyef* sometimes wishes he was better at the whole "user interface" thing. 01:56:27 Anyway, how many problems does the Python Standard Library have, practically speaking? 01:56:52 <_3b> karlw: it doesn't work with lisp? :) 01:56:58 How are you quantifying problems? 01:57:08 <_3b> (and isn't standardised) 01:57:24 nyef: you can start by reading some of the stuff by Jeff Raskin and Alan Cooper. 01:57:31 -!- anekos is now known as awayekos 01:57:40 karlw: I'd say around 37. 01:57:48 -!- disumu [n=disumu@p57A26D70.dip.t-dialin.net] has quit [Read error: 104 (Connection reset by peer)] 01:57:52 andrewy [n=irssi@cl-53.lax-01.us.sixxs.net] has joined #lisp 01:57:54 Ah, and that's the sarcastic answer I was tempted to give. 01:58:00 beach, I suppose the primary reason my class felt like a waste of time is because the professor seemed to either lecture directly out of a book, or make it up as he went. 01:58:18 karlw: I suggest you start making sense, start talking about Lisp, and start answering questions as opposed to just asking them. 01:58:18 I figured I'd start by messing around with a GUI toolkit with an end in mind. 01:58:18 I was even going to say "37", too. It's a nice prime number. 01:58:35 minion: chant 01:58:35 MORE CODE 01:58:38 chandler: heh! 01:58:54 minion: chant to karlw, please! 01:58:55 karlw: MORE CODE 01:59:43 nyef, Jef Raskin is awesome. Tog is neat too http://www.asktog.com/ 02:00:04 jli: It is actually hard to come across otherwise. The only alternative is to speak from experience, and few professors have any. I am lucky that I have worked in or for the SW industry in three different countries. 02:00:04 -!- benny is now known as bruenig 02:00:19 -!- bruenig is now known as benny 02:00:20 Oops. 02:00:41 Be wary of the interface designers who focus on efficiency of mouse movements and keystrokes instead of reducing discrete cognitive tasks necessary to use the software. 02:01:10 It's important, but not the sole focus of a user interface. If you focus on it entirely, you start to think that a text editor with incremental search is the best user interface innovation ever. 02:01:15 I managed to screw my simple debugger up on a couple fronts already. ^_^; 02:01:30 poor emacs! 02:01:37 I'll second the recommendation for Alan Cooper, though. 02:01:44 And for some reason, I can't seem to persuade SBCL to SIGTRAP. 02:01:49 after a number of attempts at programs that required pretty complex user interfaces, I'm messing around with rule-driven UIs 02:02:12 *karlw* shuts up and starts writing Mac OS X in CL. 02:02:18 "My project needs SPEEEEEEED!!!" 02:02:20 Phoodus: have you looked at the CLIM model? 02:02:59 Phoodus: CLIM presentations are a good way of managing the complexity of user interfaces. 02:03:00 I talked to a programmer once who only used C++ because he wasn't comfortable writing code that wasn't as fast as possible. 02:03:04 beach: no, I've used Tk, Java, wxWidgets, etc 02:03:07 karlw: please leave out the transparent dock; it is ugly 02:03:13 does it offer a significantly idfferent model? 02:03:24 completely, yes. 02:03:35 S11001001: did he also speak only esperanto? 02:04:06 S11001001: Did you try explaining to him about assembly language? 02:04:13 beach: howso? I'm getting a 404s from the cliki page 02:04:15 S11001001: That sounds like one of the pitiful excuses that some programmers have to avoid learning new stuff. 02:04:35 S11001001: But they hide it behind some seemingly objective goal. 02:04:38 -!- timchen1` is now known as nasloc__ 02:04:57 Phoodus: hold on... 02:05:28 yhara [n=yhara@ZC046088.ppp.dion.ne.jp] has joined #lisp 02:05:37 schme [n=marcus@c83-249-80-232.bredband.comhem.se] has joined #lisp 02:05:43 minion: tell Phoodus about clim 02:05:44 Phoodus: look at clim: The Common Lisp Interface Manager (CLIM) is a powerful Lisp-based programming interface that provides a layered set of portable facilities for constructing user interfaces. http://www.cliki.net/clim 02:05:53 nyef: no, too busy eye-rolling 02:06:01 yeah, I got that already :-P 02:06:14 Phoodus: that page works for me. 02:06:20 er, that URL. 02:06:58 yeah, I finally found a link off it that worked 02:07:08 bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has joined #lisp 02:07:12 cliki desparately needs a rating/freshness/something system 02:07:15 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Nick collision from services.] 02:07:16 schoppen1auer [n=css@unaffiliated/schoppenhauer] has joined #lisp 02:08:07 beach: I've finally decided that CLIM seems like a good system for managing complexity in user interfaces because CLIM (and it is not alone in this) adds much complexity to the user interface. 02:08:50 For systems with simple listener-style interaction, it is a good handholding framework (especially if you can use the listener itself for your interface!). But I'm not smart enough to deal with things like incremental redisplay. 02:09:16 *Phoodus* does NOT want event listener-style interfaces 02:09:26 hence, I'm trying out declarative styles 02:09:31 UI design is hard :( 02:09:36 Phoodus: that's not really what "listener" means in the context of CLIM. 02:09:43 rsynnott: "... lets go shopping?" 02:09:48 indeed 02:09:50 Phoodus: CLIM is event driven at the lowest level, but typically that level is not used. 02:09:52 Phoodus: "listener" here is a variation of a REPL for interaction with the user. 02:10:17 there are people who are good at it, and I know very well I'm not one of them so tend to avoid it 02:10:22 Phoodus: Instead it has a command loop, where commands can be invoked in various ways, including by clicking on a presentation. 02:10:32 Phoodus: An example of a "listener-style" interface would be a text adventure game, such as Zork or A Mind Forever Voyaging. 02:10:49 regardless of that, does CLIM help out with multiple overlapping states, interdependent control propogation, tec? 02:10:55 ideally, the textual command prompt is optional, and the notion of commands just exists as an organizing metaphor 02:11:12 In any event, I now think that rule-based declarative systems have a lot of promise for user interfaces, especially in terms of solving the very real problems in designing cross-platform sensible user interfaces. 02:11:34 -!- schoppen1auer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 02:11:35 chandler: well, the stratified design of CLIM lets you pick the level you work at. You don't *have* to use all the high-level features, and the only bad thing that happens to you if you don't is that you have to program the same way you would if you didn't have CLIM. 02:12:07 sz0 [n=sz@85.102.81.190] has joined #lisp 02:13:02 I never used a commercial CLIM, but with McCLIM, I was always disappointed that the multiple-argument command parsing didn't seem to work well. That seemed like a big distinction b/w CLIM and conventional UIs. 02:13:03 Phoodus: I take it you want us to say "no" so that you can brush off CLIM without the effort of reading anything about it, but I can't do that because I don't know what those terms mean. 02:13:31 rpg: what didn't work about it. I use it all the time and it works fine. 02:13:43 Any time I have a set of GUI controls with listeners, each is dependent on each other for display type, input activation, and all sorts of stuff 02:13:51 writing per-control stuff really sucks 02:14:08 think very complex graphical editors 02:14:12 not just database forms 02:14:17 Phoodus: I think you are putting too much interpretation into the word "listener". 02:14:23 beach: The low level stuff has a lot of complexity to it that I would need to carry around in my head to effectively use it. 02:14:40 I want to know how CLIM is different. There's a lot of docs to go through; what's the nutshell summary of why it's different? 02:14:43 beach: When I tried the last time I found that editing the multiple arguments (accepting-values?) didn't work properly. The editor would get all stuffed up. 02:14:46 and I am skimming the docs as I type 02:15:06 In some sense, it's a lot like Common Lisp itself, except that I don't do this often enough to actually carry it all around in my head - and the implementation situation is not as good as with Common Lisp itself. 02:15:11 -!- mejja [n=user@c-4db6e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 02:15:12 beach: that was a bit more than a year ago, so maybe it's fixed. 02:15:37 chandler: I can agree with that. 02:15:50 -!- dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 02:15:57 Phoodus: I think it is so different that giving a nutshell description of it here would be very hard. 02:16:05 well, try 02:16:11 I think if lisp is as great as we all surely think it is, we ought to be confident we can use its strange alien superpowers to stop us from getting mired in low level event driven crud, regardless of the underlying system 02:16:11 you said you give commands to a REPL 02:16:35 but that's not much different than any other system, unless the "output" back from the REPL does something 02:16:36 Phoodus: give it a go and see :) 02:16:36 (in my limited experience, it never seems worth the trouble) 02:16:45 rsynnott: that's not very helpful... 02:16:49 you can always step back gasping in horror if you don't like it :) 02:16:54 Phoodus: no, I said there is a command loop, where commands are executed in a nonempty context, which is different from event-driven GUIs. 02:16:58 -!- schme_ [n=marcus@c83-249-80-232.bredband.comhem.se] has quit [Read error: 110 (Connection timed out)] 02:16:59 seeing as all the docs claiming to give a simple overview 404 on me :-P 02:17:10 I recommend Phoodus avoid CLIM 02:17:28 I think that is better too. That avoids my having to give a summary. 02:17:32 why? 02:18:04 bakkdoor_ [n=bakkdoor@81.163.101.59] has joined #lisp 02:18:14 well, being recommended something without any basis for why except "Well, it's kind of differnet, can't explain it" tells me that people don't really understand it 02:18:14 Phoodus: because you seem to have the standard reaction of someone that doesn't *really* want to know about it, just find good reasons to go on with what you are already doing. So that's what I think you should do. 02:18:34 in terms of engineering, not API 02:18:36 Phoodus: yes, that must be it. 02:18:41 I always feel a little bad when someone new tries mcclim and is disappointed 02:18:52 Phoodus: So forget what I said. 02:18:59 beach: I already said I don't like what I'm doing, and want to change 02:19:23 beach: I'm not getting that from him. He seems curious but is somewhat stymied by the lack of links that explain the concept (other than the CLIM specification itself). 02:19:26 an inference-based approach has a long development curve 02:19:27 (or when someone less new tries it again after some extended period, and finds it hasn't really changed) 02:19:29 -!- bakkdoor [n=bakkdoor@81-163-26-99.visitor.congress.ccc.de] has quit [Read error: 110 (Connection timed out)] 02:19:35 beach: Diving in to CLIM is a lot of effort to find out what makes it different. 02:19:39 so anything already written would be faster, if it handles state better 02:19:44 -!- sz0 [n=sz@85.102.81.190] has left #lisp 02:19:52 -!- bakkdoor_ is now known as bakkdoor 02:20:14 -!- bakkdoor [n=bakkdoor@81.163.101.59] has quit [Client Quit] 02:20:28 chandler: I know, and just the way I no longer try to explain to someone how Lisp is different from what they currently use, I no longer feel like doing that with CLIM either. If that means we'll lose a user, then so be it. 02:21:03 Are you able to explain the difference in a few lines? 02:21:18 Phoodus: I don't think so no. 02:21:22 vs Can it be explained in a few lines 02:21:36 Phoodus: Are you familiar with the UNIX shell? 02:21:39 Phoodus: I don't think so, no. 02:21:42 chandler: yes 02:21:51 Phoodus: the idea of CLIM is this: you have an application frame, containing the application state, and commands which modify that state. there's a parallel type system in which you define the types of data your program uses, and commands except these as arguments. these types can be associated with graphical output, known as "presentations". 02:22:06 bakkdoor [n=bakkdoor@81.163.101.59] has joined #lisp 02:22:13 When you interact with the shell, you are supplying "commands" to it, which are then evaluated, and the results are "presented" to you by means of text output. 02:22:44 This is a very simplistic interface, though - the presentations themselves are just the names of the results, and do not link back to the objects themselves. 02:22:54 and it automatically handles propagation of all data? 02:23:23 depends what you mean, but probably no. 02:23:26 (but thank you for at least describing the architecture) 02:23:41 Imagine if you typed "ls", and instead of receiving the printed names of files and directories what was returned was the actual objects themselves - still presented to you as text or graphics, but linked back to the objects themselves. 02:24:04 so how about this 02:24:06 You could still run the output through a filter, and the filter would work on the objects themselves and not the printed representation. 02:24:07 like `slime-presentations' 02:24:19 I have an object somewhere on the display, with a name. I have a reference to it elsewhere, displaying the name 02:24:27 adeht: `slime-presentations' is derived from the CLIM concept. 02:24:36 dialtone [n=dialtone@adsl-75-17-56-28.dsl.pltn13.sbcglobal.net] has joined #lisp 02:24:37 I issue a command to change the name, do both display objects update automatically? 02:24:40 chander: yes, I thought so 02:24:53 -!- Dawgmatix [n=deep@207-237-30-94.c3-0.avec-ubr11.nyr-avec.ny.cable.rcn.com] has quit ["Ex-Chat"] 02:24:55 and (perhaps more in theory than practice) non-textual gestures can be translated to presentations. CLIM has something called a "drag and drop translator", for instance, which produces a command based on you dragging one object onto another object 02:25:10 Phoodus: "maybe". This gets into the incremental redisplay complexity I was complaining about at the start of this discussion. 02:25:21 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 02:25:54 -!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 02:26:03 And with that, I'm going to flee. hefner can probably handle this better than I can anyway - though I don't think he's really much more enthusiastic about CLIM than I am at this point. 02:26:18 also, are things like flaoting palettes displaying various properties of the current selection easy to do? 02:26:20 Phoodus: there's no data model there to do that. you can associate tags with sections of output, and invalidate them, which is incremental redisplay helps you do. 02:26:21 Phoodus: you have to tell the system how to display your model, of course. But once you have done that, you can arrange for the display function to be executed after each iteration of the command loop, which will have exactly the effect you were referring to. 02:26:39 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 02:26:53 beach: so it's not a push model? 02:27:10 Not intrinsically, though you can implement that without too much duct tape. 02:27:13 I don't know what a "push model" is. 02:27:41 beach: changes are propogated as they happen (pushed through) instead of results requeried later in their entirety 02:28:04 the latter being a "pull model" 02:28:30 funny, I'd interpret those names as doing the opposite. 02:28:36 doing, meaning 02:28:46 mqt_ [i=tran@monaco.nirv.net] has joined #lisp 02:28:46 What kind of changes? Where are they propagated? What does it mean for a change to "happen"? 02:28:46 They don't have to be requeried later in their entirety. CLIM includes most (or at least some) of the solution to this problem. 02:29:01 -!- mqt [i=tran@monaco.nirv.net] has quit [Read error: 113 (No route to host)] 02:29:08 beach: data changes, and when the data changes, those things which display the related data are given the new data to reprocess 02:29:22 Phoodus: how does it know that it has changed? 02:29:33 it starts at the UI 02:29:34 (I think of the display as pulling on my nonexistent data model) 02:29:38 Cells, man! 02:29:38 -!- m4dnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 02:29:49 m4dnificent [n=user@83.101.62.132] has joined #lisp 02:29:52 Phoodus: You can fairly easily structure your command loop so that all active presentations are redisplayed after the command has been processed. 02:29:55 Phoodus: as I explained, you have to write the code that takes the model and displays it. Once you do that, you can arrange for the changes to be displayed automatically. 02:29:57 -!- madnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 02:30:04 chandler: the UI code carefully watches all memory IO :) 02:30:06 one typical CLIM approach is to redisplay everything whenever anything could possibly change 02:30:07 madnificent [n=user@83.101.62.132] has joined #lisp 02:30:18 ..then complain on IRC that your application is rather slow 02:30:28 yeah, that doesn't sound too bright, when you've got a ton of stuff onscreen 02:31:01 Phoodus: CLIM is good though, because that simple strategy works very well for all those applications that have only a few pages to display. 02:31:10 Eclipse's EMF does a lot of what I want, but it's incredibly heavyweight, imposes its own pseudo-Java paradigms with lots of annotations, and isnt' very flexible 02:31:14 Phoodus: I can imagine solutions to this problem, but they all seem to fall into the space of actors-model languages with observers for changed data. 02:31:49 chandler: it's a solved problem ever since the advent of the spreadsheet 02:32:07 I do pine for something like this, but religious convictions prevent me from using cells, and I haven't rolled a compelling alternative yet 02:32:09 except nobody seems to actually implement it anywhere else :-P 02:32:25 hefner: have you looked at computed-class? 02:32:27 phoodus: I would think that CLIM doesn't give you that, but the means to implement that 02:32:28 Phoodus: Most of those are rather degenerate implementations of actors, really. 02:32:37 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 02:32:52 chandler: I know. but you don't need a full Actor, and it's been done properly for decades in one field, that's all I'm saying 02:32:59 luis: no, and I probably should 02:33:00 adeht: doesn't give him what? 02:33:03 hefner, what's wrong with cells? 02:33:22 jli: it's author for one thing. 02:33:37 then again, I don't know much about CLIM, though I've been meaning to snoop around in some future time 02:33:41 ah, that's not a good reason not to use it, though 02:33:48 beach: dependency management a la cells 02:33:50 anyway, he's entertaining 02:34:06 beach, I guess we'll go depth-first. what's wrong with Kenny? 02:34:23 jli: I would rather not go there. 02:34:58 adeht: ah, OK. Though I would imagine most spread sheet programs using the simple strategy of recomputing everything in each iteration. 02:35:11 I once rolled my own equivalent to computed-class, long ago, to simulate mzscheme's FrTime language. I should dig that up again. 02:35:15 beach: I doubt it 02:35:24 too expensive in some cases 02:35:30 beach: I don't think even the very first spreadsheets back in the 80s did that 02:35:33 the bit I haven't gotten my head around yet is handling both discrete and continuous inputs, and combining different evaluation strategies and push/pull in different places 02:35:36 does anyone know where I can find detailed criticism of Kenny? 02:35:38 some sort of dependency graph? 02:35:53 jli: responses to his usenet posts 02:35:56 In any event, redisplay-everything in CLIM would be great if I didn't need a Ph.D. in applied CLIMology (with a specialization in updating-output) to use incremental redisplay. 02:35:57 Phoodus: back then they probably couldn't. Now I would think it would be quite fast. 02:36:12 is Kenny a debian version? 02:36:29 This is where I quickly come to the conclusion that CLIM is just too complex - and worse, it is not an easily learnable complexity, but one which must be swallowed whole for anything but the simplest of applications. 02:36:45 jli: heh, criticism of Kenny.. reminds me of that "destructuring Kenny" thread 02:36:52 baffle them with bullshit, that's the clim motto 02:37:01 fusss: No. Didn't you ever see Toy Story? There's no Kenny there. 02:37:17 anyway, if I'm going to manage my own onscreen presences, is Tk pretty much the best solution for a "retained mode" GUI with lots of canvas objects? 02:37:59 Phoodus: have you not heard of lambda-gtk or should i smack you on the head with screenshots from three platforms for the same source? 02:38:01 -!- dkcl [n=danderse@unaffiliated/dkcl] has quit [Remote closed the connection] 02:38:03 or just use the basics of CLIM? 02:38:23 fusss: I haven't looked into much Lisp gui stuff, I'm wanting to know what sorts of things to look at first 02:38:28 though I did do a lot of ltk 02:38:54 Phoodus: talking to your gui through a pipe or a socket is crap 02:39:05 Hence, don't use X11. 02:39:23 lambda-gtk is so faithfully gtk, you can get by with the C api docs. and so lisp it feels righteous. 02:39:30 fusss: why is that? 02:39:33 fusss: I pushed a lot of behavior through as Tcl functionality, and it was pretty zippy 02:39:38 Phoodus: Unless you are going to wrench your head through the big flip into CLIM, I think you're probably happier to use tk (or possibly gtk0. 