00:00:08 plage, I don't understand what you want 00:00:22 plage, how would your proposal differ from either ASDF or XCVB ? 00:00:34 Fare: I don't either. That's why it would be interesting to think about it. 00:00:55 I suspects this discussion has split into "make CL simpler" and "make CL go fast" factions, talking past each other 00:01:07 Fare: It has nothing to do with either ASDF or XCVB, but with the semantics of compilation and how compilation is typically used. 00:01:09 plage, read the xcvb docs, and tell me what I'm missing. 00:01:32 Fare: What I am talking about has nothing to do with XCVB nor with ASDF. 00:01:40 hefner, I see no opposition between "fast" and "simple" 00:01:50 sunwukong [n=vukung@210-20-93-83.rev.home.ne.jp] has joined #lisp 00:01:54 plage, it does, actually - but you may not know it. 00:02:11 Fare: that is entirely possible. 00:02:12 Fare: sure, they're orthogonal 00:02:17 -!- jfrancis_ [n=jfrancis@66.194.68.209] has quit [] 00:02:21 -!- antoni [n=user@143.pool85-53-26.dynamic.orange.es] has quit [Read error: 60 (Operation timed out)] 00:02:33 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 00:02:38 hefner, I think there's a very high correlation 00:02:43 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:03:31 -!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 00:05:45 -!- attila_lendvai [n=ati@apn-89-223-208-211.vodafone.hu] has quit [Read error: 110 (Connection timed out)] 00:06:16 final question going back to my original stuff about optimisation.. assuming i want everything to be a double-float but there might be some integers and so on kicking around in my code, if I do (declare (optimize (speed 3))) and add in loads of declarations saying things are double-floats, what happens if there is a stray integer? Do i get a nice error / backtrace or will it just try and do floating point arithmetic and treat the inte 00:06:44 attila_lendvai [n=ati@apn-89-223-190-171.vodafone.hu] has joined #lisp 00:07:09 don't lie to the compiler. what actually happens depends on your speed and safety levels, and whether the gods like you. 00:07:23 malcolm_reynolds, you may want to insert check-type or coerce at input points to remove the stray integers 00:07:33 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:07:35 yeah, i mean i plan to change all the integers to double floats 00:07:43 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:07:46 but for the ones i miss, am i likely to get graceful failure 00:07:47 but yeah -- don't lie to the compiler, or your results will be wrong... at best 00:08:08 check-type is your friend. 00:08:27 I like to have some CHECK-TYPE and ASSERTs at the top of some routines 00:09:07 same here 00:09:10 also writing a DEFMETHOD with arguments declared as, say, floats will do sometimes 00:09:18 main entry points should have plenty of checking 00:09:36 other parts MAY have declarations... once the program is fully debugged and test-suited 00:09:53 (defmethod foo ((bar integer) (baz integer)) ... ) 00:10:22 I also like to keep SPEED 3 declarations as local as possible 00:10:41 sellout [n=greg@static-72-85-235-154.bstnma.east.verizon.net] has joined #lisp 00:10:58 there is even a LOCALLY form, where you can put in such declarations so that the effect is only in tiny parts 00:11:04 understood, but right now i'm trying to squeeze all the speed out of this that i can, which i've been informed is all about speed 3 and then type declarations 00:11:53 well, that's the simple story - the real story is more complicated 00:12:34 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:12:44 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:13:01 there is also DEBUG, SAFETY, SPACE 00:13:18 there are type specific operations 00:13:22 there is inlining 00:13:28 there are compiler macros 00:13:33 -!- rdd [n=user@c83-250-157-93.bredband.comhem.se] has quit [Remote closed the connection] 00:13:49 there are file compilers with special capabilities 00:14:27 there is the problem of tagged data 00:15:06 finally, there is the Stalin solution. 00:15:33 well, for a start there is also the block compiler of CMUCL 00:15:35 malcolm_reynolds, if you REALLY want all the RUNTIME speed... dump a scheme program and pass it to Stalin... 00:16:03 is it really that good? 00:16:28 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 00:16:50 a 'basic' introduction to optimizing Lisp code can be found in Peter Norvig's book 'Paradigms of Artifical Intelligence Programming' 00:16:54 PAIP 00:16:59 haha. it's probably going to be a while before i've got ninja skills enough to convert a bunch of highly non-functional stuff to scheme, but you never know.. 00:17:10 I would also recommend to read the manual of SBCL and/or CMUCL 00:17:21 scheme is not purely functional anyway 00:17:27 yeah i had a look at paip once but i found it wasn't that long before he could write about 15 lines of code that would almost completely baffle me 00:17:35 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:17:43 and finally, practice is the best way 00:17:45 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:17:52 stassats`: yeah but surely a scheme compiler will be optimised specifically for what scheme code normally looks like? 00:18:11 stalin is optimized for numerical code with side effects. 00:19:41 stassats`: right, practice and DISASSEMBLE 00:20:34 -!- Fare [n=Fare@ita4fw1.itasoftware.com] has quit ["Leaving"] 00:21:09 -!- ruediger_ [n=ruediger@62-47-130-138.adsl.highway.telekom.at] has quit ["Leaving"] 00:22:36 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:22:45 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:22:53 -!- attila_lendvai_ [n=ati@apn-89-223-230-139.vodafone.hu] has quit [Read error: 110 (Connection timed out)] 00:23:13 -!- sea4ever [i=sea@216.110.106.24] has quit [No route to host] 00:23:14 -!- slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] 00:25:39 -!- wbraun [n=wolfgang@p5B202536.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)] 00:27:27 -!- fgtech [n=fgtech@bnc1.shellium.org] has quit ["ZNC - http://znc.sourceforge.net"] 00:27:27 -!- desu [n=desu@unaffiliated/desu] has quit ["Read Error: Connection reset by peer pressure"] 00:27:49 wow. i appear to have stayed up working until 1.30. and they say students are getting lazier. 00:28:05 -!- Blkt [n=Blkt@dynamic-adsl-94-37-247-217.clienti.tiscali.it] has quit [Read error: 110 (Connection timed out)] 00:28:29 many thanks again for your help people, i think i'll hang out here more regularly in future as i definitely learn a lot. 00:28:30 -!- malcolm_reynolds [n=malc@78-86-4-156.zone2.bethere.co.uk] has quit [] 00:30:47 confounds [n=jackalop@CPE0013f7f0bd88-CM0013f7f0bd84.cpe.net.cable.rogers.com] has joined #lisp 00:31:50 saba_ [n=saba@212.112.40.42] has joined #lisp 00:33:30 -!- mattrepl [n=mattrepl@ip68-100-82-124.dc.dc.cox.net] has quit [] 00:33:50 -!- joast [n=rick@76.178.184.231] has quit [Connection timed out] 00:35:31 -!- plage [n=user@118.68.32.7] has left #lisp 00:35:45 joast [n=rick@76.178.184.231] has joined #lisp 00:35:46 -!- athos [n=philipp@92.250.250.68] has quit [Remote closed the connection] 00:36:37 -!- ausente [n=user7994@189-19-126-110.dsl.telesp.net.br] has quit [Read error: 110 (Connection timed out)] 00:37:18 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:37:24 -!- bobbysmith007 [n=russ@one.firewall.gnv.acceleration.net] has quit ["Leaving."] 00:37:28 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:39:39 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 00:40:03 -!- Yuuhi [i=benni@p5483C792.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:40:52 hey, that's neat. the osicat version is actually 40% faster than its sb-posix equivalent, perhaps because of these stupid sap-alien notes I never know how to fix. 00:41:24 (the sb-posix-hacked-to-include-dirent.dt_type equivalent, that is) 00:42:49 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 00:42:52 hefner: pathname completion? 00:42:59 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:45:37 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Remote closed the connection] 00:46:10 no, just playing with versions of the code that scans the library when my music player starts up. currently takes 3-4 seconds, but can clearly be cut down to a about half a second (for comparison, `find |sort` takes 1/4 second) 00:46:45 for whatever reason I have a longstanding obsession with being able to scan a directory tree fairly quickly. 00:47:03 -!- thom_ [n=thom@pool-173-51-224-238.lsanca.fios.verizon.net] has quit ["This computer has gone to sleep"] 00:47:49 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:47:58 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:48:31 fair enough 00:48:58 nvoorhies_ [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 00:49:19 out of curiosity, have you tried SB-IMPL::MAP-DIRECTORY? 00:49:38 (it's likely to become exported sooner than later) 00:50:34 nah, I'm really not into pathnames. 00:51:05 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 00:51:52 -!- mrSpec [n=NoOne@82.177.125.6] has quit [Read error: 110 (Connection timed out)] 00:52:49 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 00:53:00 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:54:54 heh, not into pathnames but directory traversal? 00:55:34 elliotstern [n=chatzill@cpe-24-93-17-195.rochester.res.rr.com] has joined #lisp 00:57:20 leo2007: Looking for me? 00:57:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 00:57:59 kami- [n=user@unaffiliated/kami-] has joined #lisp 00:59:37 attila_lendvai_ [n=ati@catv-89-132-189-132.catv.broadband.hu] has joined #lisp 01:00:04 -!- attila_lendvai [n=ati@apn-89-223-190-171.vodafone.hu] has quit [Read error: 110 (Connection timed out)] 01:01:00 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 01:01:24 -!- sepult [n=user@xdsl-87-78-25-27.netcologne.de] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 01:01:53 dysinger [n=tim@71.20.231.3] has joined #lisp 01:02:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:03:00 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:07:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:08:00 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:12:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:12:59 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:13:57 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["Leaving"] 01:15:49 ausente [n=user7994@187.34.42.81] has joined #lisp 01:16:58 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 01:17:28 fundamental [n=fundamen@24-148-122-247.ip.mhcable.com] has joined #lisp 01:17:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 01:18:00 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:21:54 drafael [n=tapio@ip-118-90-133-176.xdsl.xnet.co.nz] has joined #lisp 01:22:48 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:22:59 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:26:32 -!- Nshag [i=user@Mix-Orleans-106-4-106.w193-248.abo.wanadoo.fr] has quit ["Quitte"] 01:27:15 -!- attila_lendvai_ [n=ati@catv-89-132-189-132.catv.broadband.hu] has quit [Read error: 113 (No route to host)] 01:27:50 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:28:00 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:32:51 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:33:01 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:36:07 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit ["HULK ANGRY! HULK DISCONNECT!"] 01:36:35 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 01:37:36 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:37:46 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:38:38 -!- rread_ [n=rread@nat/sun/x-6cf19ece45ae85fd] has quit [Read error: 110 (Connection timed out)] 01:40:30 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [] 01:41:51 zu22 [n=ubuntu@unaffiliated/zu22] has joined #lisp 01:42:07 hi everyone, anyone know of a site with step-by-step instructions on installing SBCL in Linux 01:42:17 i have no lisp compilers installed at all right now 01:42:42 just download binary from www.sbcl.org and read INSTALL file 01:42:53 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:43:03 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:43:07 stassats`: oh thanks 01:43:38 -!- zu22 [n=ubuntu@unaffiliated/zu22] has left #lisp 01:44:09 should've added, "..and don't, under any circumstances, install lisp-related packages from your linux distro, especially if it's debian" 01:45:05 -!- mrsolo [n=mrsolo@nat/yahoo/x-890406b40a0dfc8f] has quit ["Leaving"] 01:45:23 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 01:46:30 -!- fundamental [n=fundamen@24-148-122-247.ip.mhcable.com] has quit ["Leaving"] 01:46:47 Why is that? 01:47:21 because they are broken and outdated 01:47:32 -!- elliotstern [n=chatzill@cpe-24-93-17-195.rochester.res.rr.com] has quit [Read error: 113 (No route to host)] 01:47:38 -!- cky [n=cky@cpe-024-211-249-162.nc.res.rr.com] has quit [Read error: 110 (Connection timed out)] 01:47:52 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:48:01 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:48:48 and common lisp controller somehow routinely wreaks havoc with newbies trying to get slime working. 01:48:59 elliotstern [n=chatzill@cpe-24-93-17-195.rochester.res.rr.com] has joined #lisp 01:49:46 cky [n=cky@cpe-024-211-249-162.nc.res.rr.com] has joined #lisp 01:50:12 -!- sellout [n=greg@static-72-85-235-154.bstnma.east.verizon.net] has quit [] 01:52:54 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 01:55:02 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 113 (No route to host)] 01:57:54 kami- [n=user@unaffiliated/kami-] has joined #lisp 01:57:58 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 02:00:21 -!- nvoorhies_ [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 02:01:59 dialtone_ [n=dialtone@adsl-67-125-22-195.dsl.pltn13.pacbell.net] has joined #lisp 02:02:16 -!- dreish [n=dreish@minus.dreish.org] has quit [] 02:02:20 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit [Read error: 110 (Connection timed out)] 02:02:54 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 02:03:04 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:03:29 Hmm. Is sxhash allowed to descend into array contents to compute a hash? Seems like it should be able to. 02:04:49 -!- Swords2 [n=kohii@pool-71-112-23-110.sttlwa.dsl-w.verizon.net] has quit [Read error: 110 (Connection timed out)] 02:05:32 -!- z0d [n=z0d@unaffiliated/z0d] has quit [Read error: 60 (Operation timed out)] 02:07:55 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 02:08:04 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:09:30 z0d [n=z0d@unaffiliated/z0d] has joined #lisp 02:10:27 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 02:12:55 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:13:05 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:17:03 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 113 (No route to host)] 02:17:53 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:18:03 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:21:08 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 02:22:16 scidan [n=dan@c-69-250-212-245.hsd1.dc.comcast.net] has joined #lisp 02:22:55 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:23:05 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:24:25 zu22 [n=ubuntu@unaffiliated/zu22] has joined #lisp 02:24:33 ok i just download latest sbcl from x86 and read install 02:24:54 it shows how to build from binary and for build from src it says u must have a supported lisp system already installed! 02:24:59 problem is i have no lisp installed 02:25:02 so what do i do?? 02:25:09 s/from/for 02:25:17 install it 02:25:22 umm how? 02:25:30 read INSTALL carefully 02:25:31 -!- danlei [n=user@pD9E2EA49.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 02:25:33 ok 02:27:03 Taggnostr [n=x@wolf.yok.utu.fi] has joined #lisp 02:27:06 -!- zu22 [n=ubuntu@unaffiliated/zu22] has left #lisp 02:27:24 -!- elliotstern [n=chatzill@cpe-24-93-17-195.rochester.res.rr.com] has quit [Read error: 113 (No route to host)] 02:27:42 -!- segv [n=mb@p4FC1CBF2.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 02:27:55 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:27:58 segv [n=mb@p4FC1C683.dip.t-dialin.net] has joined #lisp 02:28:05 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:28:47 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 02:30:16 smoofra [n=user@cthulhu.elder-gods.org] has joined #lisp 02:30:33 sup #lisp. two questions: 02:30:56 does sbcl have any nice way to pretty-print it's internal representations (refs, leafs, funs...) 02:30:57 ? 02:31:37 and how do you read (SB-C::ORIGINAL-SOURCE-START 11 1 2 1 3 0)? (what do the numbers mean?) 02:31:56 rtoym: indeed, equal => = sxhash so the sxhash may change when the contents of the array change. 02:32:12 smoofra: iirc there is an option to dump ir1 (or maybe ir2) while compiling 02:32:32 and if that's a source position, the numbers are offsets in lists 02:32:39 rtoym: often it's not done too heavily, since we want sxhash to be fast. (eg. we could take into account only the first, last and middle slot of a vector. 02:32:56 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:33:06 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:33:19 :trace-file option for compile-file 02:33:33 pjb: Thanks. That's what I was thinking---just grab a few elements and sxhash them together. 02:33:48 -!- pem [n=pem@159.226.35.246] has quit [Read error: 60 (Operation timed out)] 02:34:27 *rtoym* got O(n) performance from an equal hash table with keys of type array (simple-arrays). 02:34:40 kpreid: any idea where such an option coudl be found? would do i need to set a special var, or call some internal function..? 02:34:49 it's some special var 02:35:44 On cmucl, print-nodes will pprint some internal information. 02:36:14 print-continuation pprints continuations. 02:36:37 ManateeLazyCat [n=user@116.4.143.153] has joined #lisp 02:37:55 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:38:04 pem [n=pem@159.226.35.246] has joined #lisp 02:38:04 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:38:35 -!- saba_ [n=saba@212.112.40.42] has quit ["Lost terminal"] 02:38:42 (let ((sb-c::*compiler-trace-output* *standard-output*)) (compile nil '(lambda (f) (+ f 10)))) 02:38:48 s0ber [i=pie@220-136-227-170.dynamic.hinet.net] has joined #lisp 02:40:07 "wait, we have continuations in CL?!" 02:40:33 or as i said, :trace-file 02:41:16 ah thanks stassats` 02:41:31 -!- Taggnostr [n=x@wolf.yok.utu.fi] has quit [Read error: 60 (Operation timed out)] 02:42:47 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 02:42:57 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:43:06 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:47:30 -!- cky [n=cky@cpe-024-211-249-162.nc.res.rr.com] has quit ["Upgrading kernel"] 02:47:57 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:48:07 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:48:33 -!- ManateeLazyCat [n=user@116.4.143.153] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 02:50:19 -!- joast [n=rick@76.178.184.231] has quit [Connection timed out] 02:52:09 cky [n=cky@cpe-024-211-249-162.nc.res.rr.com] has joined #lisp 02:52:58 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 02:53:08 kami- [n=user@unaffiliated/kami-] has joined #lisp 02:53:15 elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has joined #lisp 02:56:44 rtoym: sxhash is defined in terms of EQUAL, and EQUAL is ~EQ for arrays (except for strings and bit vectors). -> can't descend. But -> can't use address-based hashing in a copying GC either. 02:57:42 -!- felipe [n=felipe@my.nada.kth.se] has quit [Remote closed the connection] 02:57:58 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 02:58:08 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:00:39 so, basically, sxhash can only work based on the array type (element type, rank, dimensions) if there isn't any way to attach auxiliary data to objects. 03:00:48 eno_ [n=eno@adsl-70-137-131-139.dsl.snfc21.sbcglobal.net] has joined #lisp 03:01:34 joast [n=rick@76.178.184.231] has joined #lisp 03:02:10 -!- confounds [n=jackalop@CPE0013f7f0bd88-CM0013f7f0bd84.cpe.net.cable.rogers.com] has left #lisp 03:02:45 -!- eno [n=eno@nslu2-linux/eno] has quit [Nick collision from services.] 03:02:52 -!- eno_ is now known as eno 03:02:58 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:02:59 -!- Adlai [n=adlai@93-173-254-22.bb.netvision.net.il] has quit [Remote closed the connection] 03:03:08 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:06:25 s0ber_ [n=s0ber@220-136-225-130.dynamic.hinet.net] has joined #lisp 03:06:42 -!- s0ber [i=pie@220-136-227-170.dynamic.hinet.net] has quit [Read error: 110 (Connection timed out)] 03:07:07 pkhuong: Oh, right. It says so right there in the entry for equal. And strings and bit vectors are excluded from that. 03:07:59 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:08:08 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Remote closed the connection] 03:08:09 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:08:52 have you had time to take a look at ? 03:09:05 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["This computer has gone to sleep"] 03:10:18 Yeah, I saw that. I don't think I'm subscribed to sbcl anymore (at least from work). 03:10:31 I agree with part 1. 03:11:08 For part 2, does Kahan have anything to say about computing that carefully? My copy is at work. 03:13:28 I was forwarded a paper on prescaling complex division; i think handling the edge cases is a bit orthogonal 03:15:08 -!- s0ber_ [n=s0ber@220-136-225-130.dynamic.hinet.net] has quit [Remote closed the connection] 03:15:14 I don't think we have to follow C99, but I think Kahan is/was involved in the C99 arithmetics. So maybe C99 is what Kahan would want. 03:16:12 _Pb [n=Pb@75.131.194.186] has joined #lisp 03:16:21 we also probably don't follow the rounding rules specified in the IEEE standard. 03:16:34 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 104 (Connection reset by peer)] 03:16:50 Why not? The rounding is in hardware, right? 03:17:57 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:18:07 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:18:19 for individual operations, not for the complex operations we build on top. That's one reason x87 was so nice. You could perform everything in long double and truncate with correct rounding at the end. 03:18:25 I know that Kahan wanted real+complex to only touch the real part, leaving the imaginary part unchanged. Only matters if the imaginary part is a signed zero. 03:18:51 There is no IEEE standard for complexes is there? 03:18:54 sea4ever [i=sea@216.110.106.66] has joined #lisp 03:19:12 I really don't feel like diving in an IEEE document. I'm told 754 (or maybe some revision) also touches on complexes. 03:19:36 There is a problem with working in long double and rounding to double. You get double rounding which affects the result in the LSB. 03:20:10 Ok. My knowledge of IEEE was the really old spec from the 80's. I haven't looked at any recent update. 03:20:21 -!- scidan [n=dan@c-69-250-212-245.hsd1.dc.comcast.net] has quit [Client Quit] 03:21:26 Oh, and with long doubles, things don't overflow doubles when they should. Or something like that. I'd have dig up my old links to find out. 03:21:55 Kahan specifically asked for that ;) 03:22:07 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit ["Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/"] 03:22:58 blargh. Hate IEEE and its pay-wall. Its only purpose is to annoy most users and make life harder for everyone else. 03:22:59 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:23:09 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:24:30 The draft is available, though. See wikipedia page. 03:25:33 mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has joined #lisp 03:25:49 Only had to go through ssh. Anyway, 754 doesn't specify complexes, yay! 03:26:18 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 03:26:45 yeah, the draft doesn't even have the word "complex". :-) 03:27:57 luis: re build process, I really liked the way Feeley draws it with the compiler underneath the component being compiled. 03:27:58 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:28:08 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:29:00 *hefner* did not know cffi had with-pointer-to-vector-data, wonders he wrote his own 03:30:04 hefner: Might be new. There's one in sb-bsd-socket too. 03:30:29 I wonder if I ripped mine off from there. I remember that coming up. 03:30:36 zu22 [n=ubuntu@unaffiliated/zu22] has joined #lisp 03:30:59 ok it looks like sbcl is installed, i used the binary, i get 3 cp errors at the end, can someone plz take a look: http://zu22.pastebin.com/f6fa8e128 03:31:16 also what do i do now to get install other lisp stuff like slime or hyperspec docs etc..? 03:32:01 the errors are normal since you didn't build the docs 03:32:14 Maybe you'd be better off doing the rest with clbuild, though. 03:33:00 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:33:10 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:33:25 neat. one minute I have "A Foreign Pointer [stack-allocated]", the next I have "BOGUS object". probably shouldn't have returned that. 03:34:45 hefner: better than pointing into random garbage. 03:35:37 The IEEE draft doesn't mention long doubles. I wonder what that means? 03:36:02 pkhuong: ok, just now i downloaded src and went into doc/manual and type make but i get error: http://zu22.pastebin.com/f645407d4 03:36:23 zu22: I said the errors were normal. 03:36:56 pkhuong: oh, I mean *I* probably shouldn't have returned that. 03:37:05 rtoym: 3.7, extended and extendable precision 03:37:42 pkhuong: ok, now how do i go about installing hyperspec, slime ..? and where do i get this clbuild you mentioned? 03:37:44 hefner: right, and I meant that the bogus object (assuming that's from tcr's modifications to the inspector) is better than potentially confusing the gc later on. 03:37:50 minion: tell zu22 about clbuild 03:37:51 zu22: have a look at clbuild: clbuild [common-lisp.net] is a shell script helping with the download, compilation, and invocation of Common Lisp applications. http://www.cliki.net/clbuild 03:38:08 cool 03:38:12 i will use clbuild :) 03:38:12 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:38:22 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:39:00 -!- zu22 [n=ubuntu@unaffiliated/zu22] has left #lisp 03:39:53 pkhuong: Oh. I only have the draft. There is no 3.7. :-( 03:41:30 rtoym: 03:42:04 Many thanks! 03:42:21 pkhuong: ISO standard were for pay ; Internet standards were free. Nowadays, the network is based on Internet standard, nobody use ISO standard anymore. 03:43:01 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:43:11 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:43:37 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 03:44:13 zu22 [n=ubuntu@unaffiliated/zu22] has joined #lisp 03:44:15 back again :) 03:44:17 salex [n=user@216.80.147.206] has joined #lisp 03:44:27 when u run the clbuild check it generates TONS of errors for sbcl 03:44:32 i installed sbcl and it seems to run fine 03:44:34 'evening 03:44:38 can someone see what is wrong: http://zu22.pastebin.com/f58a57046 03:44:48 hi salex 03:44:56 zu22: using random paste instead of lisppaste? 03:44:58 lisppaste url 03:44:59 To use the lisppaste bot, visit http://paste.lisp.org/new/lisp and enter your paste. 03:45:11 -!- fe[nl]ix [n=algidus@88-149-211-132.dynamic.ngi.it] has quit [Read error: 110 (Connection timed out)] 03:45:24 pjb: ok i will use that 03:45:42 The variable THIS is unbound. = EWRLNG 03:45:50 fe[nl]ix [n=algidus@88-149-208-244.dynamic.ngi.it] has joined #lisp 03:45:57 Error, WRong LaNGuage :-) 03:46:03 zu22 pasted "clbuild check fails on sbcl" at http://paste.lisp.org/display/82479 03:46:27 pjb: how do i fix this error? 