00:00:56 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Remote closed the connection] 00:01:19 mgm [n=mgm@m465e36d0.tmodns.net] has joined #lisp 00:01:36 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 00:03:04 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 00:04:56 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit ["Ex-Chat"] 00:11:34 legumbre_ [n=user@r190-135-38-36.dialup.adsl.anteldata.net.uy] has joined #lisp 00:11:48 -!- dwh [n=dwh@ppp118-208-241-91.lns10.mel6.internode.on.net] has quit [Read error: 110 (Connection timed out)] 00:12:41 -!- mcspiff [n=user@DC7CB.WPA.Dal.Ca] has quit [Read error: 110 (Connection timed out)] 00:19:30 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit ["+++ killed by SIGSEGV +++"] 00:24:08 -!- noptys_ is now known as noptys 00:24:18 -!- Ginei_Morioka [i=irssi_lo@78.114.177.147] has quit [Nick collision from services.] 00:24:25 Ginei_Morioka [i=irssi_lo@78.114.164.235] has joined #lisp 00:25:14 -!- chris2 [n=chris@dslb-188-098-217-225.pools.arcor-ip.net] has quit ["Leaving"] 00:26:22 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 00:28:29 -!- legumbre [n=user@r190-135-47-226.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 00:28:44 ignotus [n=ignotus@unaffiliated/ignotus] has joined #lisp 00:31:48 -!- Nshag [i=user@Mix-Orleans-106-2-4.w193-248.abo.wanadoo.fr] has quit [Remote closed the connection] 00:32:33 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 00:33:47 the_unmaker [n=g@w005.z209031033.sjc-ca.dsl.cnc.net] has joined #lisp 00:34:05 any cl-weblocks users in here? 00:34:17 the_unmaker: i was using it for a while 00:35:43 -!- hawkbill [n=pradyus@218.211.38.67] has quit [Remote closed the connection] 00:38:17 the_unmaker: what's up? 00:38:36 it's gavino, isn't it? 00:39:44 I don't know, but would like to know if that is the case 00:40:45 syamajala [n=syamajal@97-95-190-124.dhcp.oxfr.ma.charter.com] has joined #lisp 00:41:40 -!- mmondor [n=mmondor@ip216-239-92-102.vif.net] has quit ["bbl"] 00:42:35 okay n/m 00:44:38 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 00:48:05 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 00:48:24 -!- mrsolo [n=mrsolo@nat/yahoo/x-jybiubcxlfmmgrsz] has quit ["Leaving"] 00:48:27 -!- fiveop [n=fiveop@g229143083.adsl.alicedsl.de] has quit ["humhum"] 00:50:26 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 00:51:46 eno__ [n=eno@adsl-70-137-148-131.dsl.snfc21.sbcglobal.net] has joined #lisp 00:52:48 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 104 (Connection reset by peer)] 01:02:08 -!- saikat [n=saikat@c-76-21-43-94.hsd1.ca.comcast.net] has quit [] 01:02:21 QinGW [n=wangqing@203.86.81.2] has joined #lisp 01:04:34 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 01:05:55 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:05:57 newerspeak [n=newerspe@cpe-075-183-040-251.triad.res.rr.com] has joined #lisp 01:06:52 -!- spicey [n=e@87.99.83.235] has quit [Read error: 104 (Connection reset by peer)] 01:07:28 mcspiff [n=user@hlfxns0163w-142068081123.pppoe-dynamic.ns.aliant.net] has joined #lisp 01:11:37 bombshelter13_ [n=bombshel@76-10-149-209.dsl.teksavvy.com] has joined #lisp 01:13:07 gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has joined #lisp 01:13:40 -!- slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] 01:14:04 dwh [n=dwh@eth2.vic.adsl.internode.on.net] has joined #lisp 01:14:56 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:15:12 prxq_ [n=mommer@e179057009.adsl.alicedsl.de] has joined #lisp 01:17:20 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:17:37 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:17:42 -!- the_unmaker [n=g@w005.z209031033.sjc-ca.dsl.cnc.net] has quit [Remote closed the connection] 01:18:02 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:18:50 alec [n=alec@pool-71-174-175-64.bstnma.east.verizon.net] has joined #lisp 01:18:50 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:20:33 seisatsu [n=seisatsu@153.18.26.128] has joined #lisp 01:21:14 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:21:57 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:22:26 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:22:55 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:23:16 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:24:04 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:24:41 saikat [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has joined #lisp 01:26:06 -!- malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has quit [] 01:26:28 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:27:11 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:27:40 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:27:48 -!- prxq [n=mommer@e179217247.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 01:28:12 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:28:29 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:29:25 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:30:13 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:30:20 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:31:02 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:31:28 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:32:09 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:32:29 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:33:16 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:33:34 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 01:33:34 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:34:12 envi^home [n=envi@220.121.234.156] has joined #lisp 01:34:17 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:34:43 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:35:14 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:35:42 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:36:23 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:36:45 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:36:55 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 01:37:30 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:37:49 saikat_ [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has joined #lisp 01:37:49 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:40:34 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 01:41:06 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 104 (Connection reset by peer)] 01:45:11 -!- saikat [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has quit [Read error: 113 (No route to host)] 01:45:21 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit [Read error: 110 (Connection timed out)] 01:46:52 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 01:48:13 -!- syamajala [n=syamajal@97-95-190-124.dhcp.oxfr.ma.charter.com] has quit ["Leaving..."] 01:48:38 -!- seisatsu [n=seisatsu@153.18.26.128] has quit ["Ex-Chat"] 01:49:22 dialtone [n=dialtone@c-69-181-127-44.hsd1.ca.comcast.net] has joined #lisp 01:53:07 hawkbill [n=pradyus@nat/yahoo/x-rvuarkucvryshdtc] has joined #lisp 01:56:08 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 02:02:05 benny` [n=benny@i577A1775.versanet.de] has joined #lisp 02:02:10 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [] 02:02:36 -!- benny [n=benny@i577A1775.versanet.de] has quit [Nick collision from services.] 02:02:38 -!- benny` is now known as benny 02:03:52 -!- nyef [n=nyef@pool-64-222-164-136.man.east.myfairpoint.net] has quit ["G'night all."] 02:06:04 pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has joined #lisp 02:06:25 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 60 (Operation timed out)] 02:07:00 -!- hawkbill [n=pradyus@nat/yahoo/x-rvuarkucvryshdtc] has quit [Remote closed the connection] 02:07:42 Is there a standard/portable function or special operator that indicates if an object is stored atomically in the reference? Like a fixnum or a char. . . 02:08:06 atom 02:08:13 -!- fe[nl]ix [n=algidus@88-149-208-84.dynamic.ngi.it] has quit [Read error: 104 (Connection reset by peer)] 02:08:14 hey piso 02:08:20 slava!! 02:08:24 Modius: no. You could get a hint from EQ, that's all. 02:08:26 piso: atom is not it 02:08:27 Atom returns true on bignums too though (at least on lispworks) 02:08:29 fe[nl]ix [n=algidus@88-149-208-84.dynamic.ngi.it] has joined #lisp 02:08:38 piso: symbols answer t to atom but thye're not immediate in most implementations 02:08:46 Modius: its implementation-specific 02:08:56 Modius: do you mean that the object is stored literaly rather than as a memory reference? 02:09:03 Adlai : Yeah 02:09:06 oh 02:09:08 I was once tempted to make EQ on floats always return nil. 02:09:19 pkhuong: did you see my blog post? 02:09:20 yeah, as slava and others said, that's completely implementation-specific 02:09:21 hawkbill [n=pradyus@nat/yahoo/x-cinxfliccbcuowef] has joined #lisp 02:09:21 Adlai: Acknowledging that it's not safe to just assume certain types are and aren't 02:09:24 *piso* misunderstood the question 02:09:24 slava: yes, thanks. 02:09:40 Adlai: I get that it is, was just hoping for some function that was not. 02:10:46 Modius: indeed, since specialized arrays may do it for a different set of types than the rest of the system. 02:11:02 pkhuong: that would be cute. But not nice to code which is using EQ as 'cheap and useful if it works' 02:11:31 e.e. (defun florb (a b) (if (eq a b) t) ...)) 02:11:48 er, e.g. (defun florb (a b) (if (eq a b) t ...))) 02:12:11 -!- xan [n=xan@cs78225040.pp.htv.fi] has quit [Remote closed the connection] 02:12:22 xan [n=xan@cs78225040.pp.htv.fi] has joined #lisp 02:13:10 kpreid: I was more thinking of constant-folding EQ when one of the argument was known to be a float. A note to point out that EQ boxes its arguments would be more useful. 02:13:56 ah...yes. 02:14:29 pkhuong: clever 02:15:04 -!- schmx [n=marcus@sxemacs/devel/schme] has quit [Remote closed the connection] 02:15:07 schmx [n=marcus@c83-249-82-26.bredband.comhem.se] has joined #lisp 02:15:18 -!- fe[nl]ix [n=algidus@88-149-208-84.dynamic.ngi.it] has quit ["Valete!"] 02:16:23 -!- hawkbill [n=pradyus@nat/yahoo/x-cinxfliccbcuowef] has quit ["Leaving."] 02:16:33 pkhuong: I like that idea. factor inlines = as an eq? followed by a call to a generic, and if the inputs are floats the generic is inlined but as you said the eq? boxes 02:16:56 hawkbill [n=pradyus@nat/yahoo/x-ypzpwroyteddwntz] has joined #lisp 02:17:01 -!- dreish [n=dreish@minus.dreish.org] has quit [] 02:18:44 fe[nl]ix [n=algidus@88-149-208-66.dynamic.ngi.it] has joined #lisp 02:19:40 slava: sounds like you want a specialised function that is a lossy mix of eq? and eql?, without false positives. 02:20:22 I have three equality predicates, eq? = and number= 02:20:27 number= returns true if given, for instance, 3.0 and 3 02:20:39 = checks for type and value equality, and can be extended by the user 02:23:23 -!- saikat_ [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has quit [] 02:23:59 wakeup [n=wakeup@koln-5d81596d.pool.einsundeins.de] has joined #lisp 02:27:34 saikat [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has joined #lisp 02:28:22 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 02:30:56 cp2_ [n=will@please.dont.make.me.eatddos.info] has joined #lisp 02:31:54 -!- cp2 [n=will@please.dont.make.me.eatddos.info] has quit [Read error: 104 (Connection reset by peer)] 02:35:46 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 02:35:58 -!- cp2_ [n=will@please.dont.make.me.eatddos.info] has quit [Read error: 104 (Connection reset by peer)] 02:36:09 -!- wakeup_vanuber [n=wakeup@koln-5d817b0b.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 02:36:21 cp2 [n=will@please.dont.make.me.eatddos.info] has joined #lisp 02:37:55 you could write first clause as sumOf([], [], []). 02:40:17 rvirding: do you mean #prolog? 02:41:47 kpreid: yep, sorry 02:42:34 (woohoo, correct guess) 02:43:53 SandGorgon [n=OmNomNom@122.161.46.116] has joined #lisp 02:43:55 -!- hawkbill [n=pradyus@nat/yahoo/x-ypzpwroyteddwntz] has quit [Read error: 110 (Connection timed out)] 02:44:17 you will have to find a prolog interpreter in lisp and test it :-) 02:44:37 -!- hyperboreean [n=none@unaffiliated/hyperboreean] has quit [Read error: 110 (Connection timed out)] 02:44:57 -!- SandGorgon [n=OmNomNom@122.161.46.116] has quit [SendQ exceeded] 02:50:00 girzel [n=user@61.51.238.144] has joined #lisp 02:53:29 *rvirding* says goodnight 02:53:34 -!- rvirding [n=chatzill@h129n5c1o1034.bredband.skanova.com] has left #lisp 02:55:10 -!- skeptomai [n=cb@67.40.185.246] has quit [Read error: 110 (Connection timed out)] 02:55:31 -!- Ginei_Morioka [i=irssi_lo@78.114.164.235] has quit [Nick collision from services.] 02:55:39 Ginei_Morioka [i=irssi_lo@78.114.174.169] has joined #lisp 02:56:09 -!- pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 02:58:22 pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has joined #lisp 02:58:23 -!- lat [n=lat@125.167.140.159] has quit [Read error: 60 (Operation timed out)] 02:58:41 -!- ignotus [n=ignotus@unaffiliated/ignotus] has quit ["Coyote finally caught me"] 03:00:41 -!- ikki [n=ikki@201.155.75.146] has quit ["Leaving"] 03:03:42 lat [n=lat@125.167.140.159] has joined #lisp 03:04:04 SandGorgon [n=OmNomNom@122.161.46.116] has joined #lisp 03:12:24 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [Read error: 104 (Connection reset by peer)] 03:12:38 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 03:14:35 -!- rouslan [n=Rouslan@unaffiliated/rouslan] has quit [Read error: 104 (Connection reset by peer)] 03:16:53 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [Client Quit] 03:17:44 minion: tell me about cl-cont 03:17:45 Adlai: direct your attention towards cl-cont: cl-cont is a delimited continuations library for Common Lisp. http://www.cliki.net/cl-cont 03:17:50 minion: thanks 03:17:50 np 03:23:44 -!- Soulman [n=kvirc@154.80-202-254.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 03:25:48 hawkbill [n=pradyus@nat/yahoo/x-hbtpmctymageluyn] has joined #lisp 03:27:19 -!- saikat [n=saikat@c-24-5-85-71.hsd1.ca.comcast.net] has quit [] 03:27:20 -!- alec [n=alec@pool-71-174-175-64.bstnma.east.verizon.net] has quit ["leaving"] 03:28:46 is sbcl.org down? 03:29:23 www.sbcl.org, I mean 03:29:23 <_3b> dns problems, sbcl.sf.net works 03:29:36 thankyou 03:31:05 Does SBCL modular arithmetic now recognize expt or mod as well as bit-ish operations, or do so indirectly? The manual only refers to 0.8.5. 03:31:24 -!- mgm [n=mgm@m465e36d0.tmodns.net] has quit ["Leaving"] 03:34:34 It does mod for powers of two, but expt? 03:34:59 hm. 03:35:57 well, modular exponentiation is what I wanted. which is not actually a component of the sort of modular arithmetic sbcl does anyway... 03:36:00 felideon [n=user@adsl-074-186-235-232.sip.bct.bellsouth.net] has joined #lisp 03:36:04 *kpreid* grabs an algorithm to do it explicitly 03:37:38 Xof considered writing a transform to do that for his presentation. 03:37:43 -!- pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 03:38:14 *hefner* had a routine for that somewhere, but tended tended to dip into ironclad internals for that instead 03:38:43 pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has joined #lisp 03:38:48 -!- pstickne [n=pstickne@c-67-160-171-227.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 03:40:37 internals? it's exported. 03:40:50 and thanks, I already depend on ironclad... 03:41:56 it is? hrmph. 03:42:09 *kpreid* sees that ironclad uses the same algorithm WP provides but in good CL idiom 03:44:57 Modius: what sort of behavior differences might you have depending on the immediacy of objects? 03:45:21 S11001001: Weak-hash-table keys 03:46:11 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 03:53:43 -!- bombshelter13_ [n=bombshel@76-10-149-209.dsl.teksavvy.com] has quit [Client Quit] 03:57:35 saikat [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 03:58:57 cmeme [n=cmeme@boa.b9.com] has joined #lisp 03:59:15 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Read error: 60 (Operation timed out)] 04:02:49 Demosthenes [n=demo@206.180.154.148.adsl.hal-pc.org] has joined #lisp 04:04:06 -!- Yuuhi [i=benni@p5483D747.dip.t-dialin.net] has quit [Read error: 104 (Connection reset by peer)] 04:04:12 Yuuhi [i=benni@p5483D747.dip.t-dialin.net] has joined #lisp 04:05:56 Good morning. 04:06:08 morning 04:06:21 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 04:11:18 -!- felideon [n=user@adsl-074-186-235-232.sip.bct.bellsouth.net] has left #lisp 04:13:56 -!- SandGorgon [n=OmNomNom@122.161.46.116] has quit [Read error: 110 (Connection timed out)] 04:14:50 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 04:16:13 en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has joined #lisp 04:17:05 -!- newerspeak [n=newerspe@cpe-075-183-040-251.triad.res.rr.com] has quit [] 04:21:32 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 04:22:02 lookit that, cmeme is back. I don't see a working viewing site though 04:23:54 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit [Remote closed the connection] 04:30:37 pkhuong: about the eq? floats trick: I found a better way to do it 04:30:49 pkhuong: alias analysis can just check if the two inputs to eq? are in different alias classes 04:31:05 an eq? between a fresh allocation and a heap object will then always fold to f 04:31:10 not sure if this can be implemented in sbcl 04:37:22 -!- eno__ [n=eno@adsl-70-137-148-131.dsl.snfc21.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 04:37:27 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 04:39:59 -!- gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has quit [Remote closed the connection] 04:42:20 eno [n=eno@nslu2-linux/eno] has joined #lisp 04:45:37 -!- hawkbill [n=pradyus@nat/yahoo/x-hbtpmctymageluyn] has quit [Read error: 110 (Connection timed out)] 04:46:29 what terminal/shell/??? do you guys use in emacs? .. there's many; eshell, ansi-term, shell, term ... etc. 04:46:41 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 04:47:13 I use emacs in a terminal, not the other way around 04:47:40 gigamonkey [n=user@adsl-76-254-20-185.dsl.pltn13.sbcglobal.net] has joined #lisp 04:48:17 yeah, it's handy just hitting a single key and having a terminal open up in the same directory as the buffer i'm working on though 04:48:22 -!- pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has quit [] 04:48:33 pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has joined #lisp 04:50:21 hawkbill [n=pradyus@nat/yahoo/x-rjjzkcrcxzavnwhr] has joined #lisp 04:50:51 lexa_ [n=lexa_@seonet.ru] has joined #lisp 04:50:56 -!- lexa_ [n=lexa_@seonet.ru] has left #lisp 04:57:19 drafael [n=tapio@ip-118-90-133-134.xdsl.xnet.co.nz] has joined #lisp 04:58:22 -!- drafael [n=tapio@ip-118-90-133-134.xdsl.xnet.co.nz] has quit [Client Quit] 04:58:33 drafael [n=tapio@ip-118-90-133-134.xdsl.xnet.co.nz] has joined #lisp 05:03:44 -!- pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has quit [] 05:03:53 pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has joined #lisp 05:05:16 -!- beach [n=user@ABordeaux-158-1-5-93.w90-50.abo.wanadoo.fr] has quit [Read error: 104 (Connection reset by peer)] 05:08:25 lnostdal: I use bash within emacs when I need a shell, bash outside as well. I most often use it in an xterm, occasionally in rxvt term. Rarely, i use term in emacs, mostly if I wanto to check 'top'. 05:09:54 -!- hawkbill [n=pradyus@nat/yahoo/x-rjjzkcrcxzavnwhr] has quit [Read error: 110 (Connection timed out)] 05:10:42 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 05:11:01 ok, wgl .. ansi-term seems to support a lot of features .. ncurses etc. .. yeah; it's just for quick testing and messing around 05:14:07 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:14:33 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 05:22:22 beach [n=user@ABordeaux-158-1-5-93.w90-50.abo.wanadoo.fr] has joined #lisp 05:27:23 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit [Read error: 104 (Connection reset by peer)] 05:39:51 -!- nanobit [i=edargisr@gw.pulsar-zone.net] has quit ["bye"] 05:40:34 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 05:42:16 -!- gigamonkey [n=user@adsl-76-254-20-185.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 05:45:24 mgm [n=mgm@p84-72.acedsl.com] has joined #lisp 05:47:41 Cynner [n=plinka@87.110.231.67] has joined #lisp 05:48:06 -!- Cynner [n=plinka@87.110.231.67] has quit [Excess Flood] 05:55:05 -!- lat [n=lat@125.167.140.159] has quit ["Ex-Chat"] 05:55:23 lat [n=lat@125.167.140.159] has joined #lisp 05:55:42 -!- pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has quit [Connection timed out] 06:00:35 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 06:01:43 nanobit [i=nauin@gw.pulsar-zone.net] has joined #lisp 06:04:18 -!- beaumonta [n=abeaumon@84.76.48.250] has quit [Read error: 110 (Connection timed out)] 06:04:49 slyrus_ [n=slyrus@adsl-75-36-210-102.dsl.pltn13.sbcglobal.net] has joined #lisp 06:07:03 evening 06:07:52 so are we in a freeze yet? 06:09:59 no, the weather here is rather clement in fact 06:10:29 I'd say freeze, looking out the window. 06:11:19 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 06:12:13 hello slyrus_ 06:12:59 hey beach. enjoying a nice carignane from montpeyroux tonight. 06:14:19 slyrus_: Sounds good! 06:16:32 is that x silent at the end of montpeyroux? 06:16:52 slyrus_: very likely, yes. 06:17:08 and how about the S at the end of cornas? 06:17:42 slyrus_: I would pronounce it, but then I live in the south-west where more final consonants are pronounced. 06:19:12 ok, thanks! 06:19:39 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [Read error: 60 (Operation timed out)] 06:19:47 here we pronounce the final s in "plus" and "moins". 06:19:52 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 06:21:37 -!- sreeram [n=sreeram@122.174.64.194] has quit [] 06:22:20 Are there rules for that, beach? Can never figure out when to pronounce the last part s/se 06:24:06 tic: I am sure there are rules, but they are regional. 06:24:34 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 06:24:46 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit ["Ex-Chat"] 06:25:21 beach: surely theydon't pronounce the x in bordeaux? 06:25:24 -!- prxq_ [n=mommer@e179057009.adsl.alicedsl.de] has quit ["Leaving"] 06:25:25 pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has joined #lisp 06:25:33 No, they don't. 06:25:54 But I have heard the x in Chamonix pronounced. 06:26:47 Interesting. 06:27:36 beach: as an english "x" or as a "z"? 06:27:49 "x" 06:28:39 3000 meters vertical, call it whatever you want, I'm dying to go back! 06:29:00 What's stopping you? 06:29:24 Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has joined #lisp 06:30:05 time, other stuff to do. it will happen one of these days... 06:30:45 Oh, and the x in Aix is pronounced. 06:32:28 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 06:35:10 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 06:38:34 beaumonta [n=abeaumon@85.48.202.13] has joined #lisp 06:40:08 -!- cracki [n=cracki@134.130.183.101] has quit ["If technology is distinguishable from magic, it is insufficiently advanced."] 06:43:37 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 06:45:02 I'm having trouble with some CL-CONT code 06:45:30 I'm just trying to find my way around, and this snippet is causing an error 06:45:36 pasting... 06:47:33 lat pasted ".sbclrc" at http://paste.lisp.org/display/84489 06:48:04 Adlai pasted "cl-cont error" at http://paste.lisp.org/display/84490 06:49:46 Would some mcclim expert please check the above paste from lat? 06:52:15 lat: You woul have to do (asdf:operate 'asdf:load-op :mcclim) 06:53:08 anybody familiar with cl-cont? 06:53:23 lat: I build a core with McCLIM (and some other goodies in it) which I use when I start SLIME. 06:53:46 mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has joined #lisp 06:55:28 good morning 06:55:32 hello mvilleneuve 06:55:37 mvilleneuve: familiar with cl-cont? 06:56:54 Adlai: How about looking at the macro expansion? 