02:39:50 chandler: maybe the rule is really that talking to your gui through more than one pipe or socket is crap 02:39:53 Personally, what I do these days is write a web server and then script the browser for my UI. 02:40:05 IPC is broken on win32, specially for an intrisically unixy app like wish.exe 02:40:10 rpg: that sounds incredibly painful 02:40:21 rpg: depressingly, that seems to have become the easiest option 02:40:23 for form- or table-based editors, sure 02:40:27 hefner: actually, it's not bad at all. 02:40:28 (everywhere, not just for lisp stuff) 02:40:31 hefner: I'm not really in the business of writing rules, but that sounds just as fishy to me. 02:40:40 ... unless you never use SSH X forwarding :-) 02:40:44 anyway, now I really will flee. 02:40:56 -!- binarycodes [n=sujoy@59.93.192.19] has quit [Read error: 104 (Connection reset by peer)] 02:40:56 I wouldn't want to implement something like a vector animation editor over the web :-P 02:41:03 hefner: the lisp UI tools all seem to pretty much suck, and javascript isn't so painful. 02:41:37 "Please do not use this tool if you are more than 500km from the webserver, as the speed of light will mess with the framerate" :) 02:41:45 note that "pretty much suck" for me may simply be "you have to know how to use this from C or C++ and then you can use the Lisp wrapper." 02:42:20 Like the way all subversion UIs suck because they are only useful if you know how subversion works from the command-line. These may suck but still be useful. 02:44:02 rpg: I'd concede that it shouldn't in theory be so painful, and javascript is quite pleasant when it isn't implicitly coercing things and turning your debugging into an exercise in dataflow analysis, but all the web stuff strikes me as frustrating once you try to make anything work on more than one browser 02:44:41 funny, I've used rapidsvn successfully without knowing how to use subversion from the command line, and it only sucks because it's a piece of crap 02:45:39 the "web" is hard in a way no other platform is. mainly, because its UI is directly controlled by the user. Users don't change systems calls and instruction sets to fuck with a systems hacker. They can't change fonts on a gui or hexedit resources in dialog boxes to fuck with an app developer. But they all mess with a web developer. 02:46:05 well, they could if they really wanted to 02:46:25 but the vast majority of web browser users don't mess with that too much either 02:46:38 unless you mean the differences between actual browsers, which IS a problem 02:46:39 fusss: can't change fonts on a gui? I think even Windows lets you do that 02:46:51 maybe not OS X 02:46:53 Just write decent documents and rely on the magical css pixies to make it pretty 02:48:05 <_3b> S11001001: not if you want your apps to keep working :/ 02:48:12 and HTML was never made to be a pixel-precise page layout language 02:48:15 -!- beach [n=user@58.186.158.204] has left #lisp 02:48:23 you just could never guarantee how your page is gonna be laid out on Joe the Plumber's browser. Screen resolutions, colors, images size, fonts. etc. Web apps are the only ones required to run on every imaginable device. Even desktop API have "embedded" versions. 02:48:39 yhara_ [n=yhara@ZC046088.ppp.dion.ne.jp] has joined #lisp 02:48:52 Fortunately, css has pixel level control, if you really need it. 02:48:53 fusss: so do web pages these days 02:49:04 I blame web developers, for not accepting that fact and resisting the temptation to do oooh shiny layouts in absolute coordinates with fragile dependencies on text sizes 02:49:21 -!- blbrown_lt [n=blbrown@76.210.234.20] has quit ["Leaving"] 02:49:29 <_3b> hefner: desktop app devs do that too 02:49:56 (which I myself do as well, in the spirit of contributing to the eventual destruction of the web) 02:50:08 -!- yhara [n=yhara@ZC046088.ppp.dion.ne.jp] has quit [Read error: 113 (No route to host)] 02:50:31 _3b: but their target platform is tractable, compared to web apps. You can't even guarantee basic scripting to work, so you have to do "progressive enhancement". tell a desktop GUI developer he might or might not have callbacks or event processing :-) 02:50:50 that reminds me of a colleague earlier today referring to Berners-Lee as the "father of the Internet" 02:50:53 again, I never had to write a complicated web UI, but it always seemed that updating something (i.e. pushing something to the user) is a major problem.. there's Comet and stuff, but it seems to require its own http server (extensions).. and I know about the polling approach (even wrote some code to do that) but that's not exactly sane 02:51:03 <_3b> fusss: tell that to all the desktop apps that don't work when you set a correct DPI on a windows box :( 02:51:05 _3b: if they really can control things at a pixel level of precision (including text size), I think that's fine. 02:51:31 if they're just using crappy tools that encourage them to do that without really giving them full control (like your typical windows UI builder), that's not so good 02:51:36 <_3b> hefner: either they can't, or they don't :) 02:51:40 adeht: the server is the part you generally ave control over 02:51:50 so comet isn't such a problem from that point of view 02:52:17 rsynnott: any Comet extensions for hunchentoot? 02:52:17 but comet has its own issues, in that it's a nasty hack generally managed via various side-effects of browser implementations 02:52:24 not that I know of 02:52:47 given that hunchentoot uses a tthread per connection you probably wouldn't want one if it existed 02:52:53 um, it's a fairly straight-forward xml request that the server delays responding to until it feels up to it. 02:53:24 that method works most places, yep 02:53:35 mirakel [n=mikael@c-7642e353.027-10-67766c2.cust.bredbandsbolaget.se] has joined #lisp 02:53:39 only problem is that you need a new request per event 02:53:48 and some proxies will break it 02:54:20 Surely varying the url would avoid that? 02:54:51 mib_s24h4xwz [i=4400b40a@gateway/web/ajax/mibbit.com/x-706c218d73c64f1e] has joined #lisp 02:54:51 rsynnott: well, everything web seems to be composed of nasty hacks 02:54:51 l_n [n=shawn@tuxhacker/lordnothing] has joined #lisp 02:54:54 then again I never was a "web developer" so I might be ignorant of the stuff doing Right Things 02:55:00 no, I mean some proxies will silently drop requests that take too long, or send the user random text of their own 02:55:22 -!- jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has left #lisp 02:55:24 Well, that'll teach you to use broken proxies. :) 02:55:56 (and some browsers don't get rid of memory allocated for these sorts of requests properly until you leave the page, though I think that has largely been fixed now) 02:56:23 I got the latest SLIME snapshot --- the REPL buffer is gone, the inferior-lisp buffer remains, should I worry? 02:56:25 my resentment toward the web is basically that we're stuck with it forever, and for all the junk they continue to pile on that would take a decade to implement from scratch, you can't even lay out a simple page as well as you could ten years ago with a table 02:56:45 mirakel: did you reload slime.el? 02:56:48 mirakel: use (slime-setup '(slime-fancy)) 02:56:55 you could say the same of win32 api :) 02:56:55 mirakel: repl is now a contrib that you need to load 02:57:20 rsynnott: funny, I'd just assumed the windows API hadn't really changed in the past decade 02:57:37 <_3b> hefner: they add a new layer or 2 every windows version 02:57:57 and now! new! win32api, the 64bit version! 02:58:16 (see Raymond Chen's blog for some gory details about how misnamed the types are now) 02:58:43 mirakel: The REPL is unsupported now. Soon, all of SLIME will be unsupported, except for a skeletal contrib loader. Rejoice! 02:58:50 yhara [n=yhara@ZC046088.ppp.dion.ne.jp] has joined #lisp 02:59:07 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 02:59:11 *luis* puts his shiny new contrib maintainer hat on 02:59:15 Thank you! 02:59:15 \ 02:59:18 oops 02:59:25 fear not, help is on the way! 02:59:42 _3b: I don't count new layers on top as evolving the win32 api. 02:59:50 -!- yhara_ [n=yhara@ZC046088.ppp.dion.ne.jp] has quit [Read error: 113 (No route to host)] 03:00:12 I guess I shouldn't complain, software on windows actually works pretty well, if you can stand the smell. 03:00:53 I updated slime, now how do I get the swank server to load? 03:02:14 <_3b> yeah, not like i can get decent focus-follow-mouse on linux these day either :( 03:02:14 -!- ManateeLazyCat [n=user@222.212.132.169] has quit [Remote closed the connection] 03:02:27 Good morning. 03:03:16 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 03:03:28 _3b: you can get it on windows! 03:03:38 as in fvwm? 03:03:41 <_3b> rsynnott: not that i've seen 03:03:51 (or at least you could on NT4; MS Powertoys came with a thing to enable it. It was weird) 03:04:04 <_3b> rsynnott: yeah, i use that... it doesn't always get the focus right 03:04:55 <_3b> i'll probably end up dropping back to twm next time i switch to linux, last few other WMs i've tried had similar problems 03:05:46 ManateeLazyCat [n=user@222.212.132.169] has joined #lisp 03:05:59 _3b: stumpwm, the clim of window managers :-) 03:06:03 *_3b* does miss the ability to program modifier keys to move/resize windows though 03:06:57 <_3b> fusss: yeah, i may try that too 03:07:38 -!- mib_s24h4xwz [i=4400b40a@gateway/web/ajax/mibbit.com/x-706c218d73c64f1e] has left #lisp 03:07:39 is it me or do "bloggers" try a new language every week? 03:10:19 *fusss* is reluctant to post a link to today's "lisp appreciation/whinage fest" blog entry 03:10:56 fusss: ah, go on 03:11:20 http://www.jakevoytko.com/blog/2008/12/29/everything-you-need-to-get-started-with-common-lisp/ 03:11:54 too bad, this one doesn't use fib and fac, but it has good integer addition example 03:12:54 fusss: hmpf, why are you recommending asdf-install? 03:13:22 luis: please take that back. i _don't_ have a blog. 03:13:23 I don't think it's him 03:13:35 "(+ 2 2)" 03:13:35 fusss: oh, ok. 03:13:45 I like the C++ snippet, with the little-known 'evaluate' function :) 03:14:15 "(+ 2 2)" is redundant. how do the newbies know "+" isn't multiplication? lispers and their hard examples! 03:15:16 "Irony alert: I typed elisp manula into Google on my first try." 03:15:34 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 03:15:41 obviously a definition of 'irony' with which I was previously unfamiliar 03:15:49 you learn something new every day! 03:17:01 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 03:17:02 somebody on there wants an statically styped clojure, because he must have a defent FPL for java. pheeeeeeeeewww. 03:17:04 -!- mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has quit [] 03:17:14 s/defent/decent/ 03:17:34 can't he use any of the good ML dialects invented last week? 03:18:37 luis: what's wrong with asdf-install? 03:18:58 <_3b> fusss: doesn't work on windows? 03:19:09 worst system, except for all the others 03:19:18 (except maybe clbuild) 03:19:18 *fusss* is unfamiliar with all the good asdf replacements invented last week 03:19:38 <_3b> clbuild works even less on windows :p 03:19:41 funnily enough, there were two allegedly-good asdf replacements in the last month or so 03:19:45 _3b: my site-systems is a samba share, you insensitive clod :-D 03:19:53 -!- jsoft [n=Administ@unaffiliated/jsoft] has quit [Read error: 110 (Connection timed out)] 03:20:14 _3b: fortunately, few lisp implementations really work properly on windows, either :) 03:20:17 <_3b> fusss: hmm, maybe i should try that 03:20:24 too bad clbuild only build from CVS/SVN/git/darcs/whateer 03:20:27 <_3b> rsynnott: sbcl is close nough for me 03:20:34 builds* 03:20:53 <_3b> (except for the statistical profiler) 03:21:04 sbcl on win32 is freakishly good 03:21:15 for something that threatens your kittens that is 03:21:16 _3b: is threading just completely unavailable on win32 or what? 03:21:28 <_3b> Phoodus: right, no threads 03:22:22 i should hire a burglar off of craigslist to steal nyef's new ARM toy (and other non-win32 machines) 03:22:31 -!- hfoo [n=h@p5B17DA3A.dip.t-dialin.net] has quit ["Leaving"] 03:22:34 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 03:22:36 <_3b> oh yeah, the inability to kill infinite loops is annoying once in a while as well 03:22:38 mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has joined #lisp 03:22:39 also, scary interaction with any arbitrary dlls you might have around, I believe 03:22:44 though that may have been fixed 03:22:57 clozure on win32 is supposed to be getting there 03:23:04 <_3b> fusss: nah, i want ARM sbcl too :) 03:23:27 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 03:23:30 fusss: asdf-install is fundamentally flawed in that it mixes systems with projects. 03:23:31 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 03:25:39 *sigh* feeling the "What? You dare try to run code on a non-unix OS?" pinch as we speak :-P 03:25:57 dunno of clozure, but i tend to maintain exact lisp-implementation-version across platforms. makes debugging easier. 03:26:08 Allegro won't talk SLIME having downloaded the latest patches, mcclim just plain pukes on CLISP, etc 03:26:17 and l-i-type :-) 03:26:27 and ECL can't compile releases at all 03:27:04 mcclim? non-unix? are you kluding away on that hideous cygwin stack? 03:27:30 no, mcclim says it should run under CLISP 03:27:53 but the asdf load throws a few million warnings, then dies on a reader error :-P 03:28:24 the warnings being stuff like redefining methods after they've been called, not OS-specific afaict 03:28:48 try to think of the end result first. i would be glad if mcclim _didn't_ work for me on win32 clisp. 03:29:02 well, you can take your religion elsewhere 03:29:24 not religion, just common sense 03:30:21 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 03:30:54 a gui on non-thread lisp on a platform not exactly well known for its good runaway process management or easily debuggable IPC. 03:31:05 <_3b> Phoodus: sure you aren't trying to load 2 different CLX or something? 03:31:19 _3b: I have absolutely no idea what's going on 03:31:45 but my asdf paths look sensible, no obvious "Hey, this is wrong" sort of error reporting, etc 03:31:52 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 104 (Connection reset by peer)] 03:31:54 -!- kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has quit [Remote closed the connection] 03:32:17 I'd try it under ACL, but that seems to have broken 03:32:28 kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has joined #lisp 03:32:31 *Phoodus* remembers he can skip SLIME for ACL 03:32:51 you're already using allegro, look around the acldoc folder 03:35:03 -!- karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has left #lisp 03:35:49 _3b: just got a fat dvd from adobe today. cli flash hacking is sure fun, compared to this 8gig monstrosity :-) 03:36:39 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 03:39:20 -!- mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has quit [] 03:39:46 -!- l_n [n=shawn@tuxhacker/lordnothing] has left #lisp 03:40:16 saikat [n=saikat@user-12lc4ta.cable.mindspring.com] has joined #lisp 03:40:28 Phoodus: I must go off now, but for ACL, suggest you try the improved CLX that Xophe did, as modified for ACL by Mikel Bancroft. 03:41:24 Can someone remind me how to tell from whence I pulled with darcs? 03:41:40 I'm just on the free ACL since our license expired, and it's not financially sensible to renew 03:41:46 cat _darcs/prefs/defaultrepo 03:41:51 so I'm not going to lock in on ACL, I just wanted to at least _invoke_ this thing 03:42:16 but it looks like ltk is my only sensible portable option 03:42:30 Phoodus: Oh. It's just that I don't think McCLIM will work with the CLX that comes with ACL. 03:42:33 if the GUI layer is Lisp and not something else, like Erlang's Tk bindings 03:42:34 You need an updated one. 03:43:07 You can pull a new version from opensource.franz.com somehow, but I'm afraid I've lost the repo url. Ask me tmrw if you still need it. 03:43:10 meh, not worth the bother at this point 03:43:13 Otherwise, use SBCL for McCLIM. 03:43:19 not threaded 03:43:20 Phoodus: I'm inclined to agree. 03:43:23 goodnight. 03:43:44 it's amazing how difficult it is to just pull code and get it to run most of the time 03:44:02 nobody abstracts beyond their immediate OS decisions 03:44:14 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [] 03:47:37 chavo_ [n=user@c-24-118-166-62.hsd1.mn.comcast.net] has joined #lisp 03:48:29 -!- chavo_ [n=user@c-24-118-166-62.hsd1.mn.comcast.net] has quit [Remote closed the connection] 03:48:53 chavo_ [n=user@c-24-118-166-62.hsd1.mn.comcast.net] has joined #lisp 03:48:58 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 03:50:04 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 03:52:13 <_3b> Phoodus: not enough people motivated to fix things for free lisps on windows 03:52:29 the problem isnt' fixing it 03:52:48 the problem is not locking into unixisms 03:52:51 <_3b> i was counting 'port' in 'fix' 03:52:59 abstract at your application and its needs, not the OS 03:55:02 <_3b> fusss: yeah, i wouldn't want to do the artist side of the flash hacking from CLI though :) 03:58:19 yeah, but at least i can follow the flash books now with getting confused over as2/as3/flex/air differences :-) 03:58:35 -!- chavo_ [n=user@c-24-118-166-62.hsd1.mn.comcast.net] has quit [Remote closed the connection] 03:58:35 with? ;) 03:58:43 without ^_^ 03:59:34 seriously Phoodus, look into lambda-gtk. (maybe then i can bug you for help ;-) 04:00:17 OpenMCL/SBCL/CMUCL :-P 04:00:49 utterly untrue 04:00:52 -!- O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has quit ["Come alive!"] 04:00:58 clisp and sbcl, linux and win32 04:01:01 well, that's what its own webpage said, so I believed it :-P 04:01:06 -!- S11001001 [n=sirian@pdpc/supporter/active/S11001001] has left #lisp 04:01:20 -!- bakkdoor [n=bakkdoor@81.163.101.59] has quit [Connection timed out] 04:01:31 the author doesn't even have screenshots. i, otoh, made nearly every gtk widget imaginable with it. 04:02:07 it's one of those "bindings" that work so well, nobody bothers to write good documentation for it, because the C specs work just as well. 04:02:13 Phoodus: there's also a win32 port of clozure (openmcl) which is apparently okay 04:02:20 or getting there, anyway 04:02:56 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 04:04:17 srp [n=srp@122.172.3.54] has joined #lisp 04:06:04 fuss: oh man, you need to worry about explicit deallocations? 04:06:13 I guess it is just a plain C interface 04:07:45 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 04:08:07 huh? 04:09:00 _3b: sawfish is still my favorite wm for inventing funny mouse/keyboard/modifier behaviors, mostly because the configuration UI is so nice 04:09:42 <_3b> hefner: that's what i was using last time i used linux seriously, seem to remember it not getting focus right either though :( 04:11:10 <_3b> i think i was pretty much just copying twm config options for what little programming i did with it though 04:11:25 I used to have weird focus problems, then it straightened out, and now it's bitrotted in some other quirky way 04:11:52 I'm not a focus-follows-mouse guy, though 04:13:08 <_3b> yeah, i might have been better off not picking up that preference :) 04:14:23 ``Erik [i=erik@c-68-54-174-162.hsd1.md.comcast.net] has joined #lisp 04:16:12 jlf` [n=user@netblock-68-183-235-250.dslextreme.com] has joined #lisp 04:23:04 happycodemonkey [n=happycod@c-98-223-43-42.hsd1.in.comcast.net] has joined #lisp 04:23:45 -!- LiamH [n=nobody@pool-72-75-101-81.washdc.east.verizon.net] has quit ["Leaving."] 04:30:12 -!- elurin [n=user@85.104.129.185] has quit [Remote closed the connection] 04:30:42 elurin [n=user@85.104.129.185] has joined #lisp 04:32:25 -!- Hun [n=Hun@81.163.116.190] has quit [Read error: 110 (Connection timed out)] 04:34:54 -!- doxtor [n=doxtor@unaffiliated/mitja] has quit [Remote closed the connection] 04:41:15 jso [n=user@host-154-148-107-208.midco.net] has joined #lisp 04:41:36 felideon [n=user@adsl-074-186-235-232.sip.bct.bellsouth.net] has joined #lisp 04:48:23 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 04:55:45 -!- dreish [n=dreish@minus.dreish.org] has quit [] 05:00:00 Alright, who killed SLIME? 05:00:26 what do you mean? 05:01:44 I pulled the latest CVS and now I get a SBCL prompt instead of the usual CL-USER> prompt. Also, it doesn't want to autocomplete and somesuch. I've had this problem on 2 machines. 05:02:02 <_3b> REPL is a contrib now 05:02:03 add slime-repl contrib 05:02:14 Oh! Okay. 05:02:16 tiesje [n=user@202.63.242.211] has joined #lisp 05:03:32 Woo! Thanks. :D 05:04:01 everyone asking about that should be counted and present to the slime developers to tell how much inconvenience brought these changes 05:04:28 the repl is now a contrib? What's left in the core slime? 05:04:31 I guess SLIME is regressing to SLIM 05:04:42 nah, just LIME. 05:05:48 remind me what the major inconvenience of the *inferior-lisp* buffer is versus the slime repl? 05:06:26 no fancy arglist and completion? 05:06:47 yes 05:07:17 me-so-stupid [n=hooyambo@77.236.84.166] has joined #lisp 05:08:05 also, no ,command 05:08:31 bah, I never use that nonsense 05:08:47 And no nice little reassurances about which package you're in. But, the autocompletion and arglist are what really get me. 05:08:53 The SLIME user manual also lists: (1) conditions are debugged with SLDB, (2) faces distinguish printed output from return values, and (3) better prompt input handling using markers. 