03:47:02 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Read error: 113 (No route to host)] 03:47:32 Could you see more backtrace? I see no user-defined function there... 03:47:52 pjb: i pasted everything that was output from "clbuild check" 03:48:01 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:48:08 i got the latest clbuild from darcs per the website instructions 03:48:11 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:48:15 Probably you'd have to run it from a lisp image, instead of from the shell. 03:48:21 oh my 03:48:25 so i cannot use clbuild? 03:48:36 -!- legumbre [n=user@r190-135-27-47.dialup.adsl.anteldata.net.uy] has quit [Read error: 104 (Connection reset by peer)] 03:48:40 why can't i run from the shell? 03:48:45 If you cannot debug it, no. You'll have to ask to someone who is able to debug it. 03:48:46 legumbre [n=user@r190-135-27-47.dialup.adsl.anteldata.net.uy] has joined #lisp 03:48:50 everyone said clbuild was good 03:48:56 Because there's this bug. 03:48:58 but it generates this error is very bad 03:49:02 I don't know, I never tried it. 03:49:10 ok this sucks 03:49:16 i will try contact clbuild team 03:49:18 Ask later, when there's more people here. 03:49:20 Yes. 03:49:21 ok 03:51:01 sciendan [n=dan@c-69-250-212-245.hsd1.dc.comcast.net] has joined #lisp 03:51:17 is /dev/null somehow not empty on your machine? 03:51:31 -!- Jabberwockey [n=jens@port-4337.pppoe.wtnet.de] has quit [Remote closed the connection] 03:51:33 there is not even a contact address on the clbuild page to report a bug! MAJOR PHAIL 03:51:42 pkhuong: lemme check 03:52:02 pkhuong: -rw-rw-rw- 1 root root 0 2009-06-24 12:25 /dev/null 03:52:20 It's wrong, it should be: crw-rw-rw- 1 root wheel 3, 2 Jun 24 02:04 /dev/null 03:52:20 03:52:26 Um, what about `Please send patches bug reports to ' at the top of ? 03:52:26 Notice the 'c'. 03:52:59 sudo rm /dev/null ; sudo mknod /dev/null c 3 2 03:53:02 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 03:53:03 -!- joast [n=rick@76.178.184.231] has quit [Read error: 110 (Connection timed out)] 03:53:12 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:53:36 zu22: please get a clue instead of assuming that, because you can't get it right, everyone *else* is doing something wrong. 03:53:45 pjb: ok i will i try that 03:54:06 Basic functional literacy would be a good start. 03:54:48 joast [n=rick@76.178.184.231] has joined #lisp 03:55:00 pjb: ok it now has; crw-r--r-- 1 root root 3, 2 2009-06-24 12:31 /dev/null 03:55:24 pjb: but now if i run 'clbuild check' it says: ./clbuild: line 49: /dev/null: No such device or address 03:55:35 pkhuong: ok 03:56:06 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit ["leaving"] 03:57:09 zu22, probably the device numbers are wrong for your operating system (for instance, on NetBSD 5.0, the major number is 2, not 3), but if you have a screwed up /dev/null, you probably have greater problems to fix on your system before worrying about clbuild. 03:57:26 Riastradh: i use Debian squeeze 03:57:57 Riastradh: i never had any problems, this is a VPS system, i never touched /dev/null it is from the Debian image the VPS installed 03:58:02 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 03:58:06 On Linux 2.6.29, it looks like the major number is 1 and the minor number is 3. 03:58:12 kami- [n=user@unaffiliated/kami-] has joined #lisp 03:59:19 Sometimes there is a program called `MAKEDEV' that knows the map from mnemonic (ha!) names to device numbers, to be run like `cd /dev && ./MAKEDEV null'. But this is off the topic of Lisp, so I'll clam up now. 04:01:35 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit ["gbu"] 04:01:36 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 04:01:54 WarWeasle [n=brad@c-98-220-168-14.hsd1.in.comcast.net] has joined #lisp 04:03:02 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:03:12 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:04:34 -!- Lycurgus [n=Ren@pool-71-186-174-244.bflony.east.verizon.net] has quit [Remote closed the connection] 04:05:37 -!- ausente is now known as dalton 04:06:27 chavo_ [n=user@c-66-41-11-10.hsd1.mn.comcast.net] has joined #lisp 04:07:03 -!- WarWeasle [n=brad@c-98-220-168-14.hsd1.in.comcast.net] has quit [Remote closed the connection] 04:07:19 Riastradh: ok i fixed /dev/null and the 'clbuild check' works except i get 1 error at end, i looked and /usr/local/src/lisp/clbuild/systems/ is empty, it expects some files in there apparently: http://paste.lisp.org/display/82480 04:10:11 CCL's ffi reader macros (#_, #&, etc.) are neat. 04:10:35 can some someone take a look at this error at end plz: http://paste.lisp.org/display/82480 04:12:44 lks lk ur mssg dr 04:13:02 salex: huh? 04:13:03 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:13:13 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:13:16 exactly! 04:13:47 gemelen [n=shelta@shpd-92-101-140-182.vologda.ru] has joined #lisp 04:14:04 -!- ikki [n=ikki@201.155.75.146] has quit ["Leaving"] 04:14:13 zu22: did clbuild complain about anything? There's nothing in systems/ because, there are "0 system definition files registered". 04:14:37 i'm just teasing you about truncated words. not much help with clbuild as I don't use it, but that error message looks pretty straightforward. Can't find a directory or .asd files in it 04:14:48 but that's not exactly an error 04:14:52 pkhuong: yes it complains no *.asd files in systems/ this does not match the clbuild example on the web which shows no such error. 04:14:59 no, ls complains, not clbuild. 04:15:02 salex: ok i tend to worry a lot 04:15:11 pkhuong: ok so i can just ignore this? 04:15:25 yes. 04:16:00 it seems none of you use clbuild yet it was highly recommended to me, why is that? 04:17:02 zu22: I already have a working setup. I expect the same stands for salex and pjb. 04:17:07 we're old 04:17:08 ;) 04:17:10 I appreciate the assistance of everyone in here who helped me. I will pray for each of you. 04:17:16 salex: ah i am too, 32 :( 04:17:31 i mean't in terms of lisp-usage, not tree rings 04:17:39 pkhuong: do you like sbcl? 04:17:45 if you remember the growing pains of asdf and have something that works for you 04:17:55 you might not have bothered chekcing out clbuild 04:18:03 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:18:05 hell, it appeared since the last time I did much lisping 04:18:09 salex: i am young in lisp-usage and old in tree rings sadly, i'm over the hill 04:18:13 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:18:18 zu22: enough to work on it on a semi-regular basis. 04:18:53 zu22: you'll find a lot of sbcl users here, and a few of the core hackers 04:19:12 for that matter, a bunch of us have a few lines in the compiler too 04:19:23 salex: i never used asdf, i am new to lisp, i get by OK in C and can deduce the gcc compiler errors but even simple errors in a Lisp compiler message seem orders of magnitude greater in complexity and obfuscation lol. 04:19:23 at least, if they haven't bled out yet :) 04:19:52 salex: wow col 04:19:55 zu22: i just meant that clbuild is the new kid on the block 04:19:56 s/col/cool 04:19:58 real lisp hackers 04:20:04 salex: ah i see 04:20:10 so it's a pretty sbcl friendly place 04:20:18 that's great 04:20:19 not that other compilers aren't represented 04:20:36 why are lisp compiler messages so umm verbose? lol 04:20:42 that is one thing that makes me nervous 04:21:07 because they actually tell you what's happening? 04:21:14 are you referring to sbcl's warnings? 04:21:25 you can muffle them if you'd like, but they're actually very educational 04:21:56 salex: yes 04:22:15 salex: to my ignorant mind the warnings look like errors, must learn how to differentiate 04:22:46 in sbcl i type "(format t 'hi)" and it generated all this compiler output 04:22:50 i couldn't make any sense of it :( 04:22:53 (and yes, some "real lisp hackers" spend time here, myself excluded) 04:23:04 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:23:11 i wish it would just briefly tell me what is wrong in my syntax 04:23:14 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:23:19 -!- _Pb [n=Pb@75.131.194.186] has quit ["Leaving"] 04:23:29 zu22: the syntax is fine. The semantics aren't. 04:23:42 zu22: you need to rethink things a little. your syntax was ok 04:23:53 if you were missing a ")" or something, it would just tell you that 04:24:02 is there like a meta layer i can install over sbcl that will translate the warnings/errors into something more easily human parsible? :) 04:24:08 oh 04:24:26 kmcorbett [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has joined #lisp 04:24:42 -!- deylen [n=deylen@62.249.247.182] has quit [] 04:24:56 but you actually evaluated the statment, and it probably dumped you into a restart wiht a message about why 04:24:56 -!- Soulman__ [n=kae@Gatekeeper.vizrt.com] has quit [Read error: 104 (Connection reset by peer)] 04:24:56 zu22: really? I find SBCL errors to be *much* more helpful than gcc's. 04:25:03 Soulman__ [n=kae@Gatekeeper.vizrt.com] has joined #lisp 04:25:09 actually probably a very helpful message 04:25:12 Ralith: i guess i must get used to them 04:25:19 i can paste the output if anyone cares to see it 04:25:21 huh, no major incidents getting the audio and mp3 code running in CCL. If I just rip out the sb-posix gunk, I could get the music play running in it too (the minute long startup time would be painful, though). 04:25:25 but you'll have to get out of the edit->compile->run mentality 04:25:37 and into the repl mentality 04:25:41 If you execute the function, you'll get a runtime error that HI isn't a string or simple-string. If you look at the compile-time output, you'll see that the type of the value FORMAT (member format) is not a subtype of either (VECTOR CHARACTER), (VECTOR NIL), BASE-STRING (so not a string), nor FUNCTION (not a function). 04:25:51 hefner: mp3 code? what're you working on? 04:26:03 And if you read the specification for FORMAT, you'll see that the second argument must indeed be either a string or a function. 04:26:14 salex: ok 04:26:21 pkhuong: the second arg can be a function? O.o 04:26:44 pkhuong: oh i see 04:26:55 clhs format control 04:26:58 Ralith: http://vintage-digital.com/hefner/software/mixalot/mixalot.html 04:27:09 zu22: you'll need some patience to get used to sbcl errors, also you'll get used to reading lambda lists for expected arguments etc... I admit I was frustrated and confused a lot myself at first... and hi, btw 04:27:43 sometimes sbcl will also show a piece of code which was expanded, you'll get used to how the macros expand etc 04:27:58 phadthai: no, you just ask SLIME to point you to the original form. 04:27:59 Ralith: of course it can. A string or a formatter 04:28:02 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:28:12 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:28:15 see 22.2.1.3 04:28:18 pkhuong: that doesn't seem to always work for me at least 04:28:29 dysinger_ [n=tim@71.20.231.3] has joined #lisp 04:28:34 although it always works for notes 04:29:00 phadthai: oh hi :) ah 04:29:26 phadthai: i order the gigamonkey book so i am just sort of getting everything set up before it arrives: install sbcl, install slime etc. 04:29:59 i am very excited to get started on learning lisp 04:30:29 zu22: consider this. In c, if you try and pass the wrong type of argument to a function, it's syntatically caught when you compile 04:30:49 salex: right 04:32:16 when you typed (format t 'hi) though, you called format with the argments t and 'hi 04:32:46 salex: ok 04:32:55 you've already interpreted (in this case compiled) the statement, and called the funciton 04:33:04 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:33:14 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:33:48 salex: ah 04:33:56 but as format doesn't accept a symbol argument in that position, the function won't be able to do anything with that, and drops you into a restart 04:33:59 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 04:34:02 (you should read about those) 04:34:42 with a different sort of problem you may have been offered the ability to fix your input and keep going, which is kind of nice 04:34:46 salex: i see, is there a lisp version of the C manpages so i can just type 'man FORMAT' and see what it expects, that would be very helpful 04:34:50 at times, very much so 04:35:06 even better, there is a hypertex spec for the language 04:35:08 zu22: nice... any particular event that sparkled your interest about lisp? 04:35:40 zu22: so you see, not only are these things different than c, some don't even make sense there 04:35:43 zu22: yes PCL is a good starting choice 04:35:53 the quicker you change your mental maps about these things, the better 04:36:39 zu22: may I suggest: install slime in emacs, add the hyperspec, then you can easily look these things up as you code 04:37:22 interestingly, although those use a scheme, I enjoyed SICP and the SICP lectures to open my mind about functional programming, and it's also refreshing basic CS 04:38:03 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:38:12 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:39:57 cool 04:42:43 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 04:42:44 (http://groups.csail.mit.edu/mac/classes/6.001/abelson-sussman-lectures/) 04:43:05 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 04:43:07 after Practical Common Lisp you might also like On Lisp which is more advanced 04:43:15 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:43:47 s/more advanced/narrower, more opinionated/ perhaps 04:44:03 pg has some good stuff on macros in there, but it's really not the same scope 04:44:13 isn't that the one exploring macros more deeply? 04:44:20 maybe I'm mixing up with another name also 04:44:23 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 04:44:26 no, that's the one 04:44:46 but it's also stylistically weird and has many pg-ism 04:45:27 see for example "graham crackers" 04:45:38 also, his disdain for CLOS 04:46:07 so it's not really a place to look for building general common lisp knowledge 04:46:24 iirc, he doesn't like loop iether:) 04:46:44 pg also likes Scheme which is the source of several "pg-isms" 04:47:14 which have warped into arc-isms, i guess 04:47:17 I've noticed his use of recursion over iterative loops also true 04:47:28 Adamant: indeed 04:47:31 i'm not saying it's not worth reading (though some would) 04:47:35 He also dislikes parentheses for some reason, which, again, explains many arcisms. 04:47:37 just that it's quirky 04:48:02 and focussed on techniques you may not use so much, too 04:48:16 Not sure I see the point of reading PG, frankly. 04:48:34 you're in my brackets 04:48:38 then :) 04:48:51 anyway, i dunno. i learned a fair bit about macros from him 04:49:17 but it's probably a bad idea to throw newbies at them. too much "oooh, macros ... shiny" is a bad thing 04:49:50 Arc makes me wary of his other stuff. Uninteresting and uninterestingly executed. 04:50:01 i haven't actualy looked at arc at all 04:50:07 -!- dysinger [n=tim@71.20.231.3] has quit [Read error: 110 (Connection timed out)] 04:50:14 it's pretty much entirely happend while i was too busy to do much lisping 04:50:27 or looking at new things 04:50:57 so i don't really have an opinion on it, at least non-superficial one 04:51:00 *hefner* crashes CCL to the kernel debugger by daring to close an fd-stream, cheers 04:51:10 heh 04:51:36 SPAN was meant as a joke, adding as much syntactic sugar on top of delimited continuations as possible, and yet it seems less uninteresting than arc. 04:51:38 KingNatoG5_ [n=patrik@84-217-2-195.tn.glocalnet.net] has joined #lisp 04:51:54 -!- KingNatoG5 [n=patrik@84-217-1-111.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)] 04:51:55 Sandeep` [n=Agent@122.172.113.200] has joined #lisp 04:53:06 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:53:16 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:54:13 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 04:54:33 anyway, time for bed. 'night 04:54:49 -!- salex [n=user@216.80.147.206] has left #lisp 04:56:41 GN salex ;D 04:57:08 oh, hah. it closed the underlying descriptor. which was 1. oops. 04:57:37 heh 04:57:55 hefner: SBCL dies in a very slightly more elegant manner, I believe. 04:58:04 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 04:58:14 kami- [n=user@unaffiliated/kami-] has joined #lisp 04:58:58 I probably never tried this in SBCL. I do bind *standard-output* to a an fd-stream on fd 1 there, but I always quit without closing it. 05:00:30 -!- chavo_ [n=user@c-66-41-11-10.hsd1.mn.comcast.net] has quit [Remote closed the connection] 05:01:21 s0ber [n=s0ber@118-160-161-10.dynamic.hinet.net] has joined #lisp 05:03:06 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:03:16 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:04:13 salex: ok will install slime and hyperspec, is there a nice guide on getting started with slime in emacs? 05:04:41 phadthai: i came across an old lisp book at a library book sale and it seemed really different than other programming languages so i thought maybe i'd give it a try 05:06:01 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:07:35 rtoym: it is resolved now. It was about well-rng. 05:08:06 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:08:16 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:10:57 -!- mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 05:12:07 -!- SharkBra1n [n=gerard@210.48.104.34] has quit ["leaving"] 05:13:07 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:13:17 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:15:05 I'm new to lisp can someone help me ? 05:16:02 -!- konr [n=konrad@201.82.132.33] has quit [Read error: 104 (Connection reset by peer)] 05:16:15 I don't know where to start or which language to use . All I know is that I want to write a image recognition algo . 05:16:46 I'm a student willing to research on Computer Vision . 05:18:05 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:18:13 use C. 05:18:15 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:18:23 hefner : Why C ? 05:18:40 Or C++. I believe most image processing libraries are in C++ nowdays. 05:18:43 DukeOfMilan [n=DukeOfMi@adsl-76-204-181-223.dsl.pltn13.sbcglobal.net] has joined #lisp 05:18:57 Or C++. 05:19:12 -!- zu22 [n=ubuntu@unaffiliated/zu22] has left #lisp 05:19:33 pkhuong: But one of my friend was saying to me that #lisp the langugae that ppl use to write AI algorithms . 05:19:37 -!- ia [n=ia@89.169.189.230] has quit [Read error: 110 (Connection timed out)] 05:19:48 I'm new to heart surgery, can someone help me? I don't know where to start or what tools to use. All I know is that I want to do heart surgery. I'm a student willing to research on Cardiology. 05:20:13 Chris: use a scalpel! 05:20:22 hefner: cool, thanks! 05:20:29 Chris: I know someone will come up with something like this ;) 05:20:32 hefner: circular saw through the thorax first, I believe. 05:20:52 -!- existentialmonk [n=carcdr@64-252-41-17.adsl.snet.net] has quit [Remote closed the connection] 05:21:14 ia [n=ia@89.169.189.230] has joined #lisp 05:21:27 I don't have a circular saw, but I'm very patient, so maybe this belt sander will work. 05:21:46 Sandeep`: AFAICT, AI that works (e.g. machine learning) mostly happens in C++. 05:22:22 Chris : you know , even if I asked a stupid question , ppl are here to help me :D 05:22:43 Thanks pkhoung 05:23:06 -!- gemelen [n=shelta@shpd-92-101-140-182.vologda.ru] has quit [Read error: 60 (Operation timed out)] 05:23:07 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:23:17 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:23:27 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 05:23:39 I have to wonder, who's pkhoung? People seem to often confuse us with one another. 05:24:19 I made that typo today, twice, trying to find the path to my sbcl source. 05:24:36 -!- DukeOfMilan [n=DukeOfMi@unaffiliated/dukeofmilan] has left #lisp 05:25:05 pkhuong: incidentally, I've been using your sbcl version for the past week without incident, except I haven't actually done anything with complexes =/ 05:26:51 gemelen [n=shelta@shpd-92-101-133-239.vologda.ru] has joined #lisp 05:27:01 eno_ [n=eno@adsl-70-137-139-149.dsl.snfc21.sbcglobal.net] has joined #lisp 05:27:45 Don't think I've touched anything non-floaty there. Inline constants will be nice to have. Should speed up random pieces of code, e.g. abs or %negate. 05:28:07 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:28:17 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:28:49 -!- Sandeep` [n=Agent@122.172.113.200] has left #lisp 05:33:29 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 05:34:17 -!- ASau [n=user@193.138.70.52] has quit ["off"] 05:35:34 milanj [n=milan@79.101.76.232] has joined #lisp 05:38:04 -!- milanj [n=milan@79.101.76.232] has quit [Client Quit] 05:38:06 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:38:17 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:39:00 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 110 (Connection timed out)] 05:40:06 -!- elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has quit [Read error: 113 (No route to host)] 05:42:45 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 05:43:08 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:43:18 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:44:22 drafael1 [n=tapio@ip-118-90-134-212.xdsl.xnet.co.nz] has joined #lisp 05:45:52 -!- drafael [n=tapio@ip-118-90-133-176.xdsl.xnet.co.nz] has quit [Read error: 60 (Operation timed out)] 05:45:54 drafael [n=tapio@ip-118-90-134-213.xdsl.xnet.co.nz] has joined #lisp 05:48:08 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:48:18 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:48:55 any idea what slime-enable-evaluate-in-emacs does? 05:53:07 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:53:13 SandGorgon [n=OmNomNom@122.163.210.253] has joined #lisp 05:53:17 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:54:45 -!- drafael1 [n=tapio@ip-118-90-134-212.xdsl.xnet.co.nz] has quit [Read error: 60 (Operation timed out)] 05:55:36 plage [n=user@118.68.32.7] has joined #lisp 05:55:40 Good afternoon 05:57:48 Bigshot_ [n=BIG_SHOT@CPE002129abc864-CM001ac35cd4d0.cpe.net.cable.rogers.com] has joined #lisp 05:58:10 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 05:58:20 kami- [n=user@unaffiliated/kami-] has joined #lisp 05:59:58 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 06:00:54 good morning 06:03:10 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 54 (Connection reset by peer)] 06:03:20 kami- [n=user@unaffiliated/kami-] has joined #lisp 06:04:27 -!- Bigshot_ [n=BIG_SHOT@CPE002129abc864-CM001ac35cd4d0.cpe.net.cable.rogers.com] has quit ["ChatZilla 0.9.85 [Firefox 3.0.11/2009060215]"] 06:06:13 morning 06:08:10 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 06:08:20 kami- [n=user@unaffiliated/kami-] has joined #lisp 06:09:29 wbraun [n=wolfgang@p5B202543.dip0.t-ipconnect.de] has joined #lisp 06:10:58 -!- kami- [n=user@unaffiliated/kami-] has quit [Read error: 104 (Connection reset by peer)] 06:11:08 kami- [n=user@unaffiliated/kami-] has joined #lisp 06:17:17 -!- Quadrescence [n=quad@unaffiliated/quadrescence] has quit [Read error: 110 (Connection timed out)] 06:17:49 -!- X-Scale [i=email@89.180.205.13] has quit [Client Quit] 06:20:16 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 06:22:56 konr [n=konrad@201.82.132.33] has joined #lisp 06:23:10 -!- cp2 [n=will@unaffiliated/cp2] has quit [Read error: 113 (No route to host)] 06:24:28 -!- fe[nl]ix [n=algidus@88-149-208-244.dynamic.ngi.it] has quit [Read error: 113 (No route to host)] 06:24:28 lacedaemon [n=algidus@88-149-212-240.dynamic.ngi.it] has joined #lisp 06:24:53 Davse_Bamse [n=davse@130.226.210.2] has joined #lisp 06:30:28 -!- konr [n=konrad@201.82.132.33] has quit [Remote closed the connection] 06:30:32 -!- JHVH [n=jarv@dsl-217-155-101-22.zen.co.uk] has quit [Remote closed the connection] 06:30:49 -!- lacedaemon is now known as fe[nl]ix 06:31:21 sunny36 [n=somchok@203.159.92.21] has joined #lisp 06:31:58 -!- ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has quit [Read error: 110 (Connection timed out)] 06:37:15 nha [n=prefect@137-64.105-92.cust.bluewin.ch] has joined #lisp 06:38:28 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 06:38:34 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 06:38:42 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 54 (Connection reset by peer)] 06:39:18 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 06:46:16 -!- sunny36 [n=somchok@203.159.92.21] has quit [Read error: 104 (Connection reset by peer)] 06:46:42 sunny36 [n=somchok@203.159.92.21] has joined #lisp 06:46:58 anybody around 06:46:59 ? 06:47:18 who has any idea how to decipher the output of sb-c::print-all-blocks?? 06:47:24 splittist [n=dmurray@63-55.5-85.cust.bluewin.ch] has joined #lisp 06:47:27 morning 06:48:09 hi splittist 06:48:11 This_Is_My_World [i=sea@216.110.106.4] has joined #lisp 06:48:18 -!- sea4ever [i=sea@unaffiliated/sea4ever] has quit [Read error: 60 (Operation timed out)] 06:50:35 Soulman [n=kvirc@154.80-202-254.nextgentel.com] has joined #lisp 06:51:35 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 06:51:53 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 06:52:14 -!- SandGorgon [n=OmNomNom@122.163.210.253] has quit [Read error: 104 (Connection reset by peer)] 06:52:43 mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has joined #lisp 06:53:23 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit ["Leaving"] 06:53:27 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 06:53:35 drafael1 [n=tapio@ip-118-90-135-53.xdsl.xnet.co.nz] has joined #lisp 06:55:06 -!- drafael [n=tapio@ip-118-90-134-213.xdsl.xnet.co.nz] has quit [Read error: 60 (Operation timed out)] 06:58:35 mrSpec [n=NoOne@82.177.125.6] has joined #lisp 06:58:54 abeaumont_ [n=abeaumon@85.48.202.13] has joined #lisp 07:01:25 good morning 07:01:30 xinming_ [n=hyy@125.109.78.158] has joined #lisp 07:02:24 mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has joined #lisp 07:03:32 ASau [n=user@host112-230-msk.microtest.ru] has joined #lisp 07:06:27 -!- xinming [n=hyy@125.109.254.88] has quit [Read error: 60 (Operation timed out)] 07:06:32 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [""I am going to a commune in Vermont and will deal with no unit of time shorter than a season.""] 07:06:41 tombom [i=tombom@86.29.159.26] has joined #lisp 07:06:44 -!- This_Is_My_World [i=sea@216.110.106.4] has left #lisp 07:07:29 sea4ever [i=sea@unaffiliated/sea4ever] has joined #lisp 07:09:38 hello 07:09:56 SandGorgon [n=OmNomNom@122.162.134.73] has joined #lisp 07:11:35 fe[nl]ix: I checked :weak-depends-on of ASDF you suggested, but couldn't find anything related in the ASDF documentation. Would you mind pointing me to the related documentation? 07:11:43 -!- gemelen [n=shelta@shpd-92-101-133-239.vologda.ru] has quit ["I wish the toaster to be happy, too."] 07:12:28 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has quit ["Lost terminal"] 07:14:15 vy: it's in the sources :D 07:14:22 mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has joined #lisp 07:15:02 asksol [n=ask@pat-tdc.opera.com] has joined #lisp 07:17:02 -!- mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 07:17:33 legumbre_ [n=user@r190-135-21-133.dialup.adsl.anteldata.net.uy] has joined #lisp 07:18:08 mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has joined #lisp 07:18:22 mega1 [n=mega@pool-01d52.externet.hu] has joined #lisp 07:19:37 lacedaemon [n=algidus@88-149-211-230.dynamic.ngi.it] has joined #lisp 07:20:09 -!- fe[nl]ix [n=algidus@88-149-212-240.dynamic.ngi.it] has quit [Nick collision from services.] 07:20:17 -!- lacedaemon is now known as fe[nl]ix 07:21:08 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has quit ["Lost terminal"] 07:21:32 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 07:21:43 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 07:21:59 -!- antoni` [n=user@143.pool85-53-26.dynamic.orange.es] has quit [Remote closed the connection] 07:23:26 -!