06:59:24 tcr: I did, and I think I know what the problem is, but I can't tell whether it's because I'm using cl-cont wrong, or whether it's a bug. 06:59:32 I'll annotate the macroexpansion... 07:00:29 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 07:00:49 Adlai annotated #84490 "macroexpansion from hell" at http://paste.lisp.org/display/84490#1 07:01:11 I marked a line there that I think is the problem 07:02:23 beach, is there a web page that shows how to build a core like that? 07:02:26 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 07:03:26 great, tcr's computer exploded when he saw that macroexpansion. 07:03:31 lat: #+sbcl sb-ext:save-lisp-and-die #+clisp ext:saveinitmem 07:04:36 lat: Without SLIME, load what you need, then (save-lisp-and-die "clim.core") or something like that. 07:05:13 -!- saikat [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 07:06:24 Adlai, don't return 07:06:53 just: ... :do (cont:let/cc k (setf *cont* k) i))) 07:07:11 0, (funcall *cont*) => 1, (funcall *cont*) => 2, ... 07:07:14 lnostdal: I'm trying to return... the (very scant) docs of cl-cont say that return-from is supported 07:07:19 hmmm 07:07:43 how does that work? 07:08:12 won't the value i just get returned into the loop, thrown away, and the next loop iteration begins? 07:08:12 *shrug* 07:08:31 Foofie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 07:08:36 pgb and beach , thanks. I'll play around with this a little bit. 07:08:57 that's not good, there's black magic going on where it should be transparent. that's not the right behavior anyways -- that should loop infinitely. 07:09:03 Adlai: no, what is cl-cont? 07:09:44 Adlai, well, (progn before (setf *cont* k) after) .. when calling *cont*, it'll continue after the assignment there 07:10:09 mvilleneuve: evidently, a very confusing library that mangles cl. 07:10:32 prg_ [n=prg@ns.alusht.net] has joined #lisp 07:10:33 -!- fe[nl]ix [n=algidus@88-149-208-66.dynamic.ngi.it] has quit [Read error: 104 (Connection reset by peer)] 07:10:53 Adlai, that's the general idea .. i'm not sure how the context (loop) works or is interpreted; it does feel weird to use loop for this; but i dunno 07:11:20 lnostdal: yes, I know that much, but that should be an infinite loop 07:11:47 it makes -no sense- that it's returning automatically when it hits a call/cc 07:11:55 that's not how continuations work! 07:12:45 oh .. i think of them as yield in python; which also returns 07:13:45 but they shouldn't do that! uch. 07:13:59 don't use continuations .. they suck anyway .. :} 07:14:22 -!- 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.""] 07:14:40 (..or well, they might be useful in a language that supports them from the "ground up", but even there they seem to make somewhat of a mess..) 07:15:01 seriously though; for i/o there is no need for them .. it's just stupid 07:15:11 I'm not trying to use them for i/o 07:15:36 masm [n=masm@213.22.191.93] has joined #lisp 07:15:36 this is just a "toy example" 07:15:51 what else would one use them for? .. debugging is hard enough as it is .. *sigh* 07:15:54 but it's not behaving the way CL, or continuations in any language, should. 07:16:33 well, atm, I'm trying to implement Generators in CL 07:17:43 ok 07:19:16 call/cc shouldn't have the continuation immediately be returned from with-call/cc! I have no idea why it would do that. 07:20:52 -!- slyrus_ [n=slyrus@adsl-75-36-210-102.dsl.pltn13.sbcglobal.net] has quit [Read error: 60 (Operation timed out)] 07:22:06 hyperboreean [n=none@unaffiliated/hyperboreean] has joined #lisp 07:25:26 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 07:35:15 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 07:36:30 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 07:36:43 yah, stuff seems missing .. 07:38:24 lnostdal: any idea how to patch this up? 07:38:41 no, i can't even FFI with JavaScript proper .. and that's just _text_ 07:40:25 orwell .. compile Lisp to JS 07:40:59 hm? what does FFI and JS have to do with continuations? 07:41:24 you're basically compiling Lisp to something else, no? 07:41:34 (the millions of lambdas) 07:44:37 cmo-0 [n=user@92.96.20.45] has joined #lisp 07:44:55 lnostdal: yeah, I guess you could say that. But it seems that the compiler is broken. 07:45:26 or, if not broken, than badly documented with a vengeance, to the point where the documentation claims things that it can't do. 07:47:30 drewc: http://www.ccs.neu.edu/home/dorai/uwcallcc/uwcallcc.html can we have call/cc in cltl3 now pls? 07:47:36 :P 07:49:12 nha [n=prefect@137-64.105-92.cust.bluewin.ch] has joined #lisp 07:50:11 -!- _Trickster_ [i=Trickste@77.232.135.67] has quit [Read error: 104 (Connection reset by peer)] 07:50:19 asdfgh [n=asdf@d173-183-77-185.bchsia.telus.net] has joined #lisp 07:50:24 -!- asdfgh [n=asdf@d173-183-77-185.bchsia.telus.net] has quit [Remote closed the connection] 07:50:27 _Trickster_ [i=Trickste@77.232.135.67] has joined #lisp 07:53:00 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 07:59:48 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 08:01:33 Phaze [n=PhazeDK@0x5da32b16.cpe.ge-0-1-0-1104.soebnqu1.customer.tele.dk] has joined #lisp 08:04:29 Fufie [n=poff@Gatekeeper.vizrt.com] has joined #lisp 08:04:44 hm 08:05:07 minion: memo for drewc: http://www.ccs.neu.edu/home/dorai/uwcallcc/uwcallcc.html can we have call/cc in cltl3 now? ;) 08:05:08 Remembered. I'll tell drewc when he/she/it next speaks. 08:05:31 -!- dwh [n=dwh@eth2.vic.adsl.internode.on.net] has quit [Read error: 113 (No route to host)] 08:05:33 question: is there any way to make it so slime doesn't open up every single file that warned on load, when doin ,load-system? 08:05:39 sykopomp: you're so persistent, you don't even need a database. 08:05:43 -!- gh7d395pi69wd [n=asdf@unaffiliated/gh7d395pi69wd] has quit [Read error: 110 (Connection timed out)] 08:05:51 Adlai: zing 08:06:26 -!- lichtblau [n=user@port-212-202-18-15.dynamic.qsc.de] has quit [Read error: 60 (Operation timed out)] 08:06:34 ejs [n=eugen@77.222.151.102] has joined #lisp 08:06:36 easy; don't use ,load-system 08:07:18 gh7d395pi69wd [n=asdf@unaffiliated/gh7d395pi69wd] has joined #lisp 08:07:27 hefner: but the alternative is (asdf:oos 'asdf:load-op 'foo) (gag) 08:07:34 :( 08:07:40 I just use require 08:08:05 I use require on sbcl, too, but it's not the -only- implementation I use :| 08:08:19 although for mcclim, I have a shortcut function (gogogo) defined in my .sbclrc to load the various systems I usually want 08:08:25 well, that's your first mistake 08:09:06 hefner: maybe when sbcl is actually perfect, I'll use it for everything :) 08:09:29 you don't have to use it for everything, just often enough that you don't care about minor inconveniences when you don't 08:09:47 *hefner* has no useful answer, as you no doubt already gathered 08:10:12 hefner: are you familiar with cl-cont? 08:10:23 Adlai: that's not random at all. 08:10:25 Adlai: no, not at all. 08:10:27 hefner: ;D 08:10:28 -!- cmm- [n=cmm@bzq-79-179-66-150.red.bezeqint.net] has quit [Read error: 60 (Operation timed out)] 08:10:31 *Adlai* is trolling out his newfound hatred for said library. 08:10:56 I'll join you. Hating on things i know nothing about is one of my specialties. 08:11:16 it doesn't do call/cc properly! 08:11:43 it's one thing to not provide "real" continuations, and that's to be expected. But at least you could get call/cc right! 08:11:54 cmm [n=cmm@bzq-79-179-66-150.red.bezeqint.net] has joined #lisp 08:11:59 besides, there's an obvious two strikes against it: 1) it's named with the "cl-" prefix, and 2) it purports to do continuations, which everyone knows are for weenies and schemers 08:12:16 hefner: we -are- weenies, though :| 08:12:23 (and other folks who use languages where they are built in and can be reasonably expected to work) 08:12:24 -!- girzel [n=user@61.51.238.144] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 08:12:25 that's why we use lisp, and why we're so smug :) 08:13:52 -!- Foofie [n=innocent@86.80-203-225.nextgentel.com] has quit [Read error: 104 (Connection reset by peer)] 08:14:04 mcspiff` [n=user@hlfxns0163w-142068081123.pppoe-dynamic.ns.aliant.net] has joined #lisp 08:14:13 Foofie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 08:15:22 -!- mcspiff [n=user@hlfxns0163w-142068081123.pppoe-dynamic.ns.aliant.net] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- QinGW [n=wangqing@203.86.81.2] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- anekos [n=anekos@pl716.nas923.p-osaka.nttpc.ne.jp] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- ia [n=ia@89.169.189.230] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- pragma_ [n=pragma@unaffiliated/pragma/x-109842] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- rey_ [n=ikke@igwe19.vub.ac.be] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- Lucia [n=Lucia@148.Red-79-148-241.staticIP.rima-tde.net] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- lujz [n=lujz@cpe-92-37-17-30.dynamic.amis.net] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- mikezor [n=mikael@c-4fe070d5.04-404-7570701.cust.bredbandsbolaget.se] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- tessier__ [n=treed@mail.copilotco.com] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- xvx [n=user01@189.139.218.85] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- retupmoca [n=retupmoc@ppp-69-221-252-243.dsl.klmzmi.ameritech.net] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- antifuchs [n=asf@baker.boinkor.net] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- enn [n=eli@codeanddata.com] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- hdurer_ [n=hdurer@nat/yahoo/x-b883f150da830fd3] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- zbigniew [n=zb@3e8.org] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- Orest [n=Orest@hates-script-kiddies.ath.cx] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- bohanlon [n=bohanlon@TUBERIUM.CLUB.CC.CMU.EDU] has quit [simmons.freenode.net irc.freenode.net] 08:15:22 -!- seejay [n=seejay@unaffiliated/seejay] has quit [simmons.freenode.net irc.freenode.net] 08:15:37 hefner: well-implemented continuations that admit what they can't do can be expected to work within those limits 08:15:57 I'd be fine with CL continuations that couldn't break an unwind-protect, for example. 08:16:06 but cl-cont is messing up. 08:16:11 -!- xan is now known as xan-afk 08:16:14 Orest^bnc [n=Orest@hates-script-kiddies.ath.cx] has joined #lisp 08:16:29 girzel [n=user@61.51.238.144] has joined #lisp 08:16:42 enn [n=eli@codeanddata.com] has joined #lisp 08:16:44 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 08:16:48 fair enough. continuations would make some very interesting and completely hypothetical problems of mine much simpler. 08:17:10 but faking them in CL makes me nervous enough that I never seriously considered trying it. 08:17:49 amaron [n=amaron@cable-94-189-243-158.dynamic.sbb.rs] has joined #lisp 08:17:57 ejs1 [n=eugen@nat.ironport.com] has joined #lisp 08:19:23 to start out; i think you basically got to implement CL entirely, in CL .. then you got to figure out ways to not mess up (unwind-protect, debugging ...) and avoid adding a ton of overhead in cases where you do not use continuations .. random guess; it'll require massive amounts of effort .. x) 08:19:53 ..but i really don't know 08:19:58 lnostdal: I didn't say the continuations need to support -everything 08:20:51 lnostdal: see link for some stuff about mixing unwind-protect and call/cc safely 08:21:01 I'd be fine with a continuation library/hack that says, "Don't use unwind-protect, catch, throw, X, Y, or Z within the scope of a with-call/cc, or else bad juju will happen three times." as long as it actually works for everything else. 08:21:10 Adlai, i'm in the process of improving cl-cont -- can you explain where it's wrong? 08:21:46 ilitirit_: yes! finally somebody who knows the codebase. call/cc shouldn't return the continuation from with-call/cc 08:21:49 QinGW [n=wangqing@203.86.81.2] has joined #lisp 08:21:49 anekos [n=anekos@pl716.nas923.p-osaka.nttpc.ne.jp] has joined #lisp 08:21:49 ia [n=ia@89.169.189.230] has joined #lisp 08:21:49 pragma_ [n=pragma@unaffiliated/pragma/x-109842] has joined #lisp 08:21:49 antifuchs [n=asf@baker.boinkor.net] has joined #lisp 08:21:49 Lucia [n=Lucia@148.Red-79-148-241.staticIP.rima-tde.net] has joined #lisp 08:21:49 lujz [n=lujz@cpe-92-37-17-30.dynamic.amis.net] has joined #lisp 08:21:49 mikezor [n=mikael@c-4fe070d5.04-404-7570701.cust.bredbandsbolaget.se] has joined #lisp 08:21:49 tessier__ [n=treed@mail.copilotco.com] has joined #lisp 08:21:49 xvx [n=user01@189.139.218.85] has joined #lisp 08:21:49 retupmoca [n=retupmoc@ppp-69-221-252-243.dsl.klmzmi.ameritech.net] has joined #lisp 08:21:49 hdurer_ [n=hdurer@nat/yahoo/x-b883f150da830fd3] has joined #lisp 08:21:49 zbigniew [n=zb@3e8.org] has joined #lisp 08:21:49 bohanlon [n=bohanlon@TUBERIUM.CLUB.CC.CMU.EDU] has joined #lisp 08:21:49 seejay [n=seejay@unaffiliated/seejay] has joined #lisp 08:21:55 rey_ [n=ikke@igwe19.vub.ac.be] has joined #lisp 08:22:02 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Remote closed the connection] 08:22:40 Adlai pasted "this should be an infinite loop" at http://paste.lisp.org/display/84495 08:23:09 ilitirit_: that should NOT return the continuation, it should just loop infinitely. 08:23:10 redwyrm [n=nil@pool-96-249-196-59.lsanca.dsl-w.verizon.net] has joined #lisp 08:23:47 if you want it to loop finitely you need to funcall the k inside the lambda 08:23:56 s/finitely/infinitely/ 08:23:57 what does it mean if I try to start SLIME in Emacs with M-x slime, and an inferior-lisp is started but no REPL? 08:24:14 Adlai annotated #84495 "this should work fine" at http://paste.lisp.org/display/84495#1 08:24:42 ilitirit_: no, that's not how continuations work. 08:25:02 if that's how cl-cont is going to implement them... then it's not call/cc anymore, it's return/cc. 08:25:18 well it's how cl-cont works . . . is it different in scheme? 08:25:27 YES. 08:25:39 -!- ejs [n=eugen@77.222.151.102] has quit [Read error: 110 (Connection timed out)] 08:25:40 call/cc should just call the lambda, with the current continuation bound to the arg. 08:25:52 jdz [n=jdz@85.254.211.133] has joined #lisp 08:26:27 obviously, because we're in CL and we're using with-call/cc, the continuation is only captured between the call/cc and the with-call/cc, but that doesn't mean that call/cc should automatically bounce the continuation back as the return value of with-call/cc. 08:26:46 is the cl-cont:let/cc similarly different from scheme? 08:27:02 call/cc should just call the function, using the current continuation as the argument to the function, and return the return value of that function. 08:27:06 ilitirit_: yes. 08:27:48 redwyrm: you need to make slime load its repl (it's not loaded by default anymore) 08:28:09 what command do I use to do this? 08:28:10 redwyrm: add (slime-setup '(slime-fancy)) to your .emacs. 08:28:21 Adlai, i understand your point of view and being similar to scheme would be reasonable -- i wonder about changing it now though 08:28:27 ahh, how weird.... the old default was nicer 08:28:52 ilitirit_: maybe a fork then? it's quite misleading to have a call/cc which does something other than call/cc 08:29:39 tbh, my hacking on cl-cont is going to tear the guts out of it so i might start on a new library instead 08:29:52 redwyrm: *shrug* I don't remember why they made the change. Probably some plan to make lisp even harder to get into, I bet. 08:30:00 ilitirit_: I would definitely be willing to help with that. 08:30:00 sykopomp, that worked, thanks 08:30:06 redwyrm: no prob! :) 08:30:20 cracki [n=cracki@40-123.eduroam.RWTH-Aachen.DE] has joined #lisp 08:30:28 sykopomp, my mission was to get slime installed and running... with my mission accomplished, it's bed time :) 08:30:29 maybe not contribute a huge portion, but I think we should have a decent, working CPS-walker. 08:30:31 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 08:30:43 lnostdal annotated #84495 "cl-cont vs. scheme" at http://paste.lisp.org/display/84495#2 08:30:56 -!- nha [n=prefect@137-64.105-92.cust.bluewin.ch] has quit [Remote closed the connection] 08:30:59 Adlai: have you even looked at arnesi's cps-transformer? 08:31:15 Adlai, my goal is (1) more correct semantics for special forms and lexical bindings, (2) to avoid doing transforms on code that doesn't use call/cc 08:31:29 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 08:32:14 sykopomp, the arnesi cps-transformer had a horribly incorrect implementation of the semantics, was terribly slow, and buggy too (in my experience with it) 08:32:19 sykopomp: I couldn't find it in the documentation, and I haven't had time to "use the source" 08:32:22 ilitirit_: oic 08:32:33 Adlai: I just linked you to the docs :P 08:32:38 it was a few years ago though so maybe much better now 08:33:13 rdd [n=user@c83-250-157-93.bredband.comhem.se] has joined #lisp 08:34:39 -!- redwyrm [n=nil@pool-96-249-196-59.lsanca.dsl-w.verizon.net] has left #lisp 08:34:55 I imagine that it wouldn't be too hard to fix this problem with cont:call/cc though. 08:35:28 -!- cracki [n=cracki@40-123.eduroam.RWTH-Aachen.DE] has quit ["If technology is distinguishable from magic, it is insufficiently advanced."] 08:35:28 -!- en0ch [n=user@24-180-44-180.dhcp.reno.nv.charter.com] has quit [Read error: 101 (Network is unreachable)] 08:35:28 cracki [n=cracki@40-123.eduroam.RWTH-Aachen.DE] has joined #lisp 08:36:35 maybe I just don't know the codebase well enough, but it seems as though you just need to change how the form returns 08:36:53 Adlai, you could easily make your own call/cc just doing a cl-cont:call/cc and a funcall k 08:37:13 however i think it would be rather unpopular to change the behaviour of the cl-cont:call/cc as a lot of people use it now 08:38:10 boing boing 08:38:22 but it's wrong! it's not doing call/cc, it's doing some cooked-up return/cc. 08:38:59 Adlai, yes it's wrong but a lot of code is using it (e.g. weblocks, teepeedee2) -- you would have to change all that code to deal with the changed semantics 08:39:16 so we fork the code... 08:39:21 we/ 08:39:32 we/I/you/anybody who feels like using a real call/cc 08:40:35 Adlai, make macros called cl-cont:scheme-call/cc and cl-cont:scheme-let/cc or just a package cl-cont-schemy and everybody is happy 08:40:39 timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has joined #lisp 08:44:46 lichtblau [n=user@77-22-106-191-dynip.superkabel.de] has joined #lisp 08:45:40 anyone knows whats up with www.sbcl.org and b9.com? 08:46:04 antoszka [n=antoszka@unaffiliated/antoszka] has joined #lisp 08:46:18 timor: what do you mean? 08:46:35 -!- xinming_ is now known as xinming 08:46:39 they are down since yesterday or so and i wonder why nobody complains 08:46:48 I'd like to know what's up with sbcl.org as opposed to www.sbcl.org 08:46:49 sbcl.org works just fine ? 08:47:06 oh thats right, thy fixed sbcl 08:47:19 i think i mean sbcl.org 08:47:24 works just fine for me ? 08:47:36 I don't have an A record (nor AAAA) for sbcl.org. 08:47:43 b9.com has important git repos though 08:47:43 So it doesn't work for me. 08:48:06 -!- pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has quit [] 08:48:07 Tracerouting s9.com and it all seems to get stuck in some "sprintlink.net" whatever that is. 08:48:38 $ host sbcl.org 08:48:39 sbcl.org mail is handled by 10 boa.b9.com. 08:48:44 Just the MX record. 08:48:46 ahaha 08:52:18 attila_lendvai [n=ati@catv-89-132-189-132.catv.broadband.hu] has joined #lisp 08:53:07 -!- schmx [n=marcus@sxemacs/devel/schme] has quit ["leaving"] 08:53:24 schme [n=marcus@c83-249-82-26.bredband.comhem.se] has joined #lisp 08:54:26 stoopid internet. 08:54:28 ilitirit_: also, cl-cont should allow you to call return-from within the body of the lambda. 08:55:40 Adlai, yes within the body of the lambda you are in untransformed code so you cannot return-from a transformed block 08:56:29 Yuuhi` [i=benni@p5483F41D.dip.t-dialin.net] has joined #lisp 08:56:31 emacs question: is there a way to kill all buffers that match a pattern? (can I just kill all buffers that end in .lisp?) 08:56:53 masm1 [n=masm@213.22.191.93] has joined #lisp 08:56:54 ilitirit_: it should be possible, though. It makes complete sense. 08:57:15 and yes, I understand why it's not possible in cl-cont right now... but that's a bug, imo, not a feature. 08:58:18 yes, it would be possible and i agree that it is unfortunate -- not sure that it's exactly a bug though 08:58:36 there are many problems like this because the host language doesn't really support call/cc 08:59:22 i guess you could fix it by transforming the lambda which is passed the call/cc 08:59:31 ilitirit_: yes, that should be done. 09:01:44 -!- amaron [n=amaron@cable-94-189-243-158.dynamic.sbb.rs] has quit [Read error: 110 (Connection timed out)] 09:01:51 -!- benny [n=benny@i577A1775.versanet.de] has quit [Connection timed out] 09:02:51 -!- timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has quit ["Leaving."] 09:02:57 timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has joined #lisp 09:05:33 -!- timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has quit [Client Quit] 09:05:37 timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has joined #lisp 09:06:28 dwh [n=dwh@ppp118-208-241-91.lns10.mel6.internode.on.net] has joined #lisp 09:09:27 -!- Yuuhi [i=benni@p5483D747.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 09:09:55 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [] 09:10:40 -!- masm [n=masm@213.22.191.93] has quit [Read error: 113 (No route to host)] 09:11:52 mkfort_ [n=mkfort@68-189-164-209.dhcp.spbg.sc.charter.com] has joined #lisp 09:11:58 benny [n=benny@i577A26CF.versanet.de] has joined #lisp 09:13:39 frito [n=user@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has joined #lisp 09:14:53 -!- mkfort [n=mkfort@68-189-164-209.dhcp.spbg.sc.charter.com] has quit [Read error: 60 (Operation timed out)] 09:22:13 jewel [n=jewel@dsl-242-128-103.telkomadsl.co.za] has joined #lisp 09:26:11 -!- Soulman__ [n=kae@Gatekeeper.vizrt.com] has quit [Read error: 104 (Connection reset by peer)] 09:26:13 Soulmann [n=kae@Gatekeeper.vizrt.com] has joined #lisp 09:27:27 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [] 09:29:04 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 09:37:17 lexa_ [n=lexa_@seonet.ru] has joined #lisp 09:37:45 -!- lexa_ is now known as Guest25863 09:39:35 manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 09:39:43 -!- manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has quit [Remote closed the connection] 09:44:19 -!- girzel [n=user@61.51.238.144] has quit [Read error: 110 (Connection timed out)] 09:45:26 serichse1 [n=harleqin@xdsl-81-173-191-239.netcologne.de] has joined #lisp 09:48:59 -!- mcspiff` [n=user@hlfxns0163w-142068081123.pppoe-dynamic.ns.aliant.net] has quit [Remote closed the connection] 09:56:35 -!- serichsen [n=harleqin@xdsl-81-173-157-146.netcologne.de] has quit [Read error: 110 (Connection timed out)] 09:57:12 fe[nl]ix [n=algidus@88-149-208-66.dynamic.ngi.it] has joined #lisp 10:00:52 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 10:01:43 morphling [n=stefan@gssn-590e8eb5.pool.einsundeins.de] has joined #lisp 10:01:43 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 104 (Connection reset by peer)] 10:03:45 -!- Axioplase is now known as Axioplase_ 10:04:01 davelambert [n=djl@monoblock.open.ac.uk] has joined #lisp 10:05:44 stassats [n=stassats@wikipedia/stassats] has joined #lisp 10:05:59 eno [n=eno@nslu2-linux/eno] has joined #lisp 10:07:38 -!- Buganini [n=buganini@security-hole.info] has quit [Read error: 110 (Connection timed out)] 10:08:14 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 10:10:13 alinp [n=alinp@86.122.9.2] has joined #lisp 10:10:27 gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has joined #lisp 10:11:05 -!