05:09:21 some of those features are a little quirky 05:10:44 jso: this can be accomplished with (require :sb-aclrepl) 05:11:13 slime probably ought to load some reasonable default set of contribs 05:11:32 valiza1 [n=haroldo@r190-133-130-238.dialup.adsl.anteldata.net.uy] has joined #lisp 05:11:51 yeah, there is a lengthy discussion in the mailing list 05:12:09 I'm all for minimalism, but at this point any user visible change to slime is just an annoyance 05:12:38 (unless it's a bugfix, or a new feature that doesn't smack me over the head with its presence) 05:13:06 Honestly, I'm kind of surprised to tell you the truth. I didn't even know there was a contrib until now. Things just worked so well, until the latest set of changes at least. 05:13:23 Now I've got to figure out what things the contrib has to offer. :D 05:13:40 jso: try slime-fancy 05:14:05 it's a meta contrib for loading various fancy contribs 05:14:14 Will do! 05:14:30 (including repl) 05:14:51 -!- harovali [n=haroldo@r190-133-139-155.dialup.adsl.anteldata.net.uy] has quit [Read error: 60 (Operation timed out)] 05:14:59 *stassats`* didn't noticed the change because of slime-fancy 05:17:00 yhara_ [n=yhara@ZC046088.ppp.dion.ne.jp] has joined #lisp 05:17:09 -!- yhara [n=yhara@ZC046088.ppp.dion.ne.jp] has quit [Read error: 113 (No route to host)] 05:20:23 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 05:21:15 software success, sadly (?), really is reaching the point where half your users scream bloody murder whenever you change anything 05:22:34 I'd agree, but luckily this instance the users are programmers. A word on the SLIME page about the change would be nice. 05:24:38 -!- |Soulman| [n=kvirc@138.80-202-254.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 05:25:35 -!- hugod [n=hugo@bas1-montreal50-1279441402.dsl.bell.ca] has quit [] 05:32:34 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 05:33:28 morlos [n=morlos@cpe-76-94-160-6.socal.res.rr.com] has joined #lisp 05:35:21 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit ["ChatZilla 0.9.84 [Firefox 2.0.0.20/2008121709]"] 05:46:41 -!- breinded [n=nonamme@h-68-167-69-84.nycmny83.dynamic.covad.net] has quit [Remote closed the connection] 05:46:49 breinded [n=nonamme@h-68-167-69-84.nycmny83.dynamic.covad.net] has joined #lisp 05:55:25 -!- breinded [n=nonamme@h-68-167-69-84.nycmny83.dynamic.covad.net] has quit [Remote closed the connection] 06:01:30 ASau` [n=user@193.138.70.52] has joined #lisp 06:01:37 billc` [n=user@S0106001b63f442be.vn.shawcable.net] has joined #lisp 06:01:42 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 06:01:42 -!- m4dnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 06:01:42 -!- billc [n=user@S0106001b63f442be.vn.shawcable.net] has quit [Remote closed the connection] 06:01:42 m4dnificent [n=user@83.101.62.132] has joined #lisp 06:01:42 -!- madnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 06:01:51 madnificent [n=user@83.101.62.132] has joined #lisp 06:03:28 chaitanya [n=chaitany@122.162.15.105] has joined #lisp 06:14:38 Beket [n=Beket@athedsl-357902.home.otenet.gr] has joined #lisp 06:26:10 -!- Tordek [n=tordek@host112.190-137-248.telecom.net.ar] has quit [Read error: 113 (No route to host)] 06:34:17 -!- rottcodd [n=user@ppp59-167-42-106.lns2.cbr1.internode.on.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 06:38:01 S11001001 [n=sirian@pdpc/supporter/active/S11001001] has joined #lisp 06:46:54 knabb [n=hask@h45n2c1o1097.bredband.skanova.com] has joined #lisp 06:47:29 has their ever been written a real gameengine in Lisp? 3D, physics and all? 06:48:12 jak and daxter was written in a lisp 06:49:42 http://en.wikipedia.org/wiki/Game_Oriented_Assembly_Lisp 06:50:49 ok but common lisp or scheme? not a custom one 06:50:57 -!- felideon [n=user@adsl-074-186-235-232.sip.bct.bellsouth.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 06:50:59 allegro common lisp was used 06:51:11 but it was heavily modified by domain specific macros. 06:53:10 -!- ManateeLazyCat [n=user@222.212.132.169] has quit [Read error: 104 (Connection reset by peer)] 06:59:16 -!- me-so-stupid [n=hooyambo@77.236.84.166] has quit ["Leaving"] 07:01:57 -!- yango [n=yango@unaffiliated/yango] has quit [Read error: 110 (Connection timed out)] 07:04:48 slyrus_ [n=slyrus@207.189.195.44] has joined #lisp 07:06:17 -!- chaitanya [n=chaitany@122.162.15.105] has quit ["Leaving."] 07:07:44 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 07:25:08 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 07:26:52 binarycodes [n=sujoy@59.93.218.105] has joined #lisp 07:29:55 good morning 07:33:49 karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has joined #lisp 07:35:43 -!- slyrus_ [n=slyrus@207.189.195.44] has quit [Read error: 60 (Operation timed out)] 07:39:54 nicolas [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has joined #lisp 07:40:55 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 07:42:41 -!- mqt_ is now known as mqt 07:42:54 seelenquell [n=seelenqu@tmo-100-187.customers.d1-online.com] has joined #lisp 07:44:00 -!- knabb [n=hask@h45n2c1o1097.bredband.skanova.com] has quit [] 07:47:49 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 07:48:28 mrSpec [n=SomeOne@88.208.105.1] has joined #lisp 07:48:47 Morning 07:52:50 -!- tiesje [n=user@202.63.242.211] has quit [Read error: 113 (No route to host)] 07:55:54 binarycodes_ [n=sujoy@59.93.243.214] has joined #lisp 07:56:03 -!- binarycodes_ [n=sujoy@59.93.243.214] has quit [Remote closed the connection] 07:56:40 binarycodes__ [n=sujoy@59.93.243.214] has joined #lisp 07:57:13 -!- binarycodes [n=sujoy@59.93.218.105] has quit [Nick collision from services.] 07:57:17 -!- binarycodes__ is now known as binarycodes 07:59:38 good morning 08:00:09 -!- karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has left #lisp 08:00:45 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Remote closed the connection] 08:05:34 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [Client Quit] 08:14:54 -!- jso [n=user@host-154-148-107-208.midco.net] has quit [Remote closed the connection] 08:18:53 kiuma [n=kiuma@81-208-106-75.ip.fastwebnet.it] has joined #lisp 08:19:16 hello lispers 08:20:33 Aankhen`` [n=Aankhen@122.163.225.159] has joined #lisp 08:22:23 -!- seelenquell [n=seelenqu@tmo-100-187.customers.d1-online.com] has quit [Read error: 110 (Connection timed out)] 08:23:28 jso [n=user@host-154-148-107-208.midco.net] has joined #lisp 08:26:35 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 08:26:55 How long does it usually take to get approval for a cl.net project? I got a reply to my application within an hour, but nothing after that, and I'm wondering whether my reply to it got lost in the tubes. :-S 08:31:20 -!- yhara_ [n=yhara@ZC046088.ppp.dion.ne.jp] has quit ["Leaving..."] 08:35:54 beach [n=user@58.186.158.204] has joined #lisp 08:35:57 Good afternoon. 08:38:18 9:40 AM is afternoon? :-o 08:38:26 Or is that just your system clock being out of sync with where you are? 08:38:48 I think it's pretty accurate. 08:39:01 Ah. 08:39:16 hi beach 08:39:17 Well, good afternoon anyway, although here it's quite a few hours on. :-) 08:39:25 hey Krystof 08:39:58 Aankhen``: I'm in Ho Chi Minh City, Vietnam, and it's 15:40 here. 08:40:19 beach: Oh, I see. The CTCP TIME reply said 9:40 AM, hence the confusion. 08:41:11 Nice to have someone closer to my own timezone. 08:41:11 Aankhen``: I didn't reset my system time, indeed. 08:41:50 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 08:45:10 hello beach 08:45:25 Hey mvilleneuve. What's up? 08:47:19 beach: not much at the moment... How's your stay in Vietnam going? 08:47:42 So far so good :) Works starts on friday. 08:52:25 toddoon [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has joined #lisp 08:54:45 -!- saikat [n=saikat@user-12lc4ta.cable.mindspring.com] has quit [] 08:55:12 Is it true that commond lisp's standard is huge and obfuscated ? I've just read it here: http://discuss.joelonsoftware.com/default.asp?joel.3.371875.57 08:55:34 Beket: That seems like a load of crap. 08:55:54 Beket: The document has very high quality as far as specifications go. 08:56:00 <_3b> depends on how you define 'huge and complicated'... doesn't seem bad to me though 08:56:18 <_3b> *obfuscated 08:56:19 he said "obfuscated", not "complicated". 08:56:43 I see, thanks 08:56:52 Beket: Common Lisp is alternately accused of being huge and not covering enough. 08:57:11 Beket: I think it is just very popular to complain about Lisp. 08:57:39 beach: why is it popular ? 08:58:09 Beket: I don't know. I think it makes some people feel good about themselves. 08:58:29 Beket: I am guessing that this is the defense mechanism they have found so that they won't have to bother learning it. 08:59:14 Beket: see http://dept-info.labri.fr/~strandh/Essays/psychology.html 08:59:57 segv [n=mb@p4FC1E60D.dip.t-dialin.net] has joined #lisp 09:00:27 Beket: If they couldn't find any flaws with it, that would mean that their huge investement in a lesser language would have been wasted in some ways. Much better then to justify that investment by finding flaws in the other thing. 09:07:20 Beket: I personally find it hard to read. But it is a well-written exact description of what a function does (or what it isn't guaranteed to do). Therefore it may be hard, but very very usefull as a reference 09:08:19 It wasn't meant to be pedagogical, just precise. 09:09:57 Is it free to download by any chance? Or perhaps a draft of it, similar to C's standard? 09:10:14 minion: tell beket about clhs 09:10:15 beket: please see clhs: To look up a symbol in the HyperSpec, try saying "clhs symbol". For more information on the HyperSpec see http://www.cliki.net/CLHS . 09:11:17 beach: which may make it somewhat frustrating to point people to when they are asking for a function they do not know. (I'm explicitly not saying I find it bad to point them at the clhs) 09:11:44 Beket: you have been learning Lisp for two months now. How could you have missed the CLHS? 09:12:14 <_3b> madnificent: probably still easier to clarify clhs if needed than explain from scratch 09:12:37 beach: perhaps he simply didn't know it by the name in the article (didn't read it) 09:12:53 _3b: wasn't there an annotated version of it available somewhere? 09:13:02 madnificent: true, it can be frustrating, but I think it is worth getting used to the language in it as soon as possible. 09:13:14 <_3b> madnificent: not that i know of, though it has been discussed a few times 09:14:40 it could be a welcome addition, I guess. More examples could make it all the more understandable. Many people are used to things like the java api, which goes back to sticks and stones with respect to the complexity of the document 09:14:46 Beket: That whole thread reeks of pure flamebait. 09:15:06 <_8david`> what ever happened to reading CLtL2 from cover to cover first? 09:15:19 -!- jso [n=user@host-154-148-107-208.midco.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 09:15:28 <_3b> madnificent: yeah, usually copyright concerns end up derailing any proposed efforts in that direction :( 09:15:33 <_8david`> (Good morning.) 09:15:43 hello _8david` 09:17:14 <_3b> _8david`: nah, going straight from reading random PG articles to debugging lisp implementations is the way to go 09:17:23 _3b: just an idea, but maybe (just maybe) if you put the original draft in a separate (i)frame, and the annotation on the rest of the page, you could be spared of that. (I really don't know what the legal issues are though) 09:17:44 <_3b> madnificent: nobody else knows either, which is the problem :) 09:19:39 *madnificent* whishes for lawyers will finaly become active on the FOSS side of the world. Not by sueing us, but by telling us what we are still allowed to do. 09:19:58 Why would they? There's no money in it for them. 09:20:12 (Not meant to be a derogatory remark, just a statement of fact.) 09:20:35 Aankhen``: there is no money to be gained by many FOSS projects either :) 09:20:49 madnificent: What's your point? 09:20:55 Perhaps the FSF lawyer could help. 09:21:09 schaueho [i=d5a445c1@gateway/web/ajax/mibbit.com/x-71e28a99b4a44300] has joined #lisp 09:21:22 yango [n=yango@unaffiliated/yango] has joined #lisp 09:21:34 Aankhen``: why should only programmers try to help the world for free? If we're going to claim there is a community, then I may only hope it becomes more and more diverse 09:22:08 beach: do they have time for things like thisL 09:22:11 s/L/L 09:22:16 grrr, s/L/? 09:22:20 madnificent: There's no "should" or "shouldn't" involved. Lawyers will follow the money, which will lead them in the opposite direction from FOSS. 09:22:30 madnificent: I don't know. One would have to ask to find out. 09:23:07 Aankhen``: which is what many coders say before they get involved :) 09:23:27 Aankhen``: I think what madnificent is saying is that surely there must be people with law training out there who are interested in FOSS, just the way there are programmers out there like that. 09:24:20 beach: Well, I can't really argue with that. No doubt there is some small fraction which is interested in FOSS. 09:24:31 \o/ 09:24:52 madnificent: So you'll be setting up the organization to recruit them all, then? :-P 09:27:04 Aankhen``: actually, I followed one course in our law-departement here, but couldn't find a single law-student somewhat interested in computers. There were some that didn't even know FOSS existed... For most of them, computers seem to be "machines that try to slow down your work as much as possible" 09:32:44 Good times. 09:36:54 jgracin [n=jgracin@82.193.208.195] has joined #lisp 09:39:11 HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has joined #lisp 09:42:41 -!- ia [n=ia@89.169.165.188] has quit [Read error: 110 (Connection timed out)] 09:44:07 ia [n=ia@89.169.165.188] has joined #lisp 09:44:52 -!- cads [n=max@c-71-56-62-166.hsd1.ga.comcast.net] has quit [Connection timed out] 09:46:59 tiesje [n=user@202.63.242.211] has joined #lisp 09:54:22 malumalu [n=malu@hnvr-4dbb501c.pool.einsundeins.de] has joined #lisp 09:56:11 morning all 09:56:18 hello jbjohns 09:56:36 I have a wild and wacky idea, that I'd like some feedback on. :) 09:56:47 oh dear. 09:57:24 If I were to go to the trouble of making a CMS that is as easy to use as Joomla!, would other Lisp people be interested in contributing to such a thing, or no? 09:57:49 in what way is that wild and wacky? 09:58:20 :) well, given the reaction I've gotten to some more conventional ideas, I didn't know what to expect 09:58:31 "I want to make a clone of something that already exists" 09:58:48 oh? You mean Joomla! or something else? 09:59:06 Krystof: Thinking that a CMS could actually be easy to use seems wacky! 09:59:09 heh 09:59:46 beach: By easy to use, I mean in Joomla you can do everything from inside the browser. No need to mess with config files or writing code once it's done. Heck, most people don't even need to touch HTML or CSS 10:00:13 I meant a clone of Joomla!. (in response to your question, I can't speak for anyone else but I am unlikely to devote any time to implementing YACMS) 10:00:35 jbjohns: When you say "do everything from inside the browser" that makes me want to run away screaming. 10:01:03 beach: And this is the reaction I expected, hence the "wild and wacky". I figured it would seem that way to most here. :) 10:01:40 jbjohns: well, for what it's worth, i wouldnt mind such a system, but i do not have time to hack on something like that. 10:02:11 jbjohns: I suspect the only way you will be able to find out is to try it. 10:02:14 jbjohns: what framework would you use (if any), and _maybe_ I will. If you would care to elaborate what you want to build exactly and if I'd recon the code is something I'm actually willing to work in :) 10:02:29 -!- tiesje [n=user@202.63.242.211] has quit [Read error: 60 (Operation timed out)] 10:03:05 Krystof: well the thing is, I needed to get a web site up and running as fast as possible, so I looked at what CMS's exist. I was shocked to see that for what i needed to do (some web stores) PHP *still* seems to be the only legit game in town (unless you want to take the time to write code/HTML/CSS) 10:03:45 jbjohns: you can gain more attention here by writing a system that uses sexps as configuration. Then make a web-front-end to set the values of the sexps to what the users want. It's the best of both worlds :) 10:04:27 The thing is, it strikes me that such a thing could be so useful that it might be worth not releasing the source code. If no one from the Lisp community would ever be willing to contribute extensions then there is less benefit of making it open source 10:04:50 jbjohns: if you need to get something up quickly (and even considered php), then rails could be a fast option. I'm currently guessing on lisp, though I haven't found any complete framework I'm willing to work in. 10:05:14 jbjohns: and with that attitude, you lost me (and probably many others) 10:05:17 madnificent: I didn't want to write code. Just load prebuild compoents 10:05:48 emarsden [n=user@mir31-3-82-234-52-44.fbx.proxad.net] has joined #lisp 10:06:27 jbjohns: so with this joomla thing you can just click your way through the webinterface (sort of like a VB gui designer) and then push "publish" and you have a good performing functional site? 10:06:33 madnificent: Well, what I was thinking on that front was: What makes Joomla so popular is how easy it is to add extensions. So in lisp extensions could be written in the form of systems, so when you add a new system, the asd file just has to be updated 10:07:21 hypno: It seems to perform well enough, tons of people use it. So the reason I was thinking of making a clone in Lisp is (a) lisp is more powerful and (b) I would expect Lisp to be much faster 10:07:36 jbjohns: this is how I looked at web-development with minions (ahh, I'll just paste it in a pastie) 10:09:05 jbjohns: i had the same thing in mind a few months ago, which would be to "Roxen:infy" hunchentoot, but it dawned on me that a nice system within emacs would be nicer. :/ 10:09:06 madnificent: My "attitude" may not be what you think it is. I plan on having my own business, and one of the things I can offer is creating store sites for people (already have several customers lined up, even though I didn't start looking). The CMS could either be my own personal thing that I bring with me to make me fast, or I can go to the trouble of releasing it publicly 10:10:03 madnificent: If no one would ever contribute and the only thing that could possibly happen is other people take my code and use it to compete against me, then what benefit would there be? 10:10:47 (i.e. non-lisp people. I want to make this thing so simple your grandma could install it) 10:12:22 I am willing to bet a lot of money you can't make anything so simple that my grandma can install it. 10:12:32 not many lisp people are of the "ignorant or flat out dumb"-persuasion, you know. it's selling the wrong thing to the wrong people. :/ 10:12:41 I am willing to bet that talking about it on IRC is not going to achieve anything 10:12:51 hi emarsden; season's greetings 10:13:48 jbjohns: create the thing, then let folks have give it a spin on some demosite. if it's good, people will be drooling all over it (in which case you can decide to open source or go commerical). :) 10:14:58 madnificent pasted "the minions way of pages" at http://paste.lisp.org/display/72826 10:15:14 Krystof: hi! 10:15:14 -!- kuwabara [n=Kuwabara@cerbere.qosmos.fr] has quit [Remote closed the connection] 10:15:15 kuwabara_ [n=Kuwabara@cerbere.qosmos.fr] has joined #lisp 10:15:17 -!- kuwabara_ is now known as kuwabara 10:15:27 jbjohns: it was suddenly very hard to ivent a page about nothing :) 10:15:38 emarsden: Ah, you are for real! How are things? 10:16:29 jbjohns: with the right license, no competition that gets ahead of you, a tribute to continuous development with respect to clients, generally getting the world forward, spring to mind 10:16:41 hello beach! I'm on holiays, things are good, thanks 10:17:31 the channel hasn't seen the sad news of Theimo Seufer's death in a car accident 10:17:33 http://www.debian.org/News/2008/20081229 10:18:17 ah, no 10:19:06 jbjohns: minions ended up to be too slow to be practical. I had to build a workaround for the initial system that used objects for everything and nothing and I wrote my own webserver backend because I was too lazy (yes beach, I wont be so ignorant anymore) to check out hunchentoot (which I still haven't). 10:19:35 Maddas [n=Maddas@tardis-b23.ee.ethz.ch] has joined #lisp 10:19:42 jbjohns: besides that, it seems to me that a definition of pages like that would allow you to create a Joomla!-alternative rather quickly :) 10:19:53 jgracin_ [n=jgracin@82.193.208.195] has joined #lisp 10:21:08 *madnificent* just figured, that minions could work a lot better with closures, instead of objects 10:23:30 ironChicken [n=richard@79-75-110-4.dynamic.dsl.as9105.com] has joined #lisp 10:25:01 jbjohns: as for the open sourcing, clearly it is your call. AFAICT, with the lack of lispers around the world, you'll probably be in a zone in which you don't have to care about local competition. When you're talking about quick-and-easy solutions, you're probably going to have to talk to clients that aren't really computer-wizards, thus you don't have to care about competition that is further away. Allso, you can make the system 10:25:01 open (dual licence), but close your own modules for it. Effectively allowing you to see if the project helps you when you open it, and keeping most of the useable code in-house in case no one starts to co-develop. 10:25:23 beach: My grandma can't either since she isn't with us anymore, but I mean as "no hassle" as possible. 10:25:26 -!- Beket [n=Beket@athedsl-357902.home.otenet.gr] has quit [] 10:25:32 -!