- pem [n=pem@159.226.35.246] has quit [Read error: 104 (Connection reset by peer)] 07:24:56 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 07:27:05 mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has joined #lisp 07:28:06 rstandy` [n=rastandy@net-93-144-243-56.t2.dsl.vodafone.it] has joined #lisp 07:28:26 -!- rstandy` [n=rastandy@net-93-144-243-56.t2.dsl.vodafone.it] has quit [Remote closed the connection] 07:28:29 -!- rstandy [n=rastandy@net-93-144-243-56.t2.dsl.vodafone.it] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 07:28:42 rstandy [n=rastandy@net-93-144-243-56.t2.dsl.vodafone.it] has joined #lisp 07:33:25 -!- legumbre [n=user@r190-135-27-47.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 07:34:44 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 60 (Operation timed out)] 07:37:35 lclark [n=user@cpe-98-148-69-2.socal.res.rr.com] has joined #lisp 07:38:59 alinp [n=alinp@86.122.9.2] has joined #lisp 07:40:32 Harag [n=phil@wbs-196-2-98-168.wbs.co.za] has joined #lisp 07:42:07 ignas [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 07:42:25 morning 07:42:33 rdd [n=user@c83-250-157-93.bredband.comhem.se] has joined #lisp 07:42:38 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit ["leaving"] 07:43:29 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 07:44:22 so if i have a sb-c::CLAMBDA, where i find it's "implemntation"? 07:44:23 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 07:44:25 I have asked this before but I am asking again is there some one out there that has an example of using hunchentoot-vhost......or can tell me what the "host suffixes" are that is mentioned in the documentation in the code ? 07:44:27 children? 07:45:52 smoofra: IIRC, M-. should work on most sbcl-internal symbols 07:45:55 (in slime) 07:46:13 despite the sb!c/sb-c thing 07:46:13 yea it does.... 07:46:14 are there technical reasons why EuLisp never took off? 07:46:46 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 07:47:12 moocow [n=new@mail.fredcanhelp.com] has joined #lisp 07:47:40 Harag: slyrus wrote it, if he drops by, he should be able to answer your question (you may get a quicker response on the hunchentoot development mailing list, too) 07:48:16 hi lisp 07:50:04 -!- bobf [n=bob@unaffiliated/bob-f/x-6028553] has quit [Read error: 54 (Connection reset by peer)] 07:50:04 antifuchs: thanx will try 07:50:25 bobf_ [n=bob@host81-151-255-247.range81-151.btcentralplus.com] has joined #lisp 07:50:47 antifuchs: do you understand what purpose XEPs serve? 07:51:45 xtagon [n=xtagon@97-113-154-48.tukw.qwest.net] has joined #lisp 07:51:47 other than that they're external entry points? not really. 07:51:52 -!- wbraun [n=wolfgang@p5B202543.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)] 07:53:06 *younder* is annoyed by the way Rod Gatteth prances on Erik's grave 07:53:20 younder: linky? 07:53:34 xb 07:53:43 nop. comp.lang.lisp 07:53:53 danke 07:55:36 Kent Pitman wrote a very Kent Pitmanish tribute to Erik on his blog :) 07:56:16 yes, he did 07:57:01 frozsyn [n=FrozSyn@wafer.futurs.inria.fr] has joined #lisp 07:57:40 also wikipedia has been updated.. 08:00:59 In Oslo amoung lispers he is missed 08:02:25 did you meet him younder? 08:02:26 athos [n=philipp@92.250.250.68] has joined #lisp 08:02:59 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 60 (Operation timed out)] 08:03:11 Who will make me think before I write now? :( 08:03:15 He was at the university of oslo aboutthe time I was there. (he started 84, me 86) 08:04:43 so yes we met 08:05:16 but we were never friends 08:05:18 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 08:06:11 -!- xinming_ [n=hyy@125.109.78.158] has quit [Remote closed the connection] 08:06:22 xinming [n=hyy@125.109.78.158] has joined #lisp 08:07:16 holymoly [n=new@mail.fredcanhelp.com] has joined #lisp 08:07:21 I somewhat regret that, we saw eye to eye on many issues 08:07:50 oh, well. Too late now 08:07:51 -!- xtagon [n=xtagon@97-113-154-48.tukw.qwest.net] has quit ["Leaving"] 08:10:09 *younder* hates Perl 08:12:03 -!- Raiford_ [n=rai@234.157-62-69.ftth.swbr.surewest.net] has quit [Client Quit] 08:15:19 This_Is_My_World [i=sea@216.110.106.5] has joined #lisp 08:17:12 *younder* sniggers. This is your brain, this is Perl. This is your brain on Perl. 08:17:32 RIP Erik 08:23:22 -!- sea4ever [i=sea@unaffiliated/sea4ever] has quit [Connection timed out] 08:26:20 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 08:28:40 -!- fe[nl]ix [n=algidus@88-149-211-230.dynamic.ngi.it] has quit [Remote closed the connection] 08:33:36 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [""I am going to a commune in Vermont and will deal with no unit of time shorter than a season.""] 08:34:01 -!- sbahra [n=sbahra@c-76-21-209-249.hsd1.dc.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 08:34:24 kmcorbett1 [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has joined #lisp 08:34:31 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Remote closed the connection] 08:36:29 wbraun [n=wolfgang@vpn0512.extern.uni-tuebingen.de] has joined #lisp 08:36:48 sykopomp [n=sykopomp@unaffiliated/sykopomp] has joined #lisp 08:38:10 -!- dmiles [n=dmiles@c-71-197-210-170.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 08:38:20 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 08:38:57 -!- mrsolo [n=mrsolo@adsl-68-126-209-116.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 08:39:27 sbahra [n=sbahra@c-76-21-209-249.hsd1.dc.comcast.net] has joined #lisp 08:39:35 attila_lendvai_ [n=ati@catv-89-132-189-132.catv.broadband.hu] has joined #lisp 08:40:34 ignas_ [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 08:40:37 -!- A_anekos is now known as anekos 08:41:17 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit [Read error: 110 (Connection timed out)] 08:42:41 -!- younder [n=jthing@165.244.251.212.customer.cdi.no] has quit [Remote closed the connection] 08:43:00 younder [n=jthing@165.244.251.212.customer.cdi.no] has joined #lisp 08:43:38 dmiles_afk [n=dmiles@c-71-197-210-170.hsd1.wa.comcast.net] has joined #lisp 08:44:17 zu22 [n=ubuntu@unaffiliated/zu22] has joined #lisp 08:44:30 does ASDF and ASDF-install come with SBCL or i must install separately? 08:45:02 comes with 08:45:38 (require 'asdf) and (require 'asdf-install) 08:45:52 ok thanks! 08:46:08 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 08:47:00 do i stick that in my .emacs or my .sbclrc? 08:47:21 I have it in my .sbclrc 08:48:07 look at en.wikibooks.org/wiki/Common_Lisp/External_libraries/ASDF for some useful stuff 08:48:14 right, away, good luck 08:48:21 drafael1: ok thx! 08:48:24 dfox foom pragma_ azuk` kei_ PissedNumlock slyrus _3b Bootvis l_a_m leo2007 08:48:27 + [df] ineiros jkan 08:48:36 lol not sure what happened there :) 08:48:43 some crazy paste 08:51:00 -!- kmcorbett [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 08:51:15 prxq [n=mommer@e176248028.adsl.alicedsl.de] has joined #lisp 08:51:17 -!- cipher [i=weinsd0@monica.cs.rpi.edu] has quit [Read error: 110 (Connection timed out)] 08:51:18 morning 08:51:24 Dalenius [n=chatzill@lap-67.ida.liu.se] has joined #lisp 08:52:17 hi nicktastic 08:52:27 oops :-) ni nikodemus 08:54:08 Quadrescence [n=quad@unaffiliated/quadrescence] has joined #lisp 08:54:19 in cffi, how do I pass an integer by reference? The function wants a pointer to an integer, and there should be something usefull in it when it returns 08:55:59 prxq: http://common-lisp.net/project/cffi/manual/cffi-manual.html#with_002dforeign_002dobject 08:57:22 prxq: what are you up to these days? 08:58:26 jewel [n=jewel@dsl-242-129-65.telkomadsl.co.za] has joined #lisp 08:58:39 nikodemus: remember that convert-mv-call patch i sent a long time ago... 08:59:02 nikodemus: i figured out some more of what was causeing it to hit the aver 09:00:09 -!- benny [n=benny@i577A1855.versanet.de] has quit ["rcirc on GNU Emacs 23.0.94.1"] 09:00:54 -!- zu22 [n=ubuntu@unaffiliated/zu22] has left #lisp 09:02:18 -!- moocow [n=new@mail.fredcanhelp.com] has quit [Connection timed out] 09:04:07 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit ["night"] 09:05:31 krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has joined #lisp 09:07:05 blandest1 [n=blandest@softhouse.is.ew.ro] has joined #lisp 09:07:30 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [No route to host] 09:07:36 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 09:08:06 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 09:12:17 smoofra: doesn't ring a bell immediately 09:14:49 Athas [n=athas@192.38.109.188] has joined #lisp 09:15:33 mrsolo [n=mrsolo@adsl-68-126-183-252.dsl.pltn13.pacbell.net] has joined #lisp 09:17:38 -!- blandest1 is now known as blandest 09:19:47 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has quit ["leaving"] 09:22:09 mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has joined #lisp 09:22:54 ZabaQ [n=johnc@playboxgames.com] has joined #lisp 09:24:32 dys` [n=andreas@p5B314E07.dip.t-dialin.net] has joined #lisp 09:25:29 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 09:26:54 smoofra: interesting! sounds perfectly plausible to me 09:27:22 *ZabaQ* is coaxing ccl asdf-install to work with msys instead of cygwin. 09:27:42 smoofra: i'll try to find time to think about it a bit more today or tomorrow, and will probably commit then (unless someone else gets there first, of course) 09:29:11 -!- mikezor_ [n=mikael@c-36ed70d5.04-404-7570701.cust.bredbandsbolaget.se] has quit ["bbl"] 09:30:23 -!- dys [n=andreas@p5B3164A9.dip.t-dialin.net] has quit [Connection timed out] 09:32:20 ruediger [n=ruediger@62-47-155-52.adsl.highway.telekom.at] has joined #lisp 09:32:51 -!- mrSpec [n=NoOne@82.177.125.6] has quit [Read error: 113 (No route to host)] 09:33:15 mrSpec [n=NoOne@82.177.125.6] has joined #lisp 09:38:06 smoofra: are you getting more into SB-C hacking? 09:39:45 ianmcorvidae|alt [n=ianmcorv@ip70-162-187-21.ph.ph.cox.net] has joined #lisp 09:41:58 -!- SandGorgon [n=OmNomNom@122.162.134.73] has quit [Read error: 110 (Connection timed out)] 09:44:00 SandGorgon [n=OmNomNom@122.162.135.169] has joined #lisp 09:44:57 xan [n=xan@cs78225040.pp.htv.fi] has joined #lisp 09:48:32 -!- sunny36 [n=somchok@203.159.92.21] has quit [] 09:49:54 borism [n=boris@195-50-201-218-dsl.krw.estpak.ee] has joined #lisp 09:51:24 joswig [n=joswig@e177124221.adsl.alicedsl.de] has joined #lisp 09:53:03 -!- joswig [n=joswig@e177124221.adsl.alicedsl.de] has quit [Client Quit] 09:53:24 bakkdoor [n=bakkdoor@xdslgh071.osnanet.de] has joined #lisp 09:54:16 ruediger_ [n=ruediger@62-47-147-4.adsl.highway.telekom.at] has joined #lisp 09:55:22 -!- ianmcorvidae [n=ianmcorv@fsf/member/ianmcorvidae] has quit [Connection timed out] 09:55:48 Nshag [i=user@Mix-Orleans-106-2-114.w193-248.abo.wanadoo.fr] has joined #lisp 09:56:47 luis: shiny picture 09:57:53 felipe [n=felipe@my.nada.kth.se] has joined #lisp 09:59:15 it's shiny alright, but is it useful or acurate? 09:59:20 -!- wbraun [n=wolfgang@vpn0512.extern.uni-tuebingen.de] has quit [Read error: 104 (Connection reset by peer)] 10:00:22 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 10:00:42 -!- Axioplase is now known as Axioplase_ 10:04:11 -!- lispm [n=joswig@e177147004.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 10:05:17 nikodemus: I do research here at Heidelberg. Numerical methods for a variety of things. & u? 10:07:07 no dueling, huh. 10:08:42 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 10:09:02 pkhuong: got any pointers to Feeley's stuff you mentioned? 10:11:14 -!- ruediger [n=ruediger@62-47-155-52.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 10:12:01 prxq: you probably have a good handle on intricasies of floating point, then? 10:12:56 [Blay] [n=blaay@BSN-165-118-26.dial-up.dsl.siol.net] has joined #lisp 10:13:34 lol 10:14:08 That's why they have a university degree in it 10:15:35 take some advice, use the library 10:15:46 luis: it looks like an accurate summary to me 10:16:22 I guess the usual thing that people find difficult is to know which compiler (host or cross-compiler) compiles what 10:16:31 Fufie [n=poff@Gatekeeper.vizrt.com] has joined #lisp 10:16:35 you've got that in words, but that's why I did the boxes-inside-boxes thing in my paper 10:17:36 Oh well SBCL grinds the best Lisp code in objjecxt for numerical operations 10:18:24 but seems if you are a numerical analysis guy FORTRAN would be more your forte 10:18:43 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 113 (No route to host)] 10:22:07 fusss [i=73802570@gateway/web/freenode/x-1e72e7ef86e1581c] has joined #lisp 10:22:07 -!- xinming [n=hyy@125.109.78.158] has quit [Read error: 110 (Connection timed out)] 10:22:14 greetings 10:22:27 huh, how did i do that? 10:23:01 lol 10:23:17 *nikodemus* fixes 10:23:55 numerical analysis is a thorny field 10:23:59 so you can pretty much implement poor man's web continuations with hunchentoot's REFERER and REDIRECT functions .. 10:24:25 -!- BrianRice [n=water@c-98-225-51-246.hsd1.wa.comcast.net] has quit [Connection reset by peer] 10:24:27 I suppose 10:24:49 oops, numerical analysis topic in session .. hunchentoot discoveries must be /that/ way, at the kiddie table :-P 10:24:57 Not that I think that is wise 10:25:38 fusss: Your the kid here 10:25:57 you're 10:26:11 :-P 10:27:37 But if you must approach seminumerical algorithms try Donald Knuth olume II 10:29:03 younder: there might more approachable texts out there, specially if you just want something that works, yesterday 10:29:19 the much hated Numerical Recipes is actually good 10:29:19 -!- ia [n=ia@89.169.189.230] has quit [Read error: 110 (Connection timed out)] 10:29:51 fusss: I use numerical recpies in C 10:30:23 yeah, and version (1- N) is usually free online (read the license though, bit nasty) 10:30:25 good, but not exactly knuth materal 10:30:30 ia [n=ia@89.169.189.230] has joined #lisp 10:30:30 pem [n=pem@159.226.35.246] has joined #lisp 10:31:05 -!- Harag [n=phil@wbs-196-2-98-168.wbs.co.za] has left #lisp 10:31:26 prxq: i'm asking because you might want to chime in on http://thread.gmane.org/gmane.lisp.steel-bank.devel/13522 10:31:40 i reach out for my knuth volumes when i have free time, not when i need things done (not that i am good at anything programming related) 10:32:19 -!- holymoly [n=new@mail.fredcanhelp.com] has quit [Remote closed the connection] 10:33:19 xinming [n=hyy@125.109.78.158] has joined #lisp 10:34:13 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 113 (No route to host)] 10:34:50 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 10:36:52 -!- jewel [n=jewel@dsl-242-129-65.telkomadsl.co.za] has quit [Read error: 113 (No route to host)] 10:39:39 silenius [n=jl@yian-ho03.nir.cronon.NET] has joined #lisp 10:41:50 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 10:42:03 -!- This_Is_My_World [i=sea@216.110.106.5] has quit [Read error: 60 (Operation timed out)] 10:42:15 -!- Dalenius [n=chatzill@lap-67.ida.liu.se] has quit ["ChatZilla 0.9.85 [Firefox 3.0.11/2009060215]"] 10:42:18 here is an example of nice web design: http://kx.com/ 10:45:05 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 10:47:05 -!- fusss [i=73802570@gateway/web/freenode/x-1e72e7ef86e1581c] has quit ["Page closed"] 10:49:06 *Xach* steps back from fussss 10:51:18 Harag [n=phil@wbs-196-2-98-168.wbs.co.za] has joined #lisp 10:54:05 I got the "made with lisp logo" hacked to give a bit of tribute to hunchentoot/edi-ware for those of you using hunchentoot out there in the wild, I am using it on all my sites...Hopefuly by this afternoon it will have its own home ...http://196.2.98.168:8090/dx-site/lisp-edi-logo.html... 10:54:11 let me know waht you think 10:54:43 -!- authentic [n=authenti@unaffiliated/authentic] has quit ["[IRSSI]"] 10:55:26 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 10:55:56 antoni [n=user@143.pool85-53-26.dynamic.orange.es] has joined #lisp 10:56:00 blandest1 [n=blandest@softhouse.is.ew.ro] has joined #lisp 10:56:10 wbraun [n=wolfgang@hueckel.itc.uni-tuebingen.de] has joined #lisp 10:56:33 Harag: sweet - I though it was the hunchentoot logo creeping up on the lisp lizard 10:56:37 ejs [n=eugen@nat.ironport.com] has joined #lisp 10:57:01 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 10:57:19 -!- mrsolo [n=mrsolo@adsl-68-126-183-252.dsl.pltn13.pacbell.net] has quit [Read error: 113 (No route to host)] 10:58:53 authentic [n=authenti@85-127-41-26.dynamic.xdsl-line.inode.at] has joined #lisp 10:58:55 well edi ware is kick-A$$ so it had to be a serious lizard ;) 10:59:53 -!- authentic [n=authenti@unaffiliated/authentic] has quit [Client Quit] 10:59:59 there shouldn't be an "of" in there, should it? (no, i am not telling, just asking) 11:00:05 authentic [n=authenti@85-127-41-26.dynamic.xdsl-line.inode.at] has joined #lisp 11:01:10 space wise the of wsa a bit 'of' an issue ...no pun intneded 11:01:11 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 11:03:00 lde [n=user@184-dzi-2.acn.waw.pl] has joined #lisp 11:03:07 *_3b* would add the 'of' even at the expense of layout 11:04:03 -!- ignas_ is now known as ignas 11:08:35 hm, the (expt -1 integer) optimization applies just as well to -1.0 and -1.0d0, right? (if (oddp integer) -1-of-right-type 1-of-right-type) 11:08:47 or do i miss something? 11:11:32 -!- blandest1 [n=blandest@softhouse.is.ew.ro] has left #lisp 11:14:38 _3b ...next time my web designer owes me some time I will let him have a crack at adding the "off" ... 11:14:49 nikodemus: it applies, but how often does that actually happen? (expt -1 integer) happens a fair amount in transcriptions of mathematical formulae 11:17:19 I want to use a package which does not export any of its functions, how to use it and with all function defined in it available directly in my own package without prefixing package::? 11:18:56 import all of its functions into your package 11:19:03 s/functions/symbols/ 11:19:11 Xof: well, the same transform can handle all three cases 11:19:18 so it doesn't really cost anything 11:20:40 -!- sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has quit [] 11:20:56 Xof: Do you mean (import '(package::a package::b ... ))? There are lots of them:( 11:21:26 so do it programmatically 11:21:33 you're at least _trying_ to be a programmer 11:23:25 heh 11:24:05 Xof: nvm, I'll find a way around it. I am programming in cl with just knowledge of elisp. 11:24:47 -!- plage [n=user@118.68.32.7] has quit [Remote closed the connection] 11:24:49 -!- Athas [n=athas@192.38.109.188] has quit [Remote closed the connection] 11:27:01 simplechat [n=simplech@unaffiliated/simplechat] has joined #lisp 11:27:09 -!- authentic [n=authenti@unaffiliated/authentic] has quit ["[IRSSI]"] 11:27:28 authentic [n=authenti@85-127-41-26.dynamic.xdsl-line.inode.at] has joined #lisp 11:27:49 -!- seejay [n=seejay@unaffiliated/seejay] has left #lisp 11:31:07 seejay [n=seejay@plexyplanet.org] has joined #lisp 11:37:08 jao [n=jao@74.Red-80-24-4.staticIP.rima-tde.net] has joined #lisp 11:38:10 nikodemus: pong 11:40:37 ilitirit: g'day (: 11:41:07 -!- krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has quit [Remote closed the connection] 11:41:45 CyberBlue [n=yong@60.26.105.109] has joined #lisp 11:42:36 gemelen [n=shelta@shpd-92-101-133-239.vologda.ru] has joined #lisp 11:42:36 luis: not sure he ever computerised those drawings. Nothing fancy, though. He used T-shaped tetraminos, lhs was input, rhs output and bottom was the compiler. 11:42:49 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 11:44:30 yango [n=yango@unaffiliated/yango] has joined #lisp 11:44:38 mm. I cant use (last l) as a place? blast. 11:45:03 BTW, I don't understand the patch for SLEEP. What's wrong with actually doing it sort of right with a simple loop? 11:45:03 you could defun (setf last) 11:46:25 pkhuong: that was my thought too 11:46:26 Ralith: No, you wouldn't. That'd cause things like (incf (last list)) to traverse LIST twice. 11:46:32 -!- abeaumont_ [n=abeaumon@85.48.202.13] has quit [Read error: 110 (Connection timed out)] 11:46:52 pkhuong: better twice than not work at all, no? 11:47:23 Ralith: even better to understand how places are designed to avoid such idiocy. 11:47:37 *Ralith* shrugs 11:47:41 alright, how would you do it? 11:48:18 First, I'd try and understand why LAST, etc. return a CONS and not the last car. 11:48:33 Second, I'd see that defining anything on top of CL:LAST is undefined. 11:48:43 oh, last returns a cons? 11:48:45 didn't know that :P 11:48:54 spilman [n=spilman@ARennes-552-1-82-202.w92-139.abo.wanadoo.fr] has joined #lisp 11:49:10 benny [n=benny@24.121.113.82.net.de.o2.com] has joined #lisp 11:50:00 Third, I'd use define-setf-expander to make MY-LAST a place that doesn't traverse its input gratuitously often. 11:50:56 see, all those have nothing to do with understanding and everything to do with having the functions offered by cl memorized. 11:51:19 sellout [n=greg@guest-fw.dc4.itasoftware.com] has joined #lisp 11:51:37 Ralith: no. The fact that places may be defined by setf-expanders, and not only functions, is a fundamental part of places. 11:52:10 pkhuong: and the fact that last returns a cons, which makes the whole discussion rather moot, is a matter of having behavior memorized. 11:52:55 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 11:53:43 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 11:54:36 Ralith: such is the path of evolved technologies 11:56:30 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit ["+++ killed by SIGSEGV +++"] 12:00:04 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 12:00:27 timor [n=martin@w4227.dip.tu-dresden.de] has joined #lisp 12:00:42 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 110 (Connection timed out)] 12:01:08 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 12:01:39 existentialmonk [n=carcdr@64-252-41-17.adsl.snet.net] has joined #lisp 12:01:55 -!- splittist [n=dmurray@63-55.5-85.cust.bluewin.ch] has quit ["rcirc on GNU Emacs 22.3.1"] 12:02:03 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 12:04:22 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 12:05:38 antoszka [n=antoszka@unaffiliated/antoszka] has joined #lisp 12:06:26 pkhuong: ping 12:07:06 re. sleep: *shrug* -- sure, you can loop just as well 12:07:07 You pinged yesterday. 12:07:12 oh, hm 12:07:18 don't remember anymore 12:07:29 Ppjet6_ [n=ppjet@ivr94-11-88-187-39-7.fbx.proxad.net] has joined #lisp 12:09:55 mrsolo [n=mrsolo@adsl-68-126-179-209.dsl.pltn13.pacbell.net] has joined #lisp 12:10:38 -!- Ppjet6_ [n=ppjet@ivr94-11-88-187-39-7.fbx.proxad.net] has quit [Remote closed the connection] 12:10:40 -!- benny [n=benny@24.121.113.82.net.de.o2.com] has quit [Remote closed the connection] 12:14:57 stassats` [n=stassats@wikipedia/stassats] has joined #lisp 12:15:50 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 12:17:04 -!- mrsolo [n=mrsolo@adsl-68-126-179-209.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 12:19:23 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 12:21:49 (setf (last x) z) == (setf (cdr (last x 2)) (list z)) 12:22:27 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 12:23:00 dlowe [n=dlowe@ita4fw1.itasoftware.com] has joined #lisp 12:23:00 -!- blandest [n=blandest@softhouse.is.ew.ro] has left #lisp 12:23:45 And CL:LAST returns the last cons only for historical reasons. (But not so old as LISP 1.5 which hadn't that function). 12:26:36 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 12:27:21 slackaholic [i=1000@187-25-176-234.3g.claro.net.br] has joined #lisp 12:28:18 -!- slackaholic [i=1000@187-25-176-234.3g.claro.net.br] has quit [Client Quit] 12:28:20 lispo [n=SEone@cm241.delta217.maxonline.com.sg] has joined #lisp 12:31:07 -!- eno_ is now known as eno 12:31:58 envi^home [n=envi@220.121.234.156] has joined #lisp 12:36:49 LinkFly [n=linkfly@94.158.186.63] has joined #lisp 12:39:34 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-36-179.w82-125.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 12:43:32 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 113 (No route to host)] 12:44:02 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 12:46:43 http://paste.lisp.org/display/82486 bad idea? .. been trying out the latest sbcl with the fast constant-type make-instance thingy 12:48:58 -!- manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has quit [Read error: 113 (No route to host)] 12:51:15 mvilleneuve [n=mvillene@ABordeaux-253-1-140-122.w81-50.abo.wanadoo.fr] has joined #lisp 12:51:24 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 12:52:33 -!- CyberBlue [n=yong@60.26.105.109] has quit ["Leaving"] 12:52:53 -!- timor [n=martin@w4227.dip.tu-dresden.de] has quit [Read error: 110 (Connection timed out)] 12:53:35 -!- ASau [n=user@host112-230-msk.microtest.ru] has quit ["off"] 12:54:56 -!- Fufie [n=poff@Gatekeeper.vizrt.com] has quit ["Leaving"] 12:56:13 elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has joined #lisp 12:57:35 -!- SandGorgon [n=OmNomNom@122.162.135.169] has quit [Read error: 113 (No route to host)] 12:58:05 BrianRice [n=water@c-98-225-51-246.hsd1.wa.comcast.net] has joined #lisp 13:00:23 lnostdal: it generates a type-check, which is not really optimal 13:00:36 oh, .. hm .. truly-the or something? 13:01:09 neilv [n=user@dsl092-071-029.bos1.dsl.speakeasy.net] has joined #lisp 13:01:13 lnostdal: TRULY-THE would avoid that, but i'm not sure if there are really any places where (currently) having the type known to the compiler would make much of a difference 13:03:25 ASau [n=user@77.246.230.112] has joined #lisp 13:03:34 *neilv* takes c.l.l out back and shoots it 13:03:48 LiamH [n=none@common-lisp.net] has joined #lisp 13:04:42 what? .. you don't like froggy? :) 13:05:47 lnostdal: though if we had inline GF dispatch things might be different 13:06:00 LiamH must be l33t 13:09:12 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 13:10:40 -!- LinkFly [n=linkfly@94.158.186.63] has quit [Read error: 110 (Connection timed out)] 13:15:06 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit ["gbu"] 13:15:34 -!- legumbre_ is now known as legumbre 13:16:13 -!- neilv [n=user@dsl092-071-029.bos1.dsl.speakeasy.net] has quit ["Leaving"] 13:18:24 naunsins [i=charles_@120.138.100.90] has joined #lisp 13:19:55 -!- naunsins [i=charles_@120.138.100.90] has left #lisp 13:20:17 TDT [i=dthole@dhcp80ff869b.dynamic.uiowa.edu] has joined #lisp 13:20:26 kmcorbett [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has joined #lisp 13:20:30 -!- sunwukong [n=vukung@210-20-93-83.rev.home.ne.jp] has quit [Read error: 110 (Connection timed out)] 13:22:43 -!- rstandy [n=rastandy@net-93-144-243-56.t2.dsl.vodafone.it] has quit [Read error: 110 (Connection timed out)] 13:28:48 bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has joined #lisp 13:29:33 malc_ [n=malc@78-86-4-156.zone2.bethere.co.uk] has joined #lisp 13:29:46 -!- malc_ is now known as malcolm_reynolds 13:31:58 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Operation timed out] 13:32:22 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 13:33:35 -!- lispo [n=SEone@cm241.delta217.maxonline.com.sg] has quit ["leaving"] 13:34:49 -!- prip_ [n=_prip@host212-129-dynamic.52-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 13:35:00 prip_ [n=_prip@host98-135-dynamic.43-79-r.retail.telecomitalia.it] has joined #lisp 13:35:11 hm, something i don't understand .. (let ((x (the fixnum 42)) ..) that info seems to be lost .. but for (let ((x 42)) (declare fixnum x)) ..) it's there 13:35:23 i can paste some code which show this 13:36:47 just thinking it might be handy if that info was there; macros that expand to (the (make-something ..)) or so 13:36:55 ..for compiler-macros and stuff 13:37:08 xinming_ [n=hyy@125.109.248.255] has joined #lisp 13:37:30 lnostdal: 42 already is a fixnum; it's also an (EQL 42). The issue that the variable _x_ doesn't inherit the type fixnum just because of the initial value. You could later SETF that binding to an arbitrary value. 13:38:02 -!