- dmiles [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has quit [Read error: 60 (Operation timed out)] 10:12:49 drafael1 [n=tapio@ip-118-90-132-131.xdsl.xnet.co.nz] has joined #lisp 10:15:35 girzel [n=user@123.121.246.50] has joined #lisp 10:16:41 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 10:18:14 csaba [n=cgajo@193.189.163.110] has joined #lisp 10:21:43 nunb [n=user@static-217-133-104-225.clienti.tiscali.it] has joined #lisp 10:22:24 -!- QinGW [n=wangqing@203.86.81.2] has quit [Read error: 110 (Connection timed out)] 10:22:29 -!- ausente [n=user7994@187.34.44.222] has quit ["Nettalk6 - www.ntalk.de"] 10:28:06 -!- frito [n=user@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has quit [Read error: 110 (Connection timed out)] 10:28:59 hawkbill [n=pradyus@218.211.38.67] has joined #lisp 10:29:01 -!- drafael [n=tapio@ip-118-90-133-134.xdsl.xnet.co.nz] has quit [Read error: 110 (Connection timed out)] 10:31:18 ejs2 [n=eugen@77.222.151.102] has joined #lisp 10:31:44 Soulman [n=kvirc@154.80-202-254.nextgentel.com] has joined #lisp 10:36:38 nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has joined #lisp 10:37:41 good afternoon 10:38:51 hi nikodemus 10:40:36 -!- ejs2 [n=eugen@77.222.151.102] has quit [Read error: 60 (Operation timed out)] 10:40:58 ejs2 [n=eugen@nat.ironport.com] has joined #lisp 10:41:12 -!- ejs1 [n=eugen@nat.ironport.com] has quit [Read error: 110 (Connection timed out)] 10:47:38 -!- fe[nl]ix [n=algidus@88-149-208-66.dynamic.ngi.it] has quit ["Valete!"] 10:57:33 jan247 [n=jan247@222.127.86.98] has joined #lisp 11:00:09 saikat [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 11:02:34 mfk [n=salagaev@mail.star-force.com] has joined #lisp 11:07:49 e|f [n=charles@ppp121-44-84-192.lns10.syd6.internode.on.net] has joined #lisp 11:08:43 pkok [n=patrick@f102140.upc-f.chello.nl] has joined #lisp 11:11:38 spilman [n=spilman@ARennes-552-1-17-7.w92-135.abo.wanadoo.fr] has joined #lisp 11:14:08 c|mell [n=cmell@x250048.dynamic.ppp.asahi-net.or.jp] has joined #lisp 11:14:45 blandest [n=user@softhouse.is.ew.ro] has joined #lisp 11:22:33 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 11:28:55 malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has joined #lisp 11:31:03 -!- serichse1 is now known as serichsen 11:32:14 -!- e|f [n=charles@ppp121-44-84-192.lns10.syd6.internode.on.net] has left #lisp 11:36:07 -!- gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has quit [Read error: 113 (No route to host)] 11:40:52 -!- mfk [n=salagaev@mail.star-force.com] has left #lisp 11:42:40 -!- xan-afk is now known as xan 11:43:38 -!- ejs2 [n=eugen@nat.ironport.com] has quit [Read error: 110 (Connection timed out)] 11:44:45 -!- sykopomp [n=user@unaffiliated/sykopomp] has quit [Read error: 104 (Connection reset by peer)] 11:45:08 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit [Read error: 101 (Network is unreachable)] 11:50:03 -!- davelambert [n=djl@monoblock.open.ac.uk] has left #lisp 11:52:43 schoppenhauer_ [n=css@82.113.106.85] has joined #lisp 12:00:32 -!- c|mell [n=cmell@x250048.dynamic.ppp.asahi-net.or.jp] has quit [Connection timed out] 12:01:15 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 12:02:18 schoppenhauer__ [n=css@82.113.121.149] has joined #lisp 12:05:26 what is the relation of CL standard pretty printer to XP from "XP -- A pretty printer for Common Lisp"? 12:06:55 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 12:07:01 XP is an implementation wheras the CL standard pretty printer is a standard? 12:08:41 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 12:08:47 IIUC, XP was the portable pretty printing implementation that predated the standard, and was then included into CL around the time of CLtL2 12:10:41 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Connection timed out] 12:11:36 fiveop [n=fiveop@g229142147.adsl.alicedsl.de] has joined #lisp 12:14:20 kenjin_ [n=kenjin@163.152.180.247] has joined #lisp 12:14:47 -!- kenjin_ is now known as Guest76715 12:15:36 what is # in sbcl? I want to filter it, but SBCL says it's a number 12:16:57 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [Read error: 110 (Connection timed out)] 12:19:23 -!- schoppenhauer_ [n=css@82.113.106.85] has quit [Connection timed out] 12:19:59 -!- sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has quit [] 12:20:13 easyE: umm, that doesn't make sense 12:20:26 lichtblau: I thought so, but several control variables are missing from my CL 12:20:38 most importantly, *print-dispatch* 12:20:52 but I guess that was incorporated as PRINT-OBJECT? 12:21:23 Guest76715: because it is 12:21:35 NaN is a special numerical value that isn't a proper number 12:21:52 mathrick: alright. 12:22:11 easyE: I mean, XP could, or could not have gone into the standard 12:22:34 just because it's there, implemented, doesn't mean it doesn't specify an interface too 12:22:48 But the ANSI stanadard didn't have an implementation, right? 12:22:50 -!- Draggor [n=Draggor@216-80-120-145.alc-bsr1.chi-alc.il.static.cable.rcn.com] has quit [Read error: 110 (Connection timed out)] 12:23:09 Which makes my comment useless. lichtblau has the better context here. 12:26:16 -!- cracki [n=cracki@40-123.eduroam.RWTH-Aachen.DE] has quit ["If technology is distinguishable from magic, it is insufficiently advanced."] 12:26:37 <_Trickster_> what is practical different btw string base-string simple-string and simple-base-string 12:26:40 <_Trickster_> ? 12:26:54 mkfort [i=xyOiN4jf@68-189-164-209.dhcp.spbg.sc.charter.com] has joined #lisp 12:27:16 haven't checked what XP did, but there's a *PRINT-PPRINT-DISPATCH* in CL 12:27:30 Guest76715: try using REAL as the type specifier, NaN's shouldn't qualify 12:27:54 also, please change your nick to something that can be completed usefully 12:28:00 lichtblau: oh 12:28:13 _Trickster_: some libs will choke on simple-base-string's on sbcl at least 12:28:22 -!- legumbre_ [n=user@r190-135-38-36.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 12:28:38 _Trickster_: but b-s-s uses less memory and is much faster to deal with 12:28:45 *attila_lendvai* is afk 12:28:46 isn't a simple-string just a string that is also a simple vector? 12:29:19 _Trickster_: BASE-STRING is a string with only BASE-CHAR's, which might (and does at least on SBCL) preclude unicode 12:29:44 with realp , it's T 12:30:03 <_Trickster_> base-char mb equal to character 12:30:05 dlowe [n=dlowe@ita4fw1.itasoftware.com] has joined #lisp 12:30:13 _Trickster_: mb? 12:30:26 <_Trickster_> may be 12:30:33 <_Trickster_> yeah 12:30:45 then please spell it may be, we're not on twitter and we aren't SMSing 12:30:59 <_Trickster_> ok 12:31:00 _Trickster_: might be, but doesn't have to. And isn't on SBCL 12:31:10 SBCL's BASE-CHAR is iso-8859-1 12:31:53 what the standard says is that you aren't allowed to rely on it being equal to CHARACTER 12:33:14 coderdad [n=coderdad@65.67.252.194] has joined #lisp 12:33:39 <_Trickster_> ok. so the actual case when concrete type used is implementation depended 12:34:12 <_Trickster_> but simple-string used more frequently 12:34:23 <_Trickster_> i'm right? 12:34:23 rouslan [n=Rouslan@unaffiliated/rouslan] has joined #lisp 12:35:08 <_Trickster_> if so why we need so many types just for strings 12:35:30 _Trickster_: because strings are vectors of characters, and we have so many vector subtypes 12:35:54 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 12:36:02 -!- schoppenhauer__ [n=css@82.113.121.149] has quit [Connection timed out] 12:36:03 <_Trickster_> it is *inherited* feature 12:36:10 _Trickster_: you are allowed to use SIMPLE-BASE-STRING if you know you will be writing in ASCII for example. But most of the time, for dealing with the outside world you want SIMPLE-STRINGs 12:36:13 <_Trickster_> hmm 12:36:14 _Trickster_: because ANSI CL wasn't as stupid as to define characters to be 8bit (like ANSI C) 12:36:51 yeah, so SBCL gained unicode without major pains 12:37:00 although it's still pretty damn slow at handling unicode streams 12:37:29 -!- mkfort_ [n=mkfort@68-189-164-209.dhcp.spbg.sc.charter.com] has quit [Read error: 110 (Connection timed out)] 12:37:29 also, the standard tried to include already existing implementations with minimal changes, which often ran on rather different machines 12:37:48 Guest76715: so it seems NaNs aren't defined in the standard after all, so you have to hit SBCL's docs 12:38:09 <_Trickster_> p_l: for this case we just need to make character type *fuzzy* 12:38:19 npoektop [n=user@85.202.112.90] has joined #lisp 12:38:21 <_Trickster_> not to introduce 4 string types 12:38:33 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 12:38:50 what 12:38:51 Thank yo mathrick....it's so hard to program...I'm totally exhausted. 12:38:51 _Trickster_: and go with madness of fuzzy behaviour? ;D 12:38:56 _Trickster_: what do you mean by that? 12:39:14 Guest76715: 'tis okay, but please change your nick, it's very hard to talk to you this way 12:39:18 <_Trickster_> mathrick: ".... we have so many vector subtypes" 12:39:21 so? 12:39:34 they all give you and the compiler different guarantees 12:39:37 how do I change it? 12:39:45 <_Trickster_> we may constrain strings as subtype of vector 12:39:48 Guest76715: /nick newnick 12:39:52 Guest76715: /nick kenjin_che 12:39:58 <_Trickster_> and then we need only two types 12:40:11 <_Trickster_> base-string (for implementation independence) 12:40:15 _Trickster_: you fail to get it spectacularly, but thanks for playing 12:40:21 -!- Guest76715 is now known as kenjin-che 12:40:22 <_Trickster_> and implementation dependent string 12:40:23 ok 12:41:14 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 12:41:22 <_Trickster_> mathrick: why i need fill pointers on string? 12:41:51 kenjin-che: actually, I can't find any docs for that with casual googling, you might want to hit one of the knowledgeable SBCL folks here who also deal with numeric computation 12:42:00 I guess you might need to wait until they're around 12:42:21 _Trickster_: because strings are vectors of characters, and vectors are allowed to have fill pointers 12:42:33 it's really not that hard, it's a simple cross product of the types we have 12:42:44 they were just given names 12:43:15 artificially limiting wouldn't remove any complexity, but would make it impossible to talk with the compiler usefully 12:43:35 mathrick: I tried everything I can think of to filter it, nothing worked. All I need is just distinguish it from ordinary numbers. 12:43:40 _Trickster_: for your purposes, you can have your two types: SIMPLE-BASE-STRING and SIMPLE-STRING 12:43:48 you aren't obligated to use anything more 12:44:23 <_Trickster_> actually i trying to implement my own lisp :-) 12:44:32 kenjin-che: I know, but I simply can't find anything in the docs. But there are many people here who deal in numeric computations, so if you wait, I'm sure someone will know the answer 12:44:39 <_Trickster_> so i need to pay more attention 12:45:04 kenjin-che: sb-ext:float-{nan,denormalized,infinity,trapping-nan}-p 12:45:05 _Trickster_: then you should understand why it's this way. All those names give different tradeoffs on the flexibility/optimisation scale 12:45:16 <_Trickster_> i just can't figure out how to implement all this stuff 12:45:45 _Trickster_: and that's precisely we the standard gives you so many names! 12:45:49 pkhuoung: I'm newbie...can you explain it easily? 12:46:12 _Trickster_: if you have SIMPLE-BASE-STRING, you can just map it on a buffer of bytes for example, like SBCL does 12:46:24 s/we/why/ 12:46:40 <_Trickster_> mathrick: i agree with you 12:47:22 s/on/to/ 12:47:42 Draggor [n=Draggor@216-80-120-145.alc-bsr1.chi-alc.il.static.cable.rcn.com] has joined #lisp 12:47:43 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 12:47:46 kenjin-che: if you want to test whether a given value is a NaN, use sb-ext:float-nan-p. 12:47:48 kenjin-che: they are predicates that tell you whether the number is a NaN for example 12:48:07 ilitirit_: good job re ICFP! 12:48:43 -!- pitui [n=pitui@135.207.174.197] has quit [Read error: 110 (Connection timed out)] 12:49:06 <_Trickster_> mathrick: i write it on .net so i need to map all this types to one in dot net - just String 12:49:14 pkhuong: WOW Wonderful~~ Thank you it works!! 12:49:42 _Trickster_: ah, a .NET CL. I thought about doing that myself 12:49:53 and the String mismatch is widely known 12:49:55 mathrick: SBCL's base-chars are ASCII on unicode-enabled builds. 12:50:04 <_Trickster_> moreover i need to do this all the time when lisp code call inherited .net code and vice-versa 12:50:05 i was just about to say :) 12:50:11 pkhuong: it seems to be latin1 here 12:50:29 _Trickster_: you are not *required* to distinguish, you can always upgrade 12:50:35 mathrick: then you have a non-standard build, or are mistaken somehow 12:50:53 base-char-code-limit ; => 128 12:51:03 <_Trickster_> upgrade what? 12:51:07 <_Trickster_> and to? 12:51:19 _Trickster_: so you can map SIMPLE-*-STRING to just string, and box the non-simple ones into a structure with a string and fill pointer counter 12:51:34 _Trickster_: element type and the actual adjustability 12:51:42 ,clhs actually adjustable 12:51:48 HG` [n=wells@xdsled140.osnanet.de] has joined #lisp 12:51:48 oh, no glossary lookups 12:52:17 _Trickster_: http://clhs.lisp.se/Body/26_glo_a.htm#actually_adjustable 12:52:18 when moving between languages you are almost certain to want to provide for external format conversions, and in many cases you end up wanting to copy the string even if not 12:52:27 <_Trickster_> yes actually now it works just like you say 12:52:34 _Trickster_: please read the section on arrays and vectors, and then read it again :) 12:53:23 <_Trickster_> nikodemus: i thinking about hooks for conversions 12:53:23 it *might* be that boxing a StringBuilder instead of actual strings will be better for you, but I dunno the overhead of Builders on .NET 12:53:29 Post-java languages tend to only provide immutable strings, so you may very well have to convert vector-of-character <-> string in foreign calls. 12:53:37 <_Trickster_> in dot net terms is custom marshalers 12:54:48 <_Trickster_> mathrick: String more preferable because all .net methods works with it 12:54:54 <_Trickster_> not with StringBuilder 12:55:07 pkhuong: I'm inclined to think it's not actually a bad thing, but certainly a PITA when trying to make CL work on top of it 12:55:31 _Trickster_: of course you'd have to unbox into strings before you can do anything 12:56:04 CL was unfortunately not designed with .net interop and embeddability in mind. 12:56:04 it depends on whether the overhead is greater for providing mutability inside CL, or unboxing into strings for .NET use 12:56:18 <_Trickster_> wow 12:56:27 Lacrymology [n=Tomas@200.69.219.97] has joined #lisp 12:56:34 <_Trickster_> .net string actually muteable 12:56:45 <_Trickster_> one sec 12:56:46 pkhuong: I never said that, I'm just saying that immutable strings are not such a bad idea per se 12:57:18 <_Trickster_> nono a said that! 12:57:24 <_Trickster_> it can be done with pointers 12:57:34 what can? 12:57:55 _Trickster_: and I'll bet you don't *really* want to change strings behind .NET's back 12:58:15 <_Trickster_> if string was interned while buildin - yes 12:58:24 <_Trickster_> but if it dynamic string why not? 12:58:35 because code relies on them being immutable 12:58:41 _Trickster_: thread-safety 12:58:45 no-one will copy immutable data structures 12:58:57 so code that relies on the initial copy staying the same will break 12:59:02 and what p_l says 12:59:10 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 12:59:24 eh, that was phrased poorly, but you know what I mean 12:59:46 danlei [n=user@pD9E2CC29.dip.t-dialin.net] has joined #lisp 13:00:07 <_Trickster_> what a nice bit of code now obsolete 13:00:41 <_Trickster_> thanks guys 13:01:56 oops, time to run -- bye 13:01:57 -!- nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has quit ["Leaving"] 13:02:36 _Trickster_: are you writing it in C#, or are you implementing a CIL compiler in CL? 13:02:43 _Trickster_: there are many places where CL and .NET won't agree, not the least being the type and class system, and generic functions. If your goal is to make interop easy and exploit existing .NET code, you'll probably want to fork another language. If you want a CL, then you'll have to add some explicit marshalling step. 13:03:07 the floating point types will also give you headaches 13:03:09 -!- attila_lendvai [n=ati@catv-89-132-189-132.catv.broadband.hu] has quit ["..."] 13:03:11 <_Trickster_> no i'm not god im writing it in c# 13:03:20 why "god"? 13:03:36 you still need to compile down to bytecode to make it at least moderately speedy 13:03:42 <_Trickster_> i think that CIL compiler is hard task for me 13:03:49 and this way you get to work in CL 13:04:13 <_Trickster_> i'm using Reflection.Emit 13:04:25 <_Trickster_> but i'm thinking about Mono.Cecil 13:04:37 I'm quite sure implementing this part in CL would be a couple of days at best 13:04:44 or at worst actually 13:04:58 -!- Guest25863 [n=lexa_@seonet.ru] has left #lisp 13:05:06 <_Trickster_> yes i already have about 60000 loc 13:05:21 -!- ia [n=ia@89.169.189.230] has quit [Connection timed out] 13:05:29 which means you'd probably be way better off by starting in CL 13:05:41 I wonder if ECL could be retargetted to CIL 13:06:11 ia [n=ia@89.169.189.230] has joined #lisp 13:06:53 -!- drafael1 [n=tapio@ip-118-90-132-131.xdsl.xnet.co.nz] has quit ["Leaving."] 13:07:05 <_Trickster_> you mean that i should start it in CL and play with bytes by hands? 13:07:42 pretty much, yeah 13:07:47 slyrus_ [n=slyrus@adsl-75-36-210-102.dsl.pltn13.sbcglobal.net] has joined #lisp 13:07:52 you could emit either bytecode or CIL 13:08:31 I don't imagine writing a bytecode emitter for .NET would be *that* hard 13:08:45 <_Trickster_> if i'm not wrong Mr. Norvig already did that 13:08:54 then even better for you 13:09:07 <_Trickster_> of couse! 13:09:08 steal his emitter and make your implementation use that to compile 13:09:18 how do I force evaluation of an element inside a quoted list? 13:09:30 Lacrymology: what do you want to achieve? 13:09:37 <_Trickster_> he just emit CIL and then compile it with ilasm 13:09:38 it's possible that you mean to use a backquote 13:09:39 say I want to do this '(1 2 3 (+ x y)) 13:09:51 Lacrymology: you could use `(1 2 3 ,(+ x y)) 13:09:54 Lacrymology: `(1 2 3 ,(+ x y)) 13:10:06 thanks 13:10:10 sellout [n=greg@guest-fw.dc4.itasoftware.com] has joined #lisp 13:10:20 minion: give Lacrymology clhs backquote 13:10:20 Lacrymology: please stop playing with me... i am not a toy 13:10:29 hmm. 13:10:34 ,clhs backqoute 13:10:41 clhs backquote 13:10:42 Sorry, I couldn't find anything for backquote. 13:10:52 <_Trickster_> AI on stage 13:10:55 that's weird. 13:11:16 it's not actually called that in the standard 13:11:35 okay, it is? 13:11:37 http://www.ai.mit.edu/projects/iiip/doc/CommonLISP/HyperSpec/Body/sec_2-4-6.html 13:11:38 then it's weird 13:11:42 http://www.ai.mit.edu/projects/iiip/doc/CommonLISP/HyperSpec/Body/sec_2-4-6.html 13:11:43 yeah 13:12:14 ah, here's a link to the *real* hyperspec 13:12:23 Lacrymology: http://www.lispworks.com/documentation/lw50/CLHS/Body/02_df.htm 13:12:55 Lacrymology: once you get comfortable with simple backquotes, you might want to read the paper on comma calculus in nested backquotes, but it will explode your brain if you're not careful :) 13:13:13 <_Trickster_> what the best way to integrate eclipse with lisp ? cusp? 13:13:26 probably, also the only one I'm aware of 13:13:33 minion: tell _Trickster_ about eclipse 13:13:33 _Trickster_: have a look at eclipse: Eclipse is an X11 window manager written entirely in Common Lisp. http://www.cliki.net/eclipse 13:13:47 hm, the bots seem to be having an off day. 13:13:53 <_Trickster_> yep i know about it 13:14:12 Adlai: no, you seem to be asking the wrong questions :) 13:14:25 setheus [n=setheus@pool-173-74-124-37.dllstx.fios.verizon.net] has joined #lisp 13:14:29 mathrick: that too. 13:16:28 why Eclipse when you can have SLIME? ;D 13:17:07 <_Trickster_> key chords it's like love ballad 13:17:52 <_Trickster_> gently press your keys and so on 13:20:47 -!- jan247 [n=jan247@222.127.86.98] has quit [] 13:20:47 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 13:21:12 athos [n=philipp@92.250.250.68] has joined #lisp 13:21:39 -!- Phaze [n=PhazeDK@0x5da32b16.cpe.ge-0-1-0-1104.soebnqu1.customer.tele.dk] has quit ["Leaving"] 13:23:13 wtf comma calculus? 13:24:05 newbie question: if a setf-able place is passed as an arg to a lambda that setfs it, will the place be properly setf'd after the lambda returns? 13:24:25 nunb: you can only pass values as arguments, not places. 13:24:41 jan247 [n=jan247@222.127.86.98] has joined #lisp 13:24:49 pkhuong: cool, thanks. 13:25:45 mathrick: I usually factor out the second level of backquotes. avoids headaches down the line 13:25:48 bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has joined #lisp 13:27:09 more newbie questions, is there a way of passing # an expression that EVALUATES to a symbol which is the name of a function? 13:28:04 FUNCTION (#') isn't a function. FDEFINITION might be what you want. 13:32:33 -!- csaba [n=cgajo@193.189.163.110] has quit ["Ex-Chat"] 13:45:44 can you make hunchentoot drop into the slime debugger when an error has been encountered? 13:46:51 lexa_ [n=lexa_@seonet.ru] has joined #lisp 13:47:19 -!- lexa_ is now known as Guest27422 13:47:34 gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has joined #lisp 13:49:32 madnificent: yes 13:50:05 -!- Guest27422 [n=lexa_@seonet.ru] has left #lisp 13:51:18 antifuchs: how ? (or google string) 13:51:58 -!- morphling [n=stefan@gssn-590e8eb5.pool.einsundeins.de] has quit [Remote closed the connection] 13:52:15 madnificent: http://common-lisp.net/pipermail/tbnl-devel/2009-April/004688.html 13:52:41 lukjad007 [n=lukjadOO@unaffiliated/lukjad007] has joined #lisp 13:52:45 you can leave out the process-connection method if you don't want to get a debugger when a connection times out 13:53:08 *madnificent* tries 13:55:52 -!- kpreid [n=kpreid@cpe-67-249-56-101.twcny.res.rr.com] has quit [] 13:56:24 kpreid [n=kpreid@cpe-67-249-56-101.twcny.res.rr.com] has joined #lisp 13:56:57 -!- beaumonta [n=abeaumon@85.48.202.13] has quit [Remote closed the connection] 13:57:08 -!- spilman [n=spilman@ARennes-552-1-17-7.w92-135.abo.wanadoo.fr] has quit ["Quitte"] 13:57:28 Buganini [n=buganini@security-hole.info] has joined #lisp 13:57:31 -!- Buganini [n=buganini@security-hole.info] has quit [Client Quit] 13:57:43 Buganini [n=buganini@security-hole.info] has joined #lisp 13:58:45 Okay, an interesting phenomenon in 1.0.29.52: given an expression X which dumps a core, "sbcl --eval '(require :foo)' --eval 'X' 13:58:57 produces a working executable core. 13:59:15 jleija [n=jleija@user-24-214-122-46.knology.net] has joined #lisp 13:59:21 When I wrap that into a Makefile and issue make -- the resulting core segfaults. 13:59:41 linux-2.6.30/x86. 14:00:04 I'll now update to the current sbcl master. 14:01:03 (sb-disassemble:disassemble-memory ...) points at ECX dereference. 14:02:49 Naturally, a testcase would be extremely hard to reduce.. :-( 14:04:30 -!- jdz [n=jdz@85.254.211.133] has quit ["Somebody booted me"] 14:07:12 Nshag [i=user@Mix-Orleans-106-2-120.w193-248.abo.wanadoo.fr] has joined #lisp 14:09:01 -!- Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has quit ["leaving"] 14:12:09 michaelw: definitely 14:12:28 the problem is when people factor out the first level, that is just silly 14:12:44 and inexplicably Edi seems to be opposed to backquotes 14:19:23 Lacrymology: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.5.2566 <-- this is the paper I mentioned. But don't read it just yet, it'll fry your brain needlessly 14:20:57 nunb: if you really need that, you can implement your own version of "locatives", which are places you can pass around some older lisps had 14:21:10 antifuchs: doesn't seem to invoke the debugger in my slime... 