- echo-area [n=user@nat/yahoo/x-4f22e3f673adaaab] has quit [Remote closed the connection] 10:26:06 hypno: I know Lisp people aren't, and that's good because it's lisp people who would have to make extensions, but just as Joomla is very useful to people who don't know how to program, so would a Lisp CMS (hopefully) 10:27:11 Krystof: I know, talking doesn't achieve anything, but how I structure the project would depend a lot on weather I expected anyone else to want to contrib or not, but I suppose, as several of you said, I can just defer 10:28:41 -!- jgracin [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 10:28:51 madnificent: as far as getting the world forward, I'm all for it. But the best solution is to become financially independent so I'll have time to create and contribute wonderful things. :) Contributing about 4 hours per week wont get me where I want to go 10:30:15 H4ns: here? 10:30:17 as for minions, that looks interesting. I haven't thought about what implementation I would use. I'm thinking right now about the templates angle. I hate templates myself, but many more people can create HTML/CSS then can code in any language 10:31:30 though my current thoughts on the matter is that when people make templates they are really trying to create an HTML view of different objects, so having things like "foreach" be supported seems silly. Better to just describe how an object of that type looks 10:31:39 jbjohns: well, I'm going to go sort-of in your direction. I have about 3 years of studying to go, but then I'll need to become independent. The way things are going, I'm going to want to develop web-apps in lisp. So chances are, that I'll use the system you propose if it's flexible enough... I'm still searching for the perfect way to handle things 10:32:35 -!- binarycodes [n=sujoy@59.93.243.214] has quit ["WeeChat 0.2.6"] 10:32:46 ah cool. AFK a bit, have to eat or wife gets fussy. :) 10:32:48 jbjohns: build parser from html to sexps etc... The way it is defined in minions, you'd create new types of tags. These can be used just like an html-tag. You can simply read their name from the XML and let them generate the correct page :) 10:32:54 -!- bombshelter13 [n=bombshel@209-161-240-155.dsl.look.ca] has quit ["Where is the glory in complying with demands?"] 10:33:26 have fun, talk to kiuma some time soon, he was developping a framework too (rather much OO) 10:35:45 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 54 (Connection reset by peer)] 10:44:20 hypno: ack 10:44:53 guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 10:45:12 hi 10:45:40 H4ns: this svn co svn://bknr.net/trunk/ediware didnt work for me, but i think i figured it out anyway. :) you are solving my main gripe with hunchentoot with that :) 10:46:19 hypno: what exactly did not work? i'd like to fix it if i can 10:46:47 which lisp version would you pick to start? CLisp? 10:46:54 H4ns: it should be svn co svn://svn.bknr.net/svn/trunk ediware no? :D 10:47:08 guille_: clisp will work, I started out with sbcl 10:47:24 guille_: as long as it is a Common Lisp implementation, you should be safe :) 10:47:40 hypno: svn://bknr.net/svn/trunk/ediware would be it, yes. where did you find the wrong url? 10:47:47 guille_: i would recommend SBCL, CCL, Lispworks or ACL. 10:47:48 antoni [n=antoni@17.pool85-53-15.dynamic.orange.es] has joined #lisp 10:48:05 H4ns: http://common-lisp.net/pipermail/tbnl-devel/2008-December/004492.html 10:48:06 guille_: sbcl is the dominant implementation here, so if you're a beginner and hope for support, sbcl is best. 10:48:32 hypno: ah, sorry about that, i'll follow up with the correct url. 10:49:15 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 10:49:16 H4ns: btw, are you using ccl or sbcl with that trunk? 10:49:22 moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has joined #lisp 10:49:27 ok thank you, i'll try with sbcl 10:49:45 hypno: sbcl atm, but we'll be integrating the ccl work we did for ita in the coming weeks 10:51:33 guille_: what made you decide to learn Lisp? 10:51:44 -!- awayekos is now known as anekos 10:55:05 learn something about AI, i enjoy learning languages in any case, but i bought perter norvig's book so i need some lisp background; i've tried some haskell before 10:59:22 guille_: PAIP might not be great as a first Lisp book. You could try it, but there are others that might be somewhat easier in the beginning. 11:00:07 guille_: otherwise, PAIP is a great book in my opinion. 11:00:33 guille_: if you are using linux, and emacs is you favorite editor, SBCL is a really good choice. 11:00:49 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 11:01:21 yeah, i've realized it with the grammar in the first (or second) chapter, i need to learn the basics first 11:01:39 nostoi [n=nostoi@127.Red-81-38-52.dynamicIP.rima-tde.net] has joined #lisp 11:02:01 minion: please tell guille_ about PCL. 11:02:02 guille_: please look at PCL: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005). 11:03:14 -!- Aankhen`` [n=Aankhen@122.163.225.159] has quit [Read error: 54 (Connection reset by peer)] 11:03:32 jgracin__ [n=jgracin@82.193.208.195] has joined #lisp 11:03:52 chris2 [n=chris@p5B16B099.dip0.t-ipconnect.de] has joined #lisp 11:04:00 Aankhen`` [n=Aankhen@122.163.225.159] has joined #lisp 11:04:33 thanks :) 11:07:34 -!- antoni [n=antoni@17.pool85-53-15.dynamic.orange.es] has quit [Client Quit] 11:09:06 -!- guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has quit [] 11:09:20 guille_ [n=guille@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 11:10:28 good morning. 11:10:53 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 11:12:22 -!- jgracin_ [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 11:12:30 ruediger [n=the-rued@p508B7AEB.dip.t-dialin.net] has joined #lisp 11:12:34 Nshag [i=user@Mix-Orleans-106-3-144.w193-248.abo.wanadoo.fr] has joined #lisp 11:12:52 srp_ [n=srp@122.172.29.30] has joined #lisp 11:13:50 back. madnificent: I was thinking I would provide a lot of ways of creating the html "view" of given data points. You could give sexp's if you want, XML, HTML, even WEBDAV, but in the case of lists of items, I would only look at the first one for the style 11:14:12 given data items* 11:16:45 Hun [n=Hun@port-92-195-55-84.dynamic.qsc.de] has joined #lisp 11:17:58 jbjohns: mainly anything can be mapped to sexps, so using that as a backend should give you enough flexibility 11:18:20 sure, I was just talking about what I would allow users to generate templates with 11:18:53 jbjohns: allso, a nice trick to keep yourself safe, could be to open source the general system, but build a closed-source module to define pages graphically in. Just throwing ideas 11:19:44 in the back end it would all be stored as sexps. My idea in using lisp is just that with a superior language, it should be possibly to surpass the functionality of the reams of PHP that's out there 11:20:17 jbjohns: users probably want to drag-'n-drop new site-elements into their site, so creating it (and then configuring each bit), that was what I was aiming for a couple of years ago (help me god, I wrote something like that in PHP) 11:20:28 and cramming xml and other stuff in there would take away from that 11:20:42 jbjohns: cfr pail graham's success :) 11:20:47 :) 11:21:11 jbjohns: well, I told it before. If the backend is nice enough to work in, chances are that I'll join in. 11:21:30 jbjohns: you could look at it as drupal, with an actually usefull language running behind it 11:22:17 -!- srp [n=srp@122.172.3.54] has quit [Read error: 104 (Connection reset by peer)] 11:23:03 madnificent: yea, joomla is a competitor for drupal. That's exactly what I was going for. 11:23:52 well, ok. I've got to finish up this store I'm doing, then I'll start lining out some architecture and start coding. Is there any good comparisons anywhere of the given frameworks? 11:25:24 sorry, that made no sense. For the given HTTP back end servers 11:25:34 -!- mrSpec is now known as spec[afk] 11:25:51 Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has joined #lisp 11:25:52 I know about hunchentoot, and I saw a couple of others. Any comparison about which is fastest and so on? Though I guess if we're competing against PHP it can't matter that much 11:26:22 jbjohns: you'll probably get people angry if you don't use hunchentoot. I wrote a simple webserver for minions once, but it is highly inferior to hunchentoot. 11:27:52 http://www.cliki.net/web grep for server, not _that_ many to choose from 11:28:52 I have no intention of writing my own, I just wondered if anyone had really tested these, etc. 11:29:54 jbjohns: no, you can simple see the page and see what they say, it is mostly a short summary of the server. I don't know of a writeup, but maybe there is none because it's easy to discover 11:33:54 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit ["night"] 11:35:05 -!- guille_ [n=guille@164.Red-83-49-201.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 11:36:11 hortont424 [n=hortont@c-75-68-222-233.hsd1.vt.comcast.net] has joined #lisp 11:42:04 -!- Hun [n=Hun@port-92-195-55-84.dynamic.qsc.de] has quit [Read error: 60 (Operation timed out)] 11:42:04 -!- anekos is now known as awayekos 11:42:45 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 11:42:50 srp [n=srp@122.172.24.229] has joined #lisp 11:43:38 Yuuhi [i=benni@p5483CBF7.dip.t-dialin.net] has joined #lisp 11:44:50 -!- benny [n=benny@i577A2105.versanet.de] has quit [Read error: 104 (Connection reset by peer)] 11:48:48 -!- nostoi [n=nostoi@127.Red-81-38-52.dynamicIP.rima-tde.net] has quit ["Verlassend"] 11:49:22 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 11:51:15 hugod [n=hugo@bas1-montreal50-1279441402.dsl.bell.ca] has joined #lisp 11:57:21 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 11:57:25 jpcooper [n=justin@unaffiliated/jpcooper] has joined #lisp 11:59:20 -!- srp_ [n=srp@122.172.29.30] has quit [Read error: 110 (Connection timed out)] 12:00:44 blitz_ [n=julian@77.64.176.217] has joined #lisp 12:01:22 jsoft [n=Administ@unaffiliated/jsoft] has joined #lisp 12:02:33 -!- schme [n=marcus@c83-249-80-232.bredband.comhem.se] has quit ["Lost terminal"] 12:04:12 schme [n=marcus@sxemacs/devel/schme] has joined #lisp 12:05:22 antoni [n=antoni@3.pool85-53-22.dynamic.orange.es] has joined #lisp 12:15:46 I'm trying to use slime, but i'm running into a lot of weird problems. One is that M-. can't find any of my defuns, not even in the same file (but it can find builtins like push). Another is that I get undefined function errors for one function and one macro, both of which I can use at the slime repl with no problems. 12:15:52 |Soulman| [n=kvirc@138.80-202-254.nextgentel.com] has joined #lisp 12:16:06 jgracin_ [n=jgracin@82.193.208.195] has joined #lisp 12:16:53 vy [n=user@213.139.194.186] has joined #lisp 12:18:34 -!- jgracin_ [n=jgracin@82.193.208.195] has quit [Remote closed the connection] 12:20:49 reaver__ [n=reaver@212.88.102.158] has joined #lisp 12:21:52 -!- segv [n=mb@p4FC1E60D.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 12:22:02 segv_ [n=mb@p4FC1CCE3.dip.t-dialin.net] has joined #lisp 12:24:40 -!- jgracin__ [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 12:25:41 hello! 12:25:55 Hello! 12:29:12 tiesje [n=user@202.63.242.211] has joined #lisp 12:30:38 -!- hortont424 is now known as hortont424|away 12:33:36 koning_robot: I don't use slime so I'd be of little help, but it might be a problem of package. There must be a command to have slime synchronize the package in the source and in the REPL. 12:35:10 guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 12:36:01 -!- djinni` [n=djinni`@ludios.net] has quit [No route to host] 12:38:06 -!- matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 12:39:54 -!- enodran [n=brandon@207-180-130-218.c3-0.abr-ubr1.sbo-abr.ma.cable.rcn.com] has quit [] 12:41:15 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 12:41:56 -!- kleppari [n=spa@bitbucket.is] has quit [Read error: 113 (No route to host)] 12:42:04 matimago: what do you use instead? 12:44:08 -!- at [n=at@nat.garanta.ru] has quit [Read error: 104 (Connection reset by peer)] 12:44:15 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 12:45:12 The problem with the undefined functions was that I used compile-and-load instead of load (that explains the undefined macro, but how does this affect functions?)... 12:45:19 hi folks. why does this not work: (= 2 NIL) 12:45:36 Dynetrekk: because = can only compare numbers. 12:45:39 clhs = 12:45:39 http://www.lispworks.com/reference/HyperSpec/Body/f_eq_sle.htm 12:45:50 beach: clhs? 12:46:09 Perhaps you want (eql 2 nil) 12:46:14 Dynetrekk: The Common Lisp Hyper Spec. Specbot gave you the URL. 12:46:24 beach: ah, thanks :) 12:46:42 actually, I just want to compare two numbers; and one of them might be NIL 12:46:45 kleppari [n=spa@bitbucket.is] has joined #lisp 12:46:51 i.e. not a number, but still, it is not equal to 2 :) 12:46:59 Dynetrekk: then use eql as Zhivago suggested. 12:47:06 clhs eql 12:47:06 http://www.lispworks.com/reference/HyperSpec/Body/a_eql.htm 12:47:10 beach: ah, it will do the same thing? 12:47:21 Dynetrekk: no, it will work. 12:48:23 beach: touche 12:48:35 the "same thing as I want", I meant ) 12:48:45 Please try #psychics 12:48:56 Oh, that I cannot tell, for only you know what you want. 12:49:09 beach: I'm not so sure I know myself 12:49:23 koning_robot: what is your lisp? 12:49:30 SBCL 12:49:34 Zhivago: Good to see you here again. You seem to be on less often these days. 12:50:02 koning_robot: might I ask you why you use SBCL? I've only got clisp myself 12:50:04 Zhivago: You left me with all the newbies. :) 12:50:11 Yes, although largely due to a client with windows. 12:50:19 ah 12:50:24 Dynetrekk: threading 12:50:31 koning_robot: threading? 12:50:34 I have returned to a client which mixes everything together again :) 12:50:42 koning_robot: Though clisp has the threading, I believe :) 12:50:58 Dynetrekk: yes, CLISP doesn't do it afaik 12:51:11 koning_robot: I see. any other reason? 12:51:46 nope, that's the reason I switched from the one to the other 12:51:58 koning_robot: not performance, for example? 12:52:20 koning_robot: I think you need the ./configure --with-threads=POSIX_THREADS for the clisp ;) 12:52:36 I'm too early into my project to care about performance :P 12:52:50 koning_robot: I see 12:53:14 from what I understood, the optimization in sbcl is better (somehow) 12:53:35 but I'm certainly no expert 12:53:36 Dynetrekk: Sometime one is faster than the other. 12:53:40 hmm, I would've sworn clisp didn't have threads last time I checked... lemme check again 12:53:50 schme: probably 12:53:59 koning_robot: They're in the "use at your own risk" section of experimental features :) 12:54:01 do the different lisps all compile the code before it is run? 12:54:10 schme: isn't everything? :P 12:54:14 Many options are hidden to those uninitiated in the arts of literacy. 12:54:19 I guess. 12:54:44 koning_robot: I think they added the win32, posix and solaris threads just last release. 12:54:45 schme: with optimization? 12:54:54 clisp has native code jit as well, although I haven't tried it. 12:55:26 Zhivago: i.e. both clisp and sbcl do JIT compilation? 12:55:35 sbcl does not do JIT. 12:55:36 no, SBCL is aot 12:55:51 aot? that one was new 12:55:57 ebzzry [n=ebzzry@124.217.83.77] has joined #lisp 12:56:00 Dynetrekk: I'm tempted to say all lisps compile into something, but I'm sure someone will dig up some lisp that does not :) 12:56:06 'ahead of time' 12:56:07 haha 12:56:08 first seen with GCJ I think 12:56:14 Zhivago: I see 12:56:20 CL requires compilation. It may not mean what you expect, though. 12:56:30 Zhivago: it's not that I'm uninitiated in the arts of literacy, it's that I read it somewhere and I didn't bother to doubt it (but perhaps judging sources is a part of literacy). 12:56:33 Zhivago: I guess I expect it to mean some optimization as well 12:56:54 An enormous bundle of confusions you have there :) 12:57:03 I'm sure you are right :( 12:57:37 -!- ebzzry [n=ebzzry@124.217.83.77] has quit [Client Quit] 12:57:57 Dynetrekk: That is why using CL makes you a better programmer. Not any magic in the language itself, but the nice people in #lisp force you to think about what you're actually saying ;) 12:58:18 schme: aaah... annoying yet pedagogical 12:59:10 Dynetrekk: And it works both ways! 12:59:25 I guess my real question is; what lisp should I use if I want to do numerics and AI? 12:59:31 schme: the annoyance or the pedagogy? 12:59:47 Well both I guess. 12:59:53 Dyne: Probably for numeric computation you will want to use sbcl. 13:00:06 AI doesn't mean anything, so it's hard to recommend. 13:00:16 Zhivago: probably correct 13:00:37 Zhivago: sbcl is installed on the cluster I ssh to, so good thing, I guess 13:01:23 :) 13:01:53 Ok back to gaming my cold away. 13:02:04 schme: good gaming 13:03:20 benny [n=benny@i577A2105.versanet.de] has joined #lisp 13:06:12 -!- awayekos is now known as anekos 13:06:46 borism [n=boris@195-50-201-40-dsl.krw.estpak.ee] has joined #lisp 13:07:38 demmeln [n=demmeln@atradig112.informatik.tu-muenchen.de] has joined #lisp 13:08:27 -!- ruediger [n=the-rued@p508B7AEB.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 13:09:33 jollygood [n=jollygoo@129.71.215.161] has joined #lisp 13:10:14 *beach* just designed a `1' and a `2' in this Metafont-like embedded language. 13:10:29 seelenquell [n=seelenqu@tmo-100-230.customers.d1-online.com] has joined #lisp 13:10:35 jjong [n=user@203.246.179.177] has joined #lisp 13:10:50 Just another 65,000 or so to go. 13:11:03 yeah, making great progress. 13:11:40 some of them are just latin characters with diacritical marks though, so that will be fairly easy. 13:12:23 Dextor [n=NoorDext@unaffiliated/noordextor] has joined #lisp 13:12:55 lessee, at the rate of 2 per day, that will take me ... around 100 years, right? 13:13:21 -!- borism_ [n=boris@195-50-199-203-dsl.krw.estpak.ee] has quit [Read error: 145 (Connection timed out)] 13:13:34 I guess I shall need some help, or some different technology. 13:14:04 -!- JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has quit ["Leaving"] 13:14:34 beach: heh :) 13:15:26 beach: wouldn't it be possible to automatically generate those definitions from truetype fonts and some fixing? 13:15:43 mvilleneuve: That would be worth trying. 13:16:51 mvilleneuve: Though I am not sure the `fixing' part is any faster than designing it from scratch. Perhaps a little, but not a factor 10, I think. 13:17:23 mvilleneuve: The advantage is that we would have *some* glyphs right away, but they might be of low quality. 13:17:30 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 13:17:49 trebor_home [n=user@dslb-084-058-206-073.pools.arcor-ip.net] has joined #lisp 13:22:04 beach: I use basically inferior-lisp, with a few commands to evaluate forms from source buffers like eval-last-sexp. Since I mostly use clisp, there's less implementation specific deficiencies^W user-unfriendlinesses to compensate than with sbcl... 13:23:36 yango_ [n=yango@unaffiliated/yango] has joined #lisp 13:24:52 -!- tiesje [n=user@202.63.242.211] has quit [Read error: 148 (No route to host)] 13:25:05 Any Slime hackers around? 13:26:13 I've stumbled upon some old Slime code that uses ARGLIST-FOR-ECHO-AREA, but I don't know what it's called nowadays. 13:26:16 -!- Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has quit ["Leaving"] 13:28:27 -!- matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 13:28:48 Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 13:29:57 tic: I take it you are hacking Climacs :) 13:31:19 Hun [n=Hun@Hunlappy2.visitor.congress.ccc.de] has joined #lisp 13:33:24 beach, hehe, actually I'm not. Would you have any idea on what the proper message is, these days? 13:33:41 tic: 'fraid not. 13:33:53 tic: I haven't looked at that for a while. 13:33:55 Aight. Guess I'll have to wait for tcr to show up. 13:34:04 tic: or Athas. 13:34:16 beach, seems I have code from 2006. I'm afraid to merge to a later version, though. 13:34:21 beach, ah, thanks. 13:34:33 -!- antoni [n=antoni@3.pool85-53-22.dynamic.orange.es] has quit [Read error: 113 (No route to host)] 13:35:16 tic: This planned Climacs-hacking week in January might be your opportunity to introduce VIM-like command tables into Climacs. 13:36:25 djinni` [n=djinni`@ludios.net] has joined #lisp 13:36:43 tic: I would be delighted if we could offer Climacs as an alternative to Emacs for VIM users. 13:37:16 beach, intriguing. the more I hack on Slim-Vim, the more the ugly internals of Vim shows. ("What, you want a callback on network activity?!") 13:37:58 tic: Clearly, a pure CL solution is called for! :) 13:38:08 beach, the problem is to do a _full_ emulation of Vim. I guess I could make it emulate the parts I use so it'd work helpfuly, but I could never call it complete. 13:38:39 beach, yeah, so it seems. I've never gotten Climacs to work very well for me though. Anyway. When is that hacking week? 13:38:46 tic: we all have that problem. Climacs is never going to emulate Emacs completely. 13:39:12 tic: I am not in charge of it, so I don't know. 13:39:23 OK. 13:39:33 -!- yango [n=yango@unaffiliated/yango] has quit [Connection timed out] 13:39:48 beach, does Climacs run in a console yet? 13:40:37 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 13:40:58 tic: no it doesn't. And the best way to make it do so is to write a text-only backend for McCLIM (say using some curses interface). 13:41:59 such a backend would be very useful for other applications as well. 13:42:54 oh hi 13:43:01 -!- kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has quit [] 13:44:46 kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has joined #lisp 13:45:16 weirdo: hey, what's up? 13:46:40 beach, indeed. that is a requirement for me to use it as a main editor,. 13:46:55 beach, which, unfortunately, means I have to implement it. :-) 13:47:03 -!- schaueho [i=d5a445c1@gateway/web/ajax/mibbit.com/x-71e28a99b4a44300] has quit ["mibbit.com: "Reboot""] 13:47:29 Sure, but I am not so convinced about the "unfortunately" part. 13:48:02 It's more of a "so much fun to do but so little time"-unfortunately. 13:48:17 meanwhile, I'm having an argument with Swank. 13:48:31 bah, it's a matter of priorities. :) 13:49:14 binarycodes [n=sujoy@59.93.243.214] has joined #lisp 13:49:15 -!- guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has quit ["bye"] 13:50:12 tic: someone told me it was a mistake for Climacs to use SWANK, becaue it would need to be tracked, but if we don't use it, then we would have to implement all that stuff ourselves. 13:51:18 -!- ``Erik [i=erik@c-68-54-174-162.hsd1.md.comcast.net] has quit [Read error: 110 (Connection timed out)] 13:51:36 beach, just woke up. figuring out what to hack next 13:52:25 Do you need advice? 13:52:48 shmho [n=user@58.142.15.103] has joined #lisp 13:52:55 beach, I think it's a good idea to use Swank for the backend, even though it does indeed mean you have to track it. At the very least the backend. 13:52:57 Dynetrekk pasted "indentation" at http://paste.lisp.org/display/72829 13:53:14 how the hell do you format the code sensibly for something like that? 13:53:48 think i'll change my toad matcher to CLOS and then hack up a CRUD abstraction for my web framework 13:54:02 Dynetrekk annotated #72829 with "corrected" at http://paste.lisp.org/display/72829#1 13:54:03 toad matcher === pattern matcher 13:54:06 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 13:54:31 Valla [i=opera@ip4-83-240-59-144.cust.nbox.cz] has joined #lisp 13:54:51 Dynetrekk: start by using Emacs 13:54:59 malumalu_ [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has joined #lisp 13:55:00 beach: I do 13:55:10 beach: aquamacs, if it matters 13:55:17 Then you didn't do enough TABs 13:55:25 beach: but I don't like the result 13:55:32 -!- Hun [n=Hun@Hunlappy2.visitor.congress.ccc.de] has quit [Connection timed out] 13:55:39 http://straseni.kompletne.cz/baf.php?page=straseni&kolo=7&od=62 ROFL ! xD 13:56:17 -!- Valla [i=opera@ip4-83-240-59-144.cust.nbox.cz] has left #lisp 13:57:14 Dynetrekk: also, (setf bla (+ bla whatever)) is called (incf bla whatever) 13:57:24 beach: ah, thanks 13:57:26 helps a bit :) 13:57:41 Dynetrekk: You have to customize emacs a bit to make it indent LOOP sensibly. See my slime talk for the required .emacs bits. 13:57:58 tcr: ah, sounds good 13:58:23 Dynetrekk: http://common-lisp.net/~trittweiler/slime-talk-2008.pdf 13:59:23 -!- ironChicken [n=richard@79-75-110-4.dynamic.dsl.as9105.com] has left #lisp 13:59:32 dv_ [n=dv@85-127-114-35.dynamic.xdsl-line.inode.at] has joined #lisp 14:00:42 tcr, hey, do you remember at the top of your head what ARGLIST-FOR-ECHO-AREA is called in recent Swank versions? 14:01:09 it's still called that, but it's in the slime-autodoc contrib (which uses swank-arglists.lisp) 14:03:51 What do I have to do to activate it? I currently start swank with (swank:create-server). 14:04:50 (swank-loader:init :load-contribs t) 14:05:17 aiee! new slime broke custom package symbol completion 14:05:23 for the compound and fuzzy matchers 14:05:43 wfm 14:05:45 and i can't send to slime-devel from gmane. the horror, the horror 14:06:01 tcr, thanks. 14:06:58 Now I get some output, unknown function. Wrong, but better. :) 14:07:34 am i the only one with broken symbol completion on newest slime? :) 14:09:28 -!- simplechat [n=simplech@unaffiliated/simplechat] has quit [Remote closed the connection] 14:09:29 LiamH [n=none@common-lisp.net] has joined #lisp 14:12:15 -!- malumalu [n=malu@hnvr-4dbb501c.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 14:12:16 -!- HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has quit ["This computer has gone to sleep"] 14:13:55 envi^home [n=envi@220.121.234.156] has joined #lisp 14:14:04 ManateeLazyCat [n=user@222.212.132.169] has joined #lisp 14:15:51 egn_ [i=180f36f1@gateway/web/ajax/mibbit.com/x-79bbd08cee228628] has joined #lisp 14:15:55 mulligan [n=user@e178018166.adsl.alicedsl.de] has joined #lisp 14:16:27 apparently so, yes :| 14:17:46 yes 14:17:56 [15:06] tic| What do I have to do to activate it? I currently start swank with (swank:create-server). 14:18:00 Sorry. 14:18:22 -!- ManateeLazyCat [n=user@222.212.132.169] has quit [Remote closed the connection] 14:19:09 (:emacs-rex (swank:arglist-for-echo-area '("+")) "COMMON-LISP-USER" t 3) gives (:return (:ok nil) 3). That's not really what it should be telling me, is it? 14:20:06 ManateeLazyCat [n=user@222.212.132.169] has joined #lisp 14:21:41 -!- blitz_ [n=julian@77.64.176.217] has quit [Read error: 60 (Operation timed out)] 14:24:19 hi, what is the equivalent of (member 1 '(1 2 3)) for strings. ie. (member "a" '("a" "b" "c")) 14:25:00 :test #'equal 14:25:12 weirdo: it's broken for me, too. 14:26:14 that's reassuring! 14:26:25 ecraven [n=nex@140.78.42.103] has joined #lisp 14:26:41 tcr, it's broken for me for both completion and arglist 14:27:13 tic: it must be '(("+") ...), it takes a list of raw form specs 14:27:58 adeht: thank you 14:29:47 teilzeitstudent [n=teilzeit@p54897705.dip.t-dialin.net] has joined #lisp 14:32:44 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 14:33:29 -!- dv_ [n=dv@85-127-114-35.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 14:34:03 -!- vy [n=user@213.139.194.186] has quit [Read error: 110 (Connection timed out)] 14:38:03 bakkdoor [n=bakkdoor@81-163-16-118.visitor.congress.ccc.de] has joined #lisp 14:39:55 jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has joined #lisp 14:40:44 does it matter if a function is referenced before it's defined in a file? is it implementation dependent? 14:40:55 -!- toddoon [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 14:41:35 jli: no it doesn't unless 14:41:58 jli: it may be declared inline, then the compiled result may be different, the execution will stay the same 14:41:59 jli: it works ok for me except I get a warning the very first time, that's with SBCL 14:42:18 -!- bakkdoor [n=bakkdoor@81-163-16-118.visitor.congress.ccc.de] has quit [Client Quit] 14:42:26 jli: or, if it's not a function, but a macro, then it may angry (right?) 14:43:37 jgracin [n=jgracin@82.193.208.195] has joined #lisp 14:43:43 toddoon [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has joined #lisp 14:46:57 jso [n=user@151.159.200.8] has joined #lisp 14:47:17 avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has joined #lisp 14:47:50 -!- matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 14:48:40 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 14:50:28 weirdo: Have you ever used that Parenscript incf function with a loop inside it? 14:51:37 -!- ManateeLazyCat [n=user@222.212.132.169] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 14:52:47 How dare fusss threaten my linux box! 14:53:09 kzar, i haven't really use parenscript much 14:53:26 weirdo: Ah ok 14:53:53 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has left #lisp 14:53:56 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 14:53:57 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has left #lisp 14:54:02 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 14:54:31 pjb annotated #72829 with "this might do what you want" at http://paste.lisp.org/display/72829#2 14:55:18 Dynetrekk: I'm not sure of :when ... :when ... in loop, check it. Otherwise you could use and format loop that way. 14:56:05 matimago: I changed it based on some suggestions... looks a bit better 14:56:07 but thanks ) 14:56:08 :) 14:56:35 matimago: you don't use a let? 14:56:54 One less form, one less indentation. 14:57:12 matimago: good point, just haven't seen anything like this before 14:57:34 Loop is self sufficient. If you know loop, you can forget all the rest of lisp :-) 14:57:45 matimago: wow. 14:58:02 matimago: btw, any "code convention" on how to make a "behind-the-scenes" routine name? 14:58:05 tcr: your slime slides are great! (I just read the pdf for the first time) 14:59:42 -!- oudeis [n=oudeis@bzq-79-177-106-178.red.bezeqint.net] has quit ["This computer has gone to sleep"] 15:00:02 what slides? 15:00:02 milanj [n=milan@93.86.112.165] has joined #lisp 15:00:06 I have a design question. In Rhino (the JavaScript engine whose parser mine is loosely based on), there is a Parser class which parses a TokenStream. This Parser class holds on to a Decompiler instance of its own; this Decompiler instance is used to print out source for functions (e.g. `alert(fooFunction.toSource)`). My question is, do you guys think it makes sense to put the logic for the decompiler directly in the parser class, since it's rather s 15:00:06 imple and there doesn't really seem to be any value in a separate decompiler class (the parser and decompiler are coupled to a great extent)? 15:00:07 tcr, so that emacs-rex-command was broken, you say? The exact data sent to Swank was 000048(:emacs-rex (swank:arglist-for-echo-area '("+")) "COMMON-LISP-USER" t 3) 15:00:14 Whoa, that was way longer than I expected. 15:00:51 tic: No, you have to pass it '(("+")) 15:00:55 Dynetrekk: what do you mean? 15:01:12 matimago: let's say I have a function (defun calculate n) 15:01:12 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 15:01:37 Aankhen``: I'd consider keeping it separate in some fashion, as it's clearly delineated functionality not having to do with parsing per se. 15:01:54 matimago: which would be the "interface", but want to factor out the actual calculation as, say, (defun calculation-real-routine n k), and k is some sort of counter - or whatever 15:02:11 Aankhen``: indeed. I would define a macro to declare grammar rules, to which I would provide all the information needed to both parse and generate a node. 15:03:02 vasa [n=vasa@mm-56-88-84-93.dynamic.pppoe.mgts.by] has joined #lisp 15:03:11 Dynetrekk: you can just define another top-level function like that, this is the easiest to debug, or you can also use local functions introduced with FLET or LABELS. 15:03:16 -!- reaver__ [n=reaver@212.88.102.158] has quit [Read error: 104 (Connection reset by peer)] 15:03:16 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 15:03:38 matimago: okay, I see. I just know that in some languages there is a convention for the names 15:03:57 -!- joast [n=rick@76.178.184.231] has quit [Read error: 110 (Connection timed out)] 15:04:26 Dynetrekk: you can use some naming convention if you use such a top-level function. For example, using a % to denote something "private" calculate%real-routine, calculate%something. 15:04:44 Dynetrekk: in emacs, they just append a -1, -2, etc to the function name, but it's less expressive. 15:05:17 matimago: hm, do you have a code example of the first? (Agree that it is less expressive:) ) 15:05:24 Dynetrekk: I'm not sure you need any special naming convention for such "behind-the-scenes" functions, "public" functions are generally those exported by the package and that may be enough 15:05:39 Dynetrekk: in some case I would use calculate/stuff calculate/otherstuff, but in general I'd use / for public variants, possibly for variants depending on the type of the arguments. calculate/string calculate/integer. 15:05:45 mvilleneuve: I was thinking code readability as well 15:06:43 Dynetrekk: well, once you've debugged the code, if the helper function is small enough and used only in CALCULATE it's probably more readable to have it as a flet/labels. 15:06:55 matimago: Hmm, I'm not sure how to do that here. I have a stream of `jess-token' instances (with specialized `jess-identifier', `jess-number', etc. classes) already from the tokenizer; the parser would basically take those and spit out nodes at a higher level of abstraction. 15:07:07 Aankhen: What are you working on? A JavaScript to CL compiler? 15:07:11 matimago: potentially not small.... but thanks for the insight... 15:07:23 Guest53748: Not really, just a parser for now. 15:07:41 nyef: Hmm, I guess that makes sense. 15:09:24 Aankhen``: you would work where the grammar is defined. All this structure would have to be replicated to generate a generator. So instead of duplicating code/structure, you would just have the macros used to declare the grammar, generate the generator too. For this, you may have to add some parameter (eg. some formating thunk), or not. 15:09:50 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit ["Leaving"] 15:10:04 Aankhen: If you haven't done so, you should read Water's pretty printing paper, the one where he shows how to pretty print Pascal code with CL's printer. You can probably leverage the same technique to convert your parsed code back into JavaScript text. 15:10:48 tcr, Thanks! 15:11:23 Here is the URL: http://www.merl.com/reports/docs/TR93-17.pdf 15:11:32 weirdo: There's a workaround. Instead of ,!p use an explicit (in-package :foo) at the REPL 15:13:54 roger 15:13:56 thanks 15:14:38 is that workaround for slime not being able to do completion? 15:14:51 I noticed that earlier 15:15:21 it's able, it's something else that causes it 15:15:24 I'll fix that later 15:15:55 oh, hey, it's a slime developer! *hugs* 15:16:37 dys``` [n=andreas@p5B31677C.dip.t-dialin.net] has joined #lisp 15:16:39 matimago: I understand what you're saying, but I'm pretty lost when it comes to writing (or even working with an existing) parser generator. I did look at a few like META, CL-YACC, etc.; I just didn't understand how to use them for this. :-S 15:17:15 Guest53748: I don't expect the printing to be particularly difficult. I'll take a look at the paper, though, since I never expected this tokenizer to take so long either. ;-) Thanks for the pointer. 15:19:11 Ah, I'd heard about this paper before but I'd never actually read it. 15:20:51 oudeis [n=oudeis@89-139-68-237.bb.netvision.net.il] has joined #lisp 15:22:04 -!- dys`` [n=andreas@p5B3141A8.dip.t-dialin.net] has quit [Connection timed out] 15:22:14 Wow. Old-school LOOP. 15:24:21 blitz_ [n=julian@77.64.176.217] has joined #lisp 15:26:48 Aankhen``: (the JavaScript engine whose parser mine is loosely based on) / I've got a parse error here. I autocorrected it into "you have written a parser in lisp loosely based on Rhino", but now you're contradicting it... 15:26:51 dys```` [n=andreas@p5B31677C.dip.t-dialin.net] has joined #lisp 15:27:20 Aankhen``: did you write a parser? 15:27:43 alpheus` [n=user@c-98-213-176-16.hsd1.il.comcast.net] has joined #lisp 15:27:56 I've written a tokenizer so far, the next step is a parser. 15:28:11 Er, my bad, I implied the parser was done already. 15:28:54 Aankhen``: ok. Whether you use an existing parser generator, or you write the parser yourself, you have to declare the grammar. 15:29:58 Aankhen``: basically, you have to write grammar rules. Eg. (defrule expression (alt (term) (term operator expression))) for: expression ::= term | term operator expression . 15:30:01 *nyef* tends to just write recursive-descent parsers, declaring the grammar implicitly. 15:30:28 woodz [n=woodz@host81-129-70-42.range81-129.btcentralplus.com] has joined #lisp 15:30:32 -!- spec[afk] is now known as mrSpec 15:30:37 Aankhen``: so you can take this set of rules, and have them processed by a defrule macro that generates a generator, or generates a parser, or both. 15:31:46 JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has joined #lisp 15:32:18 (defmacro defrule (lhs rhs) `(progn (defmethod generate ((self lhs)) ,(generate-generator-body rhs)) (defun ,(intern (format nil "PARSE-~A" lhs)) (source) ,(generate-parser-body lhs rhs)))) 15:32:20 -!- dys``` [n=andreas@p5B31677C.dip.t-dialin.net] has quit [Connection timed out] 15:32:37 or, if you use a parser generator, your defrule macro will only generate the generator. 15:32:55 and you will process the grammar file twice, once with the parser generator and once with your macro. 15:33:48 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 15:34:40 That would pretty much mean rewriting everything I have, I think… 15:35:10 Let me give you an example of what I have, since I'm a little confused now. 15:36:59 -!- knobo` [n=user@ti0073a340-0844.bb.online.no] has quit [Remote closed the connection] 15:38:17 -!- dialtone [n=dialtone@adsl-75-17-56-28.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 15:38:44 Aankhen pasted "Mixed up?" at http://paste.lisp.org/display/72834 15:40:24 matimago: At this point I'm starting to wonder if maybe I've got the terminology confused. 15:40:29 Jacob_H [n=jacob@92.0.38.174] has joined #lisp 15:41:04 afa [n=afa@131.152.178.51] has joined #lisp 15:41:48 How it works so far is: (make-instance 'jess-tokenizer :source "source") ... then keep calling GET-TOKEN, which reads from the string and returns comment tokens, identifier tokens, syntax tokens, etc. 15:42:26 -!- alpheus [n=user@vpn.cashnetusa.com] has quit [Read error: 110 (Connection timed out)] 15:42:40 -!- dys```` [n=andreas@p5B31677C.dip.t-dialin.net] has quit [Connection timed out] 15:43:00 What the parser does (in Rhino) is to keep reading tokens and combining them to form statements, expressions, blocks, functions, etc. 15:43:18 joshg [n=josh@S01060030bd077f55.va.shawcable.net] has joined #lisp 15:44:30 HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has joined #lisp 15:44:32 anyone know what would cause slime to load up an inferior lisp and then not create a repl buffer? 15:44:37 -!- Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has quit [Read error: 113 (No route to host)] 15:44:41 Heh. ";; Never -ever- thought I'd use CADADR." -- A comment in one of my parsers. 15:44:50 *Aankhen``* chuckles. 15:44:53 Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has joined #lisp 15:44:58 -!- seelenquell [n=seelenqu@tmo-100-230.customers.d1-online.com] has quit [Read error: 110 (Connection timed out)] 15:44:59 it starts up sbcl just fine, but then doesn't give me any repl at all 15:45:01 joshg: Something going wrong with starting the swank server? 15:45:12 joshg: Do you have the `slime-repl' contrib loaded? 15:45:14 joshg: It's a contrib now. 15:45:23 Oh good grief. 15:45:24 oh huh? 15:45:29 I guess that's a no. 15:45:32 *nyef* scratches his plans to update slime. 15:45:40 joshg: You should join the mailing list. 15:45:54 joshg: It was a Christmas present to all SLIME users: the REPL was moved into a contrib of its own. :-) 15:46:03 -!- egn_ [i=180f36f1@gateway/web/ajax/mibbit.com/x-79bbd08cee228628] has left #lisp 15:46:14 I believe `slime-asdf' automatically `require's it, so if you have that loaded it should be enough too. 15:46:31 hrm 15:46:38 would be nice if they'd mention that in the manual now 15:46:39 joshg: Use `slime-fancy' 15:46:52 it's not "they" 15:47:05 I'm trying to set this up on windows, I doubt asdf is going to make my life easier just yet 15:47:17 dys`````` [n=andreas@p5B31677C.dip.t-dialin.net] has joined #lisp 15:47:23 joshg: Just use slime-fancy like he said, then. 15:47:45 use the `slime-fancy' contrib, it's a meta contrib to load most of all useful contribs 15:48:34 okay 15:50:47 -!- alpheus` [n=user@c-98-213-176-16.hsd1.il.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 15:52:42 yay it works, thanks 15:53:18 Updating SLIME from CVS without watching the mailing list is fairly risky, unfortunately. 15:53:56 a repl is fancy, now? :) 15:55:28 -!- Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has quit ["Ex-Chat"] 15:56:20 the cvs snapshot is all the main slime page links to for download 15:57:07 Yeah, there really isn't any other option. 15:57:19 -!- shmho [n=user@58.142.15.103] has quit [Remote closed the connection] 15:57:37 dys``````` [n=andreas@p5B31677C.dip.t-dialin.net] has joined #lisp 15:57:42 ruediger [n=the-rued@p508B7AEB.dip.t-dialin.net] has joined #lisp 15:58:05 joshg: think yourself lucky; until last year, I'm pretty certain there was still an antique 'release' on the main page for download 15:58:12 which caused endless confusion 15:58:16 Dynetrekk pasted "why the errors?" at http://paste.lisp.org/display/72835 15:58:24 Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has joined #lisp 15:58:34 -!- afa [n=afa@131.152.178.51] has quit [] 15:58:41 dkcl [n=danderse@unaffiliated/dkcl] has joined #lisp 15:58:56 afa [n=afa@131.152.178.51] has joined #lisp 15:59:11 Dynetrekk: My first guess would be that you need to put the DO clause before the FINALLY clause. 15:59:17 Dynetrekk: FOR var from .. 15:59:23 nyef: did that before? 15:59:25 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 15:59:35 tcr: I don't need the FOR, do I? I thought I checked that... 15:59:37 Yes, for var from. Right. 15:59:41 I think you do 15:59:47 (lovely old loop) 16:00:00 ah 16:00:05 Yeah, if you're doing iteration control, you need FOR. If you're establishing a local variable you need WITH. 16:00:06 allright... thanks guys 16:00:13 Aankhen``: that's correct. May be my parameter named 'source' confused you, it represents the stream of tokens. For each phase, you have a 'source' and a product. The tokenizer = the scanner takes a stream of characters as source, and produces a stream of tokens. The parser takes this stream of tokens and produces a parse tree. 16:00:18 rsynnott: old? well, lisp is not? :P 16:00:44 <_3b> if you don't need the iteration var, you can use REPEAT 16:00:50 matimago: Right, that makes sense to me. 16:01:13 _3b: I'll stick to FOR for now, I'm new enough to loop as it is :P 16:01:20 -!