- kmcorbett1 [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 13:38:28 ok, so (the fixnum 42) just talks about the return value "at that very point" .. heh 13:39:09 i see 13:39:12 (i think) :) 13:40:48 hm, i saw someone working on custom a LET macro and other binding forms a while back -- wrt. types 13:41:00 a custom LET* 13:42:06 bobbysmith007 [n=russ@one.firewall.gnv.acceleration.net] has joined #lisp 13:42:58 isnt the DECLARE essentially a promise to the compiler that the variable is of a certain type? 13:43:36 http://paste.lisp.org/display/82486#1 13:44:39 -!- ejs [n=eugen@nat.ironport.com] has quit [Read error: 110 (Connection timed out)] 13:45:34 -!- xinming [n=hyy@125.109.78.158] has quit [Read error: 60 (Operation timed out)] 13:46:31 ttp://lisp.ystok.ru/ylocal-time/ 13:46:47 for thos who are concerned with time 13:47:03 -!- elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has quit [Read error: 113 (No route to host)] 13:47:07 http://lisp.ystok.ru/ylocal-time/ 13:48:10 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 13:48:18 this package is based on Erik Naggum's the long painfull history of time 13:48:27 It is also efficient 13:49:23 solime [n=boot@cm241.delta217.maxonline.com.sg] has joined #lisp 13:50:00 younder: how does it relate to http://common-lisp.net/project/local-time/ ? 13:50:24 It doesn't 13:50:46 It does however fix many of the problems 13:52:43 The local-time package by onShore, Inc. doesn't seem to actually have a distribution point 13:52:43 -!- solime [n=boot@cm241.delta217.maxonline.com.sg] has quit [Client Quit] 13:53:53 ejs [n=eugen@95.135.126.183] has joined #lisp 13:54:09 but, hm, is it perhaps reasonable for the compiler to assume that the (the fixnum ..) declaration holds since X isn't SETFed? .. (i might be wasting peoples time; i have done the RTFM-thing wrt. compiler stuff ....) 13:54:25 i have _not_* .. *sigh* 13:54:26 merimus [n=wroth@nat-49.laurelnetworks.com] has joined #lisp 13:54:47 lnostdal: the compiler will figure that out, but the information isn't there yet at macro-expansion time 13:54:52 attila_lendvai_: I asked the same question on c.l.l 13:55:11 Taggnostr [n=x@wolf.yok.utu.fi] has joined #lisp 13:55:17 do the *...* thingy, SBCL is picky about that 13:55:25 the answer is: it predates that local-time by a lot 13:55:35 the ystok parser looks nicer. Uses the meta pattern matcher 13:56:13 jewel_ [n=jewel@dsl-242-129-65.telkomadsl.co.za] has joined #lisp 13:56:21 nikodemus, ah .. so this knowledge is not something a compiler-macro can take advantage, then? 13:56:37 ..i.e., it too works at macro-expansion time 13:56:50 advantage of* 13:56:56 -!- merimus [n=wroth@nat-49.laurelnetworks.com] has quit [Client Quit] 13:57:19 merimus [n=wroth@nat-49.laurelnetworks.com] has joined #lisp 13:58:27 -!- ruepel0r [n=rue@203.97.49.162] has quit [Read error: 60 (Operation timed out)] 13:58:28 grr, are initialize-instance and shared-initialize :before methods basically useless if you use initforms? 13:58:29 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 13:58:48 *Xach* uses :default-initargs 13:58:50 pinterface [n=pinterfa@knvl-static-09-0024.dsl.iowatelecom.net] has joined #lisp 13:59:10 -!- dys` is now known as dys 13:59:17 -!- Lectus [n=Lectus@189.105.42.131] has quit [Read error: 104 (Connection reset by peer)] 13:59:53 I wish I had named the local-time library something different now :p 14:00:00 but I wasn't aware of any other implementations 14:00:14 Xach: yes, I see the light now, but I can't keep subclasses from using initforms. 14:00:14 Xach: This version is revised and released today 14:00:32 -!- bakkdoor [n=bakkdoor@xdslgh071.osnanet.de] has quit ["Verlassend"] 14:00:51 it's sad. it tells it all about the cl community... i was patching a library for years and now it turns out that there was another library with almost the same name, based on the same paper, that was started about 5 years earlier... 14:00:58 -!- attila_lendvai_ is now known as attila_lendvai 14:01:07 lnostdal: pretty much 14:01:25 attila_lendvai: dimitry commercially sells lisp software, i believe 14:01:27 *younder* mutters something about misinforation 14:01:29 attila_lendvai: yes, but I'm much more grateful for your patches than they would have been ;) 14:01:37 eheh 14:01:37 Blkt [n=Blkt@160.80.125.170] has joined #lisp 14:02:19 dlowe: heh, ok, that's some relief at least... :) 14:03:20 fe[nl]ix [n=algidus@89.202.147.18] has joined #lisp 14:03:33 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 14:04:02 -!- ZabaQ [n=johnc@playboxgames.com] has quit [Read error: 110 (Connection timed out)] 14:04:10 dlowe: out of curiosity, did you know about yl-t when you started yours? 14:04:17 hello 14:04:31 benny [n=benny@i577A1855.versanet.de] has joined #lisp 14:04:31 attila_lendvai: I did not. I would have submitted patches instead if I had known. 14:05:17 salex [n=user@216.80.147.206] has joined #lisp 14:05:19 (if I thought any were necessary. it looks pretty complete) 14:05:58 danlei [n=user@pD9E2F0D0.dip.t-dialin.net] has joined #lisp 14:07:01 'morning 14:09:22 fiveop [n=fiveop@pD9E6C8E2.dip.t-dialin.net] has joined #lisp 14:09:34 can slime filter out long lines? 14:09:48 I got an array that is 5040x12 14:10:02 and it makes emacs extremely slow, I thought something was wrong 14:10:14 (setf *print-array* nil) 14:10:27 or set *print-length* to something smallish 14:11:42 KingNatoG5 [n=patrik@84-217-4-148.tn.glocalnet.net] has joined #lisp 14:12:15 thanks 14:12:30 hi attila_lendvai! I did some hackage on cl-walker 14:12:38 I'll just remember to clear the buffer 14:13:20 tcr: hi! ok, sound good... send a repo link or some patches 14:13:46 can loop do two loops? 14:13:58 attila_lendvai: I will. Got to do the laundry first, though. 14:14:03 leo2007: if you nest two loops, yeah 14:14:17 ok, that's what I am doing 14:14:31 attila_lendvai: do you use iolib's ip address readmacro ? 14:15:02 fe[nl]ix: no, i think i don't 14:15:25 fe[nl]ix: #ip127.0.0.1? 14:15:39 @127.0.0.1 14:15:45 I like mine better! 14:15:54 we don't input so many ip addresses, and we serialize them through byte arrays 14:17:09 I'm changing that according to naggum's idea of named readmacros: #/ip/127.0.0.1 14:17:19 I find that much nicer 14:18:04 sure is, and it's also scalable to arbitrary meta-data. really, really cool. 14:19:53 fe[nl]ix: but who will handle the namespace? you can unread-char only one char, so chaining the reader lambda's don't work... cl-syntax-sugar? ;) 14:20:17 krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has joined #lisp 14:20:31 attila_lendvai: the package system? 14:20:56 exactly 14:20:57 the only reason why it sucks for FORMAT is because they didn't specify a global variable which contains a default package 14:21:34 or even default to *PACKAGE* 14:22:12 Perhaps there's a reason for not defaulting to *PACKAGE*; gotta read any issue write up first 14:22:29 oh, i see... so you just funcall whatever symbol READ returns... 14:22:38 yes, pretty much 14:22:54 -!- dalton is now known as I_Like_Turtles 14:22:55 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 14:23:12 tcr: the reason is most probably that the value of *package* is taken at runtime and it can lead to some cruel bugs... 14:23:26 Oh very right 14:23:34 not the case for a reader-macro though! 14:24:08 indeed. with #// we could have nice things. 14:24:17 Seems to be good days comming for reader-macros... I'm also revamping the measures package at work 14:26:35 <_3b> is there any benefit over #. aside from saving a character or 2? 14:26:59 <_3b> never mind, i'm not thinking properly :) 14:27:41 -!- KingNatoG5_ [n=patrik@84-217-2-195.tn.glocalnet.net] has quit [Read error: 110 (Connection timed out)] 14:30:34 milanj [n=milan@79.101.76.232] has joined #lisp 14:30:53 -!- asksol [n=ask@pat-tdc.opera.com] has quit ["Be back later"] 14:31:57 postamar [n=postamar@x-132-204-253-51.xtpr.umontreal.ca] has joined #lisp 14:31:59 schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has joined #lisp 14:34:08 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 14:35:51 -!- wbraun [n=wolfgang@hueckel.itc.uni-tuebingen.de] has quit [Read error: 110 (Connection timed out)] 14:36:25 mikezor [n=mikael@c-4fe070d5.04-404-7570701.cust.bredbandsbolaget.se] has joined #lisp 14:37:36 attila_lendvai: will you be here later? 14:38:05 jleija [n=jleija@user-24-214-122-46.knology.net] has joined #lisp 14:38:31 tcr: yes, i think. there's some chance that i go out tonight though... 14:39:35 hrm, as a programmer i should have answered "yes (75%)"... :) 14:39:37 jmbr [n=jmbr@209.33.220.87.dynamic.jazztel.es] has joined #lisp 14:40:35 What advantage would a read-macro have over an 'ip' function? like (ip 127.0.0.1) ? 14:41:05 127.0.0.1 is not valid lisp read syntax 14:41:06 p8m: think again 14:41:06 p8m: what kind of data is 127.0.0.1? 14:41:08 you mean #.(ip "127.0.0.1")? 14:41:39 also the reader macro can be smart and put the read function on the symbol plist to avoid name clashes, etc... 14:41:47 p8m: being able to use it in quoted lists, for example 14:41:48 s/put/get/ 14:41:52 also, IP addresses are a bit simplistic. think regexes... #/m/foo.*bar/g (-: 14:41:53 attila_lendvai: that's what I thought too 14:41:54 -!- malcolm_reynolds [n=malc@78-86-4-156.zone2.bethere.co.uk] has quit [] 14:42:05 Ah so read-macros enhance the... reader 14:42:13 who'd have thought (-: 14:43:25 malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has joined #lisp 14:43:30 *dlowe* already has a #// regex read macro :( 14:44:01 #/foo/ism works, even 14:44:17 dlowe: Is it publicly available? 14:44:28 drforr_: on my blog, I think. It's short. 14:44:37 Address? 14:44:42 cl-bench clos benchmarks truly suck 14:44:45 heh. last entry I made: http://www.bitmuse.com/blog/ 14:45:10 Thanks. That might make its way into a perl module... 14:45:21 my make-instance improvements make no difference there, because the cases with non-constant class names also have non-constant initargs 14:45:23 dlowe: in reference to #//, I considered #/// as a regex prefix... then reconsidered (-: 14:46:13 nikodemus: hehe, time to get that optimization code again and extend the caches... ;) 14:46:27 the cl-interpol #?// isn't too bad, I guess. doesn't support the perl flags, though 14:49:48 asksol [n=ask@175.243.251.212.customer.cdi.no] has joined #lisp 14:51:43 fe[nl]ix: I've just ported release.sh to use git and get version numbers from tags instead. 14:52:09 cool 14:52:11 whre is it ? 14:53:13 http://common-lisp.net/~loliveira/tmp/release.sh 14:54:05 -!- daniel_ is now known as daniel 14:56:02 -!- drafael1 [n=tapio@ip-118-90-135-53.xdsl.xnet.co.nz] has quit [Read error: 60 (Operation timed out)] 14:58:58 -!- dmiles_afk [n=dmiles@c-71-197-210-170.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 14:59:08 dmiles_afk [n=dmiles@c-71-197-210-170.hsd1.wa.comcast.net] has joined #lisp 15:00:50 nikodemus: Perhaps didier should add the code of his paper to cl-bench? 15:02:08 would be cool if antifuchs added ccl and ecl to boinkmarks 15:02:17 ccl is hard 15:02:21 how comes? 15:02:22 slyrus_ [n=slyrus@adsl-75-55-213-251.dsl.pltn13.sbcglobal.net] has joined #lisp 15:02:26 ecl is slightly less hard and should be doable 15:02:34 ok, make-instance stuff is in cvs now 15:03:03 I just couldn't stand rewriting build scripts for every time ecl changed their build process 15:03:25 tcr: I prefer things that can be automatically built (: 15:03:43 otoh, ccl ships binaries that should be able to bootstrap in some svn repo 15:03:57 antifuchs: Surely the ecl build will settle down as the project progresses. 15:04:21 -!- ASau [n=user@77.246.230.112] has quit ["off"] 15:05:26 I am getting this package "ASDF-INSTALL" not found when trying to install packages 15:05:34 Is that odd? 15:05:45 the command I ran was (asdf-install:install :cl-ppcre)) 15:05:53 (asdf-install:install :cl-ppcre) 15:05:55 morning 15:06:02 ruepel0r [n=rue@203-97-49-162.dsl.clear.net.nz] has joined #lisp 15:06:22 antifuchs: Sure that could be automated, too. You should kindly ask on the ccl mailinglist whether they see value to be added, and if so, they should come up with some script! 15:06:26 antifuchs: same for ecl 15:07:54 slyrus: I understand that hunchentoot-vhost is your baby? 15:08:36 I tend to require builds of historic revisions, and supporting every build process ever is enormously complex 15:08:50 but yeah, let's have the original developers worry about this (-; 15:09:15 Just don't for them :) 15:10:28 -!- ejs [n=eugen@95.135.126.183] has quit ["This computer has gone to sleep"] 15:11:02 eno_ [n=eno@adsl-70-137-139-149.dsl.snfc21.sbcglobal.net] has joined #lisp 15:11:13 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["Leaving"] 15:11:31 Harag: baby is perhaps a strong word, but, yeah 15:11:37 hehehe 15:11:54 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 104 (Connection reset by peer)] 15:11:58 can you please help a simpleton with an example of its use 15:12:16 sure, but i've got to run to catch a boat shortly. later today perhaps? 15:12:52 how many hours from now is later today so I can make sure I am on...here it is already 17:12 15:13:08 hmmm... last time I built ECL it was "configure; make; make install 15:13:31 (with one small exception where it was "edit Makefile; nmake; copy shit manually) 15:13:53 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 15:15:03 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 15:15:40 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit [Read error: 104 (Connection reset by peer)] 15:16:12 ignas [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 15:16:37 slyrus: vhost is the last little bit of a puzzle I need to put my sites out there in the wild so any help would be much appreciated 15:18:44 crypto_ [n=z0d@artifact.hu] has joined #lisp 15:18:50 wbraun [n=wolfgang@p5B202543.dip0.t-ipconnect.de] has joined #lisp 15:20:32 ejs [n=eugen@95.135.126.183] has joined #lisp 15:20:53 -!- asksol [n=ask@175.243.251.212.customer.cdi.no] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- merimus [n=wroth@nat-49.laurelnetworks.com] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- lde [n=user@184-dzi-2.acn.waw.pl] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- alinp [n=alinp@86.122.9.2] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- z0d [n=z0d@unaffiliated/z0d] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- holycow [n=new@mail.wjsgroup.com] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- minion [n=minion@common-lisp.net] has quit [verne.freenode.net irc.freenode.net] 15:20:53 -!- Khisanth [n=Khisanth@pool-141-157-237-196.ny325.east.verizon.net] has quit [verne.freenode.net irc.freenode.net] 15:21:10 nyef [n=nyef@65-125-125-98.dia.static.qwest.net] has joined #lisp 15:21:28 jleija [n=jleija@user-24-214-122-46.knology.net] has joined #lisp 15:27:00 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 15:27:00 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 15:27:20 -!- wbraun [n=wolfgang@p5B202543.dip0.t-ipconnect.de] has left #lisp 15:27:57 attila_lendvai: What's the point of walked-lexical-app-form and unwalked-...? 15:30:51 -!- crypto_ is now known as z0d 15:33:33 nikodemus: that too eager *print-circle* bug in slime ... do you also have it still? 15:34:15 -!- ejs [n=eugen@95.135.126.183] has quit ["This computer has gone to sleep"] 15:34:39 *attila_lendvai* looks 15:35:11 Khisanth [n=Khisanth@pool-141-157-237-196.ny325.east.verizon.net] has joined #lisp 15:35:27 *tmh* was just bitten by NCONC. Could someone hand me a band-aid? 15:35:28 athos [n=philipp@92.250.250.68] has joined #lisp 15:35:43 attila_lendvai: Seems to be for local function comming from an &environment arg 15:36:16 is sbcl on os x uni thread by default? 15:36:22 I am getting Not supported in unithread builds. 15:36:40 leo2007: Have a look for :sb-thread on *features*. 15:36:56 I suspect that it isn't, given some of the stuff that's gone down with osx and threading. 15:37:03 not in there 15:37:08 -!- slyrus_ [n=slyrus@adsl-75-55-213-251.dsl.pltn13.sbcglobal.net] has quit [Read error: 60 (Operation timed out)] 15:37:20 Feel free to build a threaded version if you need one. 15:37:30 complex stuff committed. Hope I didn't kill anything else ;) 15:37:49 argh, forgot a NEWS entry. 15:37:54 We're late enough in the month that I'm surprised we're not in code freeze. 15:38:14 nyef: extended code freezes are bad for the buildability ;) 15:38:30 which version do you recommend? 15:38:39 No, -extending- the code freeze is bad for buildability. 15:40:05 tcr: oh, right! the unwalked-* things are coming from the lexenv... this is worth a few words of docstring... 15:41:46 Heh. Just found an 0.8.11 x86/linux build. 15:41:52 ...and unwalked- should really be non-walked- to be less confusing 15:42:35 attila_lendvai: Ok I'll do that 15:43:15 -!- rdd [n=user@c83-250-157-93.bredband.comhem.se] has quit [Remote closed the connection] 15:45:10 all my packages are built by 1.0.27, if I install 1.0.29, do I need to recompile everything? 15:45:13 sbcl 15:45:22 yes 15:46:18 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 15:46:19 leo2007: yes. but there's a trick at http://cliki.net/asdf that helps do that automatically. 15:46:24 *Xach* uses that trick everywhere 15:47:00 merimus_ [n=wroth@nat-49.laurelnetworks.com] has joined #lisp 15:47:40 nice, ok I am ready to compile the latest version 15:48:07 is the git version stable enough? 15:48:18 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 15:48:37 leo2007: ooc, why are you worried about the particular version 15:48:55 do you have a reason for wanting to be bleeding edge, let alone .27 -> .29 ? 15:49:25 *Xach* is happily using 1.0.6.37 on wigflip.com 15:49:45 it has such a long beard, i might cut it off and donate to "Beards for Love" 15:49:58 pkhuong: xFIXMEx 15:51:00 salex: no reason but I don't know which one to use? 15:53:00 -!- eno_ is now known as eno 15:53:24 leo2007: if you already have one that is working, why switch? 15:53:47 I'm not saying there is never a reason too -- but why do it in the absence of a reason? 15:53:50 there's a package that not work on unithread version 15:54:00 so I need to re-build one with thread support 15:54:44 how is sbcl's thread support for os x? 15:56:24 -!- proq [n=user@unaffiliated/proqesi] has quit [Remote closed the connection] 15:56:32 non-deterministic 15:56:54 ?? 15:57:09 heh, sorry half-jokingly; there are known problems on darwin, I think 15:57:24 :( 15:57:41 I suspected otherwise it should be on by default 15:57:42 but it mostly works, I think; you shouldn't try to run a production server on it 16:00:23 -!- Quadrescence [n=quad@unaffiliated/quadrescence] has quit [Read error: 104 (Connection reset by peer)] 16:00:50 i've run lots of things on it without trouble. it's not as solid as linux, but I can't recall what the issues are exactly (signals?) 16:00:51 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit [Read error: 110 (Connection timed out)] 16:02:18 rread [n=rread@nat/sun/x-9bac5f19348d198a] has joined #lisp 16:04:12 -!- Davse_Bamse [n=davse@130.226.210.2] has quit ["leaving"] 16:05:14 sbcl could really benefit from keeping production and development releases separate, i think. it's very easy to just get the bleeding edge, and it's not clear which of the older ones are recommended either. 16:05:47 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 16:06:02 that problem in sbcl isn't nearly as much a problem as it is in library/package land. 16:06:16 sometimes it feels like I'm building sandcastles on quicksand. ;) 16:07:36 nikodemus: is 1.0.29.11 still the safe bet? 16:07:38 true. this whole portability mess has been pretty copied from how it's done in C and its ilk. and that is what you get. :) 16:07:40 hypno: *none* of the older ones are recommended, in general 16:08:16 proq [n=user@38.100.211.40] has joined #lisp 16:08:58 -!- schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has quit ["Leaving."] 16:09:31 fe[nl]ix: heh, ok. 16:10:05 fe[nl]ix: that has to be balance with "if it ain't broke, don't fix it" 16:10:53 -!- quek [n=read_eva@router1.gpy1.ms246.net] has left #lisp 16:10:56 schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has joined #lisp 16:11:33 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 16:11:41 Xach: tech.coop still runs a 0.8.16 :D 16:12:00 =0 16:12:05 salex: there's no contradiction with that principle: "if you already have a working version, don't change it, otherwise use always the latest" 16:12:55 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 16:13:37 that said, I, being a neophile, always use the latest version :D 16:13:49 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 16:14:21 *neophiliac 16:15:06 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-140-122.w81-50.abo.wanadoo.fr] has quit ["leaving"] 16:17:11 deafmacro [n=user@59.92.135.182] has joined #lisp 16:17:34 -!- ruepel0r [n=rue@203-97-49-162.dsl.clear.net.nz] has quit [Read error: 110 (Connection timed out)] 16:18:39 ASau [n=user@193.138.70.52] has joined #lisp 16:21:57 I have got the sbcl source, how to build it with thread support on os x? 16:22:11 read the docs 16:22:45 start with the ones named README, and INSTALL 16:22:58 all will be revealed 16:23:21 i update sbcl roughly once a month... usually a few points after the release. I also subscribe to sbcl-devel so i know what i'm getting myself into ;) 16:24:09 -!- dialtone_ [n=dialtone@unaffiliated/dialtone] has quit [Read error: 110 (Connection timed out)] 16:24:29 *fe[nl]ix* starts his daily compilation of sbcl 16:24:46 ... so that I don't have to ;) 16:26:19 attila_lendvai: still here? 16:26:36 -!- willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has quit [Read error: 110 (Connection timed out)] 16:26:37 asksol [n=ask@175.243.251.212.customer.cdi.no] has joined #lisp 16:26:37 merimus [n=wroth@nat-49.laurelnetworks.com] has joined #lisp 16:26:37 lde [n=user@184-dzi-2.acn.waw.pl] has joined #lisp 16:26:37 alinp [n=alinp@86.122.9.2] has joined #lisp 16:26:37 holycow [n=new@mail.wjsgroup.com] has joined #lisp 16:26:37 minion [n=minion@common-lisp.net] has joined #lisp 16:27:35 -!- merimus_ [n=wroth@nat-49.laurelnetworks.com] has quit [] 16:28:06 thanks 16:29:57 Jarvellis [n=jarv@dsl-217-155-101-22.zen.co.uk] has joined #lisp 16:30:00 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 16:30:06 leo2007: HEAD should be a safe bet :) 16:30:37 mega1: nope 16:30:50 helmut fixed it a while ago 16:31:08 Can (and (macro-function (car form)) (macroexpand-1 form)) return NIL as secondary value? 16:31:28 nikodemus: not here it seems 16:31:34 nikodemus: 1.0.29.44, right? 16:32:00 the install file talks about customize-target-features.lisp, where to place that file? 16:32:34 tcr: for a few minutes 16:33:57 Tordek_ [n=tordek@host175.201-253-14.telecom.net.ar] has joined #lisp 16:35:03 Foofie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 16:35:07 attila_lendvai: It's ok 16:36:43 mega1: (push '(*print-length* . nil) swank:*sldb-printer-bindings*) was what caused it for me 16:37:03 attila_lendvai: Do you actually use cl-walker? 16:37:05 leo2007: in the root of the sbcl source directory 16:37:06 -!- fe[nl]ix [n=algidus@89.202.147.18] has quit ["Valete!"] 16:37:33 nikodemus: thanks 16:37:58 What are other nice features to enable for os x besides sb-thread? 16:38:02 -!- jao [n=jao@74.Red-80-24-4.staticIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 16:38:22 sepult [n=user@xdsl-87-78-168-5.netcologne.de] has joined #lisp 16:38:32 that's about the only one 16:38:54 and be adviced that threads on os x aren't really production quality 16:39:07 fe[nl]ix [n=algidus@89.202.147.18] has joined #lisp 16:39:20 tcr: yes. the js generation in cl-quasi-quote, cl-syntax-sugar, cl-delico for delimited continuations, the query compiler in cl-perec will use it, and we have an unpublished lib for partial evaluation 16:39:23 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 16:39:31 fine for private hacking, but i would not deploy a threaded sbcl app on os x -- eg. linux fine, though 16:39:53 attila_lendvai: I get reference to unknown variable for keywords 16:40:20 are the default darwin builds 32 or 64 bit? 16:40:53 attila_lendvai: i don't have a full solution to your make-instance kludges, but a partial one that speeds up my non-standard-metaclass test-case by maybe x4 16:41:08 32 16:41:16 brnhck [n=hrk@acurwa001195.adsl.ppp.infoweb.ne.jp] has joined #lisp 16:41:16 tcr: probably something above it goes on the wrong path and processes it as a variable reference 16:41:28 to build 64 just do SBCL_ARCH=x86-64 sh make.sh &> log 16:41:49 right, but he was asking "what else..." 16:42:03 and i couldn't remember the defaults. anyway, leo2007 you might want a 64 bit build 16:42:26 attila_lendvai: I reverted my changes, so you must be seeing it too. I just wonder why you didn't come across that 16:42:51 I'm not entirely sure what the non-obvious tradeoff would be 16:43:46 salex: yes i was just thinking about it. i'm running leopard on intel mac 16:43:52 can i do that? 16:44:17 tcr: send a test case or even better, darcs record a failing test 16:44:35 leo2007: yes 16:44:49 attila_lendvai: I found a couple of bugs. Sure will send test cases. 16:45:34 leo2007: well you certainly can. it's less obvious to me if you want to, is what i'm saying 16:45:52 yes, i want to 16:45:57 why? 16:46:05 i mean, do you have a compelling need? 16:46:07 tcr: ok, thanks! fyi, i've pushed a small patch... 16:46:24 i do, which is why I haven't really investigated the tradeoffs beyond the obvious ones 16:46:43 salex: not really, but it can do 64bit integer for me i guess 16:46:43 tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has joined #lisp 16:46:53 with good speed 16:47:09 willb [n=wibenton@compsci-wifi-39.cs.wisc.edu] has joined #lisp 16:47:20 -!- Tordek [n=tordek@host37.190-137-136.telecom.net.ar] has quit [Read error: 110 (Connection timed out)] 16:47:38 *Xach* hugs his immediate single-floats 16:47:46 dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has joined #lisp 16:49:05 pkhuong: did you update NEWS? 16:49:21 ooc, does anyone here know the reason darwin builds default to 32 then? 16:49:40 uname -a outputs x86 16:49:49 i haven't built 32bit anywhere for ages, and haven't followed discussions 16:50:27 leo2007: that's the mechanism, at best 16:50:36 salex: for many purposes 32 bits is faster -- less memory bandwidth. the 64 bit build also fails more tests (mainly fp exception related, iirc) 16:50:40 right 16:50:41 to enable thread support, should :sb-lutex and :sb-thread be enabled? 16:50:53 leo2007: read the files i told you to, all will be revealed 16:51:05 leo2007: just sb-thread -- the rest is automagic 16:51:19 nikodemus: right, those are the sorts of tradeoffs i'm uncertain of 16:51:22 xinming [n=hyy@125.109.244.84] has joined #lisp 16:51:30 ok, thanks 16:51:30 actually, i suppose i didn't mean builds so much as "official builds" 16:51:51 i've got a bunch of stuff that will only run 64 bit, so i haven't compared 16:51:52 salex: yes, I am reading it and then reading base-target-features 16:52:07 just a bit confused by the comments on those two features. 16:52:20 leo2007: that file tells you exactly how to enable threads 16:52:46 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 16:52:51 *luis* hands http://images.tinytags.wigflip.com/z/u/lDp4YeqB3g.png to Xach 16:52:55 including a snippit to put in customize-target-features.lisp 16:53:29 haha 16:53:44 that's cute 16:53:53 luis: heh 16:54:18 *Xach* spies the google.pt user with a firefox set to en_GB 16:56:10 I should see how well Ubuntu is translated into portuguese, yeah. 16:56:15 naeg [n=Anon@194.208.239.170] has joined #lisp 16:58:43 -!- mega1 [n=mega@pool-01d52.externet.hu] has quit [Read error: 110 (Connection timed out)] 16:58:48 hey, i got a few questions. i want to learn lisp, but i also want to create executables to share my programs. i read that compiling lisp is like 'compiling' lisp to c, and then compile/link/.. the c code, which doesn't really look efficient to me? 16:59:32 you read nonsense. or misunderstood what was being said. 16:59:54 possibly, but here what i read: http://knol.google.com/k/tom-laudeman/hello-world-in-gcl-and-clisp/3kie54vqqsra3/4 17:00:46 naeg: GCL compiles to C. Other compilers don't. 17:00:51 how can i create (small?) executables out of lisp code? couldn't find any useful information. read often that these executables are getting really big(like 20mb for a hello world) 17:01:32 naeg: clisp executables are under 10 megabytes 17:01:34 I didn't know that gcc was required to build sbcl :( 17:01:44 what compiler would you suggest? 