14:21:59 nunb: http://osdir.com/ml/lisp.sources.code/2004-06/msg00005.html 14:23:10 it's a pity there isn't a portable way to express a pointer to a pointer in CL, but nothing that can't be worked around 14:26:13 dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has joined #lisp 14:26:19 (& (& x)) 14:26:51 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 14:29:07 -!- slyrus_ [n=slyrus@adsl-75-36-210-102.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 14:30:42 mathrick: http://article.gmane.org/gmane.lisp.sources.code/17 14:31:40 *Adlai* reads that paper... 14:32:05 matley [n=matley@matley.imati.cnr.it] has joined #lisp 14:32:24 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 14:32:28 ikki [n=ikki@201.155.75.146] has joined #lisp 14:34:46 deepfire: nunb: http://osdir.com/ml/lisp.sources.code/2004-06/msg00005.html 14:36:28 -!- daniel_ is now known as daniel 14:36:36 TR2N [i=email@89.180.178.141] has joined #lisp 14:37:12 tmh [n=thomas@pdpc/supporter/sustaining/tmh] has joined #lisp 14:37:18 Greetings. 14:37:43 Ah. Haha. 14:37:58 hello tmh 14:38:43 hi tmh 14:45:20 lukego [n=lukegorr@84-75-21-37.dclient.hispeed.ch] has joined #lisp 14:45:42 -!- mgm [n=mgm@p84-72.acedsl.com] has quit [Read error: 113 (No route to host)] 14:45:45 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 14:48:28 zophy [n=sy@host-242-6-111-24-static.midco.net] has joined #lisp 14:50:04 morphling [n=stefan@gssn-590e8eb5.pool.einsundeins.de] has joined #lisp 14:50:22 fusss [i=73800949@gateway/web/freenode/x-ljnjqwpdixewjppu] has joined #lisp 14:51:05 ohai 14:53:29 -!- nunb [n=user@static-217-133-104-225.clienti.tiscali.it] has left #lisp 14:57:53 nunb` [n=user@static-217-133-104-225.clienti.tiscali.it] has joined #lisp 14:59:43 _deepfire pasted "A somewhat lousy bug report: 1.0.30.26" at http://paste.lisp.org/display/84508 15:00:15 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 15:02:11 -!- Buganini [n=buganini@security-hole.info] has quit [Read error: 113 (No route to host)] 15:02:28 -!- prg_ [n=prg@ns.alusht.net] has quit ["leaving"] 15:04:54 phf [n=phf@host.icnfull.com] has joined #lisp 15:12:26 Buganini [n=buganini@security-hole.info] has joined #lisp 15:16:09 deepfire: is your makefile setting soeme env var? 15:17:11 kpreid, it's an extremely simple makefile: no env. tweaking beyond what make will do by default. 15:18:55 _deepfire annotated #84508 "the Makefile in question" at http://paste.lisp.org/display/84508#1 15:19:52 -!- Buganini [n=buganini@security-hole.info] has quit [Read error: 113 (No route to host)] 15:20:53 _deepfire annotated #84508 "definitions of NANOSLEEP and NANOSLEEP-FAST" at http://paste.lisp.org/display/84508#2 15:21:42 -!- quek [n=read_eva@router1.gpy1.ms246.net] has left #lisp 15:22:17 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 15:24:35 Buganini [n=buganini@security-hole.info] has joined #lisp 15:29:14 I vaguely remember hearing that ECX has something to do with threads on x86. 15:31:09 postamar [n=postamar@x-132-204-252-212.xtpr.umontreal.ca] has joined #lisp 15:31:18 -!- blandest [n=user@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 15:33:56 defun inside with-alien is asking for trouble imo. 15:34:01 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit ["leaving"] 15:34:56 -!- alinp [n=alinp@86.122.9.2] has quit ["Leaving."] 15:36:01 DrForr_ [n=drforr@pool-173-55-19-85.lsanca.fios.verizon.net] has joined #lisp 15:36:17 XOR EDX, EDX / MOV [ECX], EDX is (setf (slot seq 'tv-sec) 0). 15:36:25 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 15:36:47 (and the code is really awful) 15:37:54 pkhuong, the lisp, or the assembly? :-) 15:38:38 the assembly. 15:38:54 But your code is definitely wrong. with-alien allocates for dynamic extent by default. 15:39:12 You were just lucky. 15:39:33 Ok, redoing it with MAKE-ALIEN.. 15:39:44 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 15:40:50 also, why are you using external symbols from SB-ALIEN as internal ones SB-GROVEL? 15:40:54 *in SB-GROVEL. 15:41:56 pkhuong, this is almost completely cut-and-pasted nanosleep.. 15:42:27 Accomodated, but untended. 15:42:53 sb-unix::addr is just sb-alien:addr. SB-UNIX and SB-GROVEL use SB-ALIEN. 15:43:18 Novie [n=kvirc@dsl-247.static.grp6.tnmmrl.infoave.net] has joined #lisp 15:45:13 sykopomp [n=sykopomp@unaffiliated/sykopomp] has joined #lisp 15:47:15 I see, thanks.. 15:47:42 Oh, not so simple to replace WITH-ALIEN by globals + MAKE-ALIEN. 15:49:16 -!- DrForr [n=drforr@pool-173-55-19-85.lsanca.fios.verizon.net] has quit [Read error: 110 (Connection timed out)] 15:49:51 It appears the former has a special information passing convention from the binding to the use sites, involving LOCAL-ALIEN-INFO.. 15:51:36 just move the with-alien inside the funciton 15:51:45 you'll get thread safety to boot. 15:52:42 Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has joined #lisp 15:52:58 Alien allocation was in the top of my tight loop -- I was poking x86 IO ports.. 15:53:50 Erm, I could use less obtuse english, at times. 15:54:04 Alien allocation was in the top of the profile of my tight loop. 15:54:40 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit ["HULK ANGRY! HULK DISCONNECT!"] 15:54:57 Costs about 20-30% in the end-user experience. 15:55:22 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit [No route to host] 15:56:00 good morning #lisp 15:56:00 drewc, memo from sykopomp: http://www.ccs.neu.edu/home/dorai/uwcallcc/uwcallcc.html can we have call/cc in cltl3 now? ;) 15:56:46 sykopomp: no, and i'm a call/cc user :) 15:57:02 You're worried about malloc/free around calls to nanosleep? 15:57:26 drewc: but we can have both call/cc and unwind-protect! :( 15:58:16 *sykopomp* waits for cltl4 to try again... 15:59:01 pkhuong, not sure what I'm worried about, but it was exhibited in the profile report :-) 15:59:42 sykopomp: did you read that paper? those are not full continuations ;) 16:00:01 -!- mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has quit ["Lost terminal"] 16:00:20 pkhuong, if I were to apply some intelligence before replying -- yes, I saw MAKE-LOCAL-ALIEN and blamed that. 16:01:20 drewc: optionally, in order to combine them with -one- special form. 16:03:13 sykopomp: it's not that simple, unwind-protext is not the only problem with continuations and CL. 16:03:41 sykopomp: the first is that there is no call/cc in CL, so how are you going to implement call/cc-l ? 16:04:33 sykopomp: a CPS transformer and code walker is not a solution... have you seen the code something like cl-cont generates? horrible! 16:04:53 sepult [n=buggarag@xdsl-87-78-75-197.netcologne.de] has joined #lisp 16:04:59 drewc: the point I'm trying to make is that it -is- apparently possible to implement unwind-protect and call/cc in the same language :\ 16:05:01 sykopomp: i'm becoming more an more convinced that somthing like 16:05:16 cc-let* and a DSL is all one needs. 16:05:28 sykopomp: why do you want full continuations anyway? 16:05:35 sykopomp: no .. it's not. call/cc-l is not call/cc 16:05:46 michaelw: the ying yang puzzle, of course. 16:06:44 sykopomp: Please understand that writing up a document, be it as formal as you want, which requests feature X won't be likely to change anything at all. 16:07:26 sykopomp: People have to realize that change comes from action. You have to get feature X in at least one implementation. And that most probably means that you have to do it yourself, or pay one to do it. 16:07:40 sykopomp: http://paste.lisp.org/display/77636 16:07:58 +1 tcr 16:08:06 tcr: honestly, my point was only half-serious :P 16:08:14 s/point/request/ 16:08:23 nothing is getting into cltl3 that is not already implemented, so if you want something in cltl3, implement it! 16:09:13 drewc: btw, what is the current take on networking library (sockets etc.) for cltl3? usocket? 16:09:21 p_l: iolib 16:09:45 sykopomp: Yeah, but many people who wish for a next language revision, de facto standardization do not get it. In particular they have no understanding of the costs involved. Basically they want something like a feature tracking system for Common Lisp and the Open Source Fairy to make it all happen. 16:10:22 without personal involvement 16:10:25 instantly 16:10:55 p_l: though sockets might not need to be a part of cltl3. If you have gray-streams and CFFI-SYS 'standardised', once can build a 'portable' sockets library. 16:10:56 drewc: iolib afaik didn't work outside of POSIX, so it will require some help in that case... 16:11:09 tcr: yeah. And adding anything as big as first-class continuations is probably a huge PITA for implementations, to boot, so there better be a damn good reason to even suggest it :P 16:11:11 p_l: how well do sockets work outside posix? 16:11:20 (a reason I don't have, of course) 16:11:41 -!- xan is now known as xan-afk 16:12:12 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 16:12:30 p_l: i don't know of an implementation that implements sockets on a non-posix-similar-enough system.. but if you point me to some links i'll be happy to investigate. 16:12:34 I'm with drewc. We want a good, expressive, pluggable control flow language to unite loop, iterate, series and some of screamer ;) 16:12:38 drewc: BSD sockets are not *nixy interface at all (they are more like graft from TOPS-20 iirc), so there are cases where other OSes implement BSD stack and sockets but it's not integrated at all with other IO interfaces (and I still cherish thoughts of porting SBCL to VMS) 16:13:23 p_l: what we include in cltl3 will not stop you porting SBCL to VMS regardless, so it's a non-issue. 16:13:26 iolib afaik simply didn't work on Win32, but I'm going to investigate - I'm very interested in cltl3 (and willing to try implementing) 16:14:01 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 16:14:27 There's a VMS port of SBCL on the way?? 16:15:05 last I looked at iolib code the syscalls and socket packages were separate and socket one used implementation-specific code I think but that was a little while ago 16:15:19 pkhuong: don't forget series! 16:15:38 huh? he didn't forget series 16:15:46 oh .. he didn't 16:15:49 *drewc* gets coffee 16:15:54 not sure if the multiplexer code was portable to non-posix 16:16:14 phadthai: ok, good. Need to get a working windows install again and setup a working lisp dev. environment 16:16:46 schme: Maybe. Biggest obstacle would be the need to port to a new CPU 16:17:31 though SCL does work with conversion, so it might be enough to find how they work on top of converted binary 16:17:41 p_l: Isn't there an linux/alpha port of linux already? 16:17:51 oh you are meaning ia64 vms?? 16:17:53 crazy! 16:18:26 sykopomp: i think a point that needs to be made is that the paper you linked to is about unwind-protect in scheme... not call/cc in lisp. :) 16:19:21 SandGorgon [n=OmNomNom@122.161.46.116] has joined #lisp 16:20:11 nha [n=prefect@137-64.105-92.cust.bluewin.ch] has joined #lisp 16:20:59 spilman [n=spilman@ARennes-552-1-17-7.w92-135.abo.wanadoo.fr] has joined #lisp 16:22:17 schme: but would be wonderful, don't you think? :D 16:22:46 p_l: I don't have one so I couldn't care less ;) 16:23:01 mathrick [n=mathrick@wireless.its.sdu.dk] has joined #lisp 16:23:07 schme: I can do testing for VAX and Alpha (but only baseline arch, no BWX nor other extensions), but I doubt I'll be able to go after IA-64 :D 16:23:23 this is why there is no software for IA64 16:23:47 because no-one has ever even SEEN an IA64 16:23:52 p_l: Cools. I have easy access to an alpha running vms here too. So if you get started I'll love to hand you tons of error logs :) 16:24:01 (of course the reason for that is that there is no software for them ;) ) 16:24:07 I've seen pictures of 'em... (: 16:24:31 I had a chance to buy one, but didn't have enough money (it would cost me two months rent!) 16:25:02 Hmm.. I remember browsing around HPs site and getting the impression that they cost more than two months rent :) 16:25:25 p_l: You should have bought it so we could all laugh at you (: 16:25:33 schme: it was a used rx2600 iirc 16:25:50 schme: didn't have ~£800 to spend 16:25:52 I'd like one to play with. I have heard the asm is horrid :) 16:26:16 schme: I once heard someone argue that with certain trick in assembly you get hw closures 16:26:17 the compiler is expected to do more work than would normally be the case, certainly 16:26:46 p_l: Good thinknig there with "used". You gave me the inspiration to fire up ebay.de :) 16:26:54 *rsynnott* still vaguely thinks the ia64 could take off if people get the compilers right; it's not actually a bad design 16:26:56 yes it might actually be easier to get ecl working on ia64 than sbcl since you'd not have to write the compiler magic :) 16:27:08 seems.. mostly I find "Intel Itanium 2 Inside" stickers for sale. 16:27:13 schme: be carefulyou don't get an Itanium 1! 16:27:24 the main problem with compilers is that IA64 doesn't have branch-prediction nor any kind of scheduling - explicit parallelism isn't that bad 16:27:27 You don't need magic if you using only 1/4 of the slots doesn't bother you. 16:27:41 p_l: sure... for DSP. 16:27:42 (they were the ones with wacky hardware 386 emulation) 16:28:00 maybe LLVM will evnetually help 16:28:02 rsynnott: not only that, but apparently they were incredibly slow even for ia64 16:28:02 Hmm. these used ones aren't actually that pricey. 16:28:19 jfrancis_ [n=jfrancis@66.194.68.209] has joined #lisp 16:28:34 schme: remember to get Itanium 2. Not only no hw x86, but also much more speed IIRC 16:28:49 They say itanium2 atleast :) 16:28:51 yep, also different instruction set 16:29:13 and SSE2-type stuff 16:29:37 yay. winxp preinstalled! 16:29:49 clhs #+ 16:29:50 http://www.lispworks.com/reference/HyperSpec/Body/02_dhq.htm 16:30:03 anyway, the guy who was selling that rx2600 had it with itanium2 and could equip it with various expansion cards - he was basically a reseller of unix workstations (no x86) :) 16:30:12 cools. 16:30:49 if you ever want an old sun or ten, check your local university 16:30:58 Too bad they didn't specify that (#2+key a b c) would skip over two items unless :key was in *features*... 16:31:03 rsynnott: I did, they told me "no" 16:31:14 (for reasons I'm not clear on, all universities have roughly a thousand old suns each; they periodically throw some away) 16:31:18 skeptomai [n=cb@67.40.185.246] has joined #lisp 16:31:23 -!- malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has quit [Read error: 60 (Operation timed out)] 16:31:40 rsynnott: Sun hit early with networked unix workstations in acceptable price range 16:31:55 My research center's trying to slowly get rid of its 64-way E10k (: 16:32:09 after all, the company started by branching from a PhD project for a cheap unix workstation 16:32:41 E10k... nice machines. Generate quite a gust of wind :D 16:32:47 -!- coderdad [n=coderdad@65.67.252.194] has quit [Remote closed the connection] 16:32:53 we had rooms of suns in progressive degrees of horrible oldness 16:33:33 including at least two rooms which were normally never used at all for anything 16:33:44 (they were disputed territory between the CS dept and physics dept) 16:34:04 rsynnott: My university has big chunk of networking done on Suns, CS dept. had also two 2-way Sun Ultras + several Sun Rays. They got removed because said equipment went to some research group 16:34:08 malcolm_reynolds [n=malcolm_@78-86-4-156.zone2.bethere.co.uk] has joined #lisp 16:35:18 a student society also had a weird HPPA workstation of some sort that had been discarded by the university; I've no idea where it came from 16:35:29 hadn't realised that there WERE HPPA workstations before 16:35:29 rsynnott: probably engineering 16:35:47 CAD + analysis stuff 16:36:10 (one of the main mail servers, also, was an alpha and had previously been a VAX; they had _all_ of the dead architectures :)) 16:36:15 rstandy [n=rastandy@net-93-144-28-37.t2.dsl.vodafone.it] has joined #lisp 16:36:52 my university has few HP Visualize HPPA workstations running as servers accessible over SSH and X11, with students logging in to access FLUENT and other packages 16:37:43 mrsolo [n=mrsolo@nat/yahoo/x-cgxwtyrysbvzcrqm] has joined #lisp 16:37:49 There were still a couple Apollo workstations in the lab when I entered the university. I remember doing something trivial on them a couple times. The windowing interface was slightly different, I think the shell window had a minibuffer. 16:38:03 rsynnott: pfft, until recently, I had all of the dead architectures, too 16:38:18 dmiles_afk [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has joined #lisp 16:38:18 michaelw: what, even a VAX? 16:38:23 -!- mrsolo [n=mrsolo@nat/yahoo/x-cgxwtyrysbvzcrqm] has quit [Remote closed the connection] 16:38:34 mrsolo [n=mrsolo@nat/yahoo/x-aguiddlrviqnrzav] has joined #lisp 16:38:54 there's a guy in USA that got all kinds of architectures in basement, all in working order, ready to run. Got an order from judge that he shouldn't start them all at the same time 16:39:04 *tmh* wishes there were more, varied, systems on the market. 16:39:17 last time he tried to run them all, he apparently crashed local power grid 16:39:18 rsynnott: ok, no VAX, that was around the time when I got interrupted with my collection 16:39:48 -!- lichtblau [n=user@77-22-106-191-dynip.superkabel.de] has quit [Read error: 113 (No route to host)] 16:40:02 tmh: you can have 386 or ARM 16:40:09 or PPC if you're lucky 16:41:03 rsynnott: I was thinking more along the lines of OS and interface. When everyone was kicking out their own hardware, they seemed to also be kicking out their own OS and interface. 16:41:12 Do ultrasparcs have precise exceptions? 16:41:18 Or at least more variations. 16:41:29 rsynnott: there's also quite a lot of MIPS embedded in various places 16:42:23 The first time I used a Sun workstation, it had Openwindows. That was my first real introduction to X, I liked it. 16:42:43 sepult_ [n=buggarag@xdsl-87-78-169-181.netcologne.de] has joined #lisp 16:43:24 tmh: I agree there. The lack of fun variety is ... a bit sad :) 16:43:43 woohoo! wrote my first useful bit of Lisp! 16:43:50 rsynnott: it appears that number of computers decreases roughly like (exp (- *relationship-time*)) ;) 16:43:56 *schme* hopes green arrays incs has some success (: 16:44:29 pkok: is it a web framework? 16:44:29 fusss: he said "useful" 16:44:39 cd database! 16:44:52 Portable path library 16:44:58 reddit clone? 16:45:23 Nah, only a few lines to convert from one vcard file format to another 16:45:23 i wish somebody would write a unit testing framework 16:45:35 that's actually useful 16:45:47 fusss: how are current ones not? 16:46:10 *tmh* needs lunch, wrestling with tensors in maxima makes one hungry. 16:46:13 fusss, have you read that unit testing framework overview somebody wrote about half a year ago? 16:46:19 (says me, who whipped one up himself) 16:46:23 fusss, yeah, sony ericsson stores all vcards in one file, nokia wants them in all seperate fiels 16:46:31 I think it was on the planet. 16:46:48 <_Trickster_> what is your favorite standalone lisp web server ? 16:46:48 deepfire: no, but i shopped around and settled on FORMAT 16:47:00 _Trickster: Hunchentoot 16:47:03 -!- sepult_ [n=buggarag@xdsl-87-78-169-181.netcologne.de] has quit [Client Quit] 16:47:13 <_Trickster_> ye 16:47:29 fusss: What about assert+format? 16:47:32 michaelw: dunno, my first lisp programs were mostly "emulators"; not useful at all. 16:47:55 tmh: ASSERT comes with FORMAT integrated 16:48:06 tmh: the hyperspec page for ASSERT was tl;dr 16:48:13 http://aperiodic.net/phil/archives/Geekery/notes-on-lisp-testing-frameworks.html 16:49:12 deepfire: that's exactly the paper that sent me to FiveAM 16:49:24 tcr: Do I have to be a pendant while being a smart ass? 16:49:57 -!- leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has quit [Read error: 104 (Connection reset by peer)] 16:50:09 How about being pendant in a high-wire act? 16:50:13 I remember, trying to whip myself into adding the features I use in my own framework into Stefil, and switching to that. Still TBD. 16:50:56 <_Trickster_> im had tried to install Hunchentoot on sbcl 16:51:07 <_Trickster_> and i've got an error 16:51:16 That can happen. 16:51:30 Trickster: sbcl on Linux? 16:51:34 <_Trickster_> strange error 16:51:47 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 16:51:50 fusss: http://foldr.org/~michaelw/lisp/utest.lisp ; 122 lines, no dependencies, WFM 16:52:09 *tmh* muses about STRANGE-ERROR versus SIMPLE-ERROR. 16:52:16 _Trickster_, there's so many; hunchentoot (probably the most well known), ucw, sw-http (silly test runs here from time to time; http://sw.nostdal.org/text-input ), teepeedee2 .. hunchentoot is probably the easiest to get up and running 16:52:26 legumbre [n=user@r190-135-35-185.dialup.adsl.anteldata.net.uy] has joined #lisp 16:52:30 michaelw: nice! :-) another one for the pool ;-) 16:52:55 Is there a simple pattern-matcher for lists in few hundred of lines? I need one for swank. 16:53:01 The pool needs some chlorine. 16:53:14 <_Trickster_> lnostdal: yes, i know but install manuals too strong for me 16:53:21 -!- legumbre [n=user@r190-135-35-185.dialup.adsl.anteldata.net.uy] has quit [Remote closed the connection] 16:53:22 not in a few hundreds, I think one hundred should be plenty enough 16:53:30 Trickster: what platform are you on? 16:53:40 tcr: destructuring-bind not enough? :) 16:53:47 <_Trickster_> ohoh 16:53:53 <_Trickster_> windows 16:53:56 legumbre [n=user@r190-135-35-185.dialup.adsl.anteldata.net.uy] has joined #lisp 16:54:18 <_Trickster_> i suppose on linux it will be more easy 16:54:27 Trickster: I code on hunchentoot on windows for a living. Do yourself a favor an install LispWorks, the free edition. 16:54:47 <_Trickster_> lispworks? 16:55:04 it's a commercial lisp with a free version 16:55:07 <_Trickster_> is this going to close after hour? 16:56:10 tcr: it's sad how sub-optimal library management affects our decisions, though... 16:57:13 fusss_ [n=chatzill@115.128.9.73] has joined #lisp 16:57:27 -!- sepult [n=buggarag@xdsl-87-78-75-197.netcologne.de] has quit [Read error: 110 (Connection timed out)] 16:57:42 _Trickster_: sorry i was cut off. http://www.lispworks.com/downloads/index.html http://weitz.de/starter-pack/ 16:57:52 download and install both 16:58:05 <_Trickster_> by the way! the errorThe function USOCKET::%SETUP-WAIT-LIST is undefined. 16:58:05 -!- fusss [i=73800949@gateway/web/freenode/x-ljnjqwpdixewjppu] has quit [Ping timeout: 180 seconds] 16:58:16 -!- fusss_ is now known as fusss 16:58:19 _Trickster_: bingo! :-) 16:58:29 <_Trickster_> of couse i have the latest usocket 16:58:37 <_Trickster_> yeah! 16:58:45 it's the win32 version. 16:58:52 <_Trickster_> the error! The function USOCKET::%SETUP-WAIT-LIST is undefined. 16:58:54 -!- Modius [n=Modius@24.174.112.56] has quit [Read error: 60 (Operation timed out)] 16:59:27 <_Trickster_> yeas 16:59:41 you can run hunchentoot on sbcl or even clisp, but it takes more customization; you will need to disable multithreading, and might even need to disable SSL 16:59:46 <_Trickster_> did you read my question about lisp works 16:59:50 <_Trickster_> > 17:00:04 and then, once you launch it and it starts listening, you don't get the shell back 17:00:09 <_Trickster_> is this going to close after hour? 17:00:20 _Trickster_: ~4 hours 17:00:32 <_Trickster_> i have eclipse with cusp 17:00:37 <_Trickster_> with this server 17:00:43 <_Trickster_> all works fine 17:00:55 does it run hunchentoot? 