- jlf` [n=user@netblock-68-183-235-250.dslextreme.com] has quit [Read error: 60 (Operation timed out)] 16:01:43 kuwabara_ [n=Kuwabara@cerbere.qosmos.eu] has joined #lisp 16:01:43 -!- kuwabara [n=Kuwabara@cerbere.qosmos.fr] has quit [Remote closed the connection] 16:01:45 -!- kuwabara_ is now known as kuwabara 16:02:51 kuwabara_ [n=Kuwabara@cerbere.qosmos.eu] has joined #lisp 16:02:51 -!- kuwabara [n=Kuwabara@cerbere.qosmos.eu] has quit [Remote closed the connection] 16:02:52 -!- kuwabara_ is now known as kuwabara 16:02:54 matimago: What I was thinking is… couldn't the decompiled representation of each type of token be the responsibility of the token itself? 16:03:04 -!- dys`````` [n=andreas@p5B31677C.dip.t-dialin.net] has quit [Connection timed out] 16:03:36 Aankhen``, man, how are you sending the unicode ellipses? does your client do that automatically? 16:03:51 nyef: where are some vops defined, and where are some used? 16:03:54 matimago: Along the lines of: (defmethod as-string ((token jess-token) (type (eql 'rp))) ")") 16:04:06 tcr: What sort of VOPs? 16:04:23 jli: No, it's a habit. I'm using X-Chat under Windows, the net result being that I can type in … for ellipses. :-) 16:04:26 Er. 16:04:28 That's %133. 16:04:36 nyef: found it, src/compiler/x86/... 16:05:06 Aankhen``, ho ho. okay. 16:05:18 Hmm, or maybe each token could have a slot that stores its string representation. 16:05:27 Ugh. Backtraces are going to be a pain, I can tell. 16:05:31 Even just lisp backtraces. 16:05:56 nyef: Are all VOPs interned in SB!VM? 16:06:21 tcr: Typically, yes. However, user-defined VOPs may not be. 16:06:53 (Well, some others may not be, but I think all vops defined in src/compiler/target/*.lisp are.) 16:07:29 nyef: What's an SC? 16:07:38 Storage Class. 16:07:57 That gets into register and stack allocation, and the whole debug thing. 16:08:22 See target/vm.lisp. 16:09:07 Basically, a storage base defines things like stack frames, integer-registers, floaty-registers, etc. 16:09:20 And a storage class is used to partition those further. 16:09:25 -!- ASau` is now known as ASau 16:10:44 nyef: How would you like arglists for VOP be presented? 16:11:00 VOP arglists? Hrm... 16:11:31 I really couldn't say, actually, as I rarely, if ever, mess around with %primitive or invoking VOP or VOP* directly. 16:12:03 really? I thought ir2 is really your thing :) 16:12:24 It sortof is, but I rarely need to mess with the IR2 converter. 16:12:50 And I much prefer to defknown and :translate than use %primitive. 16:13:01 Well, I'll decline messing with VOP arglists as of now then. 16:14:16 Yeah, looks like the only use-cases are for %primitive, vop, and vop*. 16:15:41 -!- yango_ is now known as yango 16:15:46 -!- jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has left #lisp 16:22:47 -!- jsoft [n=Administ@unaffiliated/jsoft] has quit ["Lost terminal"] 16:24:08 Aankhen``: exactly. The token themselves know how to print themselves. And the nodes in the parse tree know how to print themselves too. eg: (defmethod generate ((self if-statement) stream) (format stream "if ~A then~%~A~%else~%~A~%end~%" (if-condition self) (if-then self) (if-else self))) ; or whatever the syntax to generate may be. 16:24:40 jlf` [n=user@209.204.171.109] has joined #lisp 16:30:42 -!- afa [n=afa@131.152.178.51] has quit [] 16:32:23 -!- VityokOrgUa [n=user@193.109.118.130] has quit ["time to go"] 16:32:58 guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 16:34:01 nyef: Do you know why you can't macroexpand DEFINE-INSTRUCTION forms? 16:34:37 I've never tried, TBH. 16:34:45 Side-effects, maybe? 16:35:12 joast [n=rick@76.178.184.231] has joined #lisp 16:35:21 I'll file a bug anyways. 16:35:31 -!- jkantz [n=jkantz@ita4fw1.itasoftware.com] has quit ["Leaving"] 16:35:48 jgracin_ [n=jgracin@82.193.208.195] has joined #lisp 16:36:40 matimago: Great, thanks for being so patient while I sorted that out. 16:37:16 guys, is there a builtin func for summing lists? i.e. (sum (1 2 3)) => 6 ? 16:38:10 (reduce #'+ list) 16:38:55 ahaas: ah. thanks. 16:39:52 ahaas: #' means "the function also known as..." ? 16:40:29 #'x is just a shorthand for (function x) 16:40:34 right 16:40:42 so you can read about the special operator `function' 16:41:01 adeht: thanks! 16:41:09 And then you can read up on reduce, and see if it takes a function or a function designator as its first argument, and then the difference between the two... 16:42:13 (loop for item in list summing item) ; to subsume... 16:42:29 nyef: I see 16:42:30 -!- ecraven [n=nex@140.78.42.103] has quit ["bbl"] 16:43:06 matimago: what if `list' is actually a vector 16:43:15 nyef: since funcall and apply both take function designators wouldn't basically everything take a function desginator? 16:43:27 -!- jgracin_ [n=jgracin@82.193.208.195] has quit [Remote closed the connection] 16:43:35 segv_: ... No? 16:43:41 I'm getting an error here that confuses me: "don't know how to dump # (default MAKE-LOAD-FORM method called)." 16:44:10 nyef: i just don't see why a function would take a function object and not a function designator 16:44:13 This doesn't happen when I execute a function returning a register object, only in a macro which uses a register object. 16:44:22 since you'd have to explicitly add that check in 16:44:33 danderson: Sounds like you've got a literal register object showing up in the code the compiler sees. 16:44:40 -!- jgracin [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 16:44:47 adeht: a vector is a sequence, and reduce takes a sequence as input 16:44:51 So I'm guessing that I need an EVAL-WHEN somewhere, or to move something into code returned by the macro, rather than evaluated by the macro 16:45:09 danderson: And if that's actually what you wanted, have a look at the CLHS pages for MAKE-LOAD-FORM and LOAD-TIME-VALUE. 16:45:37 kuwabara: indeed.. but an `for .. in' loop clause works on lists, not sequences 16:46:10 segv_: A function could be declared to take a real function object, so the compiler need not emit an explicit conversion. 16:46:25 nyef: I'm not sure that's what I want, but if it is, I'll take a look. Right now changing my code so that the object doesn't get constructed by the macro. 16:47:26 adeht: sorry, I was mixing discussions 16:48:06 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 16:48:43 tcr: good point. 16:49:48 adeht: isn't across what you're looking for ? 16:50:23 kuwabara: adehts point was that loop for x in list summing is not equivalent to reduce #'+ sequence 16:51:27 nyef: so, if I provide an implementation of make-load-form, once the file is loaded, there will once again be a literal register object in that place, right? 16:51:35 danderson: Yes. 16:51:39 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 16:51:45 perfect, just what I need then. Thanks. 16:51:47 -!- Jacob_H [n=jacob@92.0.38.174] has quit ["Leaving"] 16:51:47 Is there any legitimate way to read that pretty printing paper by Water without subscribing to ACM? 16:51:57 I did find another PDF, but it's almost unreadable. 16:51:58 Aankhen``: it's on MERL's ftp, no? 16:52:02 xan [n=xan@62.57.40.18.dyn.user.ono.com] has joined #lisp 16:52:06 pkhuong: I think that's the one. 16:52:21 danderson: Or if you use LOAD-TIME-VALUE you can get the same effect through a different mechanism, which has different tradeoffs (and possibilities). 16:52:27 I doubt the ACM really has a better one. Many of their PDFs are scans. 16:52:50 It might be a bad PDF job. If you can find a PostScript version, try converting it yourself. 16:53:11 Aankhen``: they're all scans, I think. I don't remember the .ps being particularly bad. 16:53:27 Ah, I see. I'll keep looking. Thanks guys. 16:53:44 ahaas: oh, ok, I understand. Well, I see it as a huge error in CL spec: most functions taking a LIST as parameter should really accept any SEQUENCE. 16:54:22 nyef: in this case, the literal register appears in a quoted list representing an AST. Unfortunately, that part of the AST is generated by a macro, which is how the literal register gets to the compiler. I just need to get it to pass through as-is (or as near as possible) 16:54:38 afaict, that means I need a make-load-form that returns (make-instance 'register ...) 16:54:46 Sounds about right. 16:55:21 Or you could change your macro to generate the load-time-value form in place of the literal register. 16:55:24 Durr, I was searching for the pretty printing section in particular rather than the paper as a whole. 16:55:38 Found a PS of the entire paper. 16:55:59 nyef: ooh, that sounds better. Should have followed your suggestion to its conclusion. 16:56:02 Thanks 16:56:10 Aankhen``: where ? 16:56:35 danderson: that only works if the l-t-v form is evaluated. 16:56:40 danderson: Note that if you're expecting user code to generate register literals, then the load-time-value method could possibly be better. 16:57:55 pkhuong: does it suffice if it is evaluated at macro expansion time? 16:58:13 well, I guess I have a simple way to find out. 16:58:53 danderson: no, that won't do what you want. 17:00:47 aw, crud. 17:01:05 yeah, and the null lexenv also isn't helping. 17:01:25 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 17:01:28 well, make-load-form it is then. 17:02:13 younder [n=jpthing@062016236162.customer.alfanett.no] has joined #lisp 17:04:09 pucci [n=chatzill@82.85.85.34] has joined #lisp 17:04:31 -!- pucci is now known as caliostro 17:05:37 -!- binarycodes [n=sujoy@59.93.243.214] has quit ["WeeChat 0.2.6"] 17:06:07 fe[nl]ix: At . Even the PS seems to be the same quality, though, unfortunately. 17:06:57 Does anyone know of any projects with lisp and digital image processing? 17:07:15 I am working on a homebrew image recognition program and looking for the best language to use 17:08:22 you're at the right place 17:08:24 *grin* 17:08:34 awesome 17:09:02 there are some native CL image-format-parsing libraries, but no recognition per se being done 17:09:54 cool I'll look them up 17:10:05 lisp should be a great language to do this in 17:10:16 Luis_Byclosure [n=luis@a83-132-181-55.cpe.netcabo.pt] has joined #lisp 17:10:33 -!- Luis_Byclosure [n=luis@a83-132-181-55.cpe.netcabo.pt] has left #lisp 17:11:59 -!- mrSpec is now known as spec[afk] 17:12:20 drwhen: FWIW, I haven't seen much about image recognition, but in general you search for contours in the image (just points where a heuristic (say the color-difference between the given color and its neighbours) is high) and compare that with your reference cases. 17:12:34 valiza2 [n=haroldo@r190-133-142-218.dialup.adsl.anteldata.net.uy] has joined #lisp 17:12:40 ahaas: indeed, loop in/across is the bigest defect of loop. We will have to change that in CL2.0. In the meantime, let's use ITERATE. 17:13:59 Lol 17:14:51 is sbcl more efficient when using only one thread? 17:15:38 madnificent, how's going ? I'm writing the claw-central.org site (with claw of course :) ) 17:15:54 seelenquell [n=seelenqu@pD9E45EF9.dip.t-dialin.net] has joined #lisp 17:16:27 weirdo: than what at doing what? 17:16:47 I want to make is so to provide minimal cms capabilities 17:16:58 s/is/it/ 17:17:21 pkhuong, synchronization 17:17:33 locking overhead 17:17:45 weirdo: I guess he meant what would you do instead of using threads 17:17:49 SBCL with no threads is better than threaded SBCL at not using threads. 17:17:50 -!- oudeis [n=oudeis@89-139-68-237.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 17:18:24 for instance, there's a world-lock or a compiler-lock 17:18:33 are they noops on no threads? 17:19:25 They are no-ops on non-threaded builds. On threaded builds they still do things even when there is only one thread. 17:19:27 kiuma: I am rather curious. Talk to jbjohns too if you're willing. He seems to be on the verge of a new system too. I'm currently thinking in a lightweight approach myself (I'll show you something when/if it ever gets done). 17:20:10 but the operations the locking protect dwarf the cost of acquiring an uncontested lock. 17:20:11 however, I doubt the overhead of acquiring them when you're not actually using threads is significant. acquisition of a lock not held is pretty cheap with futexes. 17:20:13 join #lispcms 17:20:28 kiuma: I think you have to contact the people at common-lisp.net before I can show tickets 17:20:30 (if you want to talk about.... lisp cms's) 17:20:33 kiuma: that'll be for you too :) 17:20:49 Tordek [n=tordek@host14.190-137-244.telecom.net.ar] has joined #lisp 17:21:35 nyef, thank you. so swank is not the reason stumpwm is slow 17:21:57 Are you sure about that? 17:22:27 Is it still slow if you don't have swank loaded? 17:22:44 my methodology is flawed at best 17:22:44 jbjohns, great thx! 17:23:31 silly question: how to convert a (whatever-number) to an int? 17:23:33 it takes 2% cpu probably because i get some info from sysfs and procfs every few seconds 17:23:39 along the lines of (float 2) 17:23:42 Hun [n=Hun@tash.visitor.congress.ccc.de] has joined #lisp 17:23:43 and i wonder what's the cause 17:24:23 madnificent, I'd be glad to recrit coders for claw 17:24:40 Well, first is to see if it's holding the sysfs and procfs files open, and what the actual files are... 17:24:42 you can be in the list if you want 17:24:50 -!- Yuuhi [i=benni@p5483CBF7.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 17:24:56 nyef, i wrote the code myself 17:25:43 -!- avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has left #lisp 17:25:45 So... what files? 17:26:02 -!- toddoon [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has quit [Remote closed the connection] 17:26:40 /sys/class/hwmon/hwmon1/device/*_input through ffi since lunix broke posix api and sbcl hangs 17:26:55 and /proc/net/dev through sbcl 17:26:58 Umm... And are you polling those? 17:27:18 Or are you just monitoring the fds through select to see if they changed and then re-reading their contents? 17:27:56 nyef, every 5 seconds i open(2), read(2) and close(2) 17:28:02 Gah! 17:28:08 -!- valiza1 [n=haroldo@r190-133-130-238.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 17:28:15 and i use consing READ-LINE 17:28:27 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 17:28:32 every 5 seconds? READ-LINE is *not* your problem. 17:28:43 Look at read-usb-devices in http://www.lisphacker.com/temp/han14xs.lisp 17:28:58 binarycodes [n=sujoy@59.93.215.60] has joined #lisp 17:29:17 so stumpwm used 7 minutes of cpu 17:29:28 Then realize that if you hold the file open, you can use serve-event on the FD to know when it changes, and then you can re-read the content using the same trick instead of constantly re-opening it. 17:30:19 ha, so it seems musicmpd used 5 hours during 10 days 17:30:27 s/musicmpd/musicpd 17:32:31 Hun` [n=Hun@tash.visitor.congress.ccc.de] has joined #lisp 17:33:20 nyef, thanks 17:33:34 Anyway, polling sysfs and procfs files that way is just not a good idea as far as I'm concerned. Linux broke the usual semantics of select() and poll() on those files for a reason, and working with that reason seems better than ignoring it. 17:33:37 and no point in polling, since it changes all the time 17:33:42 communication blocks till you get a response from the OS, so use a seperate thread and communicate by queue 17:34:15 nyef: How do their semantics differ there? 17:34:38 tcr: The readiness state doesn't mean "there's more data", it means "data has changed". 17:35:02 decafbad [n=mehmet@88.232.13.83] has joined #lisp 17:35:03 tcr: So the usual wait-until-fd-usable on these files causes SBCL to not finish reading the file until the data has changed... 17:35:32 But reads on these files never block -anyway-. 17:35:59 nyef: doesn't select+lseek+read work ? it it necessary to reopen the file each time ? 17:36:17 fe[nl]ix: It's not necessary to re-open. It might be necessary to seek back to the beginning. 17:37:04 But, basically, when you first open the file, and then every time it comes ready via select(), just read out the entire file until it returns EOF. 17:37:20 Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has joined #lisp 17:37:49 On a normal file, if a read returns EOF and select says it is not ready for reading, there is more data to come. 17:38:11 For these files, that's not the case, so SBCL's fd-streams screw up. 17:39:16 saikat [n=saikat@user-12lc4ta.cable.mindspring.com] has joined #lisp 17:39:46 *robyonrails* football practice 17:40:37 robyonrails: I don't think we care. 17:40:39 Oh, neat. I have an example here of messing with an MMIO PCI range from SBCL. 17:40:44 I'd forgotten about this. 17:40:58 tcr: re DEFINE-INSTRUCTION, the problem is that once the disassembler tables are compiled, the metadata used to build them is cleared. 17:42:38 pkhuong: I thought something like that. I won't complain against a WONTFIX, having reported is still useful to point other people to it. 17:43:32 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 17:44:28 xinming_ [n=hyy@125.109.254.196] has joined #lisp 17:44:29 -!- Hun` [n=Hun@tash.visitor.congress.ccc.de] has quit [Read error: 60 (Operation timed out)] 17:44:36 robyonrails: I don't think we care. <-- just jealous 17:45:01 robyonrails: No. Please keep irrelevant twittering elsewhere. 17:46:29 -!- xan [n=xan@62.57.40.18.dyn.user.ono.com] has quit ["Leaving"] 17:46:42 xan [n=xan@62.57.40.18.dyn.user.ono.com] has joined #lisp 17:49:15 -!- Hun [n=Hun@tash.visitor.congress.ccc.de] has quit [Connection timed out] 17:50:05 -!- caliostro [n=chatzill@82.85.85.34] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"] 17:50:46 nyef: fd-streams screw up in using serve-event the way they do 17:50:48 -!- kzar [n=kzar@hardwick.demon.co.uk] has quit [Read error: 110 (Connection timed out)] 17:51:14 fe[nl]ix: Yes and no. 17:51:17 bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has joined #lisp 17:51:25 fe[nl]ix: they only assume they'll be used on FDs that act like normal posix FDs. 17:51:34 This is one of those places where whichever way you look at it, -somebody- loses. 17:52:24 nyef: what do you mean ? 17:53:31 I mean the interactions between fd-streams, sb-bsd-sockets, pathnames, and serve-event are intricate and have some nasty tradeoffs and gotchas. 17:53:42 -!- guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 17:53:50 -!- Davidbrcz [n=david@ANantes-151-1-65-111.w81-250.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 17:54:11 Davidbrcz [n=david@ANantes-151-1-96-175.w86-195.abo.wanadoo.fr] has joined #lisp 17:54:16 And there really isn't a way to fix it so that everything works the way everybody wants it to, because some people want incompatible things. 17:56:04 qen [n=unknown@84.92.70.37] has joined #lisp 17:56:32 pkhuong: there is at least one way to make streams work with regular and virtual files 17:57:24 -!- xinming [n=hyy@125.109.244.117] has quit [Read error: 110 (Connection timed out)] 17:58:06 -!- trebor_home [n=user@dslb-084-058-206-073.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 18:03:55 m_____a [n=ma@189.122.82.91] has joined #lisp 18:04:41 Is it possible to determine the policy a function was compiled with? 18:05:45 -!- netaust1n [n=austinsm@cpe-67-243-48-35.nyc.res.rr.com] has quit [Read error: 110 (Connection timed out)] 18:05:50 gottesmm [n=gottesmm@99-186-122-131.lightspeed.hstntx.sbcglobal.net] has joined #lisp 18:06:10 -!- gottesmm [n=gottesmm@99-186-122-131.lightspeed.hstntx.sbcglobal.net] has left #lisp 18:09:10 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit ["Sto andando via"] 18:12:15 guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 18:12:43 xvc [n=chatzill@82-69-45-88.dsl.in-addr.zen.co.uk] has joined #lisp 18:13:11 jsnell: Could you perhaps share some expertise to the disassemble output issue at the bug tracker. You recently pointed me to your pc-to-source-location frobbery which made me look confident that my wish actually could be implemented. 18:13:47 mbac [i=[LkF8Bau@panix2.panix.com] has joined #lisp 18:14:46 jgracin [n=jgracin@82.193.208.195] has joined #lisp 18:15:46 what's the library i want to use to open and rasterize a postscript file? 18:16:20 how do I test if an object is an instance of a given class? 18:16:46 or if there is a method that can be called with given arguments, that would also work for me. 18:16:47 netaust1n [n=austinsm@160.79.78.72] has joined #lisp 18:18:00 danderson: typep. but one has to wonder whether doing that is actually a good idea 18:18:08 it CAN be, but why not use a generic function? 18:18:26 (class objects, and their names, are type specifiers, so you can use typep, typecase and friends) 18:18:42 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 131 (Connection reset by peer)] 18:18:48 for general learning also see type-of and class-of. 18:18:53 well, the scenario is: I have a list of field definitions, and in my code generator I need to generate assignments for all fields that are instances of a subclass of writable-operand. 18:18:56 (the difference between them might be interesting) 18:19:09 the way I see it, I have two options 18:19:24 do you generate things other than "assignments" as well? 18:19:26 either implement a completely generic method returning nothing, along with more specialized methods 18:19:40 maybe a progn or nconc method combination would be good 18:19:41 or check the type before calling, and collect only useful output 18:20:05 "a completely generic method returning nothing" sounds good 18:20:07 I don't see what method combination has to do with things here... 18:20:28 for each operand type, there is exactly one method that generates the code needed for the assignment 18:20:41 jao [n=user@244.Red-83-40-177.dynamicIP.rima-tde.net] has joined #lisp 18:20:58 (well, one per target language type, so there is multi-dispatch going on, but within a single target, the dispatch is quite simple) 18:22:20 but in your opinion, doing unconditional dispatch and testing for nil return values is better then? 18:22:30 ...testing? 