17:01:44 naeg: lispworks too 17:02:21 naeg: well, i don't compile much to executables, and when i do, i don't care about the size (it's more for my own convenience). i really like sbcl, because it makes really fast code, it's multithreaded, supports 64 bit, unicode, etc. 17:02:33 it's very stable on linux, which is what i use on my servers 17:03:01 i think i misunderstood alot about lisp 17:03:12 -!- silenius [n=jl@yian-ho03.nir.cronon.NET] has quit [Remote closed the connection] 17:03:12 not an uncommon condition... 17:03:23 naeg: What a refreshing response. Most people get belligerent at this point. 17:03:24 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 17:03:33 i thought it would be some kind of efficient, which has high priority for me 17:03:53 tmh: take that. 17:04:19 Xach: i miss Linux 17:04:54 jfrancis [n=jfrancis@66.194.68.209] has joined #lisp 17:04:55 leo2007: it's still alive and available to you! 17:05:03 I bought a macbook just two months ago and it is having all kinds of problems 17:05:44 naeg: sbcl executables tend to be on the large side, but gzexe/bz2exe can be used to get a more reasonably sized one 17:06:00 -!- merimus [n=wroth@nat-49.laurelnetworks.com] has quit [] 17:06:10 i read that they dont get linked? 17:06:27 lisp is not C ... if you want C lisp is not for you. 17:06:29 (talking about these big executables) 17:06:32 naeg: it doesn't realy work like that. 17:06:48 naeg: i compile all the time, but not to a big executable file. 17:07:20 Xach, what does "big executable" mean for you? 17:07:59 drewc, i read alot of good things about lisp, and many comparisons between c and lisp, and lisp seemed to be great, what you can do with one elegant line etc 17:08:00 naeg: i think of it in the sbcl save-lisp-and-die context, which is the C runtime attached to a freshly dumped core image file. 17:08:14 that is often dozens of megabytes 17:08:21 elderK [n=elderK@222-152-92-36.jetstream.xtra.co.nz] has joined #lisp 17:08:36 i usually have long-running applications that start up by loading fasls, and then go from there. 17:08:40 big executable : drewc@kronos:~/src$ du -m maxclaims.bin => 175 17:08:41 17:08:51 -!- elderK [n=elderK@222-152-92-36.jetstream.xtra.co.nz] has left #lisp 17:09:00 -!- xinming_ [n=hyy@125.109.248.255] has quit [Read error: 110 (Connection timed out)] 17:09:03 won't some of the relocatable stuff being worked on be able to eventually lead to having a libsbcl.so or sbcl.dll ? 17:09:16 but, like i mentioned, clisp's executables are usually under a dozen megabytes. and lispworks can deliver GUI apps in under ten megabytes, for example. 17:09:23 (that's a debugging build ... production builds are smaller, but not much) 17:09:24 -!- alinp [n=alinp@86.122.9.2] has quit ["Leaving."] 17:09:25 i guess it depends on what your goal is. 17:10:11 nikodemus: thanks for the make-instance opts! i need to run now... 17:10:36 naeg: you want elegance and greatness... do you want that first or is C-like executables your foremost goal? 17:11:24 i dont know...i am used to these c-like executables 17:11:49 -!- jfrancis [n=jfrancis@66.194.68.209] has quit [] 17:11:54 -!- jewel_ [n=jewel@dsl-242-129-65.telkomadsl.co.za] has quit [Read error: 60 (Operation timed out)] 17:11:58 naeg: you can generate them 17:12:00 naeg: ah. well, i use lisp a lot, but don't find myself using executables very much. i usually have a running lisp system and i write & compile individual functions, and run them interactively. 17:12:10 Xach, i wanted to make mathematical stuff with lisp, but also have executables which arent few megabytes big 17:12:19 jfrancis [n=jfrancis@66.194.68.209] has joined #lisp 17:12:23 mrsolo [n=mrsolo@nat/yahoo/x-13c2e0821c57a45d] has joined #lisp 17:12:24 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 17:12:27 minion: tell naeg about pcl-book 17:12:37 naeg: Good point, disk space is so expensive these days :( 17:12:37 i know that 17:12:42 no minion? feh 17:12:49 naeg: You need to let go of generating executables. All I do is mathematical stuff, and I have never generated an executable. 17:12:58 I have an image I use that's over 400MB 17:13:01 Chris, i am not only developing on a pc ; ) 17:13:05 *drewc* will kick the bots 17:13:05 just create the executable 17:13:19 see if it's slow, then optimize or think of other ways 17:13:48 attila_lendvai: I believe you're not here anymore? 17:14:20 i just read that about common lisp...is there any kind of lisp which aims at creating executables? 17:14:53 pretty much all implementations create executables 17:15:01 -!- lisppaste [n=lisppast@common-lisp.net] has quit [Remote closed the connection] 17:15:01 -!- specbot [n=specbot@common-lisp.net] has quit [Remote closed the connection] 17:15:09 some of them are smaller, some of them are bigger 17:15:27 naeg: You don't need an executable. 17:15:40 *There is no executable* 17:15:54 *Only forms, lots of forms* 17:15:54 can we just say 'yes, there are executables if you want them, use teh google" 17:15:58 tmh, i need, cause i would share my programs, also with persons who have no idea about lisp 17:16:06 but if you're looking for numerical performance and an open source implementation you are looking at SBCL or CMUCL, both of which have big images 17:16:22 naeg: make a web app :) it's the new shareware 17:16:28 but as said, gzexe can bring them down somewhat 17:16:32 Dunno, I think ECL would do fairly good at number crunching, too 17:16:46 really, ok? 17:16:47 naeg: Ah, that's different. Still, an executable is only one way to share. Another is to dump core and use a wrapper script. 17:16:53 CLISP has fairly decent bignum performance.. does it not? 17:17:12 About the only decent performance CLISP has, isn't it? 17:17:21 naeg: Another is to simply use a wrapper script. 17:18:13 naeg: ECL will compile your Lisp program to an ELF binary. The runtime is a dynamic library, so the executables themselves are quite small. 17:18:16 nyef: no, it also performs small-ish executables :P 17:18:22 oh dude, why can't there be any language like lisp with c-like executables : ( 17:18:36 naeg: ECL compiles down to, hm, C 17:18:42 naeg: No reason. Feel free to make one. 17:18:50 naeg: There is, google LUSH. But, you'd be better off using CL> 17:18:52 naeg: there _are_ c-like executables 17:18:59 naeg: have you read a single bit of what was just told to you? 17:19:30 naeg: for various reasons, for serious native-compiling lisps it's more hard than it is worthwhile, that's all 17:19:48 almost all implementations have the ability to create executables that run on C based operating systems. Most lispers don't care, because they don't use lisp that way. 17:19:49 naeg: does it bother you that java has the same problem? 17:19:56 anyway, the java/.NET/python/INTERCAL people survive without it 17:20:10 drewc: I think the problem is that naeg is getting lots of responses, few of which make sense because how to use lisp still doesn't make sense to naeg. 17:20:14 i dont like java/python/etc 17:20:15 Most of us have learned to not need such things, anyway. 17:20:27 naeg: only LOVELY C? 17:20:40 naeg: that was false, but i also think you should expand your horizon. this is a hard step to come over, but you do not need an executable. consider your lisp environment almost as an OS itself. you run your apps in that. 17:21:20 naeg: ECL is just what you want. The only downside may be performance reason. I'd talk to its maintainer about how good it is at number crunching. 17:21:34 rsynnott, i dont really programe in c/c++, just in school when i have to. i do alot of hardwarenear things, so i use mostly assembly. but i would like to have an elegant way to do mathematical things 17:21:46 naeg: The nice thing about CL is that you can develop with SBCL, but use ECL to create your executables. 17:21:58 dysinger [n=tim@71.20.231.3] has joined #lisp 17:22:05 naeg: direct your attention towards 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). 17:22:14 what you need is a complete hardware implementation of common lisp 17:22:24 isnt there a bloody FAQ somewhere about executables? 17:22:26 (I wonder how big the microchip would have to be....) 17:22:29 there is 17:22:48 can we just say RTFM and go back to talking about lisp? :) 17:22:49 (actually, I think the FAQ for cll addresses the subject) 17:23:00 hypno: That's nice for developing, not so nice for getting other people to use your stuff. 17:23:01 drewc, i would be glad about that. then i'd not have to annoy all of you and get so many messages that i cant even read all of them 17:23:07 clearly, this channel needs a faq 17:23:10 tcr: how large is an ECL executable? 17:24:06 p_l told some days ago, I forgot. A few hundred kbs? 17:24:07 naeg: There are very many ways to skin a cat using CL, that's why you're not getting the one answer you're looking for. The answer depends on a lot of conditions specific to what you are trying to solve. 17:24:26 luis: 5.7M for ECL image 17:24:26 luis: it's 5874 bytes here 17:24:30 :D 17:24:34 -!- minion [n=minion@common-lisp.net] has quit [Remote closed the connection] 17:24:44 so i really lost the view of all the things. i'll try to make an ultimate question which solves all my problems 17:24:56 24k for executable linked to ECL image 17:25:03 naeg: Take a look at ECL. 17:25:06 *for REPL executable 17:25:16 -!- fe[nl]ix [n=algidus@89.202.147.18] has quit ["Valete!"] 17:25:44 -!- holycow [n=new@mail.wjsgroup.com] has quit [Remote closed the connection] 17:26:13 ECL can dynamically link the runtime.. non? 17:26:29 drewc: yes 17:26:41 in practice, though, you'll still have to supply it 17:26:52 not many people having ecl runtimes available 17:26:57 it can also generate separate images for different systems, so you can link LGPL code easily 17:27:12 rsynnott: true, but the same can be said of any C library that is non-mainstream. 17:27:28 rsynnott: If nuag's case is about shipping stuff to collegues, the IT people can make it available on all systems 17:27:30 and you can link libecl statically into program 17:27:34 rsynnott: the question is, does debain have a libecl package :) 17:27:40 rsynnott: Complete hardware implementation of CL? Are you thinking of the TI "Lisp Chip", used in the Explorer II and microExplorer? 17:28:09 people often talk about compiling to a single binary to ship things off to their friends, but i don't see that happen very often even in the C world. 17:28:11 no, no, complete as in a complete interpreter :) 17:28:20 (he was talking about liking assembly) 17:28:30 rsynnott: It's a microcoded bytecode interpreter, does that count? 17:28:40 *Xach* probably led sheltered life 17:29:00 Is there a page describing how to distribute lisp programs? 17:29:14 Quadrescence [n=quad@unaffiliated/quadrescence] has joined #lisp 17:29:21 No, I rarely see single-executable deliverables, and haven't really since the old Mac days. 17:29:36 rsynnott: and 6-7mbs is what people expect from applications nowadays, if not more 17:29:36 lisppaste [n=lisppast@common-lisp.net] has joined #lisp 17:29:51 abeaumont_ [n=abeaumon@84.76.48.250] has joined #lisp 17:29:53 the modern mac has lovely FAKE single-executable deliverables :) 17:29:59 (which ccl can make, actually) 17:30:12 tmh: http://stackoverflow.com/questions/25046/lisp-executable/31332#31332 17:30:15 i would like to have lisp with performant small executables, but they shouldn't get interpreted to c code and then compiled because this is ugly, i want to share the executables so that every linux user who knows how to execute binarys can execute them. possibly? 17:30:34 naeg: you can't really even do that with C, unfortunately. 17:30:47 naeg: it would be nice, but i don't think the world works like that 17:30:54 naeg: They are not interpreted to C code, they are compiled to C code 17:30:57 http://pastebin.com/d6194245a <--- listing of a whole ECL install + sizes 17:31:03 I think it could be organized into 2 main categories: (1) distribute to other lispers (i.e. just code) (2) distribute to non-lispers. (thanks luis) 17:31:10 minion [n=minion@common-lisp.net] has joined #lisp 17:31:11 naeg: you can have small, or you can have 'will run on any random machine without extra stuff' 17:31:14 not both 17:31:16 specbot [n=specbot@common-lisp.net] has joined #lisp 17:31:27 -!- attila_lendvai [n=ati@catv-89-132-189-132.catv.broadband.hu] has quit [Read error: 113 (No route to host)] 17:31:27 rsynnott: Now, see, I was thinking the lovely REAL single-executable deliverables. From back before they called the segment loader "the code-fragment manager". 17:31:42 minion: Creating Executables? 17:31:43 Creating Executables: Newcomers to Lisp often ask how to "create an executable" from their Lisp program. http://www.cliki.net/Creating%20Executables 17:31:44 naeg: And what's ugly about it? 17:31:49 rsynnott, oh, i want small and running on every linux 17:32:04 rsynnott: Not true. You can have both, but then you end up with "ugly" or "of limited use". 17:32:06 naeg: Will you keep on refusing any solution provided? 17:32:06 naeg: how would you do that in _any_ other language? 17:32:07 tcr, i dont think that the code will be really efficient 17:32:19 For application delivery, you could (a) deliver the lisp system with the program source and launch using a wrapper script, (b) deliver the lisp executable with a core and launch from a wrapper script or (c) generate an executable. 17:32:36 naeg: Because... you cannot write efficient C code? 17:32:41 -!- KingNatoG5 [n=patrik@84-217-4-148.tn.glocalnet.net] has quit [Read error: 60 (Operation timed out)] 17:32:45 c'mon 17:32:53 naeg: I wouldn't worry about the size, seriously. 17:33:06 naeg: then use something like ECL, do some simple tree-shaking on the image, maybe compile it statically, use UPX on executable... 17:33:25 ikki [n=ikki@201.155.75.146] has joined #lisp 17:33:31 even if it's a ridiculous 100MB sbcl core, it should still zip to a reasonable size 17:33:36 i am having about 16gb on my linux mobile phone. i can't allow myself such big executables 17:33:50 KingNatoG5 [n=patrik@84-217-7-60.tn.glocalnet.net] has joined #lisp 17:33:59 naeg: what do you care ECL uses C as an assembly language? You're just being picky now... 17:34:09 interesting how the requirements keep changing.... 17:34:09 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 17:34:20 naeg: Running mathematical applications on your linux phone is a little bit incompatible, isn't it? 17:34:22 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 17:34:33 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 17:34:37 tmh, why? it's a fucking smartphone 17:34:42 naeg: Don't worry about theoretical requirements, focus on the actual ones. 17:34:50 naeg: even the fastest of smartphones is very slow 17:34:54 naeg: Now you're getting belligerent. 17:35:09 Jabberwockey [n=jens@p54964BA9.dip.t-dialin.net] has joined #lisp 17:35:25 ECL ~12MB for the whole system + C compiler 17:35:26 tmh, no, i just wanted to point out that these smartphones are able to do nearly everything what noone needs 17:35:30 X-Scale [i=email@89.180.147.20] has joined #lisp 17:35:33 that's for a development install 17:35:41 *drewc* arms his anti-belligerence weaponry 17:36:37 now, lisp seems to be only nice to write, but not nice to use the written programs.(to me, for my requirments) 17:36:58 well... not really 17:37:04 naeg: how do you know that? how many lisps have you tried? 17:37:11 but you could have started off saying you wor wokring in an embedded space 17:37:17 naeg: we've given you 10 17:37:19 naeg: is a 12MB program too much? 17:37:20 How small are ACL's or Lispworks' executables? They have tree-shakers right? 17:37:31 and valued small binaries and mem footprint for that 17:37:34 rsynnott, far too much 17:37:37 I've got programs bigger than that which are actually designed for smartphones on my phone 17:37:58 (one much bigger, but that's because it's Myst) 17:38:03 and people would have steered you towards options geared more that way 17:38:15 drewc, i hoped you could tell me if there is any lisp which is good for me 17:38:33 salex, i always said that i want small executables 17:38:38 naeg: with the requirements you've given, you're probably stuck with C 17:38:45 yes, without defining what that meant 17:38:49 -!- dysinger_ [n=tim@71.20.231.3] has quit [Read error: 110 (Connection timed out)] 17:38:50 naeg: and we have... you just have not even tried to listen, not checked out any of the options, and keep changing requirements. 17:38:51 Sikander [n=soemraws@oemrawsingh.xs4all.nl] has joined #lisp 17:38:54 maybe C++, as long as you don't go using boost or anything 17:38:59 rsynnott, sad... 17:38:59 you also talked about math software 17:39:14 so it's a bit of a moving target 17:39:30 drewc, it's kinds hard to look up anything if i get few messages per second 17:39:31 naeg: So, presumably, the requirements are that you want to develop mathematical applications for Android? 17:39:46 but 10mb *is* a small exectable, by any set of default assumptions today 17:39:48 naeg: are you now complaining that you got answers to your questions? 17:39:51 tmh, for my openmoko freerunner, but also on my pc 17:39:56 *drewc* feels a naggum (rip) coming on. 17:39:58 so you can hardly complain that people were being unreasonable.... 17:40:07 drewc, c'mon, you are drilling every word i say 17:40:16 naeg: What kind of math? 17:40:23 naeg: from what you're saying, basically any language which uses a runtime other than the C one would be unacceptable 17:40:36 sounds like, yes. 17:40:54 (though I can't understand why a 12mb programme is unacceptable on a 16GB device 17:40:59 Quadrescence, everything you can imagine, i am not having a clear target yet 17:41:11 naeg: I can imagine a lot of math right now. 17:41:24 There surely is some direction 17:41:42 rsynnott, becaues the OS take already about a gb, then alot of programs(smartphone) and music, films etc 17:41:55 if you've got a film on it, you can spare 12mb 17:41:58 naeg: maybe you could take a look at factor, it's not lisp, but a concatenative language with many lispy features, and executables below 1MB 17:42:13 and some of your existing apps are probably already > 12mb 17:42:18 naeg: Your best bet then, is to use CLISP or ECL and learn the implementation specifics inside and out. Because, I have a feeling that to get your application, mathematical or otherwise, running on openmoko, you're going to have to compile against some openmoko SDK. 17:42:25 abeaumont_: Factor is portable to only a few platforms 17:42:36 rsynnott, but they do more that just echoing "hello world" 17:42:38 Quadrescence: that's true 17:43:03 naeg: ah, yes, but the lisp program won't actually grow that much when you add stuff other than hello world to it 17:43:05 Quadrescence: not sure what platform naeg is talking about, just in case 17:43:05 naeg: so? it's not hte "hello world" that's taking that space. 17:43:18 so adding complexity won't add much space to the core 17:43:22 tmh, acutally not. i am in a project to port arch linux on the openmoko phones 17:43:27 abeaumont_: Yeah. (lisp is better than factor though :D) 17:44:00 abeaumont, arm 17:44:02 Quadrescence: i won's say the opposite... not in this channel ;) 17:44:15 naeg: ok, forget about it, it's not ported to arm (yet) 17:44:26 ECL is, I think 17:44:29 rsynnott, the big thing about the 12mb isn't so much the 16gb of storage -- it's the limited ram on most of these small devices 17:44:42 abeaumont_: ;) ;) ;) (by lisp, I mean lisps in general) 17:44:44 often 128MB, 256MB if you're lucky now 17:44:57 and a good portion of that is taken by the os/system 17:45:00 as long as it has a vm system, it should probably be okay 17:45:08 phf [n=phf@host.icnfull.com] has joined #lisp 17:45:18 dcrawford, yeah, especially X takes alot 17:45:49 -!- Blkt [n=Blkt@160.80.125.170] has quit [Read error: 113 (No route to host)] 17:46:06 the phone runs x-windows? 17:46:07 dcrawford: right, so like a reasonable sized graphics card :) 17:46:08 goodness! 17:46:12 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 17:46:25 I have an openpandora on preorder -- a 256MB ram ARM system 17:46:29 rsynnott, yeah, great, isn't it? : ) 17:46:37 that would be one word for it 17:46:41 so the arm sbcl port work is of interest to me :) 17:47:04 dcrawford: what processor? 17:47:41 ti 3530 -- ARM Cortex-A8 17:47:56 http://openpandora.org/ has the basic specs on the front page 17:48:03 so, i finally think that lisp is not what i searched for, although i really like the s-expressions and the way how to write lisp code... 17:48:17 naeg: you're not making a lot of sense, really 17:48:28 the A8 is quite impressive 17:48:35 -!- krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has quit [Remote closed the connection] 17:48:36 naeg: C isn't what you're looking for for mathematical software 17:48:49 nikodemus: no, forgot to update NEWS. Next commit ;) 17:48:59 slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has joined #lisp 17:49:01 s/\ for/\ to write/ 17:49:10 i think i'm done for today 17:49:13 naeg: so far as I can see your arguments are either against all languages with a runtime of any sort of size (so should be stated that way, really), or they aren't really arguments 17:49:16 Quadrescence, i already know that, this is why i (was) interested in lisp 17:49:35 though i can see still more room for improvement in CTOR optimizations 17:49:45 naeg: various Lisp dialects are nice -- I prefer static typing and inference (for correctness, not speed) though :-) 17:49:48 roygbiv [n=blank@pdpc/supporter/active/roygbiv] has joined #lisp 17:49:48 how do you feel about fortran? :) 17:49:51 i'm not saying lisp is the *right* thing for your phone, mind, jus thtat you aren't makign a lot of sense 17:50:30 salex, this is possibly because i am exhausted and i got bombed with messages few mins ago 17:50:47 true. or maybe you're just not thinking clearly about it :) 17:50:55 naeg: No, you got fragged by the bomb that you dropped. 17:51:13 would you have prefered no answers? less answers? seems at odds with asking questions. 17:51:14 when i joined this channel an nobody was writting i hoped to get help from one particular person 17:51:15 anyway, suit yourself. There's lots of information out there if you want to learn what is actually possible rather than making illogical jumbs 17:51:18 *jumps 17:51:38 I have four fingers and a jumb! :p 17:52:01 which one particular person? 17:52:07 tmh: self/team damage is on? :( 17:52:11 me. Pick me! 17:52:13 can i do it? :) 17:52:22 i have an idea how to help :P 17:52:27 heh 17:52:27 ; ) 17:52:36 naeg: You can use lisp on your theoretical linux phone, there are multiple ways to do it, just not how you think you want to. 17:53:02 the fault there lying mostly with naeg's mental models, not with lisps 17:53:22 -!- brnhck [n=hrk@acurwa001195.adsl.ppp.infoweb.ne.jp] has quit [Client Quit] 17:53:25 pstickne: As it should be. :-) 17:53:30 i am a really (sick) minimalistic person, which would likes to have a smartphone...this makes alot of problems 17:53:31 nikodemus: inline constants around tomorrow, probably. I'll fix the test case too 17:53:52 naeg: pebkac, in that case 17:53:55 naeg: I still suspect you could manage 12mb :) 17:54:14 but if you're a real minamalist, why bother with anything as heavy as c? 17:54:16 :P 17:54:26 naeg: Based on your requirements, I'd focus my efforts on ECL or CLISP. 17:54:34 agreed 17:54:54 salex, becaues it's too much effort to write everything in assembly? 17:55:11 common lisp is certainly _not_ for minimalists though. Maybe he wants Stalin? 17:55:19 aha, so you do understand that there is a tradeoff, good. 17:55:23 now extend that a bit 17:55:29 drewc: true 17:55:39 rsynnott, yeah but it's bothering me that i could write a program in c which does the same and is much smaller(if i am not doing a really lot of things in the code) 17:55:51 *rsynnott* would have thought that telephones with X windows were not for minimalists, either 17:55:52 naeg: no, that was you being inept 17:56:01 naeg: so why use CL if you want to do C? 17:56:08 naeg: would it help that the lisp SOURCE for a given program would be much smaller? 17:56:13 and not realizing that this is an amortizable cost, and not reflective of what your limited code was doing 17:56:19 pkhuong, i dont want to code c, because i dont like the syntax etc 17:56:21 or is it just an irrational prejudice against large executables? 17:56:22 *largely amortizable 17:56:44 rather than a strict bound 17:56:47 rsynnott, no, not really, but i know that this is true in most cases 17:57:06 one solution is to make something so awesome people won't mind how big it is 17:57:14 that worked for me on the iphone version of wolf3d 17:57:25 One prerequisite is to make something that people want. 17:57:34 let's not get crazy, pkhuong 17:57:51 I want world peace. 17:57:53 *rsynnott* has a 700mb application on a phone... 17:57:56 wolf3d on iphone rules, openmoko also has duke and starcraft : ) 17:58:06 (though granted the executable itself isn't the bulk of it) 17:58:21 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 104 (Connection reset by peer)] 17:58:24 offline wikipedia? 17:58:32 yeah, evopedia 17:58:42 tcr: Myst :) 17:58:43 but..python... : P 17:58:50 android has quake :) 17:59:02 naeg: Python has smaller runtimes than most lisps, at least ;) 17:59:02 i d like to have gothic... 17:59:26 rsynnott, but believe it sucks on a 300mhz processor if everything is written in pythong...damn slow 17:59:32 rsynnott: I should hope so --- it does a lot less ;) 17:59:35 openpandora has quake3 17:59:58 dcrawford: openpandora is a game system yeah? 18:00:04 *drewc* googles, nvm 18:00:14 Did they even start shipping that yet? 18:00:21 pkhuong: iirc you had some partial hack for inline method dispatch? if you can find it, it would be nice to see it on sbcl-devel 18:00:25 They had a huge shipping issue on the open pandoras...got a GP32X because of that 18:00:36 goodness, that thing has very similar hardware to the new iphone/palm pre 18:00:46 not yet :P acutally finally starting production boards though 18:00:54 quake3 for telephones on its way, possibly :P 18:00:57 Well they're getting closer at least 18:01:04 an array 5040 x 12, needs to reorder 12 times according to the nth column 18:01:10 rsynnott: OpenPandora is said to be able to play PS1 games, easily 18:01:17 any suggestion in doing it efficiently? 18:01:17 -!- ruediger_ [n=ruediger@62-47-147-4.adsl.highway.telekom.at] has quit [Read error: 104 (Connection reset by peer)] 18:01:17 nikodemus: I think it was Xof. I had a sealing hack, but it tended to go exponential on some inputs 18:01:19 (though I can't imagine how you'd play it) 18:01:46 pkhuong: you guys have been playing with sealing again? Interesting 18:01:53 naeg: http://www.thinlisp.org/ # this is probably of interest to you 18:01:53 Yuuhi [n=user@p5483EA75.dip.t-dialin.net] has joined #lisp 18:02:00 salex: not for more than a year. 18:02:01 -!- phf [n=phf@host.icnfull.com] has quit ["Leaving..."] 18:02:04 ah, ok 18:02:07 think minicomp with gaming controls / keyboard/ touchscreen / wifi 18:02:09 nicktastic, sounds good. thanks 18:02:21 i haven't been paying attention for more than a year, though 18:02:46 dcrawford: they're too slow in releasing these things 18:02:48 schaueho [n=schauer@dslb-088-066-013-012.pools.arcor-ip.net] has joined #lisp 18:02:53 thinking of smartphones, do any of the schemes target obj-c? 18:03:08 it's in great danger of falling behind the current mobile 18:03:13 salex: you can run gambit on the iphone? 18:03:19 salex: anything which targets C would presumably be fine 18:03:28 i don't know. that's why im asking 18:03:39 rsynnott: For OpenPandora? there's 2 digitial pads, number of keys...it would be fine. One of the challenge of this and the GP32X are external joysticks and support for those. 