17:01:50 <_Trickster_> yes 17:02:09 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit [Read error: 60 (Operation timed out)] 17:02:21 <_Trickster_> but version of server is 0.3 17:02:28 <_Trickster_> if i'm not wrong 17:02:35 no 17:02:51 hunchentoot is 1.0.0, previous was 0.15.7, IIRC 17:02:54 <_Trickster_> The current version is 1.0.0. 17:02:59 <_Trickster_> so i need to update 17:03:05 michaelw: If if it was better, I think swank is a bit of a special case 17:03:11 michaelw: *Even if 17:03:12 that's the latest version 17:03:27 <_Trickster_> may be i have the 0.15 17:03:49 Modius [n=Modius@24.174.112.56] has joined #lisp 17:03:54 <_Trickster_> any way when i try to run it in standalone mode it not working at all 17:04:09 completely different, API-incompatible version 17:04:38 <_Trickster_> i've got USOCKET::%SETUP-WAIT-LIST is undefined. 17:04:41 it's usually better to say *what* is wrong, instead of just "something is wrong" 17:04:43 -!- legumbre [n=user@r190-135-35-185.dialup.adsl.anteldata.net.uy] has quit [Remote closed the connection] 17:04:51 in LispWorks? 17:05:03 <_Trickster_> sbcl 17:05:19 we established that win32 sbcl is not very hunchentoot friendly 17:05:35 <_Trickster_> sbcl win32 17:06:03 you still have many other options, i recommended the freebie LispWorks, but it's not the only other option. 17:07:21 <_Trickster_> i'm will install linux under VM 17:07:56 ok 17:09:05 <_Trickster_> i have a project i think it can be done at least partly on lisp 17:09:24 3 17:09:41 mattrepl [n=mattrepl@129.174.97.34] has joined #lisp 17:09:43 <_Trickster_> and this part will be web server. actually it will work under heavy load 17:10:13 Does that mean 10,000+ requests per second? 17:10:22 <_Trickster_> so i playing with google to grab the information 17:10:27 sepult [n=buggarag@xdsl-87-78-169-181.netcologne.de] has joined #lisp 17:10:42 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 17:10:46 <_Trickster_> Zhivago: yes i already readed the project page 17:10:47 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 17:10:58 _Trickster_: have you done this before? the web server under heavy load thing, that is 17:11:13 Just trying to work out what you mean by heavy load. 17:11:43 <_Trickster_> 10.000+ req per second 17:11:47 _Trickster_: what is your native language? 17:11:57 <_Trickster_> how this requests was processed? 17:12:09 <_Trickster_> fusss: russian 17:12:29 _Trickster_: You tried CCL? 17:12:31 _Trickster_: you're lucky enough to be able to read the nginx documentation, then 17:12:32 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 17:12:47 Lacrymology1 [n=Tomas@200.69.219.97] has joined #lisp 17:12:47 <_Trickster_> my application will do expensive calculations 17:12:52 <_Trickster_> on each request 17:13:06 Hmm, I hope they don't take more than 1/10000 of a second to do. 17:13:10 <_Trickster_> fusss: english docs for nginx exists as well 17:13:35 <_Trickster_> Zhivago: but i have a db 17:13:50 what db? 17:13:51 _Trickster_: you might also try ClozureCL, but linux in a VM will be probably easier (as you can use clbuild then) 17:13:54 <_Trickster_> Modius: CCL no i didn't 17:14:38 <_Trickster_> linux in a VM is solution only for developer's machine 17:14:59 <_Trickster_> it will be *hard* server for each task 17:15:02 legumbre [n=user@r190-135-35-185.dialup.adsl.anteldata.net.uy] has joined #lisp 17:15:36 _Trickster_: good luck! also, read up on fastcgi just to be safe, in case you need to fall back on whatever other language you know better 17:15:56 Davidbrcz [n=david@ANantes-151-1-2-71.w83-195.abo.wanadoo.fr] has joined #lisp 17:16:31 <_Trickster_> thank you, fusss! 17:16:36 actually, CLISP has a native fastcgi module; you can let nginx do the request processing and you do your calculations in Lisp. win win. 17:16:49 <_Trickster_> other way is to use apache with lispmod 17:16:56 <_Trickster_> i will try it as well 17:17:14 <_Trickster_> clisp lucks multithreading 17:18:38 go for mod_proxy instead .. any mod_proxy; both apache, lighttpd and nginx have reverse proxy modules usually named mod_proxy or so .. they all work with hunchentoot and other lisp web servers 17:18:50 i.e., no need for mod_lisp 17:19:46 web server optimization; how come no two people ever have the same "fastest" configuration? 17:19:51 (they tend to support load balancing and stuff too .. round robin .. detection of servers that are down etc.) 17:20:20 <_Trickster_> yes, now i'm running nginx + HUNCHENTOOT + asp.net + tomcat 17:20:40 -!- Lacrymology [n=Tomas@200.69.219.97] has quit [Read error: 110 (Connection timed out)] 17:20:40 lnostdal: lighttpd out performs apache on nearly every category, at least according to its docs 17:20:55 <_Trickster_> i'm developing a web service so i must be sure that all things are cross platform 17:22:39 careful not to make the wrong thing "fastest"; I gained some performance by removing debugging messages early. 17:22:57 _Trickster_: soon clisp will have threads completed. 17:23:08 You can already beta test them. 17:23:17 _Trickster_: holy clustermonkey 17:23:33 did you just say you are running nginx + hunch + asp.net + tomcat? 17:23:36 woooooooooooooooooow 17:23:38 dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has joined #lisp 17:23:57 i mean, thats just cool on a whole different scale :) 17:24:12 <_Trickster_> i just say that 17:24:49 holycow: he is running them on his machine; i.e. each is serving an "it works" page on a different port 17:24:56 not that all four are doing something 17:25:12 syamajala [n=syamajal@140.232.183.111] has joined #lisp 17:25:15 <_Trickster_> fusss: why not? 17:25:17 oh, i thought he kind of hooked up some sort of frankenstein interdependent infrastrcture 17:25:19 hehe 17:25:28 man, that would be something to see :) 17:25:47 <_Trickster_> ok remeber it 17:25:52 <_Trickster_> i show you late 17:25:52 <_Trickster_> r 17:26:14 <_Trickster_> i hope in next month 17:26:40 _Trickster_: please be very careful if this is for work or school 17:26:53 -!- syamajala [n=syamajal@140.232.183.111] has quit [Client Quit] 17:27:14 bighous1 [n=bighouse@modemcable239.238-202-24.mc.videotron.ca] has joined #lisp 17:27:19 just make sure you have something that works ready first, before you make it fast 17:27:26 <_Trickster_> why? 17:27:28 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 17:27:41 <_Trickster_> o i know it 17:27:51 <_Trickster_> all works just fine 17:28:03 <_Trickster_> i need redraw ui 17:28:29 <_Trickster_> this project for university 17:28:38 <_Trickster_> it is my Ph.D. wokr 17:28:44 <_Trickster_> work 17:29:11 -!- spilman [n=spilman@ARennes-552-1-17-7.w92-135.abo.wanadoo.fr] has quit ["Quitte"] 17:29:12 <_Trickster_> or Ms.C. i don't know realtion between our degrees and USA ones 17:29:14 it's been a while since we had the free software debate on c.l.l . I find myself siding with the GNU followers less and less every time... i must be getting old. 17:29:53 drewc: peace of mind is something worth paying for, imo 17:29:57 <_Trickster_> i must paying taxes 17:30:03 <_Trickster_> ) 17:31:03 <_Trickster_> what i just say? 17:32:27 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 17:33:51 drewc: there's some purposes for which the GPL is quite useful. 17:34:10 <_Trickster_> BSD or MIT IS free 17:34:15 fusss: it's like the GNU-crowd thinks there is only one closed-source commercial software company... so any time the debate swings to software quaility, they all say 'but windows is crap' and think they've won the debate. 17:34:50 meh. "it builds better software" is the wrong angle, imo :) 17:34:58 misses the point. 17:35:37 sykopomp: i'm sure there is, and i in fact licence some of my _commercial_ software under GPL... specifically because it takes away some of the freedoms of my clients. 17:35:53 not to mention that in particularly GNU software is nothing to write home about... 17:35:53 drewc: exactly 17:35:53 jfrancis__ [n=jfrancis@66.194.68.209] has joined #lisp 17:35:55 s/ly// 17:36:29 sorry... didn't mean to spill the debate over here too... lets talk about lisp! :) 17:36:47 fe[nl]ix [n=algidus@88-149-208-66.dynamic.ngi.it] has joined #lisp 17:37:20 drewc: this is sort of relevant -- one big reason why I have a lot of MIT-licensed stuff is because the GPL sucks for lisp libraries, and because MIT makes a wider variety of users happy. 17:38:15 sykopomp: can I infer then that you have a lot of lisp libraries? 17:38:22 for languages and libraries, unless you are a GNUhead, permissive libraries make more sense 17:38:37 or trying to make money off them, depending on business plan 17:38:44 sykopomp: and on the other side of things, compare GCL to Allegro ;) 17:39:02 drewc: I'd rather use ACL ;) 17:39:26 pkhuong: i can't think of any sane person who'd choose otherwise :) 17:39:27 _Trickster_: Specialist degree is pretty close to a Master's. 17:39:49 takes the same amount of time, except in some cases where Master's take two or three years 17:39:53 michaelw: for small values of a lot. Out of 11 repos on github, 7 of them are lisp libraries. 17:40:12 soon, you'll be the next Edi 17:40:27 hello 17:40:30 michaelw: nah. Unlike Edi, I'm incompetent :) 17:41:08 phax [n=phax@unaffiliated/phax] has joined #lisp 17:41:43 _Trickster_: until the recent "reforms" to get in line with Bologna model, Russia was one of the few countries to require more years in college for the basic full degree than the US 17:42:47 <_Trickster_> Adamant: i'm an aspirant 17:43:00 <_Trickster_> not student 17:43:24 you mean you want to get into college, or you're a grad student 17:43:42 <_Trickster_> grad 17:43:46 ok, np 17:44:05 <_Trickster_> i end university last year 17:44:21 with a Ph.D? 17:44:25 sweet. 17:45:05 Elench [n=jarv@dsl-217-155-101-18.zen.co.uk] has joined #lisp 17:45:28 <_Trickster_> no 17:45:28 mark doesn't seem to know what he thinks, drewc .. heh .. FOSS is bad .. no, FOSS only bad because govrn. money is stuck in universities .. etc. x) 17:45:37 <_Trickster_> currently i'm a specialist 17:45:57 <_Trickster_> but now i'm studing a next russian degree 17:46:07 ah. 17:46:08 <_Trickster_> "kandidat nauk" 17:47:00 good luck. :) 17:47:10 lnostdal: FOSS itself isn't bad. There's pretty good FOSS projects out there. The problem is assuming that slapping the GPL on something will simply make it good :) 17:47:21 right. 17:47:28 lnostdal: it's not the best debate on the issue c.l.l has ever had, that's for sure. At least KMP isn't involved this time. 17:47:39 yeah everyone knows that it's not good unless you put the GPLv3 on it. 17:47:52 yeah, but then nobody can use it 17:47:58 which makes it even better 17:48:01 :P 17:48:03 :) 17:48:18 foom: don't forget to buy the shirt! 17:48:39 s/buy shirt/drink kool-aid 17:48:51 :P 17:49:19 drewc: and the cost of commercial lisps is pretty much a non-issue 17:49:42 fusss: it's a non-issue because I will refuse to pay that much! :) 17:50:00 so I won't even remotely consider running my code on ACL/LW 17:50:00 -!- Elench is now known as JHVH 17:50:02 LW seems reasonable. 17:50:08 SCL too. 17:50:13 make something useful with it and bill your first client for it 17:50:21 lw charging per user/per platform seams reasonable? 17:50:42 i don't think so 17:50:51 xristos: most lisp shops are one-person shops, unless i am missing something 17:51:06 fusss: i don't know about that 17:51:15 -!- jfrancis_ [n=jfrancis@66.194.68.209] has quit [Read error: 110 (Connection timed out)] 17:51:15 -!- jewel [n=jewel@dsl-242-128-103.telkomadsl.co.za] has quit [Read error: 113 (No route to host)] 17:51:30 fusss: I think the big issue here is that commercial lisps involve a huge barrier of entry to the language. I'm not paying hundreds of dollars just so I can seriously try out a language. 17:51:53 xristos: I'm sure if you approached LispWorks with an order for X>>1 number of licenses, you could haggle. 17:51:55 and there's no way in hell I'm paying even $5 just to grab a copy of a commercial lisp to port libraries to, for portability. 17:52:08 i'm not much of a SBCL hacker, but i find the ability to go M-. M-. M-. and end up inside SBCLs' guts interesting and potentially useful 17:52:28 tmh: maybe so but i don't find any value in lw/allegro 17:52:35 no, you try it out and master it elsewhere. commercial lisps are just an option on the table for when you wann get paid for your work, in return: the fact that there is an established entity you can point at for support is a huge win 17:52:54 sykopomp: you can try out LW for free 17:52:55 in fact up till recently, they did not even support threads without global lock 17:53:03 sykopomp: Both LW/ACL have evaluation versions that are sufficient for exploring the language. 17:53:12 michaelw: doesn't LW give you a crippled version, like ACL? :) 17:53:18 theyr development environments are atrocious 17:53:22 sykopomp: enough to try it out 17:53:22 tmh: I don't want a crippled version of a language. I want the language. 17:53:23 i'd rather use emacs/slime 17:53:30 the gui stuff for lispworks looks great but there are a good few packages which make trying out CL free and relatively easy to install (no setting up required for slime) 17:53:32 i don't really need commercial support, so what am i paying for 17:53:37 "have my cake and eat it, too"? Sure. I can do that with CCL and SBCL 17:53:38 the lisp-in-a-box or whatever it's called 17:53:42 sykopomp: I thought the use-case was someone trying to master CL. 17:53:50 xristos: that's your opinion, I think others disagree 17:54:14 I do most of my development on SBCL+SLIME on linux, but I run windows in a Xen HVM and have a copy of LispWorks there for delivering stuff. It's not that big of a deal. 17:54:19 pkhuong: the two use-cases are "newbie wanting to really try out CL" and "library author trying to write portable code" 17:54:51 sykopomp: I think PCL's interpretation of "portable" is most useful. 17:54:57 I'm not satisfied with being given an artificially crippled version of a language. I'd rather just grab CCL (and get commercial support,to boot) 17:54:57 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 17:55:02 malcolm_reynolds: you were doing the logfile parsing, right? 17:55:29 referring to the email i sent to sbcl-devel? yeah it was parsing web requests in a log 17:55:33 pkhuong: how will I get a library working under LW if I don't have a working copy of LW? 17:55:38 by any definition? 17:55:40 malcolm_reynolds: IIRC, you mentioned hours-long runtimes for your 8M records. what's the limiting factor there? the parsing or the processing? 17:55:42 sykopomp: How does the evaluation version of LW cripple the language? (This is a legitimate question, I'm really not following that argument) 17:55:49 try to hire an "entertainer" for your barbecue party next time; a juggler, a belly dancer, a magician, or a fire swallower would each cost you more than a LW license. I like to think that my peers' work is worth at least just as much ;-) 17:56:16 michaelw: probably the processing. I was getting a list of all the fields (parsed from csv) and selectively processing some of them to be doubles, ignoring some, etc 17:56:17 fusss: What about a fire swallowing belly dancer? 17:56:36 tmh, that'd be the Vista Ultimate edition, no? 17:56:56 sykopomp: what's the point of making something work in LW if you don't use LW? Hence "portable: can be ported". 17:57:00 tmh: those can be sourced from your guests after a few taquilas 17:57:10 fusss: neither LW nor entertainer are in my (or my employer's) budget :D 17:57:30 tmh: 5 hour limit for each session, closes without warning and without saving work. Heap size limit, exits without warning when limit is reached. No init file loading. No image dumping. 17:57:31 hmm, I continue to be humbled when I look at my 5+ year-old code, no matter how often I do this... 17:57:32 now it's even worse because I'm building an association list, as I want to be able to at some point down the road say "put the data through a perceptron using these 5 features" and remembering what index each feature at is much harder than just saying '(:requesttime :requestlength) etc 17:58:00 malcolm_reynolds: might be time to stop using lists. 17:58:00 there's room for improvement but I'm only going to need to process all this once 17:58:09 michaelw: that never seems to end does it... my code from 5 years ago was written by an idiot! 17:58:26 drewc: yeah :) 17:58:43 carrot-on-a-stick .. you know you're really screwed when you stop noticing that, i guess :P 17:58:44 pkhuong: you mean stop using assoclists for the output? i'm constrained that lists is what the csv-parser gives me for input 17:58:47 pkhuong: -someone- uses it, so I generally like putting a little extra effort into trying to get something to run on someone else's platform. Depending on the amount of effort it takes. If it would only take me an hour to patch things up so that a library is supported on clisp, I'll go ahead and do it. 17:58:56 and i'm iterating along the list popping as I go, so no linear lookups into the list 17:59:03 pkhuong: it means more potential users. 17:59:17 drewc: initially, I had hoped for some kind of convergence, but no. 17:59:48 sykopomp: are you complaining the the trial version of the commercial LW is limited? That seems like an odd complaint... that they give you a trial version at all is commendable. 18:00:06 i considered hash tables for the output, but i'm dumping to text files and I quite want to be able to read them rather than having to use some cl-store or something 18:00:49 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit [Remote closed the connection] 18:00:50 drewc: it's not satisfactory for learning lisp. 18:01:01 ausente [n=user7994@187.34.44.222] has joined #lisp 18:01:02 sykopomp: Closing without warning is annoying, but none of the other limitations should hinder one's ability to study CL and learn enough to decide if buying a seat is justified. 18:01:07 drewc: it's probably good enough if you want to toodle around in the REPL, if you're an experienced lisper. 18:01:30 tmh: i find it obnoxious enough that I wouldn't touch it with a 10-foot pole. 18:01:39 sykopomp: why is it not satisfactory for learning lisp exactly? 18:02:02 because it could give the wrong impressions 18:02:16 xristos: that's nonsense 18:02:17 drewc: I don't like being given artificially crippled versions of stuff, and I'm not going to fork over hundred (or thousands) just to have a proper compiler. 18:02:40 sykopomp: ok, but how does that prevent others who are not so prejudice from learning lisp via LW? 18:03:00 it does not prevent them learning lisp 18:03:02 sykopomp: you artifically limit yourself, LW does not do this for you. 18:03:04 sykopomp: You're not forking it over for the compiler, you're forking it over for the IDE, CAPI, CLIM, etc. 18:03:08 it makes the whole process harder than it needs to be 18:03:41 unless of course that someone intends to go into developing lisp applications using commercial implementations 18:03:48 xristos: nonsese .. dowload trial package, run lisp.. learn. Is sbcl somehow easier? 18:03:48 sykopomp: The ability to generate an executable and call for support when it's not doing what you want. 18:04:03 what xristos says is the best I can come up with, really. I don't develop commercial software -- if I did, I would probably mind this less. 18:04:48 drewc: someone learns lisp using lispworks trial, has no plans to buy a commercial license 18:04:51 sykopomp: I remember when I came into Lisp from being a Devstudio user. I wrote off Lispworks because I couldn't figure out how to make a "project" and add source files haha. 18:04:56 but pretty much all the code I've written is open source. I have no interest in dealing with commercial lisps. 18:04:59 -!- fusss [n=chatzill@115.128.9.73] has quit ["ChatZilla 0.9.85 [Firefox 3.0.12/2009070611]"] 18:05:00 he will have to get familiar with opensource implementations eventually 18:05:06 xristos: bullshit. 18:05:08 and probably emacs/slime 18:05:13 lispworks only makes the process harder 18:05:22 xristos: how can you possibly speak for 'someone'? 18:05:33 i dont speak for anyone, merely giving an impression 18:05:48 you made clear that you do not see things the same way 18:05:51 xristos: a prejudiced and subjective impression, yes. 18:06:03 drewc: the moment a lisp image shuts down on me because I reached a 5 hour limit is the moment I rm -rf that shit and move to something that isn't retarded. 18:06:15 drewc: you are a lw licensee yes ? 18:06:17 that's -obnoxious- 18:06:19 please note that i don't use lispworks, trial edition or not. 18:06:32 xristos: no, i use SBCL almost exclusivley. 18:06:32 allright 18:06:37 sykopomp: ACL afaik only limits the heap, not run-time 18:06:45 sykopomp: Agreed, that is obnoxious, even in a trial version. 18:06:56 -!- mattrepl [n=mattrepl@129.174.97.34] has quit [] 18:07:02 p_l: but ACL is prohibitively expensive :\ 18:07:19 sykopomp: is it? you've spoken with franz sales? 18:07:30 Spyderco [n=nash@194.45.110.65] has joined #lisp 18:07:30 sykopomp: corman has much nicer limits - they are placed only on the IDE, and the price is much more acceptable. also, with full license comes redistributable license for generated images 18:07:34 Athas` [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 18:07:38 drewc: last I heard, it was thousands of dollars, although it sounds like franz is open to negotiation. 18:07:50 drewc: they certainly won't give me a full working copy with full permissions for $20, though. 18:08:22 they might if you promised to pay a tithe of your startup's earnings FOREVER! 18:08:30 sykopomp: they might.,, 18:08:31 p_l: what I want to know is why I should even worry about what kind of commercial lisp I should use when I can just grab SBCL or CCL, and not worry about any of that at all. 18:08:40 yeah, what rsynnott said :) 18:08:45 (A Franz person insinuated as much during the discussion of reddit moving off lisp 18:08:45 rsynnott: oh yes, that. 18:08:46 no tx 18:09:26 specially when it comes to CCL -- doesn't Clozure Associates provide commercial support? 18:09:28 sykopomp: if corman had better unicode support, I'd have bought a license already, just for apps that I might want to deploy on windows 18:09:38 sykopomp: LW is in business and still selling compilers... so there are some that might see value.. non? 18:09:38 clearly they need to move to a modern revenue model, like obnoxious tricky dual GPL/commercial licensing 18:09:53 p_l: ccl is usable for win32 these days, no? 18:10:03 -!- kenjin-che [n=kenjin@163.152.180.247] has quit [Read error: 110 (Connection timed out)] 18:10:21 milanj [n=milan@93.87.181.140] has joined #lisp 18:10:44 drewc: google "shareware" and see how many people still make a business out of selling wav->mp3 converters. You can milk plenty of stuff for money. 18:10:51 Can anyone offer any insight as to why a long list printed out with (format t "~S~%" list) is truncated, ie only the first dozen or so elements are printed so it comes out like (a b c d e f g h i j k ...) 18:11:02 -!- Lacrymology1 [n=Tomas@200.69.219.97] has quit ["Leaving."] 18:11:10 malcolm_reynolds: *print-length* 18:11:10 surely the point of ~S is to give you something you can #'read back in? 18:11:13 ah 18:11:24 this was working before somehow.. is it possible turning on optimisations broke that? 18:11:26 drewc: not to mention, that doesn't tell me why -I- should care about LW. Like I said, I don't write commercial software, and if I did, I can get commercial support for CCL. 18:11:33 -!- SandGorgon [n=OmNomNom@122.161.46.116] has quit [Read error: 60 (Operation timed out)] 18:11:55 michaelw: it's currently nil. Should I set it to something really big? 18:12:15 no, nil is fine. hmm 18:12:16 sykopomp: that is a fallacious argument... LW is not a shareware mp3 converter. 