18:22:41 what would you do as a result of that test? 18:22:51 typep is tricky, you probably want subtypep 18:23:20 younder: um, no. typep is not nontransitive 18:23:41 kpreid: well, discard nils. I'm accumulating (string) return values into a list, for concatenation 18:23:50 I suspect CONCATENATE won't like the nils 18:23:51 danderson: have your GF return a list 18:23:58 either nil or a list of one element 18:24:05 then just concatenate all the bits 18:24:16 well, actually 18:24:27 (concatenate 'string "foo" nil "bar") works just fine 18:24:38 unless you want a single string as a return value, in which case you can return an empty string. 18:25:00 heh. sure, you could just as well...what pkhuong said 18:25:07 concatenate doesn't care what kind of sequences its arguments are 18:25:50 kpreid: The point is you don't know the type at run time. If you ask for integer and it is fixnum you get false. Or you can ask for fixnum and the system reads bit. Therefore check for the root of te type tree which is relevant. 18:26:10 well, apparently it does care whether it can actually make nil into a string type, but empirically, it does work 18:26:32 younder: (typep 1 'integer) is true whether or not (type-of 1) is BIT 18:26:46 younder: typep respects subtype relationships 18:26:53 not portably 18:27:03 false 18:27:13 typep is accurate. subtypep is sometimes inaccurate 18:27:38 (typep a b) is true whenever (subtypep (type-of a) b) is 18:28:50 chaitanya [n=chaitany@122.162.15.105] has joined #lisp 18:29:08 Oh, about that.. is a type subtypep of itself? 18:29:12 yes 18:29:30 I think abcl doesn't implement that little detail correctly. 18:30:50 -!- blitz_ [n=julian@77.64.176.217] has quit [Read error: 60 (Operation timed out)] 18:32:19 can gengc run in constant space? 18:32:33 -!- mgr [n=mgr@psychonaut.psychlotron.de] has quit [Read error: 145 (Connection timed out)] 18:32:36 weirdo: what does that mean? 18:37:18 -!- jao [n=user@244.Red-83-40-177.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 18:37:20 jao [n=user@244.Red-83-40-177.dynamicIP.rima-tde.net] has joined #lisp 18:37:45 weirdo: Could you run M-x slime-run-tests for me, please? 18:39:29 -!- woodz [n=woodz@host81-129-70-42.range81-129.btcentralplus.com] has quit [] 18:40:28 tcr, on HEAD? 18:40:28 ok 18:41:47 no, not on head 18:43:12 josemanuel [n=josemanu@193.1.222.87.dynamic.jazztel.es] has joined #lisp 18:43:27 pending RPCs or open debuggers 18:44:06 how w8 18:44:07 s/how // 18:44:12 basically every test failed? 18:44:40 no 18:44:44 i had some open debuggers 18:44:53 which i couldn't close, since they popped up immediately 18:44:57 start a fresh emacs for running the tests 18:45:04 weirdo, yes: the number of bytes used by gen[c]gc is bounded by 2^64 (or 2^32 if you are lucky enough to have a 32-bit machine). 18:45:16 sometimes slime tries to READ stuff with backquote ,-commas 18:45:31 Riastradh, that's reassuring 18:46:45 tcr, could you fix breaking on quasiquote, btw? :P 18:47:15 tests running 18:47:17 -!- xvc [n=chatzill@82-69-45-88.dsl.in-addr.zen.co.uk] has quit ["Chatzilla 0.9.75.1 [SeaMonkey 1.1.14/0000000000]"] 18:48:03 You could report a test case to the mailing list. 18:48:21 Riastradh: 2^48, for now. 18:50:07 tcr, can the ML support posting from gmane? 18:51:00 okay 18:51:04 got a reproducible case 18:51:40 weirdo: Yes, it does. I post through gmane, too. 18:51:50 weirdo pasted "how to make sldb appear" at http://paste.lisp.org/display/72839 18:52:19 this is 2008-12-07 18:52:37 tcr: do you recommend updating SLIME to HEAD? 18:52:52 tcr, failed on 33 tests 18:53:00 0 expected 18:53:11 same version, 20081207 18:53:37 You should report it to the mailing list. I do not have time, nor inclination to fix random bugs that do not affect me immediately. 18:53:54 pkhuong: No, there's still a major bug causing completion and arglist display fail. 18:54:05 I'm going to fix it. 18:54:34 weirdo pasted "slime tests" at http://paste.lisp.org/display/72840 18:54:40 thank you. 18:54:46 -!- H4ns [n=hans@p57A0E7A8.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 18:57:14 avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has joined #lisp 19:02:54 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 19:04:07 mjf [n=mjf@r6y163.net.upc.cz] has joined #lisp 19:04:24 Ok I fixed it in CVS. Wait half an hour, then you can update. 19:05:30 -!- chandler [n=chandler@opendarwin/developer/chandler] has quit [Remote closed the connection] 19:06:36 -!- m_____a [n=ma@189.122.82.91] has quit [Client Quit] 19:08:02 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 19:11:00 dankna [n=d@ool-43516bc9.dyn.optonline.net] has joined #lisp 19:16:00 -!- jajcloz [n=jaj@pool-98-118-118-197.bstnma.fios.verizon.net] has quit [] 19:17:18 <_3b> tcr: how extensible is slime arglist display? 19:19:47 which is a more specific type for nil? list or symbol? 19:19:54 demmeln: null. 19:22:32 pkhuong: but nil is of type list and symbol aswell 19:22:36 rsynnott: is your article about working with cl-facebook still online? all the links google exposes seem dead. 19:23:09 which of those 2 is more specific. or where can i find general information about the hirachy of the nil type 19:23:20 i'm a bit confused about that 19:23:23 !clhs nil 19:24:04 demmeln: neither is more specific. 19:25:31 sledge: No leading exclamation point on clhs queries. 19:26:05 so if i have two methods specialising on type list and symbol 19:26:07 demmeln: the intersection of the types LIST and SYMBOL is exactly the type NULL, which only contains the value NIL. However, not all all symbols are lists and not all lists are symbols, thus they're incomparable. 19:26:13 which one will be called? 19:26:43 if i pass nil 19:27:42 clever question. no clue. 19:27:54 <_3b> is 'specific' defined anywhere? 19:27:55 oh i found this line in the clhs for the type null 19:28:01 Class Precedence List: null, symbol, list, sequence, t 19:28:16 so it should be symbol 19:28:24 if i interpret is right 19:28:44 _3b: yes, in the context of effective method selection. 19:28:47 demmeln: right. 19:29:17 demmeln: a third explicit method for NULL might be clearer. 19:30:11 pkhuong: so this is just the (arbitrary) way it is defined in the spec. might aswell have been the other way around... 19:30:21 demmeln: right. 19:30:50 pkhuong: yeah sure thing. i just wanted to find out which method is called to find a bug. turns out it was the one specified for SYMBOL 19:30:56 thanks guys 19:31:28 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 19:31:48 schme_ [n=marcus@c83-249-80-232.bredband.comhem.se] has joined #lisp 19:32:12 demmeln: you could have used COMPUTE-APPLICABLE-METHODS. 19:32:21 jlf [n=user@209.204.171.109] has joined #lisp 19:32:51 <_3b> pkhuong: which section has the definition? 19:33:30 tomsw [n=user@mail2.siscog.pt] has joined #lisp 19:33:41 clhs 7.6.6.1.2 19:33:41 http://www.lispworks.com/reference/HyperSpec/Body/07_ffab.htm 19:34:08 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [Read error: 145 (Connection timed out)] 19:34:27 -!- pjb [n=t@81-66-196-92.rev.numericable.fr] has quit [Remote closed the connection] 19:34:35 <_3b> pkhuong: ah, thanks 19:36:13 pkhuong: ah cool thanks for that tip. didn't know that. 19:36:40 -!- JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has quit ["Leaving"] 19:37:12 blitz_ [n=julian@77.64.176.217] has joined #lisp 19:37:16 -!- jlf [n=user@209.204.171.109] has quit [Client Quit] 19:40:17 rpg [n=rpg@fw.henn.dunn.pcspeed.com] has joined #lisp 19:44:02 -!- schme [n=marcus@sxemacs/devel/schme] has quit [Read error: 110 (Connection timed out)] 19:46:48 jgracin_ [n=jgracin@82.193.208.195] has joined #lisp 19:47:41 ikki [n=ikki@201.155.75.146] has joined #lisp 19:48:36 grrrr... buggy clim inspector.... 19:49:47 -!- malumalu_ [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 19:50:21 -!- ruediger [n=the-rued@p508B7AEB.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 19:51:17 -!- dfox [n=dfox@r5cv134.net.upc.cz] has quit [Remote closed the connection] 19:51:22 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 19:55:56 -!- jgracin [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 20:00:52 -!- Aankhen`` [n=Aankhen@122.163.225.159] has quit ["“You think you can *stop* me with a BUBBLE?” [boom] “It’s a really good bubble.”"] 20:01:42 gottesmm [n=gottesmm@99-186-122-131.lightspeed.hstntx.sbcglobal.net] has joined #lisp 20:02:09 -!- gottesmm [n=gottesmm@99-186-122-131.lightspeed.hstntx.sbcglobal.net] has left #lisp 20:02:21 moghar [n=user@unaffiliated/moghar] has joined #lisp 20:04:40 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 20:04:45 Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has joined #lisp 20:05:44 -!- valiza2 [n=haroldo@r190-133-142-218.dialup.adsl.anteldata.net.uy] has left #lisp 20:05:54 -!- decafbad [n=mehmet@88.232.13.83] has quit ["Leaving"] 20:08:01 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [] 20:08:57 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 20:09:20 greetings! 20:12:01 greetings fusss 20:13:51 huh, sad news on planet.lisp 20:15:11 anything new in the world of hunchentoot and html generation macros that pprint their results readably? :-P 20:16:44 *fusss* is looking for a firefox plugin that might be able to do that 20:17:20 <_3b> you mean indenting the html nicely? 20:18:05 -!- chaitanya [n=chaitany@122.162.15.105] has left #lisp 20:18:20 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 20:18:28 milanj: fuck. man. I'm shocked how much that knocked the wind out of me for a guy I never met. T_T 20:20:06 html "IDEs" haven't changed since CoffeeCup, circa 1997 20:20:12 -!- mulligan [n=user@e178018166.adsl.alicedsl.de] has quit [Remote closed the connection] 20:20:50 I was actually wondering about him a few hours ago -_- 20:23:16 -!- dkcl [n=danderse@unaffiliated/dkcl] has quit [Remote closed the connection] 20:25:31 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 20:25:58 -!- josemanuel [n=josemanu@193.1.222.87.dynamic.jazztel.es] has quit ["Saliendo"] 20:26:50 Jacob_H [n=jacob@92.0.250.208] has joined #lisp 20:27:36 -!- jso [n=user@151.159.200.8] has quit ["going home."] 20:28:10 optikalmouse [n=user@bas1-toronto10-1279556315.dsl.bell.ca] has joined #lisp 20:29:35 -!- kiuma [n=kiuma@81-208-106-75.ip.fastwebnet.it] has quit ["Bye bye ppl"] 20:29:37 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 20:31:42 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [SendQ exceeded] 20:32:23 -!- jpcooper [n=justin@unaffiliated/jpcooper] has quit [Remote closed the connection] 20:33:04 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 20:34:05 :-/ 20:35:22 pretty much. 20:37:33 -!- ASau [n=user@193.138.70.52] has quit [Read error: 104 (Connection reset by peer)] 20:38:37 -!- nicolas [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has quit ["Lost terminal"] 20:38:48 from now on I will make sure every line of code I write somehow relies on symbol preceding list in null's CPL 20:39:56 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Remote closed the connection] 20:42:46 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 20:44:07 decafbad [n=mehmet@88.232.131.111] has joined #lisp 20:44:28 [ot] my cat was distressed by the sound coming from the headphones lying on top of a cabinet. so i showed him my pressing a keyboard button pausing and restarting the sound, that stopped his being distressed 20:45:02 weirdo: Do you know that video of the cat examining a theremin? 20:45:13 no, what's a theremin? 20:45:34 my cat's a very smart purring fuzzy creature, i blinked to him through a mirror and he blinked back 20:45:40 The very idea of a cat messing with a theremin is amusing. 20:45:54 weirdo: An electronic instrument from the 1920ies which is the only musical instrument that is played by not touching it. 20:45:59 gottesmm [n=gottesmm@66.60.241.7] has joined #lisp 20:46:08 That theremin from the video is only ~$20, IIRC. 20:46:09 nyef: once used to the noise, they don't mind them much 20:46:24 bah, you should get a Good theremin, like a Moog 20:46:27 weirdo: You have two antennas. Get one of your hands closer to one of the antenna and the tone gets louder. Get the other hand closer to the other antenna and the pitch goes up. 20:46:34 my cat's a very smart purring fuzzy creature, i blinked to him through a mirror and he blinked back 20:46:38 now do the same thing but don't blink 20:46:42 he'll still blink :-) 20:46:48 ahaas: Yes, unfortunately, it's just an annoying toy :) 20:46:59 ahaas: But I loved the puzzled look of that cat :) 20:47:03 sohail, why? 20:47:23 blinking means 'everything cool for me, how it for you?' 20:47:40 Jabberwockey: iirc, it's not the only instrument not touched 20:47:43 Man, there are now even more videos of cats with theremins... 20:47:51 laserharps and other oddities probably qualify 20:47:59 caliostro [n=chatzill@78.14.8.214] has joined #lisp 20:48:05 lucca: Ah, true... 20:48:18 my cat also pressed the space bar and chased the cursor on the screen 20:48:49 but now he's very careful as not to touch the keyboard 20:49:20 there's no point in scolding cats, but begging, pleading and bribing works well 20:49:34 *LOL* 20:49:38 I wonder if that works on IRC 20:49:42 weirdo: Or let other cats do the punishment. 20:50:09 weirdo: A friend of mine had three male cats. One of them tried to pull the "poor little kitty" number when we were around, after it had one leg amputated. 20:50:25 weirdo: The other two cats quickly showed him with their claws what they thought of that. 20:50:38 weirdo: No more "poor little kitty", so he got the message, I guess. 20:50:44 weirdo: http://basicinstructions.net/?p=539 20:50:50 wdouglas [n=user@c-24-21-34-206.hsd1.wa.comcast.net] has joined #lisp 20:51:20 Borat^ [i=cracked@AVelizy-152-1-34-85.w82-120.abo.wanadoo.fr] has joined #lisp 20:51:51 -!- teilzeitstudent [n=teilzeit@p54897705.dip.t-dialin.net] has quit ["Leaving"] 20:53:54 hefner: what sort of climadness have you been up to lately? 20:53:59 Jabberwockey, what's a "poor little kitty"? 20:54:15 and why did the poor cat have one leg amputated? 20:54:43 -!- wdouglas [n=user@c-24-21-34-206.hsd1.wa.comcast.net] has quit [Client Quit] 20:54:54 -!- blitz_ [n=julian@77.64.176.217] has quit [Read error: 148 (No route to host)] 20:55:25 weirdo: Car accident. My friend had a cat flap so that the cats could come and go as they pleased. 20:55:50 that comic is exemplary of animal cruelty 20:55:52 weirdo: And one day one of the cats came back with one paw almost detached and a pretty fierce head wound 20:55:57 scolding cats i animal cruelty 20:56:05 Jabberwockey, :( 20:56:08 cars should be banned 20:56:14 i'd scatter glass on the road 20:56:18 or pins, or razors 20:56:34 weirdo: And "poor little kitty" is the strategy of a cat to induce pity in the humans to be treated to more treats and all. 20:56:38 "that's for Morris, you sadists!" 20:57:06 weirdo: Essentially, the cat was very fast on his three legs. But the moment he appeared in the door, he started limping. 20:57:14 my cat's name is  20:57:34 ahem 20:57:44 #cat, please (unless cdiggin already registered that). 20:57:45 he's male, but i really like the russian word. hope it doesn't insult Him in any way 20:58:23 pkhuong, Krystof: aren't cats typical to Lisp? 20:58:24 pkhuong: I haven't really done any CL lately (except for just now, as I'm hacking up something to compare directory trees and figure out what to restore from backup) 20:58:32 s/typical/topical 20:58:48 weirdo: No, cats aren't, but cars are ;-) 20:58:50 weirdo: Only kittens of death, sorry. 20:59:08 Even very long cars like this one: caaaar :) 20:59:09 since you clearly marked your initial statement with [ot], you even know the answer. Last warning 21:00:37 egn [i=tux@nodes.fm] has joined #lisp 21:00:38 Hrm... Nothing obvious on cliki for ELF or DWARF... 21:00:58 nyef: What's ELF / DWARF? 21:01:10 nyef: I seem to remember something on cmucl-imp years back to parse ELF 21:01:12 binary-types should make that one easy... 21:01:14 executable format / debugging info format 21:01:20 Jabberwockey: ELF is an executable linking format of some sort. DWARF is debug information format. 21:01:42 Ah, THAT ELF. 21:01:58 I thought that there might be some lisp-package of said name. 21:02:04 Krystof: I'm more interested in the DWARF side of things, actually. 21:02:07 Krystof: SBCL doesn't pass the tests on darwin/x86-64. Bisecting... 21:02:48 I'm going through my simple debugger and trying to figure out what is needed to make it minimally useful. 21:03:50 pkhuong: OK, thanks 21:04:04 apart from that, I have the patch to make it work on sparc from Juho 21:04:05 anything else? 21:04:21 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 21:04:28 1.0.24 release cycle? 21:04:43 tomorrow 21:04:52 Ah. Cool. 21:05:01 unless pkhuong or someone else tells me not 21:05:18 Only thing in my pending queue can easily wait, and probably will for months. 21:05:56 oudeis [n=oudeis@bzq-79-177-106-178.red.bezeqint.net] has joined #lisp 21:06:09 (And that's the sub-{access,set}-debug-var-slot unification.) 21:06:22 Krystof: I'll wait for a release before committing my :conditional patch set of doom (: 21:06:35 nyef, FWIW freebsd has a manpage on ELF but you probably already know the format 21:06:59 nyef: Is libgdb2 too crappy? 21:07:04 weirdo: Yeah, I'm not really worried about figuring out the format... 21:07:14 pkhuong: libgdb2? 21:07:36 it's FFI, FFI is evil, therefore libgdb2 is evil 21:07:45 Well, yes, there is that. 21:07:57 nyef: and you'd fight with gdb for the signals. 21:08:14 Actually, my position is that gdb itself is evil. 21:09:16 what's the licensing status? GPL or LGPL? 21:09:22 nyef pasted "simple-debug" at http://paste.lisp.org/display/72847 21:10:54 That's about where I am right now. And, as far as it goes, it works, too. 21:11:38 fwiw, i found the various libelfs crap, for manipulating assembly written binaries 21:11:42 I can even call (register-value process ) for the basic x86-64 GPRs. 21:11:53 clhs symbol 21:11:53 http://www.lispworks.com/reference/HyperSpec/Body/t_symbol.htm 21:12:14 all the little binaries from the "whilwind tutorial on teensie elf executables" pretty much choke most libelfs 21:13:32 -!- sjbach [n=sjbach__@ita4fw1.itasoftware.com] has quit [Read error: 110 (Connection timed out)] 21:13:33 -!- jollygood [n=jollygoo@129.71.215.161] has quit [Read error: 110 (Connection timed out)] 21:14:22 merAch` [n=none@c-71-199-20-205.hsd1.co.comcast.net] has joined #lisp 21:14:29 Jasko2 [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 21:15:46 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 21:15:50 -!- jgracin_ [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 21:15:52 moghar` [n=user@157.185.jawnet.pl] has joined #lisp 21:16:38 -!- moghar [n=user@unaffiliated/moghar] has quit [Nick collision from services.] 21:16:41 -!- moghar` is now known as moghar 21:16:59 m4dnificT [n=user@83.101.62.132] has joined #lisp 21:17:14 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 21:17:46 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- jlf` [n=user@209.204.171.109] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- madnificent [n=user@83.101.62.132] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- m4dnificent [n=user@83.101.62.132] has quit [orwell.freenode.net irc.freenode.net] 21:17:46 -!- specbot [n=specbot@common-lisp.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- djkthx_ [n=yacin@glug.id.iit.edu] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- phadthai [i=mmondor@ginseng.pulsar-zone.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- nyef [n=nyef@64.222.164.52] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- tltstc [n=tltstc@cpe-76-90-92-154.socal.res.rr.com] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- merAch [n=none@c-71-199-20-205.hsd1.co.comcast.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- kidd [n=kidd@70.Red-83-36-89.dynamicIP.rima-tde.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- Riastradh [n=rias@pool-141-154-217-174.bos.east.verizon.net] has quit [orwell.freenode.net irc.freenode.net] 21:17:47 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [orwell.freenode.net irc.freenode.net] 21:18:01 madnificent [n=user@83.101.62.132] has joined #lisp 21:18:12 what's so wrong with ffis, anyway? it's not like every C library implicitly asociates singal handlers 21:18:30 phadthai [i=mmondor@ginseng.pulsar-zone.net] has joined #lisp 21:18:43 and finalization data can be attached to FFI values 21:19:08 Riastradh [n=rias@pool-141-154-217-174.bos.east.verizon.net] has joined #lisp 21:19:09 djkthx [n=yacin@glug.id.iit.edu] has joined #lisp 21:20:03 nyef [n=nyef@64.222.164.52] has joined #lisp 21:20:34 -!- caliostro [n=chatzill@78.14.8.214] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"] 21:20:44 FFI is wrong because it's an extra thing to keep track of. 21:21:16 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 21:21:18 -!- netaust1n [n=austinsm@160.79.78.72] has quit [Read error: 145 (Connection timed out)] 21:21:22 I don't like CFFI, for example, partly because it's an extra dependency. 21:21:32 maybe also because stuff has to be converted to alien values, but that's me, always worrying about consing 21:21:37 -!- AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has quit [Remote closed the connection] 21:21:40 moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has joined #lisp 21:21:53 nyef: i second that. in the same vein, i bet you are a big fan of hunchentoot then? :) 21:22:11 nyef, library written for NIH value is an extra dependency too :) 21:22:13 hypno: I'm currently a big fan of static web sites. :-P 21:22:19 heh. 21:22:20 tltstc [n=tltstc@cpe-76-90-92-154.socal.res.rr.com] has joined #lisp 21:22:26 AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has joined #lisp 21:22:34 an NIHer who replaces one library with another is not a true NIHer 21:22:53 (or maybe they are, and I shouldn't want any association with them) 21:23:07 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 21:23:14 malumalu [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has joined #lisp 21:23:24 Well, take libusb for example. 21:24:17 I had a choice: Bind against libusb when I can't find a single example of a program that uses libusb that works on my system, or bind directly against the kernel interface. 21:24:30 dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has joined #lisp 21:24:33 _3b annotated #72839 with "hackish fix" at http://paste.lisp.org/display/72839#1 21:24:43 <_3b> weirdo: see that ^ 21:25:00 To bind the kernel interface, I had to lay out a handful of structures and figure out how to define some constants. 21:25:02 <_3b> (unrelated to current topic) 21:26:05 jlf [n=user@209.204.171.109] has joined #lisp 21:26:07 And the whole ended up becoming lh-usb, a library. 21:26:38 it's not a library until someone else uses it ;) 21:27:04 kidd [n=kidd@70.Red-83-36-89.dynamicIP.rima-tde.net] has joined #lisp 21:27:11 That's a fairly arbitrary definition. 21:27:27 -!- gottesmm [n=gottesmm@66.60.241.7] has left #lisp 21:28:04 nyef: not even lsusb works on your system ? 21:28:22 fe[nl]ix: I thought that worked in terms of /proc/bus/usb/devices? 21:28:37 Hunh. Guess not. 21:28:52 See, I tried usb-robot, and that sucked horribly. 21:29:16 -!- replaca [n=tom@76-191-193-111.dsl.dynamic.sonic.net] has quit [Read error: 60 (Operation timed out)] 21:29:49 no, usbutils use libusb 21:30:18 it opens stuff in /dev/bus/usb/ 21:30:26 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 21:30:59 That's... neat. 21:31:23 *nyef* wonders what the difference is between the "regular" files in /proc/bus/usb/ and the device files in /dev/bus/usb/... 21:31:41 bombshelter13 [n=bombshel@209-161-240-155.dsl.look.ca] has joined #lisp 21:32:23 the file system of /proc/bus/usb/ is usbfs, the files in /dev/bus/usb/ are created on the fly by udev 21:32:43 not a great explanation, I guess 21:33:58 -!- xan [n=xan@62.57.40.18.dyn.user.ono.com] has quit [Read error: 104 (Connection reset by peer)] 21:34:32 -!- dcrawford [n=dcrawfor@ita4fw1.itasoftware.com] has quit [Read error: 113 (No route to host)] 21:34:40 replaca [n=tom@76-191-193-111.dsl.dynamic.sonic.net] has joined #lisp 21:34:40 -!- bfein [n=bfein@ita4fw1.itasoftware.com] has quit [Read error: 113 (No route to host)] 21:34:54 xan [n=xan@62.57.40.18.dyn.user.ono.com] has joined #lisp 21:35:33 Yeah, but... 21:36:15 The files in /proc/bus/usb/ are "special", in that because they're on a synthetic filesystem they can have custom behavior to an extent. 21:36:29 The files in /dev/bus/usb/ are "special" in that they are device nodes. 21:36:40 And thus can have custom behavior to an extent. 21:36:56 The question is, to what extent are they the same, and to what extent do they differ? 21:37:07 dv_ [n=dv@85-127-114-35.dynamic.xdsl-line.inode.at] has joined #lisp 21:37:41 is debian the prefered distro for linux if i want a solid sbcl? 21:37:52 (freebsd 7.0 do not quite seem to be up for it?) 21:38:30 I think any sane linux distribution would work. 21:39:00 -!- albino [n=albino@69.12.222.214] has quit [Read error: 54 (Connection reset by peer)] 21:39:17 heh, ok. 21:39:29 hypno: if you want up-to-date distro-provided SBCL, go for Gentoo 21:39:32 albino [n=albino@69.12.222.214] has joined #lisp 21:39:50 or even Archlinux 21:41:44 i'll give debian a try first, but noted. 21:42:01 You might also consider just managing your SBCL installation yourself. 21:42:12 yeah, i will do that. 21:42:19 Is there a way to list the slots of an object, together with the slot's value? I'm trying to make a make-load-form more generic, to fit a whole class hierarchy, but for that I need to generically grab all slots/values from the original object. 21:42:42 describe-object seems to have that information, I just can't find a function that will give it to me. 21:42:46 SBCL works well enough for me (more than darwin) on fbsd/amd64. 21:42:53 i am mainly looking for stable threads/system really. netbsd is out. solaris/sparc is out. freebsd seems a bit wacky. solaris/x86 might work well too, perhaps? 21:43:10 Why are those systems `out', hypno? 21:43:38 hypno: you could try using SCL. 21:43:47 cads [n=max@c-71-56-69-62.hsd1.ga.comcast.net] has joined #lisp 21:43:54 Riastradh: because the system has crashed on me repeatedly? (and neither of them have threads, except for fbsd) 21:43:56 lhz [n=shrekz@c-db43e455.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 21:44:38 pkhuong: heh, actually i am using scl. but i'll go with sbcl and paserve first and buy safety if needed. SCL has been pretty darn excellent btw. :) 21:45:08 paserve!? 21:45:30 yes, well, portable allegroserve that is. 21:46:04 i can only reiterate. 21:46:05 danderson: it's likely that you have some `class-slots' function (apropos for it) and you can get the value of a slot with `slot-value' 21:46:19 -!- dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has quit [] 21:47:25 phadthai: well, SCL has a paserve port which uses its backend http-library, and also, everything comes properly prepackaged. why do you consider this odd? 21:47:29 adeht: that's the problem, searching for functions containing 'slot', I find none that could be relevant 21:47:48 danderson: what implementation are you using? 21:47:59 oh, well, I just found exactly what I needed in another form: make-load-form-saving-slots 21:48:08 adeht: SBCL 21:48:14 danderson: sb-mop:class-slots 21:48:31 ah, an SBCL specific. Thanks. 21:49:39 hypno: because just about everyone is using hunchentoot (which is actively maintained) for new development nowadays. 21:50:23 mejja [n=user@c-4db6e555.023-82-73746f38.cust.bredbandsbolaget.se] has joined #lisp 21:51:25 -!- Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has quit [Remote closed the connection] 21:51:33 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 21:52:15 pkhuong: well, sure. but im used to paserve and pretty much ships with everything i need. paserve will be the easiest way for me. 21:52:23 dfox [n=dfox@r5cv134.net.upc.cz] has joined #lisp 21:52:25 Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has joined #lisp 21:53:22 dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has joined #lisp 21:54:59 -!- hortont424|away is now known as hortont424 21:56:10 -!- LiamH [n=none@common-lisp.net] has quit ["Leaving."] 21:58:46 -!- dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has quit [] 21:59:59 eno__ [n=eno@adsl-70-137-133-89.dsl.snfc21.sbcglobal.net] has joined #lisp 22:00:15 dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has joined #lisp 22:00:21 daniel [i=daniel@unaffiliated/daniel] has joined #lisp 22:00:32 -!- eno__ [n=eno@adsl-70-137-133-89.dsl.snfc21.sbcglobal.net] has quit [Client Quit] 22:00:38 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 131 (Connection reset by peer)] 22:02:47 hypno: paserve has been good to me 22:03:21 i have it running on sbcl 0.81! several years and still going strong 22:03:30 Dynetrekk pasted "why does this fail?" at http://paste.lisp.org/display/72848 22:03:37 xan_ [n=xan@62.57.40.18.dyn.user.ono.com] has joined #lisp 22:03:53 -!- xan_ [n=xan@62.57.40.18.dyn.user.ono.com] has quit [Read error: 54 (Connection reset by peer)] 22:04:54 fusss: sounds great! on linux or what? 22:05:05 yeah, slackware 22:05:05 <_3b> Dynetrekk: CONS doesn't modify its arguments 22:05:27 _3b: ah... but does that explain the error messagE? 22:05:31 fusss: ah, ok. 22:05:50 <_3b> Dynetrekk: no, that is from (pass-list fail-list) which looks like a function call 22:05:55 ah 22:05:58 I see 22:06:02 hypno: what do you wanna run it on? os x? 22:06:05 <_3b> Dynetrekk: (list pass-list fail-list) might do what you want 22:06:10 I should put a '(pass-list fail-list) 22:06:13 ah, okay 22:06:16 probably better 22:06:16 -!- daniel_ [i=daniel@unaffiliated/daniel] has quit [Read error: 60 (Operation timed out)] 22:06:44 <_3b> '(pass-list fail-list) would create a list containing the symbols pass-list and fail-list... probably not what you want 22:06:53 _3b: but as you say, cons does not update the list, so... :P didn't help either 22:06:55 but thanks 22:07:11 <_3b> you might be looking for PUSH 22:07:17 *fusss* is embarrassed he has never written a line of LOOP in his life 22:07:22 _3b: ah, right. I'll read up on that 22:07:32 fusss: is it possible? to do without it? :P 22:07:37 fusss: actually, i am using scl/solaris/sparc right now, but i'll switch to a free system. 22:07:55 -!- hortont424 [n=hortont@c-75-68-222-233.hsd1.vt.comcast.net] has left #lisp 22:08:00 scieneer? 22:08:06 yupp. 22:08:08 <_3b> Dynetrekk: it is just a macro, which is build on lower level constructs, you could use them directly, or use some other macro 22:08:27 _3b: ah, I see 22:08:51 <_3b> Dynetrekk: or you could use higher order functions like MAPCAR 22:09:01 _3b: yeah, I like that one 22:09:19 <_3b> in scheme, tail recursion would be another option 22:09:21 _3b: I'd like it if it was slightly more like haskell's map though :P 22:09:35 _3b: I guess recursion can always solve the problem somehow 22:09:51 <_3b> not without TCO, which is required in scheme, but not CL 22:10:03 mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has joined #lisp 22:10:05 there is also Iterate, for pompous fools with functioning taste-buds :-) 22:10:09 -!- dakeyras [n=dakeyras@66-154.126-70.tampabay.res.rr.com] has quit [] 22:10:49 *_3b* actually sort of likes the non-lispyness of LOOP, separates the iteration from the actual work more clearly 22:10:57 fusss: 0.8.1 ?? 22:11:14 fe[nl]ix: yeah, circa '03 22:11:33 _3b: I think loop is ok, but it seems to be a different language somehow :P 22:11:51 yes, the download page says 2003/06/24 22:12:04 <_3b> Dynetrekk: it is, which is what i was referring to 22:12:32 _3b: macro => possibly "new language" ? 22:13:09 fe[nl]ix: i froze the whole thing because subsequent sbcl started requiring 2.6 kernels; the app is just for call center management; notes, lead tracking and task assignment. 22:13:27 iirc, threads where added to 0.8.5 22:13:45 lol 22:13:46 <_3b> Dynetrekk: yeah, writing new languages is one of the more powerful uses of macros 22:13:55 you're running it on 2.4 ? 22:14:05 _3b: (push 3 (loop for i from 1 to 10 collecting i)) <- why is this wrong? I don't understand the function spec it seems :s 22:14:31 Dynetrekk: if it were right, what would that do? 22:14:38 <_3b> Dynetrekk: (loop ...) isn't a place 22:14:58 hefner: I thought it would make the list longer 22:15:09 why not use (cons 3 (loop ...)) ? 22:15:10 daniel barlow helped me out so much on that project; where is the guy anyway? 22:15:22 _3b: so this is due to the "destructive update" property of push? 22:15:32 hefner: we already discussed why not: it does not update my list 22:15:49 <_3b> Dynetrekk: right, you need to pass it something it can update 22:16:23 _3b: ah, I see. is there a "push at the other end", though? this method reverses my list :P 22:16:47 <_3b> Dynetrekk: easiest would be to just use LOOP to build the list, since it deals with that for you 22:16:48 -!- Jasko2 [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 22:17:02 <_3b> Dynetrekk: next best would be to nreverse the list before returning it 22:17:02 _3b: maybe, but I don't see how... 22:17:11 _3b: ah, sounds ok 22:17:38 <_3b> if (funcall ...) collect item into pass-list else collect item into-fail-list 22:17:48 -!- dfox [n=dfox@r5cv134.net.upc.cz] has quit [Remote closed the connection] 22:18:20 -!- bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has left #lisp 22:18:21 <_3b> 3rd option would be to actually add it to the end, but that would either be inefficient or more code 22:18:25 bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has joined #lisp 22:18:45 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 22:19:24 minion: Chant! 22:19:24 MORE CODE 22:19:29 _3b: right, I'd have to run along the whole list to find the end 22:19:31 -!- spiderbyte [n=dcl@unaffiliated/dcl] has left #lisp 22:20:01 <_3b> Dynetrekk: right, or keep track of it by hand. 22:20:22 _3b: "store the address" somehow, I guess 22:20:59 Dynetrekk annotated #72848 with "_3b: want to elaborate?" at http://paste.lisp.org/display/72848#1 22:22:19 -!- emarsden [n=user@mir31-3-82-234-52-44.fbx.proxad.net] has quit [Remote closed the connection] 22:23:51 dfox [n=dfox@r5cv134.net.upc.cz] has joined #lisp 22:24:10 _3b annotated #72848 with "like this" at http://paste.lisp.org/display/72848#2 22:27:11 lenst [n=user@host-217-214-129-135.mobileonline.telia.com] has joined #lisp 22:29:10 -!- Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has quit [Remote closed the connection] 22:29:20 -!- rpg [n=rpg@fw.henn.dunn.pcspeed.com] has quit [] 22:29:35 _3b: you just moved the finally? how does it matter? I guess the LOOP syntax got to me :P 22:29:55 <_3b> moving FINALLY was just for readability 22:30:20 <_3b> i used LOOP's IF clause instead of DO (if ... .. ..) 22:30:20 _3b: sorry, you removed the let. so, the loop can't modify a list created outside itself? 22:30:25 Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has joined #lisp 22:31:00 _3b: oh, there is a different one for loop... I see... still don't quite understand though :P 22:31:02 <_3b> right, it creates its own bindings for INTO 22:31:20 -!- malumalu [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has quit [Read error: 104 (Connection reset by peer)] 22:31:31 _3b: ah. so then, if a LET tries to make a list first, there's a problem somehow with the into 22:31:51 <_3b> no problem, it would just ignore the outer one 22:31:53 slash_ [n=Unknown@p4FF0B827.dip.t-dialin.net] has joined #lisp 22:32:13 _3b: hm, then I don't quite understand why it didn't compile like I wrote it 22:32:28 -!- Davidbrcz [n=david@ANantes-151-1-96-175.w86-195.abo.wanadoo.fr] has quit [Remote closed the connection] 22:32:41 <_3b> you can't use LOOP clauses inside lisp forms within the LOOP 22:32:41 malumalu [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has joined #lisp 22:32:58 <_3b> you tried to do DO (IF ... COLLECT ...) 22:33:15 _3b: ah, not inside the DO, because then you're back to "ordinary lisp" ? 22:33:15 <_3b> but LOOP won't look into the IF, so it can't find the COLLECT 22:33:27 hm, okay 22:34:12 -!- fe[nl]ix [n=algidus@88-149-212-85.dynamic.ngi.it] has quit ["Valete!"] 22:34:40 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 22:36:51 is there another name for "pretty print"? trying to google for a firefox plugin that might pprint generated html 22:37:17 -!- mjf [n=mjf@r6y163.net.upc.cz] has quit ["dew on the telephone lines"] 22:37:24 fusss: tidy? 22:37:37 or "html tidy" 22:37:45 yeah 22:37:53 -!- tomsw [n=user@mail2.siscog.pt] has left #lisp 22:38:41 fusss: Why does it have to be a firefox plugin? 22:39:03 what should i do then? 22:39:05 -!- optikalmouse [n=user@bas1-toronto10-1279556315.dsl.bell.ca] has quit [Remote closed the connection] 22:39:14 pprint on the server before it is sent out 22:39:17 ? 22:39:34 What problem are you trying to solve? 22:39:45 <_3b> doesn't cl-who have an option to generate nice html? 22:40:25 too many moving parts thoug; parenscript, ht-ajax, cl-who, etc. 22:40:33 yeah, it can add the proper indenting, if you want. 22:42:56 -!- decafbad [n=mehmet@88.232.131.111] has quit ["Leaving"] 22:44:38 hrm. every time I hit the debugger in slime, no matter what I select, the repl buffer seems to lose its connection to sbcl 22:44:49 this is odd 22:45:21 tripwyre [n=sathya@117.193.160.171] has joined #lisp 22:47:35 -!- dv_ [n=dv@85-127-114-35.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 22:48:28 Good morning. 22:48:59 <_3b> joshg: sbcl + win32? 22:49:20 yeah 22:49:28 Hello beach. 22:50:57 -!- hypno [n=hypno@impulse2.gothiaso.com] has quit ["leaving"] 22:51:24 <_3b> joshg: does http://article.gmane.org/gmane.lisp.slime.devel/7991 sound like the same thing? if so try the fix at the end 22:52:33 fullets [n=user@robotines.co.nz] has joined #lisp 22:53:16 mgr [n=mgr@psychonaut.psychlotron.de] has joined #lisp 22:53:36 <_3b> joshg: (actually i guess the problem description was earlier in the thread) 22:53:57 yeah, that's what I was just figuring out 22:54:48 -!- Jacob_H [n=jacob@92.0.250.208] has quit ["Leaving"] 22:54:52 I'll give this a try 22:55:50 -!- vasa [n=vasa@mm-56-88-84-93.dynamic.pppoe.mgts.by] has quit ["I am not vasya, i am vasa"] 22:55:52 froog [n=user@81-234-149-249-no94.tbcn.telia.com] has joined #lisp 22:56:32 is there a magic incantation to let me open my file with funny latin1 name in sbcl? 22:56:42 Is a clim guru around? 22:57:13 can i print some text on a pane with a specified _background_ color? 22:57:26 just for there area where the text is printed 22:57:31 I'll answer that question in exchange for an answer to my question 22:57:50 ^^ 22:58:01 if i know 22:58:13 are you using mcclim? 22:58:29 yes 22:59:23 if you just want the color to be under the text (and not fill the pane) you can use surrounding-output-with-border around the printing 22:59:40 -!- tripwyre [n=sathya@117.193.160.171] has quit [] 22:59:40 there are various undocumented mcclim extensions which allow you to do what you want 22:59:47 tripwyre [n=sathya@117.193.160.171] has joined #lisp 22:59:49 hmm 22:59:50 demmeln: you would have to draw a rectangle under the text. 22:59:59 where do i find those? 23:00:10 or do they ship with mcclim, but are just not documented? 23:00:25 something like (surrounding-output-with-border (stream :shape :rectangle :background +gray60+) (format t "Hello!")) 23:01:02 demmeln: mcclim/Examples/bordered-output-examples.lisp demonstrates various combinations of arguments 23:02:39 hefner: thanks a lot!! this is exactly what i was looking for 23:02:58 Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 23:03:02 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 23:03:03 got to go. catch train. thanks again. good luck with your funny latin-1 file 23:04:26 dkcl [n=danderse@unaffiliated/dkcl] has joined #lisp 23:04:33 mathrick [n=mathrick@195.116.35.13] has joined #lisp 23:04:36 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 23:04:41 hefner: I /think/ binding some default-internal-format variable to the external format that you want used for your pathname might work 23:04:50 of course, then you also need the C library and the kernel to agree 23:05:09 -!- demmeln [n=demmeln@atradig112.informatik.tu-muenchen.de] has left #lisp 23:06:27 gottesmm [n=gottesmm@99-186-122-131.lightspeed.hstntx.sbcglobal.net] has joined #lisp 23:07:20 -!- guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 23:08:13 Polarina [n=polarina@freecode-project/mentor/polarina] has joined #lisp 23:10:40 oh, interestingly, it's sb-alien::*default-c-string-external-format* (which I already bind whenever calling into sb-posix, but didn't think to try here) 23:11:07 decafbad [n=mehmet@88.232.105.79] has joined #lisp 23:14:57 jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has joined #lisp 23:14:58 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [] 23:18:14 -!- Borat^ [i=cracked@AVelizy-152-1-34-85.w82-120.abo.wanadoo.fr] has quit [] 23:18:15 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Connection reset by peer] 23:18:28 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 23:24:46 -!- Nshag [i=user@Mix-Orleans-106-3-144.w193-248.abo.wanadoo.fr] has quit ["Quitte"] 23:25:03 kzar [n=kzar@hardwick.demon.co.uk] has joined #lisp 23:25:20 -!- decafbad [n=mehmet@88.232.105.79] has quit ["Leaving"] 23:28:52 fe[nl]ix [n=algidus@88-149-212-85.dynamic.ngi.it] has joined #lisp 23:29:41 guille_ [n=guille_@164.Red-83-49-201.dynamicIP.rima-tde.net] has joined #lisp 23:30:18 -!- lenst [n=user@host-217-214-129-135.mobileonline.telia.com] has quit [Read error: 54 (Connection reset by peer)] 23:30:48 -!- milanj [n=milan@93.86.112.165] has quit ["Leaving"] 23:35:36 smolyn [n=smolyn@S01060016cbc4b572.vc.shawcable.net] has joined #lisp 23:36:10 -!- cads [n=max@c-71-56-69-62.hsd1.ga.comcast.net] has quit [Read error: 60 (Operation timed out)] 23:37:46 -!- malumalu [n=malu@hnvr-4dbb5b7e.pool.einsundeins.de] has quit ["Verlassend"] 23:45:04 sjbach [n=sjbach__@ita4fw1.itasoftware.com] has joined #lisp 23:46:31 blitz_ [n=julian@77.64.176.217] has joined #lisp 23:48:02 dcrawford [n=dcrawfor@ita4fw1.itasoftware.com] has joined #lisp 23:48:14 bfein [n=bfein@ita4fw1.itasoftware.com] has joined #lisp 23:48:36 schme [n=marcus@c83-249-80-232.bredband.comhem.se] has joined #lisp 23:51:05 -!- saikat [n=saikat@user-12lc4ta.cable.mindspring.com] has quit [] 23:54:08 -!- mejja [n=user@c-4db6e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 23:55:10 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 23:55:19 -!- spec[afk] [n=SomeOne@88.208.105.1] has quit ["C Ya!"] 23:56:05 xnixan [n=xnixan@unaffiliated/xnixan] has joined #lisp 23:57:27 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 23:57:58 -!- Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has quit [Remote closed the connection] 23:59:20 Jabberwockey [n=jens@dslb-082-083-100-144.pools.arcor-ip.net] has joined #lisp