18:03:39 (though if you wanted an obj-c bridge, that'd be another story) 18:03:44 but yes, that would seem ot be an easy-ish path to lispy iphones 18:03:55 *drewc* notes that clojure is said to almost run on android 18:04:04 nikodemus: I don't think I can contribute much to that issue. Usually the CL point of view has been to favor correctness [i.e covering all edge cases] over efficiency. That's why you get that note on python not being able to inline sin or cos because of the size of the argument. 18:04:06 Problem with the iPhones is that mostly only ObjectiveC is supported..unless you jailbreak it 18:04:09 I think I saw someone mention that they were using ECL on iphone a while back 18:04:22 TDT: no, that's the point. it woudl be objc 18:04:24 though there would potentially be issues with Apple's app store rules 18:04:34 ah here we are google knows all: http://jlongster.com/blog/2009/06/17/write-apps-iphone-scheme/ 18:04:40 you certainly couldn't distribute an ecl repl 18:04:48 nikodemus, " translated into C code"..like a punch in my face 18:04:54 -!- simplechat [n=simplech@unaffiliated/simplechat] has quit ["Leaving"] 18:04:55 but it's less clear whether you could write an app in ecl and distribute it 18:05:12 rsynnott: according to that posting you actually could distribute a repl 18:05:27 so long as it can't download code at runtime 18:05:36 how you convince apple of the latter I dunno 18:05:37 naeg: who gives a fuck? 18:05:40 ah, maybe they changed it 18:05:51 naeg: so, you want c-like size, c-like executables, c-like speed... but compling to C is out of the question. WTF?! 18:05:56 until recently they were very unhappy about the idea of user-accessible interpreters 18:05:56 naeg: you do realize that's sort of moronic, don't you? 18:06:03 oh noes, it targets c? 18:06:06 drewc: that's a bit unfair. 18:06:07 and inconsistent about non-user-accessible interpreters 18:06:21 drewc: the system naeg wants seems to be way better than C. 18:06:29 (or anything else i can think of) 18:06:30 naeg: plenty of perfectly respectable compilers target C 18:06:38 i just dont think that this translated code would be good 18:06:39 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 18:06:43 Xach: lol, true enough. 18:06:54 naeg: that's because you don't know much about compilers, apparently 18:07:02 naeg: just because you think something doesn't make it true. 18:07:13 rsynnott: yeah, just going by the posting, nothing authoratative 18:07:14 naeg: in clisp, it targets a VM. In sbcl it targets x86 or whatever. In ecl it targets C 18:07:31 you could think of C as a rather high-level low-level backend if you wanted to 18:07:42 i know, i know. but how can a general translation of anything be good? its like if you translate a book from english to any other language with google translation 18:07:44 there was an article somewhere on why c is not a good assembler 18:07:52 or as portable assembler :-P 18:08:00 i think it was on c.l.l. 18:08:03 prxq: see stalin 18:08:05 naeg: you could call C code compiled to assembler a 'general translator' 18:08:11 *translation 18:08:11 naeg: clearly, the only good way to execute a language is right in the silicium. 18:08:17 naeg: some talented people manage to translate notes on a page into music with decent results. 18:08:19 granted, C probably is not the best target language 18:08:24 (for a compiler) 18:08:26 but it ain't bad 18:08:32 salex: I don't think that proves anything either way 18:08:36 rsynnott: sure is convenient 18:08:40 The Eiffel people seem to survive it :) 18:08:48 rsynnott: well, if you want c-like everything... it's not a bad target is it? :) 18:09:00 prxq: it, and many other examples, prove you can generate decent code this way 18:09:11 which may not be optimal in some global sense 18:09:18 *drewc* prefers compilers that target hardware for most tasks, but can see the utility in compiling to portable C . 18:09:20 drewc, i dont want c-like everything, i just realized that i really like the executables 18:09:24 sure, you can also write great software in cobol if you want. So? 18:09:29 you do have to wonder how well they thought the name through, though 18:09:40 but if you've got a targettable c compiler for a platform, it makes sense 18:09:40 make's clisp's menorah look like a non-even 18:09:43 *event 18:09:55 salex: that's another thing entirely :) 18:10:00 rsynnott: qobi doesn't give two shits. 18:10:10 prxq: not at all. that's the exact point 18:10:32 (relative to naeg not likeing the idea of intermediate c code) 18:11:03 how is stalin not evidence this can be done practically and well? 18:11:24 of course, if you think you can do low level stuff like scheduling or selection better than the C compiler, targetting C might not be such a good idea, but over all, it's usually better to leave such concerns to someone else. 18:11:59 particularly if your concern was to get higher level 18:12:01 ruediger_ [n=ruediger@62-47-157-234.adsl.highway.telekom.at] has joined #lisp 18:13:27 if i can put a lisp on a smartphone, i might even have to get one :) 18:13:54 salex: GCL, CLISP and various schemes have been running on winmo smartphones for a while 18:14:46 yeah, i was just musing 18:14:55 not suggestng it's never been done 18:15:01 no interest in winmo though 18:15:21 oh man, now i'm stressed 18:15:24 you might actually get better results going through C than you could reasonably expect to get on your own 18:15:36 especially something peculiar like an itanium 18:15:51 i'm flying out for interviews next week, just got the schedule and the first thing they've got me doing is interviewing with the dean 18:16:07 salex: pretty sure people have done it on jailbroken iphones 18:16:12 rsynnott: nice :) 18:16:18 though emacs might be a bit of a stretch, for UI reasons :) 18:16:22 *p_l* is trying to find a solution for moving from aberdeen to coventry and fails... 18:16:23 salex: Kawa is said to run on android 18:16:29 -!- seejay [n=seejay@unaffiliated/seejay] has quit ["Coyote finally caught me"] 18:16:30 i think i will try TL, even if i dont really like this translation thing, at least when i write something where easy-to-write>"performance/efficiency" 18:16:34 rsynnott: chording might be a bithc 18:16:42 rsynnott: i use emacs through ssh on android daily :) 18:16:53 naeg: can you articulate a real reason for not likeing targeting c 18:16:57 other than "ew" ? 18:16:57 salex: You can make the argument that by having to do so much to generate decent code (iirc) it basically makes the point that C is not a good assembler rather well. 18:16:57 salex: sticky keys, like in Office 2007 18:17:02 drewc: you have a real keyboard on that phone, though. 18:17:09 pkhuong: ahh ... good point. 18:17:30 prxq: i suppose. there are other examples. But my point wasn't that it was easy, but that it was doable 18:17:34 drewc: iphones don't have keyboards 18:17:35 i keep forgetting that most smartphones are without a real keyboard, as all mine have had one. 18:17:41 counter to naeg "point" that it woudl be bad 18:17:43 prxq: everything you have to do to generate decent C code, you'd have to for just as decent assembler. 18:17:44 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 18:17:45 salex: with that i can agree 18:17:55 Hi I am trying to get the file-descriptor of a stream using sb-posix:file-descriptor but sbcl is complaining saying "The symbol "FILE-DESCRIPTOR" is not external in the SB-POSIX package". Is the interface not being exposed anymore because the documentation seems to suggest otherwise. 18:17:58 (though they should be able to support bluetooth ones, at least on an application level, now) 18:18:02 salex, as i said i dont really believe that this general translation is very performant/efficient 18:18:08 -!- schaueho [n=schauer@dslb-088-066-013-012.pools.arcor-ip.net] has quit ["Leaving"] 18:18:09 i did see a link somewhere recently to a guy who got his (broken) iphone to talk toa bluetooth keyboard 18:18:16 but it seems to be the best solution 18:18:17 pkhuong: I don't remember the article in question, but the upshot was that no, that that is not true. 18:18:19 naeg: believe away. 18:18:27 naeg: but you seem to be saying this based on no knowledge at all 18:18:28 afaics 18:18:32 seejay [n=seejay@unaffiliated/seejay] has joined #lisp 18:18:41 naeg: but there is evidence that it works perfectly well 18:18:49 naeg: you're beliefs are not really relevant, especially when the evidence is otherwise. 18:19:03 there's a strange echo in here :) 18:19:08 what's the evidence? 18:19:08 I don't remember the exact reasons, I mean. But things like returning multiple values where hard to do efficiently in C, but are easy in asm 18:19:18 salex: yep, as of OS3 legit applications are allowed support whatever bluetooth hardware they feel like, though 18:19:22 naeg: you haven't been paying attention, have you 18:19:23 deafmacro: sb-sys:fd-stream-fd returns a file descriptor for a stream, maybe. 18:19:36 salex, i am sorry, but i really couldn't read everything 18:19:37 deafmacro: the sb-posix symbol is not related to normal lisp streams 18:19:59 naeg: we gave you examples of high performance compilers that ... wait for it... target c. 18:20:06 naeg: i suspect you are simply not very experienced with software development in general, so are coming from a position of ignorance and making assumption based on faulty reasoning. You won't get far if you're not willing to read and learn. 18:20:08 prxq: this used to be true. Not so much these days. In any case, that's not relevant to the analyses one has to perform to emit decent C code. 18:20:50 prxq: that's a somewhat orthogonal issue ot the one we're discussing, really 18:20:59 rsynnott: still, you'd need drivers for some of the stuff, which probably aren't allowed to run on non-jailbroken stuff 18:21:13 Sure, C is sometimes too high level. That's a frequent problem with high level things, but you can always escape to inline assembly in the worst case. 18:21:14 p_l: apparently, they are 18:21:22 but as part of the application, not the operating system 18:21:22 nice! 18:21:26 Xach: I can go ahead and use sb-posix:open but I am just curious to know the reason why sb-posix:file-descriptor is not being exposed anymore. The documentation says shows a stream returned from with-open-file returning a fd int 18:21:38 which sort of breaks the keyboard idea until apple deign to support it properly 18:21:50 but do open the door to games and such 18:22:03 ah, i see 18:22:05 deafmacro: i don't know about the history, sorry. 18:22:24 deafmacro: sb-posix is an internal package. Use it at your own risk. 18:22:51 Xach: ah. thanks never the less will try out sb-sys:fd-stream-fd 18:23:24 pkhuong: you're thinking of sb-unix, maybe. 18:23:30 thinking back to threaded sbcl issues 18:23:33 or sb-sys? 18:23:48 did i tell you guys someone is giving me a rack full of suf fire x2200's? 18:24:00 so it'll be time to do some stress testing :) 18:24:05 *sun 18:24:14 Xach: It is a contrib. Not internal then! 18:24:47 Xach: sb-sys:fd-stream-fd works cheerfully. Thanks a lot. 18:24:47 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 18:24:54 salex: heat/noise testing? 18:24:55 salex: now I really have to work on binary serialisation ;) 18:25:18 Xach: meh, it'll be in a big machine room 18:25:40 but i'll get at least 8 dual-quads to play with uninterupted 18:25:49 so that'll be fun 18:26:22 can anomalies in the sbcl manual be reported on the sbcl mailing list? 18:26:29 deafmacro: sure. 18:26:32 there's a 100 node version coming online too, but that's crowded 18:26:32 salex, i will learn lisp and create a executable with the help of TL, after that i will write it in c and see what is more performant/smaller/etc. i'd like to really convince myself of this 18:27:01 pkhuong: cool. thanks. 18:27:14 naeg: this has nothing particularly to do with lisp, and i'm obviously not saying that any compiler that targets c will generate good c 18:27:14 naeg: and I'll write assembly. What would the point of the experiment be? 18:27:32 i am outa here now, cu all and thanks, i hope i didn't annoy you too much 18:27:32 what i'm saying is that your reaction is not founded on an knowledge of hwat has been done before 18:27:52 since there are in existence compilers that od this successfully 18:28:06 to which I would heretically add that that is a normal stage in the path to enlightenment 18:28:18 pkhuong: i may have to play around with some MPI stuff for CL, even 18:28:26 i've mostly hacked around it previously 18:29:11 and the new MPI package that's been blogged about on planet lisp seems good so far. 18:29:13 pkhuong, of course will the assembly be better, but its one step forther down with TL, its like assembly>c>lisp(tl) (just depends on how far these steps are) 18:29:31 pkhuong: yeah, i could give it a good bashing at any rate 18:29:46 maybe get it out to play on a couple hundred nodes too, not sure 18:29:46 mattrepl [n=mattrepl@ip68-100-82-124.dc.dc.cox.net] has joined #lisp 18:29:59 anyway, cu all 18:30:04 but the small cluster at least I can abuse at will 18:30:27 and you can do a fair bit with 64+ cores, after all.... 18:30:57 I've been fairly happy with the 8 cores under my desk ;) If only I had CPLEX licenses to go with it 18:31:23 -!- naeg [n=Anon@194.208.239.170] has left #lisp 18:31:30 I had a 48 node cluster to play with these last couple of years, but it's been dying 18:31:50 so this is a welcome rejuvination 18:32:41 there's another dozen dual quads here that I can use, but it's shared with the entire department so I can't do anything too crazy with them 18:33:04 whereas we had several month-long computations on the 48node chugging along 18:33:29 so it will be nice to have our own sandbox to play in again 18:33:37 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["This computer has gone to sleep"] 18:34:13 -!- seejay [n=seejay@unaffiliated/seejay] has quit ["Leaving"] 18:34:26 I had to use another uni's cluster for that sort of job a couple times. A friendly sysadmin over there gave me a ton of low-priority tasks. 18:34:42 yeah, that can work well 18:35:12 at [graduate uni] we had a 512 node and then a bigger one, but scheduling can be a pain because everyone wants on them for big simulations 18:35:32 the nice thing about having a small cluster all to ourselves is not having to worry about that 18:36:05 and also if i want to do something like roll out a version of sbcl across the cluster, I can just do it 18:36:16 not put in a request 18:36:17 I seem to remember when I was in university some postgrad setting up a lab full of desktops as a cluster, without telling anyone, and running something on it for weeks 18:36:25 heh. 18:36:30 it's been known to happen 18:36:41 (the machines were slow anyway; people thought they'd just gotten slower) 18:36:49 sometimes it's hardly worth the trouble though, if they're heterogenous enough and/or small enough memory 18:37:14 identical machines, in this case, but very little memory 18:37:17 seejay [n=seejay@plexyplanet.org] has joined #lisp 18:37:33 it was a lab of old ultrasparcs kept for running weird software which likes old ultrasparcs 18:37:33 when i was in undergrad i did a summer research term and my colleague used to go around setting up a parallel job on everyones desktop machines every friday night, for running over the weekend 18:37:37 cluster on the cheap 18:37:43 Ever try building a mini beowulf system? There was a nice howto on building one, that wasn't really too expensive in the end. 18:38:02 i helped put one together with a couple hundred boxes 18:38:07 pretty straightforward 18:38:10 *Xach* has a shortage of interesting questions for which the answer is "make your own computer cluster" 18:38:24 so not big, but not "mini" either 18:38:26 (the whole thing was ultimately replaced with one large sun machine and a bunch of X terminals) 18:38:28 how about "got too few computer clusters?" 18:38:35 interesting, andersca 18:38:37 err, antifuchs 18:38:46 Xach: that animated text thing you had on your blog? 18:38:47 Xach: I, on the other hand, have a plethora of questions for which the answer is "need a bigger cluster" 18:38:47 salex: Yeah, I help put together a 30-node setup once, but..the miniwulf one is 4 nodes 18:38:57 Let me see if I can find the article, it's pretty interesting. 18:39:29 rsynnott: hmm, possibly 18:39:44 *Xach* could assign a frame per node 18:39:50 -!- sepult [n=user@xdsl-87-78-168-5.netcologne.de] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 18:39:54 Xach: finding diffeomorphisms on time series of largish volumetric data is my current issue 18:39:57 need more memory :) 18:40:14 salex: more memory or more bandwidth? 18:40:22 ljosa [n=ljosa@gm3b6-dd5.broad.mit.edu] has joined #lisp 18:40:24 memory first, then bandwidth 18:40:49 it's always something though 18:43:14 claint [n=utku@85.105.94.127] has joined #lisp 18:43:14 the current hardware is just memory starved, is all 18:43:19 Jacob_H [n=jacob@92.6.218.36] has joined #lisp 18:43:57 hmm.can't find that article. What do you use for distributed and parallel programming on these systems? 18:44:24 -!- erk [n=MrEd@about/apple/iPod/BeZerk] has quit [K-lined] 18:44:36 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 18:45:00 -!- benny [n=benny@i577A1855.versanet.de] has quit [Connection timed out] 18:45:55 SandGorgon [n=OmNomNom@122.162.130.77] has joined #lisp 18:47:30 you mean the systems I've got now? 18:48:00 Yeah 18:48:25 torque/maui mostly 18:48:44 but with lisp stuff i've often hacked around it 18:49:04 in problem specific ways 18:49:08 mostly 'cause i'm lazy 18:49:37 so i haven't been using MPI with it yet, for example 18:49:43 Ah ok. 18:49:57 but probably should 18:50:10 I've been toying with the idea of creating a miniwulf for awhile now 18:50:20 they aren't that expensive to make, I really should try it sometime. 18:50:29 i typically let the scheduler handle shopping the jobs out, but the jobs themselves sychronize work 18:50:37 manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has joined #lisp 18:50:59 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 110 (Connection timed out)] 18:51:01 rdd [n=user@c83-250-157-93.bredband.comhem.se] has joined #lisp 18:51:04 but this is bit of a quirk of the sorts of problmes I've been doing, where I can just specify jobs in a small text file, for example, and treat it like a stack 18:51:10 cheap hack, really 18:51:17 jao [n=jao@94.Red-88-6-161.staticIP.rima-tde.net] has joined #lisp 18:51:20 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 18:52:56 -!- SandGorgon [n=OmNomNom@122.162.130.77] has quit [Read error: 60 (Operation timed out)] 18:53:53 *rsynnott* thankfully does not have any problems for which such a thing would be useful 18:54:15 salex: you could at least use a message queue :P 18:54:53 (Amazon now sell a message queue service, presumably on the basis that there are people too lazy to set up their own message queues) 18:55:51 rsynnott: yeah, i've done that too. Often though, it's not even needed (for what i'm doing) 18:56:10 sometimes the parallelization is really trivial. 18:57:06 cp2 [n=will@unaffiliated/cp2] has joined #lisp 18:58:14 oh great. Constant fold add/sub of 0 away. 18:59:31 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 60 (Operation timed out)] 19:00:51 merimus [n=wroth@nat-49.laurelnetworks.com] has joined #lisp 19:02:39 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 19:03:06 -!- Foofie is now known as Fufie 19:06:54 TDT: have you seen lunchbox cluster? 19:07:32 p_l: No, I haven't, but will google that 19:08:37 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 19:08:52 http://www.linuxjournal.com/article/8177 <-- that' spretty cool, it's similar to the article I was looking for 19:09:14 But the guy had 4 nodes, on maybe a 4-feet tall, 2 feet wide and long "box" of sorts 19:09:27 -!- Nshag [i=user@Mix-Orleans-106-2-114.w193-248.abo.wanadoo.fr] has quit ["Quitte"] 19:10:00 Everything was exposed, and he cut and built the case himself out of Plexiglas or something like that 19:10:15 Nshag [i=user@Mix-Orleans-106-2-81.w193-248.abo.wanadoo.fr] has joined #lisp 19:10:16 there's some company currently trying to sell a ready-made cluster made out of some low-power chip 19:10:22 (ARMs, I think) 19:10:35 and a big welcome back for the transputer! 19:11:12 rsynnott: That's kinda what the article above is using, I believe ARM chips, in a very small case. I guess since ARM processors generate less heat, it can handle it. 19:12:14 -!- joast [n=rick@76.178.184.231] has quit [Read error: 110 (Connection timed out)] 19:12:17 For what I think I'd want to do, it'd be say core 2 duos, not the highest chip..but get some cheaper ones, 4 chips - then buy cheaper motherboards and lots of RAM - should have onboard ethernet, but make sure it's gigabit...I think that'd probably be the most simple, anyways. 19:12:26 Probably could be built for less than 2k as well 19:12:59 -!- Dave2 [i=Dave2@freenode/staff/dave2] has quit [Read error: 104 (Connection reset by peer)] 19:13:04 If you get your hands on MIPS from SiCortex and manage to put those modules into lunchbox, it would work nicely as well 19:13:12 Dave2 [i=Dave2@freenode/staff/dave2] has joined #lisp 19:13:24 well i should hope so. you can get dual quads for $1500 ish prebuilt. 19:13:34 (the 2k thing) 19:14:14 joast [n=rick@76.178.184.231] has joined #lisp 19:14:45 If you're on a budget, going AMD is probably smarter. Even more so if you want large amounts of RAM. 19:15:11 yup 19:15:30 dual quadcore opterons seem to be the sweet spot these days 19:15:31 TDT: put SiCortex PDS into lunchbox format, instant 100GFlop minicluster 19:16:33 so you can get 8 cores per 1RU with say 1gb/core and stay under $1500 19:16:50 more if you want more memory, obviously. gigabit ethernet, etc. 19:16:59 -!- deafmacro [n=user@59.92.135.182] has left #lisp 19:17:22 are AMD chips still cheaper for a given performance? 19:17:46 p_l: unfortuntely, SiCortex has exploded 19:17:56 it swaps around. when you're building up racks of this sort, yes I think so at the moment 19:18:02 (hence sweet spot). 19:18:06 rsynnott: or more like got their floor removed from under them 19:18:13 legumbre_ [n=user@r190-135-26-198.dialup.adsl.anteldata.net.uy] has joined #lisp 19:18:25 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["This computer has gone to sleep"] 19:18:36 and dual quads these days, for best bang/buck 19:18:36 Also comes down to overall price, but also the needs...a small cluster is good enought o learn off of. 19:18:56 -!- malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has quit [] 19:18:57 I wonder what is the price of HT bridges for Opterons 19:19:08 Like HT <-> Infiniband 19:19:37 dunno. i'd have to ask our cluster guys 19:19:39 Infiniband switch + several nodes - SSI 256 core cpu 19:19:47 I belive you can pick up an infiniband switch for 1-2k 19:19:48 You're looking at huge proprietary form factor 4 or 8 way mobos then, I believe. So Very Expensive. 19:19:59 EnglishGent [n=EnglishG@ai-core.demon.co.uk] has joined #lisp 19:20:22 hmm... probably, Horus-enabled mobos are probably 4-way per node 19:20:27 Quad chips are pretty cheap, overall. 19:20:33 the only thing I've done vaguely related to this is a large erlang app running over a pile of dual-processor server 19:20:38 hmm.... I thought tyan had a 4x mobo that fit in an eatx 19:20:42 http://www.pricewatch.com/browse/cpu/core_2_quad_q9650 19:20:51 inclined to become very unhappy when the cheap switch acted up 19:21:43 rsynnott: my acquaintance crashed university's cluster by flipping a light switch ;-) 19:22:18 there was only *one* switch for the whole thing 19:22:28 p_l - was it labelled 'magic' & 'more magic' ? :) 19:22:35 TDT: that's still double the opterons though 19:22:55 we used to have that problem at work, due to the building having been rewired by the company founder over a weekend 19:23:02 fortunately, the servers were elsewhere 19:23:03 which i think is why things like the M2 1U racks come in at under 1500 with 8gb.... 19:23:11 EnglishGent: no. But lightning was on the same circuit as backup switch. Primary switch wasn't working for last few months :> 19:23:14 depending on what you are doing gigE will be good enough though 19:23:21 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 19:23:21 -!- legumbre [n=user@r190-135-21-133.dialup.adsl.anteldata.net.uy] has quit [Read error: 104 (Connection reset by peer)] 19:23:31 merimus: yup, can be 19:23:42 eg: http://www.sun.com/servers/x64/x2200/ 19:23:45 Myrinet is nice, since you can use 10GbE fabric with it 19:24:01 not saying you can't do better than that, just it's a baseline 19:24:12 for compact format prebuilt stuff 19:24:21 which matters if you want a couple hundred of them :) 19:24:37 sun always did make a point of having more memory than everyone else 19:24:48 8gb isn't a lot for a dual quad 19:25:01 but they'll do up to 2tb /node which is nice 19:25:09 yea, tyan has quad socket motherboards for EATX form factor 19:25:26 there are rumours that Oracle will kill their hardware division, sadly 19:25:29 bah... sgi has more memory than anybody else :) 19:25:32 (they've already killed Rock, it seems) 19:25:40 rsynnott: Opterons for long time had lead on memory in x86 market 19:25:46 merimus: don't talke to me about sgi memory 19:25:48 don't know how i7 fares 19:26:04 salex: I only said they had lots of it :) 19:26:17 p_l: ECC DDR3 is still much more expensive and power-hungry than DDR2. 19:26:18 i laughed them out of hte office once because in maybe 97 or 98 they were quoting me $18,000 for 256Mb 19:26:23 -!- Jarvellis [n=jarv@dsl-217-155-101-22.zen.co.uk] has quit [Remote closed the connection] 19:26:24 or maybe it was 512 19:26:41 sure, it was sprightly... but ... well.. they died. 19:26:53 holycow [n=new@mail.wjsgroup.com] has joined #lisp 19:27:04 salex: imagine trying today to buy 1G for AlphaServer 255 19:27:08 ah, you can get infiniband on that sun for a mere thousand euro or so 19:27:09 salex: they have one system with a max memory config of 128TB ram 19:27:14 I gave up in my search -_-; 19:27:34 p_l: i've done that too, actually 19:27:47 they gave me a nice digital shirt you can almost still read 19:27:51 -!- postamar [n=postamar@x-132-204-253-51.xtpr.umontreal.ca] has quit [Connection timed out] 19:27:53 salex: for AS 255? 19:27:54 jewel_ [n=jewel@dsl-242-129-65.telkomadsl.co.za] has joined #lisp 19:28:23 128MB simm modules? 19:28:40 Chips and computer stuff is dirt cheap any more.... 19:28:41 hmmm, now I'm trying to remember which model 19:29:09 I've seen 128MB simms *once* 19:29:09 TDT yeah the problme iwth the 18k was that x86 equivalent was two orders cheaper 19:29:30 not that old stuff was expensive 19:30:06 p_l: i had a 1gb AS under my desk for a while but now I'm trying to remember the config 19:30:07 http://www-e.uni-magdeburg.de/urzs/asgard/ <--- btw, the only photos of PDS' internals I've found 19:30:11 salex: i vaguely remember shopping for sun ethernet cards, finding a few for hundreds of dollars. 19:30:26 salex: I just remember building my PC years ago, and spent so much money on it 19:30:46 (the problem withthe sgi's was that they were being specced for a product that should have stayed under 20k complete 19:30:50 salex: I have old AS 255 - 233 MHz EV45, 32 MB FP simm w/ parity -_-; 19:30:51 sgi just had no clue at all 19:30:59 salex: Just thinking I can build one heck of a computer, dual quad with good amounts of memory and all for a small portion of the overall cost that I spent on my current comp 7 years ago... 19:31:14 tis the nature of the beast, TDT 19:31:24 computers have gotten wonderfully cheap 19:31:30 oldtimers talke about pooling grants to buy a 20,000 5mb hard drive 19:31:36 $20k i mean 19:31:38 (except laptops with macos x on them, sadly) 19:31:40 -!- laynor_ [n=laynor@93.107.1.23] has quit [Read error: 60 (Operation timed out)] 19:31:51 Now if only some parts where more clued... 