18:12:38 sykopomp: you should care about LW because you care about lisp. 18:12:54 i care about lw, thats why i test my libraries there 18:13:02 that doesn't mean that i *must* see value in it 18:13:24 malcolm_reynolds: also *print-level* if the list is contained in some other data structure 18:13:26 rsynnott: yes, and when I get my win64 platform running again, I'm going to do some work to make it more "accessible" for newbies 18:13:27 malcolm_reynolds: if you want reliable printing, wrap it in with-standard-io-syntax 18:13:49 xristos: that's a little circular .. if you see value in having your libraries run on LW, then you see value in LW... non? 18:13:55 michaelw: just wrap the format statement I assume? I'll try that, thanks 18:13:57 rsynnott: getting a not-crashing mcclim demo AND mcclim on win32 would be the best, though :) 18:14:13 drewc: i don't see enough value to make me fork over the $$ 18:14:19 tcr: the list in question is the only argument to format, but i'll take a look at that too 18:14:48 fusss [n=chatzill@115.128.9.73] has joined #lisp 18:14:59 has the crap flood stopped? 18:15:16 sheesh, some of us are on metered 3G 18:15:38 haha 18:15:47 *p_l* always ran over 18:17:00 okay, haven't yet tried the with-standard-io-syntax, but it appears my problem goes away if I turn off optimisations. should that happen? does this need reporting to sbcl-devel? 18:17:31 xristos: you might feel different if you wanted to, say, deliver a small dll or executable on win32 .. especially if you might be getting paid for it. I have a client who might want a win32 version of his app, and would be more than happy to shell out the $$ .. it's cheap compared to what he pays for my time, after all. 18:17:53 the optimisations in question were (declaim (optimize (speed 3) (safety 3))) for the two files of code 18:18:24 malcolm_reynolds: you are likely misdiagnosing 18:18:24 malcolm_reynolds: are you sure your code is safe? 18:18:24 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 18:18:24 oh .. safety 3 .. 18:18:31 i'm not reading things well today. 18:19:08 michaelw: i know, it doesn't really make a lot of sense. hmm. well once I set this going again there's another ~8 hour computation, so I guess in the meantime I'll see if I can put together a small snippet to reproduce the behaviour 18:19:08 drewc: agreed 18:19:29 malcolm_reynolds: cut down your data set for testing 18:19:57 well at the moment i'm just starting it off running and having it print everything to command line as well, then doing C-c C-c 18:19:59 not pretty but it works 18:20:01 xristos: so, that's the value i see in LW, even though i'm not a user... LW could potentially make me money, and that's what i use lisp for after all :) 18:20:17 -!- Spyderco [n=nash@194.45.110.65] has quit [Read error: 104 (Connection reset by peer)] 18:21:08 -!- matley [n=matley@matley.imati.cnr.it] has quit [Read error: 110 (Connection timed out)] 18:21:14 drewc: same here but different; LW to host a webserver on win32 boxes :-) 18:21:34 fusss: yeah, mine is a web based app too :) 18:21:37 if you sell a server app, at least half of the people will want to run it on windows 18:22:06 zoldar [n=zoldar@kyh66.internetdsl.tpnet.pl] has joined #lisp 18:22:15 if you really wanna low ball it you could even run it with Corman 18:22:31 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Success] 18:23:13 Can you reach sun.com? 18:23:30 tcr: doesn't seem so 18:23:59 <``Erik> not working for me, either 18:24:01 PING sun.com (72.5.124.61): 56 data bytes / 10 packets transmitted, 0 packets received, 100% packet loss 18:24:44 fusss: LW is not all that expensive really, for a niche product. 18:25:01 drewc: agreed; 1500 18:25:18 ``Erik: I am updating java as speak 18:25:33 They're moving over to oracle? 18:25:47 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Read error: 104 (Connection reset by peer)] 18:25:55 sun is unreachable! 18:26:02 sykopomp [n=sykopomp@unaffiliated/sykopomp] has joined #lisp 18:26:13 wow, end of an era 18:26:22 blame Larry. 18:26:28 Does this mean my computer's java won't autopatch itself every other day? 18:26:29 pjb: that would explain sudden outages. My Oracle experience (second hand, fortunately) suggests that it's not the most reliable platform :P 18:26:32 mattrepl [n=mattrepl@129.174.112.5] has joined #lisp 18:26:50 PING symbolics.com (64.71.33.143): 56 data bytes / 10 packets transmitted, 10 packets received, 0% packet loss 18:26:56 Symbolics wins! 18:27:07 haha 18:27:14 pkhuong: schwartz, rather 18:28:02 "#1=#1# cannot be printed reliably" means i'm somehow attempting to print a recursive structure, right? 18:28:24 #1=#1# is nothing. 18:28:30 *``Erik* recently was involved in buying 2 new sun machines at work... thinks those're the last new sun machines he'll ever see :/ 18:29:04 ``Erik: do not unwrap them! 18:29:14 malcolm_reynolds: nope, #1# could be anything, but it can't be written readably. 18:29:20 return to eBay ASAP! 18:29:27 <``Erik> heh, they're already racked and running, srry :) nothing fancy about 'em 18:29:36 the only sun machine worth buying was built by Archimedes 18:29:46 <``Erik> <-- works for the gubmint, they get real pissy when ya ebay the new toys 18:30:31 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 18:31:48 at my old employer we also always bought mostly Suns (and some IBM pSeries) 18:31:52 ``Erik: a relative of mine gave be a bunch of ultras through the back loading gate. SAIC ;-) 18:32:23 gustavo [n=gustavo@201-75-224-81-am.cpe.vivax.com.br] has joined #lisp 18:32:47 -!- gustavo [n=gustavo@201-75-224-81-am.cpe.vivax.com.br] has left #lisp 18:32:58 <``Erik> nice, I have an old u5 sitting in my basement, a friend found himself in possession without an idea of what to do with it :) probably a similar story 18:33:06 gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has joined #lisp 18:33:39 -!- saikat [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 18:34:24 okay I appear to have found some combination of settings which has optimisation on, prints the full list, and doesn't complain about #1=#1#. Joy. 18:34:49 #1=#1# is nothing. It cannot be printed. 18:35:01 yeah but I've no idea why it was coming up in the first place 18:35:25 *** - READ from # : #1= #1# is illegal 18:35:39 all i'm dealing with is a load of associations lists which contain exclusively :keywords, strings, and doubles. 18:36:23 #1='#1# is fun though. :) 18:37:05 I'll go with michaelw's interpretation. Your program probably interacts with the runtime (especially the debugger) in ways you don't expect which lead you to misdiagnosing the root problem. 18:38:00 probably not wrong. but the data that's currently coming out the other side looks good, and this a fairly throwaway bit of data mangling. if it was happening in the middle of my gnarly tree manipulations then i'd panic 18:39:01 not an entirely principled approach, maybe, but i have a dissertation deadline to worry about 18:39:20 ``Erik: I've got my Alpha in a funnier way - it was going to be dumped out, some guy that worked for the moving company found it and tried to sell it for parts (he didn't knew anything about it). Once I bought it I found it was in near-mint condition and that dumping it (or selling it in first place) was a crime xD 18:39:43 If I were worried about deadlines I'd probably just change my research topic (: 18:40:40 pkhuong: haha. well i did kind of consider switching the data i'm working on when the company busted my ass yesterday for revealing their utmost internal secrets in the form of... some variable names! 18:41:00 o_O 18:41:41 <``Erik> heh, alpha 26xxx's were neat beasts... if only so many weren't sullied with winNT O:-) 18:41:43 yeah. hot tip: when posting code snippets to paste.lisp.org, don't include all of a) your name b) the company whose data you are working with c) variable names of their database schema 18:41:48 (setf *we-cook-our-books* t) 18:42:03 -!- nunb` [n=user@static-217-133-104-225.clienti.tiscali.it] has quit [Read error: 110 (Connection timed out)] 18:42:36 -!- nanobit [i=nauin@gw.pulsar-zone.net] has quit ["bye"] 18:42:40 it tends to end up as the top match on google quite quickly 18:42:57 Tordek [n=tordek@host163.190-137-177.telecom.net.ar] has joined #lisp 18:42:59 malcolm_reynolds: i got a good laugh out of that, at your expense... sorry ;) 18:43:01 Bah, getting acquainted to an API is so much easier in Lisp, and Slime. 18:43:03 malcolm_reynolds: i peruse lisppaste just for that sometimes. even showed it to a few #lispers. 18:43:06 envi^home [n=envi@220.121.234.156] has joined #lisp 18:43:10 -!- gko [n=gko@122-116-15-138.HINET-IP.hinet.net] has quit [Read error: 60 (Operation timed out)] 18:43:11 checkout #php for lulz 18:43:34 drewc: haha it's fine, i'm not in any lasting trouble as I got the paste pulled 18:43:39 malcolm_reynolds: you can boot a private lisppaste on your intranet. 18:43:46 malcolm_reynolds: glad to hear it :) 18:44:08 i want lisppaste on my fridge 18:44:16 re-arrangable magnetic parens ftw! 18:44:26 although there's a certain beautiful irony that a company took from march until 3 days ago to provide the data they wanted me to work on, but they caught a minor breach of the nda in 2 hours 18:44:26 -!- alexbobp [n=alex@ppp-70-253-87-206.dsl.austtx.swbell.net] has quit [Read error: 60 (Operation timed out)] 18:44:36 they've got their priorities sorted, for sure 18:44:51 -!- envi^home [n=envi@220.121.234.156] has quit [Client Quit] 18:45:10 *they* caught it? ouch! 18:45:57 nanobit [i=nwats@gw.pulsar-zone.net] has joined #lisp 18:46:09 Sun Oracle Systems is back 18:46:10 oh for sure. i posted it, the url had only gone as far as sbcl-devel, and went into uni to meet my supervisor. by the time i got there there was an email to both of us saying "contravention of nda". essentially you could google " vector" and it was top match 18:46:29 they must have automated stuff to check on me or something 18:46:51 heh, NDA. The only NDA I ever signed was probably one of the most scary NDAs I have seen outside of security-clearance-needing jobs :D 18:47:32 technically my university signed this one 18:47:37 -!- gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has left #lisp 18:47:45 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 18:47:47 most NDAs I guess don't include "minimum 2 years in jail on criminal charges" as penalty for breach 18:47:55 but i'm planning to stay here for a phd so it's not really about messing around 18:48:05 p_l: haha yeah that's a pretty serious one 18:48:21 p_l: yakuza NDA? 18:48:37 fusss: personal data protection law - I was working as junior sysadmin at a telco 18:48:56 it really isn't hard to get personal data, billings etc. if you can get root access on DB cluster 18:49:54 gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has joined #lisp 18:50:00 -!- gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has left #lisp 18:50:07 *p_l* wasn't that high in hierarchy to get the access, but I had accesscard to some of the server rooms 18:50:59 best job I had, though 18:51:26 server room monkey? ouch man 18:51:31 was it the thrill of potential impending prison that you liked? 18:51:46 fusss: not really server room monkey - rather imagine sysadmin job without lusers :-) 18:52:15 and when actual physical visits to server room were needed, it was what geeks like - assembly and disassembly of hw :3 18:52:55 surreptitiously slipping one HDD into your trousers to give to the KGB later on 18:53:10 all in properly cared-for server rooms with adequate facilities, the kind that makes BOFHs smile contently at seeing their job appreciated :3 18:53:14 i do programming/admin when i am not doing well; sales and import/export rock! 18:53:35 I just found out last weekend that a wind turbine blade design that I acquired while working at a university, that is the basis for many public reports, can only be presented in public research if the research is performed for a certain government agency, otherwise you're violating the companies IP> 18:53:45 excuse me for my dilettantism, what is generally faster: common lisp or scheme? practically: sbcl or plt scheme? 18:53:56 Totally asinine. 18:54:00 that's a kettle of fish right there 18:54:16 ost [n=user@217.66.22.68] has joined #lisp 18:54:32 I had plans to use it as the basis for some research. 18:54:37 -!- Davidbrcz [n=david@ANantes-151-1-2-71.w83-195.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 18:55:02 tmh: define 'acquiring' a design 18:55:03 npoektop: I'd guess hard to compare - it will heavily deepen on the code you have running, and with two different languages and two completely different implementations... apples to oranges doesn't cover it, I guess :P 18:55:17 npoektop: you'll have to define "faster". Both must fall at approximately the same speed in void. 18:55:40 npoektop: languages are not fast or slow, only implementations.... and the implementations you mentioned are different languages... so hard to compare. 18:55:51 how fast is call/cc in CL? how fast is CLOS in plt? 18:55:51 malcolm_reynolds: We were doing research for the government agency, received the full design data for the blade, then performed modifications of the design, published publicly. 18:56:18 tmh: i see. but you can't do anything else with that design, including expanding on your previous modifications, without their permission 18:56:19 drewc: actually, easy to compare. There's an intersection language of R5RS and Common Lisp, so you can write a program that runs on both. 18:56:32 npoektop: I don't know about plt scheme except that it uses compiling to C, I know that SBCL generates code directly from CL and has for example full native threads 18:57:07 PLT Scheme's Scheme->C compiler is not really seriously used by anyone as far as I know; most PLT Scheme code runs on a byte-code interpreter. 18:57:09 p_l: PLT has a JIT on x86[-64]. 18:57:09 (intersection r5rs common-lisp) http://paste.lisp.org/display/84252 18:57:37 malcolm_reynolds: exactly. It's pretty ridiculous, because the blade is published all over the place, it's just that it has to be published through the correct people. And, it's not even really a blade that is state of the art, the designs only value is the associated data set. 18:57:37 pkhuong: oh, they added JIT? haven't heard, though I keep a drscheme install on my lappy :) 18:58:12 i want to make a very simple database with some supporting scripts. which is lighter for this: sbcl or plt? 18:58:14 tmh: and quite possibly a company outside your country has the exact data and can act on it being public domain ;P 18:58:21 npoektop: sqlite. 18:58:39 pjb: i suppose that's fair, but the comparison isn't of much use IMO :) 18:58:55 p_l: Heh, I can almost guarantee you that the Indians probably have a copy of the design. 18:59:09 i ment: i want to use sqlite with some supporting scripts. which is lighter for this: sbcl or plt? 18:59:11 npoektop: http://www.reddit.com/r/lisp/comments/93ofv/how_fast_does_your_lisp_compute_10000_digits_of_pi/ 18:59:16 drewc: it would still be funny to write a benchmark in this subset! :-) 18:59:33 for just one use case 18:59:42 tmh: if it's published all over the place, then certainly polish intelligence agency has it, the question is when the archivist will find it (seriously) 18:59:57 pjb: indeed :) 19:00:45 fusss: It's a one-liner with CLISP's arbitrary precision floats. 19:01:21 p_l: Well, my plan is to do some more research for the government on this blade, get some interesting results, then try to convince them that it would be to everyone's benefit to make it public. 19:01:26 npoektop: why not try both? 19:01:54 PLT will most probably result in a smaller (in bytes) executable. 19:02:05 drewc: cause my lifetime isn't infinite 19:02:30 tmh: sounds good. just don't upload any code snippets of the research, oh boy... 19:02:30 npoektop: well, trying to get a straight answer out of #lisp is not exactly a time-saving maneuver ;) 19:02:32 npoektop: it doesn't take a life time to evaluate to lisp implementations 19:04:00 everytime i see a winky I do the Sarah Palin "gotcha" wink with the tongue click 19:04:47 npoektop: first question: do you prefer lisp or scheme? If you don't know, then it's probably a good idea to evaluate both languages... they are very different (save for pjb's glorious hack there) 19:04:54 pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has joined #lisp 19:05:47 yeah. even with the overlap, surely scheme implementations are way more optimised for functional solutions to problems, and cl is more geared towards imperative (at heart) solutions 19:06:41 -!- dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has quit ["Verlassend"] 19:06:59 remember folks, we still don't know what kind of problem npoektop is trying to solve fast 19:07:01 i knew this is a bad idea to ask such a question ) it was obvious to try both and do not bother you ) 19:08:11 npoektop: I like CL because of Slime. 19:08:20 npoektop: well, regarding databases - there you have some rather interesting libraries like clsql, postmodern, perec, rofl etc.. and of course, there's slime 19:08:31 malcolm_reynolds: I don't know about more optimised, but scheme certainly worries a lot less about assignments. 19:08:55 PLT in particular warns against repeatedly re-assigning. 19:09:03 p_l: don't forget rucksack, allegrocache, and elephant! 19:09:32 fusss: common lisp looks huge for me, scheme seems light. but i do not know anything about their performance 19:09:32 Scheme also has friendly people. a cluster of those can study SICP in 6 hours. 19:09:36 slime used to run on scheme48 IIRC 19:09:43 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [No route to host] 19:10:00 npoektop: scheme is 'light' in the sense that the base language contains very little. 19:10:07 npoektop: PLT is as big as CL 19:10:23 (SLIME ceased to run with Scheme48 when SLIME's undocumented network protocol incompatibly changed. (This is more a reflection on SLIME than on Common Lisp or Scheme.)) 19:10:31 npoektop: if anything "seems" anything to you, it's probably way too early to worry about performance. ANSI CL is 300 or so pages and teaches Lisp fairly well. I learned from Winston and Horn. 19:10:47 minion: tell npoektop about that-dead-sexy-book 19:10:48 npoektop: have a look at that-dead-sexy-book: 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). 19:11:18 why dead sexy book? 19:11:25 ask Xach. 19:11:31 npoektop: ask Xach 19:11:32 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has left #lisp 19:11:39 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["If technology is distinguishable from magic, it is insufficiently advanced."] 19:12:26 no, better not i guess 19:13:04 jao [n=jao@21.Red-83-43-33.dynamicIP.rima-tde.net] has joined #lisp 19:13:36 i'm reading PCL, and on 200th page i'm still far from writing practical code 19:13:59 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 19:14:11 i'm reading the web version. what chapter is page 200 in? 19:14:13 npoektop: lies. What about the database?! 19:14:24 npoektop: one cannot simply read pcl .. one needs a REPL as well! 19:14:30 sykopomp: How is that "practical"? 19:14:42 tcr: do not question the s-exp db! 19:14:59 it's a hell of a lot more practical than anything i've seen in chapter 3 of any other programming book 19:15:11 except maybe paip but that doesn't really count 19:15:21 PAIP doesn't count? 19:15:38 paip? 19:15:43 minion: paip? 19:15:44 paip: Paradigms of Artificial Intelligence Programming: Case Studies in Common Lisp by Peter Norvig. http://www.cliki.net/paip 19:17:07 drewc: paip isn't an introduction to programming / lisp in the same way. it pretty much gets straight in there with the hardcore stuff, teaching lisp is mostly a by product. or that was my take on it anyway. 19:17:35 i should go to M-x doctor now 19:17:46 npoektop: considering how PCL actually teaches you a significant chunk of lisp, and assumes you're starting from ground zero, it gets practical fairly quick. I think it's a good compromise between teaching you the language properly, and giving you examples of how itt's applied. There's also chapters here and there before the big chunk of practicals that is the last 1/3 of the book, such as 19:17:46 http://gigamonkeys.com/book/practical-a-portable-pathname-library.html 19:18:41 francogrex [n=franco@158.224-247-81.adsl-dyn.isp.belgacom.be] has joined #lisp 19:18:55 -!- xan-afk is now known as xan 19:18:56 does anyone remember a website/project that was translating code samples from a Perl book to various languages? 19:19:49 It could have been Python or PHP as well. OCaml was the first language to have the full example set translated, IIRC. 19:20:13 no idea.. seems like a fairly wacky idea to translate from perl to ocaml 19:20:14 ufff, good night 19:20:19 malcolm_reynolds: well, i learned more about Lisp from PAIP than any other book, but you're right in that it doesn't teach lisp (syntax, semantics, macros etc) all that much... rather it teaches how to use lisp. 19:20:28 fusss: "higher order perl" 19:20:33 iirc 19:20:34 drewc: no way 19:20:50 ah, higher order perl would make sense to translate to ocaml i guess 19:22:11 drewc: yeah I do plan on finishing it some day. it's just kind of intimidating how good norvig is at coding - about a third of the way in there was an 8 line function that did so much stuff i pretty much decided i wasn't ready for what might be in the rest of his book 19:22:37 that's the spirit. 19:22:41 although his conditional debug statement thing from near the beginning, i've expanded that idea and it's all throughout my current project. i'd be sunk without it. 19:22:50 here 19:22:52 http://pleac.sourceforge.net/ 19:23:08 oh, that one. 19:23:10 great site to get quick ideas of what "practical" code looks like in various languages 19:23:27 haha oh god, translating perl to nasm and R 19:23:47 the Common Lisp one is horribly unidiomatic, btw. 19:24:10 fuss: really? if the rest of the code is the quality of the CL code, i'm sceptical 19:24:21 ah, you got to it before me 19:24:21 :) 19:26:10 near the bottom here: http://pleac.sourceforge.net/pleac_commonlisp/arrays.html 19:26:13 it's up to lispers to fix it then; who ever started this did a great job. 19:26:21 ;;The array now has 10001 elements 19:26:22 ;;The index of the last element is 9999 19:26:22 ;;Element #3 is 0 19:26:31 that's literally a whole barrel of wrong 19:27:06 this is maybe for #emacs but maybe someone (who'se using emacs to run common lisp in inferior lisp) knows. There is emacs syntax highlighting with font-lock that for example highlights "defun" in blue etc... but it's specific for emacs lisp (for example setf is not highlighted etc). 19:27:38 you mean for a buffer of common lisp code? 19:27:44 is there some library that is loadable into emacs that is specific for CL syntax highlighting? 19:28:04 i dunno, I'm using lisp-mode and my setf's aren't highlighted. i hadn't really given it much thought 19:28:20 malcolm_reynolds, yes that sort of stuff 19:28:43 look up the documentation for lisp-mode i guess 19:28:59 i assumed it would be guesstimating what lisp dialect you're using based on file extension 19:29:06 francogrex: maybe someone added keywords to highlight. I certainly do have defuns etc. hiligihted in SLIME/Lisp-mode, but I'm not sure if there are any specifics like setf etc. 19:29:16 good nite 19:29:20 -!- fusss [n=chatzill@115.128.9.73] has quit ["ChatZilla 0.9.85 [Firefox 3.0.12/2009070611]"] 19:30:09 p_l, yes defun defstruct etc are highlighted by defzault but i wonder why not setq and setf; I'll ask in #emacs 19:30:49 To prevent angry fruit salad 19:31:02 but seriously people, any ideas on how successive calls to (length *people*) and (1- (length *people*)) can return 10001 and 9999 19:31:06 gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has joined #lisp 19:31:12 -!- gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has left #lisp 19:31:40 grouzen [n=grouzen@91.214.124.2] has joined #lisp 19:32:06 -!- TR2N [i=email@89.180.178.141] has left #lisp 19:32:38 Weak conses. 