19:31:52 rsynnott: what, $1k is expensive now? 19:32:05 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Read error: 54 (Connection reset by peer)] 19:32:13 rsynnott: but with the added software installed, it is almost worth it, isn't it? 19:32:13 reasonably 19:32:24 compared to the price of other computers 19:32:38 *rsynnott* paid it :{ 19:32:40 *:P 19:32:49 (well, rather more, actually; MBP) 19:33:01 but it would be nice if apple did a low-end laptop 19:33:02 *p_l* has bought a new computer *once* 19:33:20 Fare [n=Fare@ita4fw1.itasoftware.com] has joined #lisp 19:33:34 p_l: you know, now i think of it i can't remember if that AS was an EV56 or a 21164 19:33:36 it was the laptop that currently runs Windows Server in prestigious role of fileserver and router 19:34:05 rsynnott: watching the low end laptops die around here, i'm not so sure it would 19:34:13 salex: well, 21164 = EV5* 19:34:20 my laptop = $120 on ebay 19:34:20 yes, cheaper version 19:34:53 salex: EV56 was modernised version released sometime later, with main difference being BWX, iirc 19:35:02 i think it was a EV56 though. We were thinking of buyign 21164's for product, but couldn't get a good enough price in bulk 19:35:05 oon the plus side, it should last me half a decade or so 19:35:20 good for us though, i guess, as digital didn't last all that much longer 19:35:30 laynor [n=laynor@93.107.70.207] has joined #lisp 19:35:45 p_l: what was BWX ? 19:35:48 p_l: i'm thinking of 21164a vs 21164pc i think 19:35:58 are alphas still relevant today? 19:35:59 Fare: set of instructions to access 8bit bytes 19:36:16 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 19:36:33 Fare: so you could operate on 8bit quantities, including 8bit load/store 19:36:41 and 16bit, iirc 19:36:53 -!- younder [n=jthing@165.244.251.212.customer.cdi.no] has left #lisp 19:36:55 8bit and 16bit, iirc 19:37:09 younder [n=jthing@165.244.251.212.customer.cdi.no] has joined #lisp 19:37:12 it was probably a bad idea to leave that out of the original alpha, in a byte-oriented world 19:37:15 previously 32 or 64 only 19:37:25 Fare: yes 19:37:44 damn that was a fast workstation for the time though 19:37:53 i was a little sad we couldn't make it work with them 19:38:03 a great toaster, too, considering the heat dissipation 19:38:15 alpha is, to date, still my favourite arch. 19:38:16 ended up with dual proc intels for the product 19:38:16 make what work with them? 19:38:24 embedded product 19:38:28 I had an alpha multia once. It was very nice 19:38:29 -!- Nshag [i=user@Mix-Orleans-106-2-81.w193-248.abo.wanadoo.fr] has quit [Remote closed the connection] 19:38:30 Fare: Alpha's afaik weren't that hot 19:38:32 i was working on back then 19:38:39 e.g. 97 19:38:43 *i.e. 19:38:56 *Fade* still has a multia 19:39:04 awesome little machine. 19:39:15 mine worked for some time without fan, only because I didn't notice it was missing from case 19:39:18 166MHz. *giggle* 19:39:19 SandGorgon [n=OmNomNom@122.162.130.77] has joined #lisp 19:39:28 I think I left my multia in an attic when I left college. 19:39:29 I think mine is 233mhz 19:39:46 I haven't powered it up in a long time. 19:39:53 *p_l* is on lookout for working Alphas 19:40:00 i dumped all my olld gear liek that a while ago 19:40:05 I'm trying to get some SPARCs as well 19:40:10 some sparc IIs, the alplha, a bunch of stuff 19:40:14 p_l: that should be easy tho? 19:40:20 i move to much 19:40:29 I have a pile of 64bit sgi gear; i'd happily trade it for equiv DEC alpha gear. :) 19:40:32 p_l: I recommend searching the attic of 316 High Street in Harrisonburg, VA. 19:40:33 hypno: SPARCs, yes. Except when your cash reserves dropped to 0 long time ago :D 19:40:43 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Read error: 60 (Operation timed out)] 19:40:44 oh, i had an O2 kicking around, too 19:40:46 sellout: I'm in UK, unfortunately 19:40:54 p_l: i pretty much get paid to pick em up, reinstall them and put them out in production again. *g* 19:41:04 There's a lot of SGIs being sold in Poland, iirc 19:41:28 there's some Sparcs, lately I found more RS6000 and later models, Alpha's are rare as hell 19:41:30 the mips >= r10k is mostly useless. :( 19:41:52 didn't amd64 machines incorporate most of the arch goodness from the alphas, anyway? 19:41:54 Fade: Well, I'd gladly pick up a SiCortex MIPS64 ;-) 19:41:55 unless, of course, you're running mips. 19:41:56 salex: mm. that sounds sexy. i was very long thinking about a free O2 give-away at the university, but decided against it. the 4DW environment looks so cool. :) 19:42:05 s/mips/irix 19:42:13 Fare: not ISA-wise. But K8 looks like it's real daddy was EV7 :D 19:42:15 (including the ability to run OpenGenera) 19:42:27 amd64 incorporated the memory arch from alpha. 19:42:36 Well the ISA is just a layer of code translation, anyway. 19:42:47 to be exact, K8 incorporates EV7 design 19:43:03 shagou [i=user@Mix-Orleans-106-1-179.w193-248.abo.wanadoo.fr] has joined #lisp 19:43:05 Who cares if the instruction encoding has weird 8080 legacy? 19:43:56 Fare: it also includes various stuff regarding caches that is legacy of in-order, no-cache cpu that was running at the same speed as it's expansion bus 19:44:04 brb 19:44:04 I mean, we program in high-level languages that shield us from instruction encoding 19:44:05 p_l: the trinity college computer society has a large ancient alpha which is never used; you could steal it :) 19:44:06 hypno: i gave that away too, to some deserving undergad CS types who promised to treat it right 19:44:16 same place most of my gear went, really 19:45:06 p_l: in order and strongish cache coherence is often a nice property for debugging signal-based / concurrent programs. 19:45:06 I did have a dozen alpha stations for a while at work, but most of them went back to digital 19:45:10 i think we kept 2 19:45:18 and ran linux on them 19:45:25 :) 19:45:35 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 19:45:39 salex: heh. 19:45:59 I mean, if your signal is delivered at a random point in a block, that makes the job of the code-generator / signal handler way more "interesting" than most people would like 19:46:28 and/or make you insert barrier instructions like crazy, that eliminate any desired performance improvement. 19:46:32 (I'd converted that company from a solaris shop to a linux shop by then) 19:46:41 i have to say the IBM Power machines has impressed me a lot tho. very fast. PA-RISC is pretty darn snappy too. i have a B132 lying around. 19:46:54 what's a B132? 19:47:10 Fare: HP PA-RISC machine. 19:47:12 stassats [n=stassats@wikipedia/stassats] has joined #lisp 19:47:15 *rsynnott* has never even seen a HPPA 19:47:20 which was how we ended up with dual pentium intel boxes in product, for that matter 19:47:35 it feels very quick for 133mhz that's for sure. 19:47:36 i had a couple of those too (the HP's) at the same tome 19:47:49 they were even more out to lunch than sgi iirc 19:48:11 price wise, on bulk 19:48:39 heh. at the same time 19:49:30 benny [n=benny@i577A0A6B.versanet.de] has joined #lisp 19:49:45 it was kind of sad actually. I would rather have worked with almost any of those companies than intel 19:49:56 but it's like they didn't want to think about what was possible 19:50:15 (and by intel, i mean intel actually built the boxes we ended up using, not just the cpus) 19:51:22 why? 19:51:52 cost. they couldn't get their heads around what we were trying to do 19:51:59 or do you mean, why did intel build them? 19:52:25 no I mean what was so bad about intel in that time 19:52:29 the company 19:52:52 yeah, volumes. i guess Sun has pretty much the same problems even today, even though their "recent" stuff has at least been somewhat impressive. 19:53:11 oh. nothing terrible, but the architecture just wasn't nearly as nice as alpha, or sgi 19:53:20 so it wasn't so much anti-intel 19:53:32 *tmh* is annoyed by the return value of FMAKUNBOUND. 19:53:33 as pro- these guys 19:53:38 oh I thought you were talking about bad experience with the company 19:53:44 also, we were a unix shop, and so were they 19:54:03 but in the end, we had to go with linux on intel because they couldn't even get in the ballpark 19:54:27 and that was why i was sad about it, because i thought at the time "this is why you will all die" and i was mostly right 19:54:28 I would like to see FMAKUNBOUND a boolean indicating FBOUNDP so that I know I actually passed a bound symbol. 19:54:40 *return* 19:54:45 -!- smoofra [n=user@cthulhu.elder-gods.org] has left #lisp 19:54:58 SGI had incredible hardware... if you were willing to pay the price 19:55:10 it was pretty good yes 19:55:16 -!- Jacob_H [n=jacob@92.6.218.36] has quit ["Leaving"] 19:55:21 alpha walked all over it in our space, though 19:55:24 I've still got a 500mhz 21264 alpha 19:55:25 tmh: it should signal an error if the symbol is not fbound 19:55:36 if you needed brute cpu then alpha was better 19:55:40 tmh: err scratch that 19:55:42 -!- Dave2 [i=Dave2@freenode/staff/dave2] has quit [Read error: 104 (Connection reset by peer)] 19:56:00 I actually need to get rid of that alpha 19:56:08 however .... they were clueless. i gave them a rough outline of what we wanted to embed and how we could do it for maybe 2500 in quantity. 19:56:23 they came back with a "super deal" at around 15000 19:56:34 it was a joke 19:56:36 heh 19:56:37 tcr: I'm thinking of wrapping it. Hmm, think I'll paste for feedback. Just a minute. 19:56:58 they were expensive... but for what they were good at... no one could tough them 19:57:12 they were pushing 11M poly's when everyone else was doing 600k 19:57:18 merimus: it didn't matter 19:57:26 when I saw that quote, i knew they were absolutely fucked 19:57:31 c|p [n=will@69.163.33.38] has joined #lisp 19:57:37 and they only lasted a few years after that 19:57:42 salex: then you didn't need that kind of machine 19:58:01 -!- schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has left #lisp 19:58:05 merimus: actually, i did need something like there O2's or O200s 19:58:05 -!- manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has quit [Read error: 113 (No route to host)] 19:58:17 and i like the infinite reality as much as anyone 19:58:18 tmh: (defun fmakunbound* (x) (prog1 (fboundp x) (fmakunbound x))) 19:58:23 i thik the same is true even today. the people that buy IBM are in it for the long term with solid cash flow. the people that buy Sun are in it for Oracle with similiar needs. important systems for important applications. 19:58:47 tmh: or even better: (defun fmakunbound* (x) (prog1 (and (fboundp x) (fdefinition x)) (fmakunbound x))) 19:58:52 the point was, they had no idea what was going on elsewhere, who their competitors were going to be, what sort of market it would be 19:59:09 salex: well... they got overrun with comodity hardware 19:59:10 they made a belated stab at it a year or so later, but it was already too late 19:59:11 pjb: Thanks, I'm wanting a little more than that. I want to be able to enter a new symbol if the one I provided is not bound. 19:59:23 -!- c|p [n=will@69.163.33.38] has quit [Client Quit] 19:59:30 pjb: why invoke fmakunbound on a symbol which is not fbound? 19:59:39 hypno: I worked at company that made a point of keeping nearly everything to real server hw from IBM or Sun (because HP killed alpha which was big part of the backbone before) 19:59:39 merimus: but that was the point. I was at the bleeding edge of that, and saying "look, if you can't do this in a year or two, your done" 19:59:42 although even today there are some things that no other machine can do 19:59:44 which is what happened 20:00:05 because the niche where they were untouchable was too small for a company 1/10th their size 20:00:10 anyway, old history 20:00:11 tcr: right. (defun fmakunbound* (x) (when (fboundp x) (prog1 (fdefinition x) (fmakunbound x)))) 20:00:23 they were just impressively clueless 20:00:33 hypno: to be exact, that was the "mission critical" part. Office support ran on x86 with windows 20:00:37 sun, otoh, was impressively cynical, and sent us hott blonds to talke about a deal 20:00:38 I would be happy to have a choice OTHER than x86 20:02:01 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 20:02:02 What you get when you let women design robots: http://www.youtube.com/watch?v=OnWZJgR6ZcE 20:02:13 p_l: heh. yeah. these brands are "consulting" friendly mainly i guess. ime, the oracle installations our customers have would do just as fine on postgres and x86. but it's the management and "serious" factor i guess. 20:02:40 c|p [n=will@69.163.33.38] has joined #lisp 20:02:43 Why do robots get all the fun? 20:02:45 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 20:02:48 pjb: well... I can do that 20:02:55 hypno/pl: well, it can be about service contracts too, at that level 20:03:34 hypno: yeah, we had oracle, but one of the point of said systems was also how they were built 20:03:35 I do wonder what will happen to IBM when everyone abruptly realises that Lotus Notes is terrible 20:03:37 Let's work on telepresence :-) 20:03:39 salex: indeed, and in many places it's just stupid to got with some cheap ass x86 hardware of course. but i get the feeling that that niche is getting smaller every day. 20:03:56 (it seems to be basically impossible for anyone to escape once they buy stuff from one of these big companies) 20:04:20 hypno: actually for the contract I'm referring to above, we could only go with intel because they would build the whole box for us with some guarantees 20:04:21 rsynnott: is it really terrible? I only hear positive opinions about it. 20:04:29 not unlike going with ibm/sun/whoever 20:04:45 most of the PC market was (and still is) totaly irrelevant 20:04:48 Like, it brings more bread than lisp... 20:04:51 for that purpos 20:05:17 I'd like a x86 system that wasn't tied to IBM PC ideas 20:05:42 *Xach* (very slowly) produces http://xach.com/tmp/zx.gif with vecto and skippy 20:05:43 -!- cp2 [n=will@unaffiliated/cp2] has quit [Read error: 113 (No route to host)] 20:06:06 Jasko2 [n=tjasko@75-149-33-105-SFBA.hfc.comcastbusiness.net] has joined #lisp 20:06:10 cute! 20:06:28 -!- ljosa [n=ljosa@gm3b6-dd5.broad.mit.edu] has quit [] 20:06:33 it's a little jaggy part way through... could you use a higher order spline? 20:06:48 jwz commented that he couldn't imagine doing daliclock with , but i think you could make something interesting if not identical... 20:07:07 hmmm. 20:07:37 What is the equivalent of fboundp for macros? 20:07:50 salex: there's a lot of room for cleverness in the quest for a better appearance, for sure 20:08:07 tmh: (eq (macroexpand-1 foo) foo) 20:08:27 -!- Jabberwockey [n=jens@p54964BA9.dip.t-dialin.net] has quit [Remote closed the connection] 20:08:32 Xach: that wasn't meant to sound too complainy ... it's neat 20:08:37 Fare: Thanks, and ugh. 20:08:41 i'm jsut thinking about interpolating smooth curves today 20:08:47 except it may have side-effects, including borking because of wrong number of arguments (I smell ignore-errors) 20:08:56 tmh, FBOUNDP works for macros. 20:09:07 Riastradh, is it guaranteed to? 20:09:11 Yes. 20:09:21 salex: yeah, but i took as many shortcuts as possible to get the basic idea into motion. 20:09:35 `fbound ['ef,band] adj. (of a function name) bound in the function namespace. (The names of macros and special operators are fbound...)' 20:09:40 maybe I was thinking about symbol-macros 20:09:47 Riastradh: Okay, I couldn't divine that from the hyperspec, I'll read again. 20:10:14 Oh, well, it was buried in the examples. 20:10:24 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [] 20:10:31 jmbr_ [n=jmbr@253.32.220.87.dynamic.jazztel.es] has joined #lisp 20:10:42 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 20:10:46 -!- jmbr [n=jmbr@209.33.220.87.dynamic.jazztel.es] has quit [Read error: 60 (Operation timed out)] 20:11:11 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 110 (Connection timed out)] 20:12:43 However, it is not always the case that if FBOUNDP is true then you can FUNCALL what you get out of FDEFINITION. 20:13:08 Xach: that's v.neat 20:13:32 (and with symbol-macros, you can't get the number of arguments wrong) 20:13:32 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [Client Quit] 20:13:35 I think you can do that iff (1) FBOUNDP is true, (2) SPECIAL-OPERATOR-P is false, and (3) MACRO-FUNCTION yields nil. 20:14:17 Riastradh, poor you -- your mind is polluted with too much knowledge of the CLHS. 20:14:19 that sounds right 20:14:27 erm, that to Riastradh, not to Fare! 20:14:40 No, Fare is just as right. 20:14:47 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit ["ChatZilla 0.9.85 [Firefox 3.0.11/2009060215]"] 20:14:52 sure, but i wasn't responding to Fare 20:15:04 but agreeing with your analysis 20:15:04 An excessive knowledge of Marxism is a sign of a misspent youth. -- John McCarthy (who didn't guess the same would apply to Lisp!) 20:16:00 my misspent you gave me excessive knowledge of motorcyles, whiskies, and several highway routes 20:16:03 jointh the revolutionth comdradeth! 20:16:04 :) 20:16:06 so it's all tradeoffs 20:16:09 An exccessive knowledge of Marxism is a sign of a misspent Lisp? 20:16:11 Fare: the lisp designer or the evil politician? 20:16:31 oh, he was Joe not John 20:16:38 rsynnott: Joseph 20:16:48 ah, you caught it 20:17:15 elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has joined #lisp 20:17:27 nyef: you generalize interesting variables. 20:17:36 I try. 20:18:04 will you be able to lightning talk in Boston any time soon? 20:18:16 *Fare* is looking for a 2nd lightning talker next monday 20:18:18 Not on Monday, I'm afraid. 20:18:47 (What with not being scheduled to leave North Dakota until the following thursday evening at the earliest.) 20:19:26 next Monday: http://fare.livejournal.com/144600.html 20:19:46 nyef: you can speak in July, August, or whenever you want 20:20:56 I was so hoping to be able to drive all weekend, wind up in the Boston/Cambridge area around midday Monday, hit up the Bromfield Pen Shop with my souvenier from South Dakota, and then attend the meeting in the evening, but it was not to be. 20:21:04 tmh pasted "FUNBIND - Feedback please" at http://paste.lisp.org/display/82510 20:22:06 `providing a continuation error'? 20:22:39 I know what I meant. :-) 20:22:59 What's your use-case for this? 20:23:10 Clearly, it's for interactive use, so...? 20:23:12 erm, what nyef said 20:23:19 Jasko3 [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 20:23:46 nyef: I'm trying to get into the habit of keeping my lisp image running when I make changes, like factoring code. 20:23:50 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:23:51 -!- merimus [n=wroth@nat-49.laurelnetworks.com] has quit [] 20:23:57 fe[nl]ix [n=algidus@88-149-211-230.dynamic.ngi.it] has joined #lisp 20:24:06 tmh: it's called a "continuable error" 20:24:08 Hunh. 20:24:48 -!- sellout [n=greg@guest-fw.dc4.itasoftware.com] has quit [] 20:24:53 For a while, when I made changes sufficiently invasive as to require reloading stuff I'd blow my fasls, delete-package my loaded instance, and reload. It at least leaves the -lisp- running... 20:25:03 'course, it tended to leak memory abominably. 20:25:34 nyef: what about existing threads, connections, etc? 20:25:37 hello 20:25:47 Didn't have any, so no loss there. 20:25:48 nyef: Okay, I'm trying avoid that, not knowing about the memory leaking. I'm trying to get into the habit of managing things at a finer level. 20:26:10 -!- [Blay] [n=blaay@BSN-165-118-26.dial-up.dsl.siol.net] has quit [Read error: 110 (Connection timed out)] 20:26:30 FMAKUNBOUND is particularly annoying because I'm never confident that I've actually deleted anything. It's feedback is worthless. 20:26:51 managing things at a finer level and not knowing about memory leaks 20:26:55 ?? 20:27:00 The other thing is that the old function definitions don't tend to cause problems. 20:27:06 So why bother unbinding them? 20:27:13 younder: You're thinking of a different kind of manage. 20:27:48 tmh: why do you want to _delete_ anything? Leave that up to the garbage collector! 20:27:48 nyef: They cause problems if I've missed replacing their use somewhere. I want an error generated. 20:27:50 perhaps 20:28:07 "deleting" is awkwardly low-level... 20:28:26 I can see that, I guess... But that's what who-calls is for, or grep. 20:28:39 s/delete-package/unregister-package/ there's a problem of terminology too... 20:29:00 who-calls and so fourth 20:29:08 Yes, a stale direct reference could be a problem, but, being familiar with your own deisgn, you should know when those are likely. 20:29:26 tmh: i'm missing something.... so you don't like it if you've called fmakunbound on a symbol that isn't fbound because why? 20:29:32 there should be NO MEMORY LEEK 20:29:34 i can understand ethe other way around 20:29:59 i.e. you've got refrences to old versions of functions around before you "refactored", so you aren't running what you thought you were 20:30:01 unless you f***ed up 20:30:06 For those of you that think I'm managing memory, we are talking passed each other. 20:30:09 although, what ney fast about who-calls 20:30:20 wow. what nyef said 20:30:35 how the hell that turned into what ney fast, I have no idea 20:31:10 salex: Are we going to have the keyboard layout discussion again? 20:31:33 heh 20:31:48 Heh. "memory leek" sounds like a vegetable for the treatment of amnesia. 20:31:50 -!- SandGorgon [n=OmNomNom@122.162.130.77] has quit [Read error: 110 (Connection timed out)] 20:31:53 no,no,no 20:31:53 i'm trying to do actual work while procrastinating here 20:31:53 -!- c|p [n=will@69.163.33.38] has quit [Client Quit] 20:32:00 failing, but trying 20:32:01 cp2 [n=will@unaffiliated/cp2] has joined #lisp 20:32:02 no keyboard layout 20:32:18 i'm so needing to get a talk together it's not funny though 20:32:22 yet here I am 20:33:00 Fare: do you know of a more readable version of AITR-874 than that pdf you mentioned in the xcvb paper ? 20:34:12 -!- EnglishGent [n=EnglishG@ai-core.demon.co.uk] has quit [Read error: 60 (Operation timed out)] 20:34:14 no -- but if you find it, I'll link it. 20:34:40 damn 20:35:05 -!- ikki [n=ikki@201.155.75.146] has quit [Remote closed the connection] 20:35:08 that pdf is so blurry that my eyes hurt trying to read it 20:35:26 fe[nl]ix: What are you reading it in? 20:35:31 what?? 20:35:38 -!- rdd [n=user@c83-250-157-93.bredband.comhem.se] has quit [Read error: 104 (Connection reset by peer)] 20:35:55 tmh: what do you mean ? 20:36:09 Acroread, evince, xpdf, etc? 20:36:56 evince 20:37:05 wonder if it's worth printing and rescanning with a bit of sharpen. that's pretty nasty 20:37:15 Jacob_H [n=jacob@92.6.218.36] has joined #lisp 20:37:23 Yeah, I've had problems with that for some documents. Sometimes they display better in acroread. 20:37:43 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 20:37:44 sometimes they just aren't scanned well 20:37:51 salex: see for yourself: ftp://publications.ai.mit.edu/ai-publications/pdf/AITR-874.pdf 20:37:56 yes, i did 20:37:58 it's bad 20:38:09 it looks ok zoomed out 20:38:20 The one I'm always struggling with is the Norvig and Pitman "Tutorial on Good Lisp Programming Style". It's barely readable in evince and I use it frequently. 20:38:24 but i don't know how fixable. it's good enough that might work 20:38:38 ikki [n=ikki@201.155.75.146] has joined #lisp 20:38:58 Ugh, is it a scan? 20:39:46 fe[nl]ix, are you interested in building systems? 20:40:06 hardly surprising for a 1985 report, thm 20:40:08 tmh 20:40:10 and the history of system builders? 20:40:21 salex: yes? 20:40:28 adobe invented pdf, do you perhaps imagie it would display better in one of their products? 20:40:33 duh 20:40:39 build isn't THAT relevant today 20:40:42 tmh: i was correcting my typo 20:40:54 younders, the pdf is just a convert from ps though 20:40:55 Fare: I'm interested in adjusting ASDF a little 20:41:06 and no, sometimes adobe doesn't do such a great job 20:41:08 the main idea is that you should be declarative as to who provides or uses what. But CL doesn't help much with that. 20:41:23 fe[nl]ix, why not join xcvb instead? 20:41:29 salex: Ah, okay, I'm opening it in acroread to see if it is any better. 20:41:50 -!- Jasko2 [n=tjasko@75-149-33-105-SFBA.hfc.comcastbusiness.net] has quit [Read error: 110 (Connection timed out)] 20:41:52 Fare: ATM I dislike xcvb 20:42:05 And that would be a big fat nope. It's a bad scan no matter how you look at it. 20:42:28 acroread does a bad job with some things, fwiw 20:42:34 fe[nl]ix, OK -- what should I improve about it to make you like it? 20:43:19 (I have a long TODO list, but you might help me prioritize it) 20:43:36 tmh: there's nothing for acroread to help. essentially these things are raster 20:43:38 (or you might dislike something *essential* about it) 20:44:11 for that matter, pdf display is sometimes worse than .ps for things with embedded fonts of the wrong type, etc. 20:48:01 Fare: the integration with make is enough to make me cringe 20:48:41 fe[nl]ix, so you'd like another backend? 20:49:29 Fare: also, I *like* having a central list of dependencies 20:49:52 fe[nl]ix, that's not compatible with incremental compilation when said list changes. 20:50:03 fe[nl]ix, or with maintainability 20:50:15 so i had a look at that document... and you can probably improve readability by importing it to a raster tool like gimp and unsharp masking. bit of a trudge though, for 50 pages or wahtever, and i fyou overdo it it will just be worse 20:50:17 I don't care about incremental recompilation 20:50:20 Sure it is, Fare. Also, can't xcvb do that now? 20:50:31 do what? 20:50:37 Have central lists of dependencies. 20:50:47 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 20:50:47 still won't be beautiful, but help with the wobbly eyes 20:51:04 each module has its own dependencies. The main BUILD files has its own. 20:51:11 (disguised as generated files with no generation dependencies) 20:51:12 or OCR it 20:51:17 (that may transitively include other stuff) 20:51:19 that too 20:51:34 or ask nice, maybe someone has the TeX sources.... 20:51:49 maybe not, considering it looks tob e a scan 20:51:53 salex: use imagemagick to do mass alteration of images 20:52:03 dlowe: either way 20:52:16 it's about the same, and you'll have to fiddle with the params a bit 20:52:22 OK, so generated files currently have their dependencies listed in the central BUILD file. Hum. That's true. A lot of things should depend on the BUILD files. 20:52:27 hence gimp suggestion 20:52:27 If the dependency list for some module changes, then you probably need to recompile that module. A simple approximation is that if the central file of dependency lists changes, then everything it describes must be recompiled, but you can do better approximations. 20:52:45 At least, generated files should depend on them, and xcvb.mk itself. Hum. 20:53:20 dlowe: i'm overall mixed on imagemagick 20:53:38 salex: there's not much better for command line image munging 20:53:41 (not that i'm a huge fan of gimpe either) 20:53:49 pkhuong, salex: the article on C as an assembler is . 20:53:55 dlowe: oh, it's useful 20:54:08 it just has some design quirks etc. 20:54:21 i do a *lot* of image processing, and hardly ever use it 20:54:27 C isn't assember 20:54:35 *Fare* adds a few things on the TODO.rest 20:54:37 but it's very good for some use cases, sure 20:55:00 not good with big images though, doens't handle 16bit represnetations, et.c et.c 20:55:03 particularly in this paralell prrocessing world! 20:55:05 (well, it does with a recompile) 20:55:24 unless they've fixed that 20:55:34 salex: that's what xargs is for :) 20:55:42 ? 20:55:50 xargs 20:55:53 ? 20:56:08 find . -name '*.jpg' | xargs -P 8 -n 1 convert -geometry 640x480 20:56:17 Quick, informal poll, when you are making modifications to your code, at what point do you delete the package, fasl's and reload? That used to be may practice for even small changes, but lately I've been wanting to maintain things by hand with, for example, makunbound and fmakunbound. 20:56:21 the command, xargs? 