19:33:15 threads :) 19:33:31 Cosmic rays 19:33:47 M-x butterfly 19:33:53 perhaps when you translate code from perl everything you do is implicitly cursed 19:34:05 vector-pop 19:34:05 -!- ost [n=user@217.66.22.68] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 19:34:17 Hah, "The perl won't wash off." 19:34:19 alexbobp [n=alex@75.55.116.42] has joined #lisp 19:34:51 You can take the code out of Perl, but you can't take Perl out of the code. 19:36:32 i love how a perl cookbook includes such awesome neat tricks as resizing an array 19:37:32 naggum was pretty right regarding perl 19:38:12 naggum was right about a lot of things. 19:38:29 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 19:38:58 i mention only this because i have lot of experience with perl :) 19:39:02 hi 19:39:07 erik naggum? was that the guy who died recently? 19:39:20 yeah 19:40:31 i have to confess I hadn't heard of him before he died. all i've managed to ascertain since is that at times he could instigate some pretty spectacular arguments 19:41:03 Davidbrcz [n=david@ANantes-151-1-2-71.w83-195.abo.wanadoo.fr] has joined #lisp 19:41:16 he had a pretty ascerbic style in the old usenet fashion. 19:41:17 -!- r0bby [n=wakawaka@guifications/user/r0bby] has quit [Client Quit] 19:41:29 -!- Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has quit ["Lost terminal"] 19:41:40 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 19:41:47 r0bby [n=wakawaka@guifications/user/r0bby] has joined #lisp 19:42:27 -!- postamar [n=postamar@x-132-204-252-212.xtpr.umontreal.ca] has quit [] 19:43:03 yeah. not that i'm against that kind of style, but surely it's counterproductive in terms of more people finding, using and enjoying lisp 19:43:15 of course his goal may well not have been that at all... 19:43:24 that wasn't his goal, for the most part. :) 19:43:30 envi^laptop [n=envi@220.121.234.156] has joined #lisp 19:43:32 natch 19:43:38 -!- envi^laptop [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 19:43:46 yes, style matters, not just point you want to make 19:44:28 i prefer dry technical style 19:44:45 yeah. if hardened lispers withdraw from the community because they are sick of the flames, then a valuable resource is lost 19:44:51 that is something you cant find on c.l.l. though 19:45:09 milanj: that's not true, but it's rare 19:45:23 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 19:45:30 -!- sepult [n=buggarag@xdsl-87-78-169-181.netcologne.de] has quit ["leaving"] 19:45:32 tcr, yes, i know, but who wants to look through all of the posts 19:45:57 -!- zophy [n=sy@host-242-6-111-24-static.midco.net] has quit [Remote closed the connection] 19:46:14 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 19:46:37 jthing [n=jthing@165.244.251.212.customer.cdi.no] has joined #lisp 19:46:41 i haven't really spent any significant time there. anything i might have considered going to a newsgroup for I'd now rather come here for. this place has the right combination of banter and people who know their stuff, and i haven't witnessed any epic flamefests yet 19:46:49 milanj: You have to know the people 19:47:26 Happy 10th Annual Sysadmin Appreciation Day everyone! http://sysadminday.com 19:47:58 saikat [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has joined #lisp 19:49:32 garg. the repo for md5 is clotting up clbuild. :P 19:50:15 tessier__: heh, it's also my birthday today :) 19:50:32 tessier__: yay! (even if I'm not employed as a sysadmin right now!) 19:50:38 drewc: happy birthday :) 19:51:44 drewc: Slainte! 19:51:50 -!- saikat [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has quit [Remote closed the connection] 19:52:03 saikat [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has joined #lisp 19:52:05 *malcolm_reynolds* wonders if it's okay to still feel slightly sketchy on the details after reading the "conditions and restarts" chapter of pcl twice in a row 19:52:20 it's a big topic. 19:52:50 thanks Fade, p_l :) 19:52:55 yeah. hopefully it will turn up enough in the practical chapters that i'll be fine 19:52:56 malcolm_reynolds: I found that I didn't understand them until I actually used them. 19:52:57 In my experience restarts are rarly used. 19:53:05 oh and happy birthday drewc! 19:53:15 *Adlai* notices the balloons. 19:53:22 drewc: Happy Birthday! 19:53:35 *Adlai* sings. 19:53:41 malcolm_reynolds: the condition system takes a bit of getting used to. 19:54:21 thanks guys! 19:54:34 yeah. i mean i conceptually get it that it's similar to Java/ C++ stuff but a) you don't unwind the stack and b) high level code can choose amongst the recovery strategies provided by low level code 19:54:37 malcolm_reynolds: in PCL, Peter Seibel does use the condition system a bunch of times later though, both for error handling and for signaling conditions, so you'll at least have examples. 19:55:06 but it's still a bit confusing. oh well i guess it could be worse, i could be trying to understand call/cc 19:55:09 -!- skeptomai [n=cb@67.40.185.246] has quit [Connection timed out] 19:55:21 call/cc is straight forward. ;) 19:55:42 malcolm_reynolds: There's a selling point. "CL: you don't need to understand call/cc" 19:55:42 Oddly anything you are familiar with is.. 19:55:49 malcolm_reynolds: one thing that took me a while to understand is that the bits of code in handlers are all closures, which can be executed -just fine- without unwinding the stack to where they were established. 19:56:01 jthing: tru'tru 19:56:04 *Adlai* recommends Continuations for Curmudgeons 19:56:31 http://www.intertwingly.net/blog/2005/04/13/Continuations-for-Curmudgeons 19:56:35 *drewc* recommends the LAMBDA papers 19:56:54 it seems like call/cc is "simple" in the same way that writing anonymous recursive functions using the Y combinator is "simple".. but more so. 19:57:21 anybody here responsible for the git.b9.com repos? 19:57:57 malcolm_reynolds: maybe, but that site I linked uses a different approach from the ones that I've usually read, and I think it's clearer than "Here's the entire theoretical background behind continuations, and here are a few puzzles. See?" 19:58:03 haven't seen kmr here in years Fade 19:58:07 salex [n=user@216.80.147.206] has joined #lisp 19:58:10 'afternoon 19:58:16 any sbcl hackers about? 19:58:28 pong, I guess. 19:58:57 oh hi pkhuong! how goes? 19:59:34 btw, i checked your sse additions on that m-brot stuff i was doing a while ago. As you suspected, about a 25% perf. improvement 20:00:17 Bummed that I have to rewrite boring code to use < 4G ram ;) 20:00:23 anyway, re sbcl: i'm just curious if there is anything besides an in-place build I need to do to replicate a "binary package" I can relocate 20:00:28 pkhuong: oh, that's no fun 20:01:03 Lacrymology [n=Tomas@200.69.219.97] has joined #lisp 20:01:27 I believe tcr has a patch that could be useful. Failing that, might it be enough to have a shell script set SBCL_HOME correctly? 20:01:42 iow, if i just build a package locally and send it off to our hpcc guys to propogate across clusters, will anything get borked? 20:01:55 yeah, i guess as long as the environment pickst ath up, all is happy 20:02:01 pkhuong: Is this SBCL stuff so that it works on 32bit, or is it lack of access to a 64bit machine? 20:02:16 (SBCL_HOME , I meant) 20:02:18 tmh: not SBCL-related at all. 20:02:39 tmh: Just that I wasn't representing my very sparse DAG sparsely enough. 20:02:54 Hagbard__ [n=rob@c83-252-78-71.bredband.comhem.se] has joined #lisp 20:03:39 salex: I use an executable core. Less variables that way, and Gb ethernet doesn't care about 50 MB. 20:03:47 pkhuong: I have RHEL5.3, 8 cores and 16G, I'd be willing to create an account and give you SSH access. 20:03:48 pkhuong: were you using an adjacency matrix? 20:04:25 If working on RHEL5.3 is acceptable. (It's surprisingly annoying to me.) 20:04:28 pkhuong: that sort of thing tends to happen to us when we say "ok, now what happens when we go from 2D to 3D" scribble scribble ... estimate .. no problem, asl long as we have 32Gb :( 20:04:29 malcolm_reynolds: no. 20:04:38 pkhuong: exec core is maybe a good idea 20:04:50 yikes.. that must be a big graph to fill 4 gigs then. how many nodes? 20:05:01 tmh: you probably don't have enough CPLEX licenses ;) 20:05:03 i'm just trying to avoid either having to explain to hpcc guys how to build it, or have them redo 20:05:38 pkhuong: Mmmm, let me check (cplexp tmh) => nil 20:06:18 on the upside, I have a new stack of dual quadcores to play with and abuse at will 20:06:26 tmh: rewrite isn't a problem really, I just have to whip myself and code it up. 20:06:38 malcolm_reynolds: 90/300/700 nodes at each level. 20:06:41 salex: I have a patch which looks for the core file at where the runtime binary resides. 20:07:09 pkhuong: okay. not a neural net by any chance? 20:07:13 pkhuong: Good luck. 20:07:14 not at all. 20:07:25 tcr: gotcha. can I grab it form somewhere? 20:07:46 salex: Sure from the mailing list. It's two weeks old, I think. 20:07:54 thanks. i'll poke around 20:07:55 heh. i'm clearly a bit too into machine learning, i'm seeing neural nets in fairly arbitrary places.. 20:08:15 there's a lot more to ml (most of it) than neural nets, malcomlm 20:08:25 salex: If it's useful, send a reply to the list. 20:08:39 k. 20:09:00 salex: haha don't worry I know, it's what I do all day. but a 3 level DAG with expanding numbers of nodes just seemed to fit.. 20:09:23 this new cluster is a little mem starved at the momement (only 1gb/core) but i can't complain too much.... 20:09:33 -!- HG` [n=wells@xdsled140.osnanet.de] has quit [Client Quit] 20:09:45 huh. "momement" that's an interesting one 20:10:12 malcolm_reynolds: ah. what sort of ML are you doing? researcher? student? dev? 20:10:32 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Connection timed out] 20:10:35 finishing up my MSc right now, about to start a PhD which will be Computer Vision but fairly heavy on the Machine learning side 20:10:46 right now it's fast semi supervised learning on a graph 20:11:55 malcolm_reynolds: ah, great ... lots of good stuff ahead of you then. Who are you studying with (if you don't mind)? What aspects of vision are you interested in? Using any lisp? 20:12:14 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 20:12:14 -!- milanj [n=milan@93.87.181.140] has quit ["This computer has gone to sleep"] 20:12:29 no worries, I'm happy to discuss it. my supervisor currently is Mark Herbster at UCL, and my PhD supervisor will be Gabriel Brostow also at UCL 20:12:50 and yeah I've done all the code for my project (counts for half of my Msc) in CL 20:13:08 I'm hoping to carry on with exclusively lisp (or as close as possible) for my PhD stuff 20:13:47 fighting the overwhelming ML bias towards matlab is tough.. but it's got to be done. 20:14:09 although that said I might be teaching matlab next year to other Phd students. but hey, gotta buy beer somehow. 20:14:13 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 20:14:13 ah, i don't know that it's really overwhelming. lots of poeple using python + libs 20:14:22 matlab is useful, if limited 20:14:27 usually unless there is involvement from industry, you can get away with using anything you like 20:14:40 that was certainly the case with me 20:14:52 i don't know either of your supers, but not surprising as a) UK, and b) CS vs. maths 20:15:02 lots of people arrive on my MSc course without having done any programming and apart from a little mathematica everyone is just taught matlab 20:15:24 this becomes a problem when we get onto genetic programming and people realise matlab isn't really built for tree manipulation 20:15:27 xristos: yes, but there is momentum that leads many to use matlab for some things, particularly for people whitoug much background 20:15:42 sure that shouldn't influence someone who l 20:15:48 *knows lisp and wants to use it for his phd 20:15:55 and to be fair, it has some very practical strenghts and a lot of useful libs 20:16:00 exactly. it's perfect for people who don't know/care about static vs dynamic languages, don't know about compilers or interpreters, etc 20:16:07 yeah i think i'll have free reign 20:16:13 but no, that shouldn't affect anyone who wants to use somethign else 20:16:20 do they teach any lisp at ucl ? 20:16:26 i was nearly going to do a phd in diffusion mri stuff which would probably have been mostly java. 20:16:30 hmm.. not really sure 20:16:43 i don't know much about the undergrad syllabus, or any of the other msc courses 20:16:49 dv_ [n=dv@85-127-110-139.dynamic.xdsl-line.inode.at] has joined #lisp 20:16:52 anyway, i did all the code for mine in lisp, for what it's worth 20:17:27 they definitely do some functional programming courses 20:17:40 i'd hope so! 20:17:42 i did haskell/prolog 20:17:54 but i haven't heard of any universities in the uk teaching lisp 20:18:00 it was really upsetting when i checked out their functional programming exam. i could answer it all without having done the class, but i wasn't eligible to take the exam either.. 20:18:11 yeah cambridge definitely doesn't in any undergrad courses 20:18:43 they are bringing in a MSc in Advanced Compsci but I don't know if they have decided on the syllabus yet. knowing them it'll include a whole lot of denotational semantics 20:19:19 some more industry-focused unis here in the uk, stopped teaching java and started teaching c++ as main programming language 20:19:42 poor students :/ 20:19:43 which tells a lot about what they are really trying to do 20:20:18 yeah. i heard variously bad things about ucl's undergrad syllabus, in terms of the coding they teach being pretty much whatever the industry asks for 20:20:20 most of our grad students come with no programming capability at all, so starting of with mathematica and/or matlab is pretty practical. opposite problem to having ones who would like to use something else 20:20:38 ucl has a lot of industry links.. cambridge is still holding out and teaching ML first, which is good. 20:20:57 the lecturer wrote the book (literally) so I'm sure he won't be giving up his royalty stream any time soon 20:21:29 it's pretty rare to make anything from that, honestly. not impossible, just mostly doesn't happen 20:21:40 what, make any money? 20:21:44 yes. 20:22:00 vast majority of text books make no money (or negligable) for the authors 20:22:08 unless you are Stewart or the like 20:22:10 i suspect cambridge university professors publishing things through cambridge university press make sure the deal is relatively sweet for them 20:22:23 I'd be surprised 20:22:47 It's inherently low margin. Might do a bit better than usual. 20:23:01 -!- cmo-0 [n=user@92.96.20.45] has quit [Remote closed the connection] 20:23:04 they can keep print runs / distribution relatively low because most of the people in the world buying ML for the working programmer will buy it from one shop in the centre of town, ad 20:23:12 *and all at about the start of october 20:23:34 Everyone I know who has written one has not done it with any expectation of making money out of it, and they've all been right 20:23:41 particularly once you factor in the time 20:23:51 it's a nice way to get your notes in good shape though. 20:25:12 i guess it depends on how long the interest in your book lasts 20:25:29 jleija [n=jleija@user-24-214-122-46.knology.net] has joined #lisp 20:25:35 if it's a long term demand thing it'll be worth it in the end 20:25:48 hence setting the book syllabus for your lecture course 20:26:27 malcolm_reynolds: you might be grossly underestimating the price a professor can put on consultancy. 20:27:01 Maybe. Graduate texts are particularly bad. Can help with tenure & promotion. One person I know figured at the end (after 10+ year run) the time spent on his relatively successful text had been paid at about $1.50/hour 20:27:01 in Aberdeen we have to suffer Java throughout the first year, and probably second as well. third year has OS design course that apparently uses ANSI C and fourth year has "alternative programming models" course taught in Haskell 20:27:23 so almost anything else he had done with the same time woudl have been "better" in some sense 20:27:33 so this is why nobody does it for the money 20:27:38 p_l: no functional programming in CS until the fourth year is A Bad Time 20:27:50 unless you are trying to write the next 1st year calculus or whatever (and that's pretty much sown up) 20:27:56 *sewn 20:27:58 ? 20:27:59 bah. 20:27:59 there's some CLIPS coding in AI course, but AI course likes to shook people with a practical in Prolog in first year (without textbook) 20:28:11 unless they do some theoretical courses starting with lambda calculus. but even so. 20:28:59 malcolm_reynolds: 6.001's replacement is supposedly still rather close to SICP just using Python instead of Scheme 20:29:14 p_l: java is not bad as an introductory language 20:29:34 compared to C for example 20:30:01 I think something non-imperative should be taught as soon as possible, even if they then move onto other things 20:30:02 I think Python would be a nice first language, except for it's ad-hoc scoping rules 20:30:13 tcr: javascript! 20:30:33 The best introductory programming language is a physics class followed by machine architecture followed by assembly. 20:30:43 pkhuong: Perhaps, I've never used it. 20:30:45 xristos: I heard opposite claims even from "industry", saying that students didn't learn to manage memory to the point where they disregarded memory usage completely, thus creating lots of leaks even in Java 20:30:50 Or approaching from the other end you start with Microsoft Word. 20:30:51 yeah, python's probably a better choice than java. but i still reckon it's a shame mit switched away from scheme.. it was such a legendary course 20:31:05 *tessier__* is also bummed MIT switched from scheme 20:31:07 xristos: any program a student writes must include ~8 magic lines of code. Not really a recipe for instigating intellectual curiosity. 20:31:14 But they decided they want to control robots or some such. 20:31:18 yeah, not to mention Order of the Kinghts of Lambda Calculus 20:31:19 Darn ADD generation, ya know 20:31:27 kniiggits? 20:32:09 *Knights 20:32:17 yeah I remember the first java lectures (I had a basic understanding of the language at this point, thankfully) where he has to say "okay, so this stuff here that says public static void main string args... we'll come to that later!" 20:32:22 alright, back to work. thanks pkhuong, tcr 20:32:25 -!- salex [n=user@216.80.147.206] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 20:32:38 Controlling robots is a very cool way to make simple programming tasks sexy. 20:32:57 depends on the robots/tasks 20:33:04 world domination. 20:33:14 Controlling robots is a very cool way to make sexy things look like simple programming tasks. 20:33:19 Yeah I think so, too. I took a course using Lisp on Lego mindstorms 20:33:20 haha 20:33:30 logo, that's where it's at people 20:33:38 Logo embodies it 20:33:47 I don't think lisp would fit on my generation 1 control brick. :) 20:34:25 I'm not sure what interesting things a robot can possibly do. 20:34:42 xristos: blinkenlites >>> System.Out.println. 20:34:42 solving rubics cube! 20:34:55 aside from not destroying itself accidentally 20:35:39 thanks to whoever recommended continuations for curmudgeons.. i'm still slightly in the dark, but the idea of the stack being a tree of frames instead of a linked list of frames (and these tree elements being subject to garbage collection) is definitely a new ray of light on the whole thing 20:35:43 hefner: explore mars? 20:35:57 malcolm_reynolds: no probs :) 20:36:03 hefner: one of my crazier plans included AI-controlled unmanned blimps equipped with simple weaponry (mostly chemical) to use against robbers in park 20:36:20 I like the milling robots that can carve engines out of aluminum billets. 20:36:42 yesterday I read about some drone that can go into humans veins 20:36:53 pkhuong: my msc was half robotics, i don't want to see another robot in my life 20:37:05 pretty terrifying 20:37:05 tcr: seriously? 20:37:09 none of these wonderous things sound feasible for a first year course. 20:37:25 Vacuum cleaner? 20:37:37 hefner: that was beside coilgun computer-controlled artillery (actually simpler than the blimps) 20:37:59 malcolm_reynolds: yeah it'd move by magnetics from the outside, so they could drop accelerator etc. to make it very tiny 20:38:14 tcr: that's highly worrying 20:38:28 -!- mattrepl [n=mattrepl@129.174.112.5] has quit [] 20:38:34 I'd like to build a robotic squirrel-shooting turret, but I don't think the neighbors would approve. 20:38:37 *p_l* plans on usage of such intravenous agents 20:38:49 what does this little guy do? i mean you're set up for a lot of internet meme references like "im in ur heart messin with ur valves" but i presume that wouldn't get funding 20:39:25 hefner: you only need a couple of pylons, it's easy 20:39:36 malcolm_reynolds: it's supposed to clean clutter from the veins 20:39:53 prevent heart attacks 20:40:08 reprap machines are technically robots 20:40:09 haha wow, so if people have veins filled with crap from smoking or high cholesterol, this just pushes it out? 20:40:29 -!- _Trickster_ [i=Trickste@77.232.135.67] has quit [Read error: 110 (Connection timed out)] 20:40:44 malcolm_reynolds: yeah, it's theoretically a better solution than doing bypass surgery. 20:40:53 Sets it free so that it all flows to some other constriction, clogs, and kills you. 20:41:16 i guess. but yeah, i can't help thinking of the logistics of how they can push it out or whatever 20:41:24 gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has joined #lisp 20:41:47 maybe just collect it all in your appendix or something, then take the person's appendix out 20:41:54 the vein clearing is pretty much a freebie 20:42:06 I imagine that if it's just high-density lipids, it could be enough to shake them up so that they don't clog anymore. 20:42:49 ie, they'd then get absorbed by the body, assuming that the person had a sufficient metabolism (which seems highly unlikely since they had that problem in the first place... this is where doctors come in handy) 20:43:01 but as you might be able to tell, I'm just guessing here :) 20:43:04 haha 20:43:07 uh. 20:43:17 interesting though. i look forward to being prescribed robots. 20:44:00 ... or assigned robots. 20:44:02 *Adlai* shudders. 20:44:15 Is there some nice thing to look up mcclim stuff from emacs.. like hyperspec lookup? 20:44:17 *tmh* is happy with a simple roomba. 20:44:30 schme: It's going to come 20:44:42 schme: For the time being, use specbot 20:44:46 tcr: But.. but... hurry up :P 20:44:58 switching over to irssi is a bit of a hassle. But thanks :) 20:45:10 malcolm_reynolds: it might just as well break down those lipids into small chunks unlikely to clog 20:45:12 schme: It's all written, just not approved 20:45:21 tcr: I approve :) 20:45:34 *p_l* wants to replace his whole body with nanomachinery 20:45:35 tcr: That's very good news. 20:45:46 schme: there's some elisp in the mcclim repository for that 20:45:57 oh. I'll go look there. 20:45:58 p_l: pssh. they need to go back to the drawing board until they've got something that will teleport the fat outside the patient's body 20:46:01 then we'll talk 20:46:06 whether my shoddy attempt classifies as "something nice" is another matter 20:46:38 hefner: I'm not so picky. "something that somewhat saves me the trouble of poking around w3m" :) 20:46:51 *hefner* never looks anything from emacs anymore, ever since "SLIME changed all the damn keys" 20:46:53 schme: perhaps Xach's l1sp.org supports mcclim 20:47:04 -!- dv_ [n=dv@85-127-110-139.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 20:47:13 tcr: Good thinking. I'll give it a whirl. 20:47:25 malcolm_reynolds: the people who need to go back to the drawing board, in my opinion, are those who got that fat there in the first place -- our brave new world's fantastic fast food companies. 