20:56:22 The -P runs multiple processes 20:56:28 i know what xargs is. i don't understand what you are responding to 20:56:31 Riastradh, sure -- you could have a cache of which dependencies were used during the compilation of each file, store that in the obj/ directory and update when needed. Lots of infrastructure, but doable. 20:56:43 younder: particularly in this paralell prrocessing world! 20:56:58 I think I conflated conversations 20:57:17 it's very very slow for some simple things too, because it always uses an intermediate representation 20:57:28 anyway, 'nuff about that 20:57:37 tmh: Typically, my decision depends on how much internal state would be broken with my change and how hard it would be to repair it vs. simply reloading. 20:58:35 If I change a struct, definition, for example, I'm going to reload. 20:58:40 nyef: Yes, internal state, that's what started motivating me to manage the symbols by hand. 20:58:48 ejs [n=eugen@206-249-124-91.pool.ukrtel.net] has joined #lisp 20:58:50 tmh: very seldom, typically only if i've done extensive gf refactoring and want make sure i don't forget to remove some old methods 20:58:54 ... Where'd that first comma come from? 20:59:20 nyef: It's a pause, I was thinking. 20:59:22 while i do think your environment should really cover all of this, why is delete-package not sufficient for most development needs? 20:59:23 it you nuke your package, it's almost the same as restarting the image 20:59:34 I refactored lisppaste's paste storage and new-paste pipeline without reloading it, for example. 20:59:54 Riastradh, but the "each file has its own" allows for more decentralized use of files. 21:00:07 without requiring heavy infrastructure. 21:00:10 most often i use slime inspector to remove methods i don't want, but i almost never remove functions explicitly 21:00:29 or heavy metadata 21:00:32 hypno: It's not desired for a live system, plus a full reload cycle is extra time in your edit-compile-test cycle. 21:00:40 nikodemus: I really don't mean to pester, but were you planning on taking a look at that runtime executable path patch before the next release? 21:00:52 Contrariwise, Fare, browsing a source tree organized with a central collection of dependency lists is easier without extra infrastructure than browsing a source tree in which the dependency information is scattered. 21:00:56 joshe: yes, tomorrow or saturday 21:01:01 I understand that you're probably busy and it's not a high priority 21:01:10 ah, thanks 21:01:14 unless someone else gets to it first, of course 21:01:25 Riastradh, well, you can easily reconstruct the global dependency graph. XCVB does that for you, actually! 21:01:28 When are we entering code-freeze? 21:01:34 sunday/momday 21:01:39 monday. even 21:01:50 I was going to say, wasn't that last month or so? 21:02:01 -!- salex [n=user@216.80.147.206] has quit [Remote closed the connection] 21:02:05 (plenty of committers who at least in theory could merge it as well as me) 21:02:20 declarative programming is all about local syntactic changes, global semantic changes. 21:02:38 nfma [n=nuno@ip-87-86-105-3.easynet.co.uk] has joined #lisp 21:02:44 So I could possibly land the tale of two once-onlies in time for the new release, but probably shouldn't. 21:03:07 nyef: ah, i can see that for a live production system, but is there any tools really to do /safe/ (100%) system modifications in a running lisp? i've worried a lot about that before, and my response has basically been to disable interrupts, compile and load and hope it all worked out. 21:03:08 cyb3r3li0g [n=eguzman@63.100.62.66] has joined #lisp 21:03:29 hypno: There are a couple techniques for making some changes safely. 21:03:44 fork, load, test, kill original. :) 21:03:59 hypno, there are no safe 100% tools for much anything as general as a full programming environment. 21:04:17 Another possibility is to have a light enough load and fast enough responses that nobody really notices when you break something. 21:05:11 On the other hand, I'd like to see something that would allow applications to execve() a new version of themselves, preserving connections and restoring state from a dump. 21:05:24 Fare: well true, but i've got the feeling that failure modes and system maintenance are not as explicit as in say Erlang. this is bad for the "hot swapping" argument of course. 21:05:34 Fare: that's easy enough, but not really what you want to do 21:05:35 hypno: yes indeed. 21:05:43 foom: what do I want to do? 21:05:54 Fare: you really want to fork/exec the new server, make sure it's working, then send the open connections over. 21:06:03 -!- ejs [n=eugen@206-249-124-91.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 21:06:13 foom: probably. 21:06:16 you don't want to drop the connections just because your new server didn't startup properly 21:06:17 levy [n=levy@apn-94-44-11-86.vodafone.hu] has joined #lisp 21:07:00 so, just "start a different process, handshake then migrate connections over with cmsg ?" 21:07:19 but, doing the execve-and-pick-up-connections was something that was standard practice in MUDs oh...at least 10 years ago. :) 21:07:21 -!- anekos is now known as A_anekos 21:07:37 clhs cl-user 21:07:37 Sorry, I couldn't find anything for cl-user. 21:08:20 foom: Heh. I remember copyover. 21:08:40 -!- nfma [n=nuno@ip-87-86-105-3.easynet.co.uk] has quit [Remote closed the connection] 21:09:05 is there an easy way to get the source (using SBCL) for a function or a method of a generic function? 21:09:06 clhs common-lisp-user 21:09:06 Sorry, I couldn't find anything for common-lisp-user. 21:09:08 prxq: nobody will contest a claim that C is sometimes uncomfortably high level. The actual topic was whether it's possible to write a good compiler that targets C. 21:09:36 nyef: indeed. :) 21:09:56 actually I would like to get the source of CL functions, generic functions, each method, etc. so the source for SBCL itself 21:10:24 what's your C-k bound to? kill-sexp or something else? 21:11:11 For the mud I ran I added a SIGSEGV handler that reran main() if it crashed, and if that failed, saved its state and re-exec'd itself preserving connections. That worked nearly all the time. :) 21:12:56 jonathon [n=user@208.89.21.37] has joined #lisp 21:12:57 prxq: Duane is also somewhat attacking a strawman there, with a weird goal of fully portable C. In the real world, C backends tend to use C as a simpler to use and optimise assembler. 21:13:15 krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has joined #lisp 21:13:29 levy: perhaps swank-backend:find-source-location ? 21:13:34 foom: ugly 21:13:40 levy: ask SLIME. 21:14:02 actually I'd like to play with partial evaluation and make-instance 21:14:05 foom: how did you identify which open fd's where what? 21:14:21 Fare: it wrote it to a file and read it in at startup. 21:14:23 for that I need to be able to inline the call tree of make-instance 21:14:57 and see what I can get when say the class and the initargs are constants 21:15:03 levy: there is no single source form for make-instance. Specialised constructor functions are compiled at runtime. 21:15:03 foom: did you also fork and dump core to have some debug state? 21:15:21 I know, but there is compute-applicable-methods and friends 21:15:30 so it is possible to build that single source 21:15:38 from the sources of all methods etc. 21:15:53 What open source projects would you guys suggest that demonstrate good use of macros and closures and other more advanced lisp techniques? 21:16:09 -!- krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has quit [Remote closed the connection] 21:16:16 jonathon: cl-ppcre 21:16:27 Fare: yeah, it did fork(), called abort() in one child, and waitpid(); exec() in the other. 21:16:36 krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has joined #lisp 21:17:06 fe[nl]ix: I'll check it out.... 21:17:07 what would waitpid() do in a child? Or you mean, the parent? 21:17:16 sorry, right. 21:17:17 fe[nl]ix, is swank:find-source-location something new? I don't have it 21:18:54 *Fare* is reminded me of the hacker who was using syssigreturn as his "tail call into coredumped process state" instruction so he could coredump & restore long computations across system reboots and/or periods when his process was using too much resources for the shared VAX. 21:19:38 DragonFly BSD has OS-level support for that, IIRC 21:19:50 (he'd just reopen fds before to syssigreturn) 21:20:06 p_l: I used to run DFly a couple years ago... cool system. 21:20:12 p_l: any link to that? 21:20:44 heck, emacs does that on every UNIX out there. :) 21:20:46 The clustering support Matt is working on for DFly is really slick. 21:20:54 how do you think it starts up so fast! 21:20:54 Fare: no links, but search around DFly website 21:21:01 foom: Heh. For the mud I was coding on, I ended up building a structured exception handling system in C based on setjmp/longjmp and some truly nasty stack frame manipulation. 21:21:06 (why do you need to waitpid in the parent? What if there's a double-fault or horrible thing happening in the coredumping child?) 21:21:16 I haven't checked recently, but does sbcl build on DFly? 21:21:19 levy: in your slime, I think it's find-definitions-for-emacs 21:21:35 Fare: to keep it from hanging around as a zombie process, I guess. 21:21:52 (won't init take care of it?) 21:21:59 -!- jonathon [n=user@208.89.21.37] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 21:22:02 pkhuong: right, but anyways, ten years later, there remains only a single processor to target, so there's no portability advantage to C anymore. (Well, two processors, ix86, and arm). 21:22:07 (does the exec thing inherit children from the parent?) 21:22:32 my favorite features I added to my MUD were ssh support and vt100 support. 21:22:35 pjb: like I said, the point of targetting C isn't really (processor-) portability. 21:22:54 http://paste.lisp.org/display/82486#2 .. this's probably go wrong in one way or another .. :} 21:22:55 foom: what were they? 21:23:00 foom: Ooh. One of the things I was thinking of possibly adding was X support for the admins. 21:23:04 foom: as in, fancy vt100 graphics? 21:23:05 ssh support was done using a hacked up openssh which translated ssh into telnet. :) 21:23:23 pkhuong: Remain the other points, that you can write a better compiler than C compilers (you may deal directly with the good features of the processor). 21:23:30 Never got past the "that would be cool to see" stage, though. 21:23:32 Fare: that trick won't necessarily work. glibc(and others) are starting to use O_CLOEXEC. I'm not sure about BSDs 21:23:40 Fare: "ssh -p 4005 ar.static.net" if you want to see for yourself. 21:23:44 wouldn't just a simple entry in the .ssh/config have done that? 21:24:00 fe[nl]ix, you are right, but it seems that method just returns some snippets not the whole source 21:24:13 pjb: sure, you might. You could also build a 1:1 eiffel tower with matchsticks. 21:24:15 Fare: you need to get past logging in before it starts vt100 stuff. 21:24:15 pjb: I can assure you that MIPS and PPC feel fine in embedded space 21:24:19 -!- elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has quit [Read error: 60 (Operation timed out)] 21:24:21 foom: ar thinks that é is not a letter! 21:24:28 Fare: sorry. :P 21:24:40 fe[nl]ix, of course it returns file positions but that is somewhat more difficult to make it work, I just thought there's some SBCL build switch to remember source forms 21:24:46 levy: that's correct. you then need to open the file, and browse it according to what f-d-f-e returns 21:24:48 and no I'm not gonna fix it, I haven't touched that code for 5 years. :) 21:24:52 You can also sometimes use the __asm__ escape hatch. 21:25:16 *p_l* did some MIPS debugging once and has MIPS-based router at home 21:25:32 mips and ppc aren't going anywhere. 21:25:39 levy: I don't think so. what you could do is modify sbcl to make it keep sources in some global hash table. shouldn't be too hard 21:25:42 fe[nl]ix, after all that is not that hard 21:25:53 :) 21:25:54 hell, now that sparc is a free design, it'll probably proliferate. 21:26:08 Fade: SPARC has been free since birth. 21:26:39 didn't sun open up their lastest sparc64 designs last year? 21:27:21 Fade: CoolThreads is available under GPL 21:27:40 both T1 and T2 21:28:10 one of the best parts was when I realized that most vt100 emulator writers couldn't read a spec if it bit them on the nose. 21:28:26 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["This computer has gone to sleep"] 21:28:30 and only make it vt100-enough to work with whatever sequences terminfo and emacs can output 21:30:22 *p_l* doesn't need emulator, he keeps a genouine VT510 at home 21:30:42 windows telnet and gnome terminal were the worst. 21:31:32 somehow I'm not surprised 21:31:34 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit ["Leaving."] 21:32:52 sea4ever [i=sea@216.110.106.39] has joined #lisp 21:33:25 manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has joined #lisp 21:33:33 -!- tombom [i=tombom@wikipedia/Tombomp] has quit ["Peace and Protection 4.22.2"] 21:34:00 -!- cp2 [n=will@unaffiliated/cp2] has quit ["leaving"] 21:34:22 cp2 [n=will@unaffiliated/cp2] has joined #lisp 21:34:24 regarding SPARCs, there's also single core, independent variant based on T1, with some stuff to make it easier for experimenters (try fitting T1 into FPGA). There's also LEON 21:34:54 dv_ [n=dv@85-127-129-32.dynamic.xdsl-line.inode.at] has joined #lisp 21:35:35 slash_ [n=Unknown@whgeh0138.cip.uni-regensburg.de] has joined #lisp 21:36:51 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 21:37:04 -!- asksol [n=ask@175.243.251.212.customer.cdi.no] has quit [] 21:37:51 -!- cp2 [n=will@unaffiliated/cp2] has quit [Client Quit] 21:38:08 btw, PDS goes for $15k on university promotion ;-) 21:38:12 cp2 [n=will@unaffiliated/cp2] has joined #lisp 21:38:41 only in USA, though 21:39:50 -!- prxq [n=mommer@e176248028.adsl.alicedsl.de] has quit ["Leaving"] 21:40:33 dreish [n=dreish@minus.dreish.org] has joined #lisp 21:40:47 -!- Fare [n=Fare@ita4fw1.itasoftware.com] has quit ["Leaving"] 21:44:36 -!- bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has quit [] 21:46:07 [Blay] [n=blaay@BSN-142-113-43.dial-up.dsl.siol.net] has joined #lisp 21:46:31 qbg [n=qbg@65-73-93-20.dsl1.mnd.mn.frontiernet.net] has joined #lisp 21:46:59 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 21:47:06 -!- nyef [n=nyef@65-125-125-98.dia.static.qwest.net] has quit ["Leaving the office."] 21:48:19 -!- LiamH [n=none@common-lisp.net] has quit ["Leaving."] 21:48:51 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit [Read error: 104 (Connection reset by peer)] 21:49:01 asksol [n=ask@062249178132.customer.alfanett.no] has joined #lisp 21:49:23 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 21:50:59 nfma [n=nuno@ip-87-86-105-3.easynet.co.uk] has joined #lisp 21:55:43 dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has joined #lisp 21:57:13 -!- willb [n=wibenton@compsci-wifi-39.cs.wisc.edu] has quit [Read error: 110 (Connection timed out)] 21:59:43 rdad [n=rdad@cpe-67-247-62-86.nyc.res.rr.com] has joined #lisp 21:59:54 Hello all 22:00:24 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 22:00:45 Does anyone have a link to up to date documentation on slime? 22:00:53 fundamental [n=fundamen@24-148-122-247.ip.mhcable.com] has joined #lisp 22:02:08 I'm trying to get the REPL up 22:02:24 -!- gemelen [n=shelta@shpd-92-101-133-239.vologda.ru] has quit ["I wish the toaster to be happy, too."] 22:02:27 herbieB: Hi, I'm almost afraid to ask: did you by chance have a look at the escaping of \\ symbols? Or could you point me where in the source I'd have to muck about to fix it? 22:02:51 rdad: (slime-setup '(slime-fancy)). 22:03:10 pkhuong: Ok, I will try that 22:03:49 -!- claint [n=utku@85.105.94.127] has left #lisp 22:03:58 -!- legumbre_ is now known as legumbre 22:05:11 Thank you, Jesus 22:05:16 antoni` [n=user@143.pool85-53-26.dynamic.orange.es] has joined #lisp 22:05:39 Adlai [n=adlai@93-173-254-22.bb.netvision.net.il] has joined #lisp 22:05:53 archypetro [n=hellequi@cpc2-oxfd4-0-0-cust85.oxfd.cable.ntl.com] has joined #lisp 22:06:00 -!- TDT [i=dthole@dhcp80ff869b.dynamic.uiowa.edu] has quit ["Leaving."] 22:06:19 What does it mean that Allegro CL "links against libpthread on Unix platforms that have it"? 22:06:23 I spent hours trying to get that up. The state of documention in the lisp world is killing me. All the tutorials, docs and books are more that 5 years old 22:07:11 Adlai: that it links against libpthread when it's there? I don't understand your question. 22:07:37 What does it mean to link against libpthread? 22:07:41 that it does use native threads? 22:08:10 This_Is_My_World [i=sea@216.110.106.24] has joined #lisp 22:08:41 IIUC, it uses native threads to implement green threading, somewhat like Python's GIL. 22:08:53 salex [n=user@216.80.147.206] has joined #lisp 22:09:07 Adlai: it means it uses pthreads so libs that use those interfaces will work as well... 22:09:10 pkhuong: python threads are OS threads 22:09:20 fe[nl]ix: exactly like I said. 22:09:20 pthreads doesn't necessarily mean native threading, after all 22:09:57 pkhuong: Why doesn't slime REPL buffer start up by default? Is it considered old? 22:09:58 -!- milanj [n=milan@79.101.76.232] has quit ["Leaving"] 22:10:00 Adlai: ACL does green threading, but will cooperate with OS threads in FFI 22:10:12 ok 22:10:48 rdad: make sure you have 'slime-repl started, like (slime-setup '(slime-repl)) or with slime-fancy 22:11:06 pkhuong: python has nothing like green threads 22:11:09 rdad: SLIME has been slightly rearchitected to make it more modular. The REPL is now a contrib. 22:11:25 -!- fe[nl]ix [n=algidus@88-149-211-230.dynamic.ngi.it] has quit ["Valete!"] 22:11:37 p_I: I do, thanks to pkhuong 22:11:37 fe[nl]ix: show me code that runs python threads from the same process on two different processors. 22:11:44 pkhuong: I see 22:12:53 felipe: how come according to clbuild iolib depends on lispbuilder-sdl? 22:13:12 fe[nl]ix [n=algidus@88-149-212-242.dynamic.ngi.it] has joined #lisp 22:13:19 s/felipe/fe[nl]ix 22:13:24 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 22:14:11 luis: that's a mistake 22:14:13 pkhuong: I think stackless does that 22:14:27 pkhuong: and of course native calls from different threads 22:14:37 yeah, I figured 22:14:59 fe[nl]ix: sure, python threads are OS threads. It only means that it's a fairly lossy implementation of green threading. 22:15:42 pkhuong: I suspect they didn't want to do green threads, but didn't make sure to write proper support for concurrency :D 22:16:00 willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has joined #lisp 22:16:24 -!- sea4ever [i=sea@unaffiliated/sea4ever] has quit [No route to host] 22:16:25 fe[nl]ix: can you build iolib's or osicat's documentation? 22:16:43 fe[nl]ix: I'm getting "fatal error before reaching READ-EVAL-PRINT loop: bad toplevel option: "#\\#" 22:16:53 pkhuong: by green threads I mean independent units of execution which are not scheduled by the OS, but by some VM underneath 22:17:15 pkhuong: by that definition, python's are not green threads, just lossy OS threads 22:17:41 fe[nl]ix: no in-language concurrency -> green threads. Using OS threads to implement coroutines is only an implementation accident. 22:19:07 luis: iolib's builds 22:19:26 fe[nl]ix: where did you get texinfo-docstrings from? 22:19:32 -!- qbg [n=qbg@65-73-93-20.dsl1.mnd.mn.frontiernet.net] has quit [Remote closed the connection] 22:20:16 -!- spilman [n=spilman@ARennes-552-1-82-202.w92-139.abo.wanadoo.fr] has quit ["Quitte"] 22:20:17 luis: I'm using a private hacked copy :) 22:20:20 forgot about that 22:20:35 ah, cool, can I pull from it? 22:21:58 -!- Jacob_H [n=jacob@92.6.218.36] has quit [Remote closed the connection] 22:22:43 -!- antoni [n=user@143.pool85-53-26.dynamic.orange.es] has quit [Connection timed out] 22:25:14 -!- This_Is_My_World [i=sea@216.110.106.24] has quit [Read error: 54 (Connection reset by peer)] 22:26:11 luis: git://lycia.ath.cx/texinfo-docstrings 22:29:03 sea4ever [i=sea@216.110.106.17] has joined #lisp 22:32:25 when I get home I'll try to propagate the changes and ask lichtblau to get you access to the repo. 22:33:21 luis: actually lichtblau told me to revert all his changes because they were incomplete 22:33:36 oh 22:33:39 I only fixed a couple of minor issues, nothing fancy 22:34:11 ok, I'll have to look into this more carefully then 22:36:48 -!- krumholt [n=krumholt@port-92-193-55-248.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)] 22:36:55 still doesn't work for me. 22:37:05 krumholt [n=krumholt@port-92-193-48-220.dynamic.qsc.de] has joined #lisp 22:37:21 I blame cl-launch. 22:37:29 possibly 22:38:04 -!- levy [n=levy@apn-94-44-11-86.vodafone.hu] has quit ["..."] 22:43:57 -!- slash_ [n=Unknown@whgeh0138.cip.uni-regensburg.de] has quit [Client Quit] 22:45:12 -!- fundamental [n=fundamen@24-148-122-247.ip.mhcable.com] has left #lisp 22:50:23 Lectus [n=Frederic@189.105.43.162] has joined #lisp 22:50:45 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 22:51:22 mark`` [n=user@74.196.9.26] has joined #lisp 22:53:05 -!- roygbiv [n=blank@pdpc/supporter/active/roygbiv] has left #lisp 22:53:40 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit ["leaving"] 23:01:04 elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has joined #lisp 23:05:09 rstandy [n=rastandy@net-93-144-204-143.t2.dsl.vodafone.it] has joined #lisp 23:06:36 -!- krumholt [n=krumholt@port-92-193-48-220.dynamic.qsc.de] has quit [Remote closed the connection] 23:08:06 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 23:08:54 Is there a lispy way to read from/write to a serial port? 23:09:32 Sikander, in linux you can use the device file 23:09:32 -!- dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 23:10:16 joga: Does that work correctly using with-open-file? How about setting rates, flow control etc? 23:10:33 dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has joined #lisp 23:10:36 Sikander, I used it to control a roomba, at least :) 23:10:44 joga: in lisp? Cool! 23:10:55 yeah 23:11:03 Sikander: through FFI? 23:11:03 -!- pinterface [n=pinterfa@knvl-static-09-0024.dsl.iowatelecom.net] has left #lisp 23:11:04 it was over bluetooth though but it's no different 23:11:11 no 23:11:26 err, scratch that, thought you asked me 23:11:47 -!- elliotstern [n=chatzill@cpe-67-240-162-85.rochester.res.rr.com] has quit [Read error: 60 (Operation timed out)] 23:11:49 stassats: yeah, figured it would need to be over ffi... 23:12:11 Sikander: it's OS-specific thing. On *nix, Plan 9 and NT, you open a device file. Setting options depends on OS: unix uses ioctl (sometimes also xattrs), Windows probably has some weird IOCTL-like equivalent and Plan9 has eiactl 23:12:12 joga: but how did you set all the parameters like rates, flow control, etc? 23:12:14 I think I just set the serial port settings from the command line 23:12:24 then wrote there from lisp 23:12:41 I'm on linux, and have most experience with serial ports in python (via pyserial) 23:12:42 I have the code somewhere though, it's fugly though, I'm just a beginner :) 23:13:04 spilman [n=spilman@ARennes-552-1-82-202.w92-139.abo.wanadoo.fr] has joined #lisp 23:13:10 I know that you always open the device file, but it would be nice to have a lispy interface. 23:13:30 you can write one 23:13:35 joga: I would still like to see that code; messing with a roomba is cool 23:13:51 stassats: yep, that's what the plan is now; I was wondering if there was already a "standard" 23:14:08 stassats: "standard" lib that is, but couldn't find anything on cliki 23:14:13 Sikander, I'll give you it 23:14:39 joga: great, I'm very interested! 23:14:49 -!- cYmen [n=cymen@squint.a-oben.org] has quit [Read error: 60 (Operation timed out)] 23:14:52 cYmen [n=cymen@squint.a-oben.org] has joined #lisp 23:14:57 |Soulman| [n=kvirc@154.80-202-254.nextgentel.com] has joined #lisp 23:14:59 -!- joga [i=joga@unaffiliated/joga] has quit [Read error: 60 (Operation timed out)] 23:15:04 joga [i=joga@rikki.fi] has joined #lisp 23:15:07 -!- mornfall [n=mornfall@kde/developer/mornfall] has quit [Read error: 60 (Operation timed out)] 23:15:12 Foofie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 23:15:16 sbcl has sb-posix:ioctl 23:15:49 stassats: yeah, I saw that. thanks 23:15:55 -!- Drakeson [n=user@69-165-133-213.dsl.teksavvy.com] has quit [Read error: 60 (Operation timed out)] 23:16:08 Drakeson [n=user@69-165-133-213.dsl.teksavvy.com] has joined #lisp 23:16:23 stassats: I'll probably end up writing a with-open-port type of thing 23:16:27 -!- erg [n=erg@li13-154.members.linode.com] has quit [Read error: 60 (Operation timed out)] 23:16:39 erg [n=erg@li13-154.members.linode.com] has joined #lisp 23:17:07 mornfall [n=mornfall@anna.fi.muni.cz] has joined #lisp 23:18:06 -!- mrSpec [n=NoOne@82.177.125.6] has quit [] 23:20:51 -!- nha [n=prefect@137-64.105-92.cust.bluewin.ch] has quit [Remote closed the connection] 23:21:11 -!- Soulman [n=kvirc@154.80-202-254.nextgentel.com] has quit [Read error: 60 (Operation timed out)] 23:21:32 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit [Read error: 60 (Operation timed out)] 23:21:32 -!- luis changed the topic of #lisp to: Common Lisp, the #1=(programmable . #1#) programming language . New: Osicat 0.6.0, CFFI 0.10.5, trivial-features 0.5, ABCL 0.15.0, SBCL 1.0.29.1, Hunchentoot & Drakma 1.0.0, usocket 0.4.1, cl-net-snmp 5.19, yason-0.1, series 2.2.10, CMUCL 19f. 23:21:51 -!- Lectus [n=Frederic@189.105.43.162] has quit [Read error: 60 (Operation timed out)] 23:22:12 Let the flow of patches begin: . 23:22:13 -!- jewel_ [n=jewel@dsl-242-129-65.telkomadsl.co.za] has quit [Read error: 60 (Operation timed out)] 23:26:37 -!- salex [n=user@216.80.147.206] has left #lisp 23:27:39 -!- I_Like_Turtles is now known as Michal_Jackson 23:27:52 -!- Michal_Jackson is now known as Michael_Jackson 23:29:53 nikodemus: I have a bugfix for signed zeros in float (especially complex) arithmetic. NEWS entry at the same time. Hopefully, I can also squeeze inline constants in before the freeze. 23:31:06 -!- sea4ever [i=sea@216.110.106.17] has quit [No route to host] 23:31:23 harleqin [n=harleqin@xdsl-81-173-180-122.netcologne.de] has joined #lisp 23:31:31 hello 23:32:13 -!- ruediger_ [n=ruediger@62-47-157-234.adsl.highway.telekom.at] has quit ["Leaving"] 23:32:52 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 23:34:30 hmm, I feel dumb for not sharing the forementioned roomba code here, I got a lot of help from here anyway: http://www.cs.helsinki.fi/u/strommer/roomba-seven.tar.gz 23:37:12 vsync [n=vsync@174-24.202-68.tampabay.res.rr.com] has joined #lisp 23:37:17 forgot who pasted me a very useful piece of code then, but I think it was mathrick :) 23:38:18 joga: can't recall, but if it was me, glad to have helped :) 23:38:46 I guess there were two pasters http://paste.lisp.org/display/51313 :) 23:39:05 thanks for those, they really helped 23:40:52 {df} [n=df@bspencer.plus.com] has joined #lisp 23:42:14 -!- |Soulman| [n=kvirc@154.80-202-254.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 23:42:24 -!- [df] [n=df@bspencer.plus.com] has quit [Read error: 113 (No route to host)] 23:43:28 pkhuong: you mean unboxed constants in code-objects? 23:45:17 -!- cp2 [n=will@unaffiliated/cp2] has quit [Remote closed the connection] 23:45:52 -!- {df} is now known as [df] 23:47:37 cp2 [n=will@unaffiliated/cp2] has joined #lisp 23:49:02 plage [n=user@118.68.33.117] has joined #lisp 23:49:05 Good morning 23:49:40 keithr [n=user@ip68-13-249-183.ok.ok.cox.net] has joined #lisp 23:52:02 plage: I thought you were in EU? :D 23:52:07 lacedaemon [n=algidus@88-149-208-97.dynamic.ngi.it] has joined #lisp 23:52:08 good morning vietnam 23:52:29 hi slyrus_ 23:52:39 -!- fe[nl]ix [n=algidus@88-149-212-242.dynamic.ngi.it] has quit [Nick collision from services.] 23:52:41 -!- lacedaemon is now known as fe[nl]ix 23:52:57 -!- Jasko3 [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 23:54:01 -!- rdad [n=rdad@cpe-67-247-62-86.nyc.res.rr.com] has quit [] 23:59:15 rread_ [n=rread@192.18.41.196] has joined #lisp