20:47:29 It's.. a bit of a distraction to look everything up "by hand". 20:47:53 Hello all ... I've installed SBCL 1.0.30 on a new machine and get a bunch of memory faults. Is there anything in particular I should be looking for that is known to cause this? 20:47:56 tcr: it does support mcclim :) 20:47:56 enn pasted "memory fault" at http://paste.lisp.org/display/84530 20:47:57 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:48:09 (also sorry to distract you all from robot madness) 20:48:43 connecting with slime in particular triggers a slew of these errors 20:49:23 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit ["Ex-Chat"] 20:49:33 enn: declaim (safety 3) in your .sbclrc, clear all fasls, try again 20:50:07 platform? 20:50:09 yo, S11001001! What's up? 20:50:29 -!- phax [n=phax@unaffiliated/phax] has quit ["Leaving"] 20:51:01 pkhuong: Linux (Arch distribution), 2.6.30 20:51:10 -!- zoldar [n=zoldar@kyh66.internetdsl.tpnet.pl] has quit ["Ex-Chat"] 20:52:18 hey tcr, making my money off Python these days 20:52:42 trying to get another Weblocks project off the ground, but it's slow going :/ 20:53:18 S11001001: hmm, with (safety 3) I get a type error from some of the CLSQL postgres stuff, maybe that's it 20:53:43 enn: you can get a patch for the postgres thing from my clsql repo 20:54:34 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 20:54:35 S11001001: ah, where is your repo? 20:54:49 http://repo.or.cz/w/clsql/s11.git 20:55:10 if you prefer, cherry 1c926ff 20:55:53 Question: I just recompiled SBCL on my Mac and when I ran it, I got this error: # was compiled for implementation X86-64, but this is a X86. 20:56:07 -!- jfrancis__ [n=jfrancis@66.194.68.209] has quit [] 20:56:20 rpg: cool 20:56:26 Seems like somehow I messed up my environment so that install.sh thinks that my Mac is not an x86_64... 20:56:40 Any idea what ham-fisted maneuver could have caused that? 20:57:10 Only thing I did differently was run make.sh from inside emacs. Could that have messed it up? 20:57:19 rpg: SBCL builds for x86 by default. 20:57:27 rpg: SBCL_ARCH not set? 20:57:34 You want SBCL_ARCH=x86-64 ./make.sh on OS X. 20:58:02 well, FSVO"want" 20:58:27 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 20:59:05 what exactly is the deal with OS X and 64 bit? it seems to skirt round the issue in a way that linux and and windows don't. is it just 64 bit if you have a Core 2 duo and 32 bit on a core duo? 20:59:18 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 20:59:39 i read some article which explained how they were making more bits of the OS run in 64 bit mode.. sounded like voodoo 21:00:07 malcolm_reynolds: it doesn't have the same library coverage in 64bit code as in 32bit 21:00:18 for example Carbon didn't get 64bit version 21:00:43 and Core doesn't do x86-64. 21:00:52 michaelw, pkhuong: Not sure how I managed to get this to work before, and break now. But thanks. Setting ENVAR first seems to work. 21:01:02 core libraries, which is basically all BSD code + IOKit was ported to 64bit long time ago (back with G5, I think) 21:01:14 tompa [n=thomas@h59ec273c.sehjjak.dyn.perspektivbredband.net] has joined #lisp 21:02:25 -!- phf [n=phf@host.icnfull.com] has quit ["Leaving..."] 21:03:24 -!- fiveop [n=fiveop@g229142147.adsl.alicedsl.de] has quit ["humhum"] 21:03:27 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 21:07:10 -!- timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has quit [Remote closed the connection] 21:08:04 -!- tompa [n=thomas@h59ec273c.sehjjak.dyn.perspektivbredband.net] has quit ["leaving"] 21:08:15 tompa [n=tompa@h59ec273c.sehjjak.dyn.perspektivbredband.net] has joined #lisp 21:09:10 beach, therep 21:10:03 -!- morphling [n=stefan@gssn-590e8eb5.pool.einsundeins.de] has quit [Remote closed the connection] 21:10:17 -!- pizzledizzle [n=pizdets@pool-96-250-220-91.nycmny.fios.verizon.net] has quit [Connection timed out] 21:11:57 TDT [n=user@vs1202.rosehosting.com] has joined #lisp 21:13:53 S11001001: thanks ... I'm still getting this NIL is not of type (ALIEN (* T)) error with the clsql fork but I might be able to chaes it down now. 21:14:03 skeptomai [n=cb@67.40.185.246] has joined #lisp 21:15:25 syamajala [n=syamajal@c-75-68-227-231.hsd1.ma.comcast.net] has joined #lisp 21:21:13 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit ["Ex-Chat"] 21:22:56 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 21:22:57 danlei` [n=user@pD9E2F528.dip.t-dialin.net] has joined #lisp 21:23:14 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 21:23:27 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit [Client Quit] 21:23:45 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 21:24:08 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 21:24:21 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit [Connection reset by peer] 21:27:57 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [] 21:28:49 -!- sellout [n=greg@guest-fw.dc4.itasoftware.com] has quit [] 21:29:41 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 21:30:10 milanj [n=milan@93.87.181.140] has joined #lisp 21:30:16 -!- bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has quit [] 21:31:40 amaron [n=amaron@cable-94-189-243-158.dynamic.sbb.rs] has joined #lisp 21:32:30 leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has joined #lisp 21:35:52 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 21:35:55 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 21:37:03 -!- skeptomai [n=cb@67.40.185.246] has quit [Read error: 104 (Connection reset by peer)] 21:39:34 -!- danlei [n=user@pD9E2CC29.dip.t-dialin.net] has quit [Read error: 101 (Network is unreachable)] 21:43:24 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit [Read error: 60 (Operation timed out)] 21:43:46 -!- schme is now known as schmx 21:45:34 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit ["Leaving."] 21:45:45 skeptomai [n=cb@67.40.185.246] has joined #lisp 21:48:16 kleppari_ [n=spa@bitbucket.is] has joined #lisp 21:52:00 bombshelter13p_ [n=bombshel@24.114.232.37] has joined #lisp 21:53:11 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 21:53:29 pstickne [n=pstickne@24.21.76.57] has joined #lisp 21:53:38 -!- cmeme [n=cmeme@boa.b9.com] has quit [Read error: 104 (Connection reset by peer)] 21:53:48 -!- antifuchs [n=asf@baker.boinkor.net] has left #lisp 21:53:52 antifuchs [n=asf@baker.boinkor.net] has joined #lisp 21:54:59 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 21:55:25 bombshelter13p__ [n=bombshel@24.114.232.35] has joined #lisp 21:55:52 bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has joined #lisp 21:56:54 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 21:57:13 -!- bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has quit [Client Quit] 21:58:01 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 21:59:53 cmeme [n=cmeme@boa.b9.com] has joined #lisp 22:00:59 kglovern [n=kglovern@bas1-london14-1096624743.dsl.bell.ca] has joined #lisp 22:01:34 serichse1 [n=harleqin@xdsl-81-173-146-16.netcologne.de] has joined #lisp 22:04:31 -!- Davidbrcz [n=david@ANantes-151-1-2-71.w83-195.abo.wanadoo.fr] has quit [Read error: 113 (No route to host)] 22:06:10 -!- pkok [n=patrick@f102140.upc-f.chello.nl] has quit ["Leaving"] 22:07:03 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 22:10:05 -!- mathrick [n=mathrick@wireless.its.sdu.dk] has quit [Read error: 110 (Connection timed out)] 22:11:59 -!- serichsen [n=harleqin@xdsl-81-173-191-239.netcologne.de] has quit [Read error: 110 (Connection timed out)] 22:12:48 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 22:14:47 -!- bombshelter13p_ [n=bombshel@24.114.232.37] has quit [Connection timed out] 22:21:00 -!- kglovern [n=kglovern@bas1-london14-1096624743.dsl.bell.ca] has quit ["Lost terminal"] 22:21:10 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 22:22:38 -!- bombshelter13p__ [n=bombshel@24.114.232.35] has quit [Connection timed out] 22:22:46 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Read error: 104 (Connection reset by peer)] 22:29:21 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 22:30:23 saikat_ [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has joined #lisp 22:33:13 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 22:35:42 amaron_ [n=amaron@cable-94-189-243-158.dynamic.sbb.rs] has joined #lisp 22:37:50 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 22:38:07 -!- saikat [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has quit [Read error: 60 (Operation timed out)] 22:40:14 -!- Lacrymology [n=Tomas@200.69.219.97] has left #lisp 22:40:31 saikat [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has joined #lisp 22:41:59 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 22:42:16 -!- serichse1 is now known as serichsen 22:45:42 -!- seb- [n=seb@li30-51.members.linode.com] has left #lisp 22:47:10 -!- francogrex [n=franco@158.224-247-81.adsl-dyn.isp.belgacom.be] has quit ["Leaving"] 22:47:21 -!- amaron [n=amaron@cable-94-189-243-158.dynamic.sbb.rs] has quit [Read error: 110 (Connection timed out)] 22:49:53 morphling [n=stefan@gssn-590e8eb5.pool.einsundeins.de] has joined #lisp 22:50:33 brianj_otter [n=brianj@wsip-24-234-246-213.lv.lv.cox.net] has joined #lisp 22:50:46 -!- cp2 [n=will@please.dont.make.me.eatddos.info] has quit [Read error: 104 (Connection reset by peer)] 22:50:48 cp2_ [n=will@69.163.33.38] has joined #lisp 22:51:01 (cos 100000000000000000000000000000000) returns an incorrect value. is that a bug? 22:51:04 -!- a-s [n=user@92.81.46.25] has quit [Connection timed out] 22:51:37 Does anyone know how to get SBCL on Xp Pro to do the equivalent of what LC_CTYPE does on UNIX to enable Unicode handling? 22:51:43 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit [Client Quit] 22:51:49 nurv101 [n=nurv101@bl14-67-238.dsl.telepac.pt] has joined #lisp 22:52:42 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 22:53:12 leo2007: probably, you should report it 22:53:33 -!- kleppari_ [n=spa@bitbucket.is] has quit ["Lost terminal"] 22:54:43 envi^home [n=envi@220.121.234.156] has joined #lisp 22:55:01 -!- Hagbard__ [n=rob@c83-252-78-71.bredband.comhem.se] has quit ["Ex-Chat"] 22:55:10 -!- cp2_ [n=will@69.163.33.38] has quit [Read error: 60 (Operation timed out)] 22:55:13 -!- gugamilare [n=gugamila@201-75-224-81-am.cpe.vivax.com.br] has quit [Read error: 104 (Connection reset by peer)] 22:55:43 JAS415 [n=jon@ip24-250-13-137.ri.ri.cox.net] has joined #lisp 22:55:51 kleppari_ [n=spa@bitbucket.is] has joined #lisp 22:57:52 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 22:58:25 cp2 [n=will@please.dont.make.me.eatddos.info] has joined #lisp 22:59:26 -!- leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has quit ["rcirc on GNU Emacs 23.0.60.1"] 22:59:32 brianj_otter: You need unicode enabled in your build, and then set proper encoding on I/O 22:59:53 Unicode is in the *features*. 23:00:13 brianj_otter: then how does it not work with unicode? 23:00:15 I'm using trivial-utf-8 and it's telling me that the encoding is wrong. 23:00:16 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 23:00:25 So, I'm not sure how to set the encoding for what it needs on XP Pro. 23:00:32 brianj_otter: maybe you have data in UCS-2 23:00:47 -!- saikat_ [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has quit [Connection timed out] 23:01:10 -!- ikki [n=ikki@201.155.75.146] has left #lisp 23:01:14 (defvar *kanji* (make-array 3 :element-type '(unsigned-byte 8) :initial-contents '(229 132 170))) 23:01:29 That's UTF-8 encoding for a Kanji symbol \u512a. 23:01:58 ikki [n=ikki@201.155.75.146] has joined #lisp 23:01:58 NT uses either 8bit codepage/whatever local encoding system is, or UCS2 or UTF-8 (with UCS2 mostly in internally) 23:02:05 leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has joined #lisp 23:02:28 Yes, UCS2 internally in most cases. 23:02:31 Which is fine ... it should hold the decoded string properly. 23:02:57 But, sbcl is complaining about the external-format being cp437, which is the stock XP Pro default code page. 23:03:28 So, I'm guessing there's either an environment var or global in SBCL I can set to tell it to use utf-8. 23:03:36 -!- rread_ [n=rread@nat/sun/x-tgbllwqrzvljbgaz] has quit [Read error: 110 (Connection timed out)] 23:04:09 brianj_otter: you have to set encoding on the stream you read that data from 23:04:10 use it where/when? shouldn't you choose encodings according to what you're doing? 23:04:28 I've got an array of bytes, no stream. 23:04:42 And, I'm pretty sure trivial-utf-8 is using a stream and "relying" on the native encoding being set right. 23:04:54 In Windows, how do I tell SBCL to use UTF-8 for the stream? 23:05:10 sure it isn't blowing up when you try to print it? 23:05:42 Actually, it's dying on UNIX to, so setting LC_CTYPE per google doesn't work either. 23:06:17 actually, if anyone knows how to control SBCL's choice of encoding on the default streams at startup, I'd love to know that too, if only to ensure it never picks utf8 23:06:27 And, hefner, I'm pretty sure it IS dying on printing in UNIX. 23:06:31 I'll check widnows. 23:06:31 brianj_otter: why not simply use unicode directly, without that quirky array code (which looks wrong to me, regarding types at least) 23:06:56 *p_l* has no problems with printing #\U512A 23:07:24 The 'quirky array code' is simply an easy way to ensure there's a consistent test case. 23:07:44 But, our data isn't coming from a stream anyway. It's coming in encoded, and being decoded into bytes, which become strings. 23:07:48 *drewc* can print that fine, linux en-ca.utf8 23:09:22 SB-IMPL::*DEFAULT-EXTERNAL-FORMAT* > :UTF8 FWIW 23:10:18 Setting that doesn't change the failure, unfortunately. 23:11:31 what exactly is the failure? 23:12:02 I think hefner's comment on the error is correct. I think it's dying on output. 23:12:12 If I assign the result to a value isntead of let it leak to the repl, there's no error. 23:12:33 Basically, I've got a string that blows up the stream because the stream can't display it. 23:12:36 so your terminal is not setup for utf8 23:12:52 I'm sure that's correct. 23:12:59 That the terminal is not utf-8. 23:13:12 However, it's not passing data back from the server either. 23:13:32 how are you speaking to this server? 23:13:42 base-64 encoded text pipes. 23:13:55 Since we can't get binary pipes to work portably. 23:14:04 hah. 23:14:06 Between clisp, sbcl, and abcl. 23:14:31 what kind of 'text' are these text pipes? utf8? 23:15:01 They are whatever the default is on the machine running them, since base-64 doesn't require anything beyond low ascii. 23:15:25 WHich is why there's no stream to read/write unicode too. 23:15:39 so what does that have to do with your mysterious string of death? 23:15:43 now i'm confused.. what exactly is the problem? 23:15:43 caoliver [n=oliver@75-134-208-20.dhcp.trcy.mi.charter.com] has joined #lisp 23:15:53 (notice the work /exactly/ again) 23:15:56 word* 23:16:34 Adding the trivial-utf-8 library to encode/decode the bytes causes hte server to cease responding. 23:16:42 At the repl it dies because it can't print the result. 23:17:03 maybe your data can't be represented as utf-8. it is, after all, braindead. 23:17:10 So, I'm trying to get the repl to work again (slime/xemacs is pretty dead the minute Unicode enters hte picture). 23:17:31 so, don't use xemacs .. gnu emacs has not problem with unicode 23:17:51 The other languages handle the symbol. 23:17:57 *hefner* wonders if Rube Goldberg programmed any lisp 23:18:02 ha 23:18:14 So, where there may be issues with surrogates or such later, this at least encodes/decodes to UTF-8. 23:19:08 *hefner* hates unicode under most circumstances, can't sympathize with anyone foolish enough to trifle with it 23:19:51 Hefnet -- I'd love to disagree with you. 23:19:58 Hefner, even ;) 23:20:24 But I can't. 23:20:25 brianj_otter: why does the application need to print utf8 to a non-utf8 terminal? seems like an odd requirement to me 23:20:32 The app doesn't. 23:20:52 maybe there's some unrelated error, but just popping the debugger nukes his whole session because of some string in the backtrace 23:20:53 But it's sure a ton easier to test with the repl then to write entire clients to test lines of code against. 23:21:02 -!- cmm [n=cmm@bzq-79-179-66-150.red.bezeqint.net] has quit [Remote closed the connection] 23:21:02 hefner: if everything supported unicode then there wouldn't be issues 23:21:08 there's too much broken old shit 23:21:15 cmm [n=cmm@bzq-79-179-66-150.red.bezeqint.net] has joined #lisp 23:21:19 The nice thing about standards: there's so many to choose from and some are even followed in some implementations. 23:21:29 brianj_otter: so, use emacs and slime, or a terminal capable of using utf8... 23:21:50 seems like a no-brainer to me ... use tools that support the encoding you are apparently using. 23:22:01 -!- nha [n=prefect@137-64.105-92.cust.bluewin.ch] has quit [Remote closed the connection] 23:22:11 -!- JHVH [n=jarv@dsl-217-155-101-18.zen.co.uk] has quit [Remote closed the connection] 23:22:13 And in Windows Xp Pro 64, with sbcl, that's what I asked ;) 23:22:25 Windows itself obviously handles unicode fine. 23:22:58 -!- ikki [n=ikki@201.155.75.146] has quit ["Leaving"] 23:23:01 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Connection timed out] 23:23:57 Ah ... there's output going to a string stream. 23:24:13 I bet I can setup that stream to be utf-8 in it's setup. 23:24:24 slava: maybe. I find it intolerable that it isn't a bijection converting between UTF-8 and random bytes. But software often has no business dealing with characters. 23:24:33 brianj_otter: I had no problems with SLIME & unicode on win64, though I had used CCL and setup utf-8 in slime (though I;m not sure...) 23:25:15 hefner: the unix culture not distinguishing between binary and text files is a problem too 23:25:19 I get pretty spectacular confusion in slime the minute a unicode string appears. 23:25:37 slava: if they made that distinction, do you really think the world would be less rather than more complex? 23:25:40 However, it's very possible that my slime doesn't have the proper settings. 23:26:07 hefner: it would prevent the issue of programs treating bytes as utf8 and vice versa at least 23:26:46 Maybe a different solution: is there a way to tell the lisp NOT to fail if it can't show a character but to show #\? instead? 23:27:05 slava: believe me, typed file I/O is not that great. There's more problem with char being defined as a single octet 23:27:13 I don't see anything wrong with treating utf8 as bytes, unless you want to have separate text/binary versions of several dozen trivial unix utilities. 23:28:08 p_l: but an octet is not enough to express a char 23:28:25 hefner: that's the thing, all those unix utilities like grep, head, tail need to be aware of character encoding 23:28:29 unless everything standardizes on utf8 23:28:34 and cat would need separate text and binary modes 23:28:35 slava: then just make all text UTF-8, like Plan9 23:28:36 ertc 23:28:41 p_l: that would be great 23:28:51 isn't that what we pretend is the case already? 23:29:03 yeah, but a lot of older stuff assumes latin1, etc 23:29:05 or ascii 23:29:24 Or dies whent he utf-8 leads to a character not present in the code-page for presentation. 23:29:30 standard-character heh 23:30:05 text processing is a gateway to insanity anyway. strings are for humans. 23:30:10 make-string-output-stream doesn't give me a way to provide encoding. 23:30:25 -!- envi^home [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 23:30:50 brianj_otter: that's because strings don't have an encoding 23:30:54 I've got a 'remote' repl that sets *standard-output* to (make-string-output-stream) so when run eval the results are captured and transmitted back to the caller. 23:31:05 they are encoded once you write them somewhere via an external format 23:31:14 brianj_otter: use flexi-streams or babel, this square-peg-round-hole stuff is going to cause nothing but headaches 23:31:32 So, if attempting to "display strings" requires codepage, then doing (format t "~a" string-with-unicode) that won't print on the server's system will die there. 23:32:25 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit ["leaving"] 23:32:45 It looks like any attempt to render the string _to any stream_ is going to die even though the stream is virtual. 23:33:22 latin1 for everybody! 23:33:54 antifuchs: they do. sbcl uses ucs4 internally when compiled with :sb-unicode 23:34:21 fe[nl]ix: a technicality. you're not going to get at the bytes unless you go through internals 23:35:11 and even then you have to treat the bytes as bytes, and not as characters. 23:35:34 mogunus [n=marco@173-9-7-10-New-England.hfc.comcastbusiness.net] has joined #lisp 23:35:53 anyway. sleep. 23:36:55 rread_ [n=rread@nat/sun/x-qagbgjartgtrkxvh] has joined #lisp 23:37:12 mogunus pasted "cl-couch issues" at http://paste.lisp.org/display/84538 23:37:42 Hi all. I'm trying to compile cl-couch with (require 'cl-couch), I have the latest versions of cl-couch and all deps, and I get that error 23:38:21 *brianj_otter* will look at babel 23:39:39 No, that won't fix it. 23:39:40 -!- ausente is now known as ademir 23:39:44 *brianj_otter* checks flexi-streams 23:40:53 just stop using strings. nothing good will come of it. 23:42:12 When accessing lisp via a repl on another machine, no strings makes things shockingly hard. 23:42:35 Since even (+ 2 2) would end up havings its result converted into a string to be shown. 23:43:34 don't be silly. 23:44:23 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 23:44:33 seb- [n=seb@li30-51.members.linode.com] has joined #lisp 23:44:35 Giving up textual output for humans in their native language would be tantamount to saying "No thanks, we don't serve your kind here." 23:45:32 Seems it is easy to implement Lambda Calculus in Lisp but not the inverse......e.g. eq and quote don't seem to be *trivial* in Lambda Calculus right? 23:46:19 only way i read about is to implement lambda calculus in lambda calculus so you can access syntax of the "upper" implementation 23:46:27 inputs 23:47:07 brianj_otter: on the other hand, attempting to use their native language as an internal representation would be absurd, don't you agree? 23:47:40 -!- Athas` [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 23:48:19 The internal representation is UCS4 in SBCL, so the problem doesn't seem to be internal. It seems to be rendering for external consumption, which IS what the need is. 23:48:34 sure, you have to fix your broken slime setup. 23:48:50 a-s [n=user@92.81.135.104] has joined #lisp 23:49:46 -!- milanj [n=milan@93.87.181.140] has quit [Read error: 110 (Connection timed out)] 23:49:59 In production, this runs in a server. I'm pretty sure that the make-string-output-stream that I'm creating and assigning to *standard-output* is being treated as the server's native encoding. 23:50:32 strings don't have encodings.. 23:50:35 So, even though the server's locale and fonts can't show the data, it's never _actually_ being shown on the server. But the rendering to string is using that server local. 23:51:12 The string doesn't have an encoding ... but the stream does. 23:51:22 which stream? 23:51:27 make-string-output-stream. 23:51:33 That gives a stream that will dump into a string. 23:52:22 I do (let ((*standard-output* (make-string-output-stream)) .... ) 23:52:46 why should it have an encoding? you write characters to it, not bytes, and you get a string out. 23:53:03 So, (format t "~a~%" (eval source-from-user)) causes the locale to get checked. 23:54:21 And it's normal for the result of the eval to have foreign characters. Because that's providing real-time access to the running system, and the data includes strings. 23:58:39 -!- seb- [n=seb@li30-51.members.linode.com] has left #lisp