00:00:04 <oerjan> So I shouldn't mention how tomorrow is forecast to be above freezing here for the first time in a while?
00:00:10 -!- FireFly has left (?).
00:00:17 <ehird> it was actually pretty neat
00:00:28 <ehird> yes, it ran on flash too
00:00:32 <ehird> it was smalltalk in a browser
00:01:29 <oerjan> Which, paradoxically, will of course only ensure that we get _more_ ice and dangerous roads.
00:02:31 <oerjan> if "wut" = "i seeee" then i think "i seeee" must mean something other than i think
00:03:05 <oerjan> or are "what", "vut" and "wut" not equivalent?
00:05:11 <ehird> oerjan: not equivalent
00:05:18 <ehird> also, I seeee is kind of sarcastic
00:05:27 <ehird> "what" doesn't mean "what", it means "wow, that was completely incomprehensible"
00:05:36 <ehird> "wut" means "I see your LSD there."
00:05:43 <ehird> "vut" means "huh?"
00:07:28 <ehird> no. that does not work.
00:08:30 <ehird> project euler problem #1: 1 1000 [a,b) [ [ 3 mod zero? ] [ 5 mod zero? ] bi or ] filter sum
00:08:53 <ehird> (find the sum of all multiples of 3 and 5 below 1000)
00:10:10 <ehird> 25*:*25**1-00p010p>00g 3% #v_10g00g+10pv
00:10:20 <ehird> >00g5%#v_10g00g+10p^
00:15:34 <AnMaster> INIT: store 1000 in (0,-1)... write 0 in (0,1)
00:16:31 <AnMaster> (read from 0,0) % 3,compare to 0.. branch...
00:17:06 <AnMaster> if 0 then check 5... compare to 0... branch...
00:18:04 <AnMaster> if 0 it adds it to another variable
00:18:41 <ehird> well, it works, says the author
00:18:47 <ehird> it is befunge-93, I think
00:18:56 <AnMaster> that is all that is left to figure out
00:19:17 <AnMaster> read from 0,0, substract 1, write to 0,0
00:19:52 <AnMaster> and it stored 999 to 0,0, not 10000 to 0,-1
00:20:04 <AnMaster> ehird, indeed the program should work
00:21:07 <oerjan> why is it 25*:*25** rather than simply 25*::**
00:22:12 <AnMaster> oerjan, you could golf that code quite a lot
00:22:35 <oerjan> hm right project euler isn't a golf site
00:22:55 <AnMaster> I mean using swap instead of getting from funge space all the time
00:37:59 -!- Mony has quit ("Quit").
00:59:01 -!- dean has joined.
01:08:16 -!- dean has quit (Remote closed the connection).
01:08:54 -!- Sgeo has joined.
01:15:03 -!- metaphysician has quit ("Oh Noes!").
01:33:18 -!- oerjan has quit ("Good night").
01:44:03 -!- GreaseMonkey has joined.
02:39:07 -!- fizzie has quit (brown.freenode.net irc.freenode.net).
02:39:18 -!- fizzie has joined.
02:47:48 -!- Slereah_ has joined.
02:51:40 -!- oklopol has quit (Connection timed out).
02:51:56 -!- oklopol has joined.
02:57:55 -!- Slereah has quit (Read error: 110 (Connection timed out)).
03:41:59 -!- Slereah_ has quit (Read error: 60 (Operation timed out)).
03:45:15 -!- Slereah has joined.
04:16:08 -!- GreaseMonkey has quit ("Client Excited").
04:30:35 -!- Dewi has joined.
05:01:15 <GregorR> ehird: I see you posted something about WSGIScriptAlias in .htaccess ... did you ever find a workaround?
05:01:24 <GregorR> (Unless this is a different ehird :) )
05:13:02 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
05:21:13 -!- sebbu2 has joined.
05:39:17 -!- sebbu has quit (Read error: 110 (Connection timed out)).
07:10:09 -!- MigoMipo has joined.
07:22:43 -!- Slereah_ has joined.
07:28:19 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
07:35:01 -!- Slereah has quit (Read error: 110 (Connection timed out)).
07:55:44 -!- Deewiant has quit (Remote closed the connection).
07:56:22 -!- Deewiant has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
09:29:31 -!- ais523 has joined.
09:36:27 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
09:36:30 -!- ais523_ has joined.
09:41:58 -!- ais523_ has changed nick to ais523.
09:43:21 -!- ais523 has quit (brown.freenode.net irc.freenode.net).
09:43:21 -!- sebbu2 has quit (brown.freenode.net irc.freenode.net).
09:43:21 -!- psygnisfive has quit (brown.freenode.net irc.freenode.net).
09:43:25 -!- Slereah_ has quit (brown.freenode.net irc.freenode.net).
09:43:25 -!- rodgort has quit (brown.freenode.net irc.freenode.net).
09:43:27 -!- ehird has quit (brown.freenode.net irc.freenode.net).
09:43:27 -!- GregorR has quit (brown.freenode.net irc.freenode.net).
09:43:28 -!- lifthrasiir has quit (brown.freenode.net irc.freenode.net).
09:43:31 -!- Leonidas has quit (brown.freenode.net irc.freenode.net).
09:43:33 -!- AnMaster has quit (brown.freenode.net irc.freenode.net).
09:43:34 -!- bsmntbombdood has quit (brown.freenode.net irc.freenode.net).
09:43:34 -!- Asztal has quit (brown.freenode.net irc.freenode.net).
09:43:34 -!- lament has quit (brown.freenode.net irc.freenode.net).
09:43:35 -!- Deewiant has quit (brown.freenode.net irc.freenode.net).
09:43:35 -!- fizzie has quit (brown.freenode.net irc.freenode.net).
09:43:36 -!- Dewi has quit (brown.freenode.net irc.freenode.net).
09:43:36 -!- comex has quit (brown.freenode.net irc.freenode.net).
09:43:38 -!- chuck has quit (brown.freenode.net irc.freenode.net).
09:43:38 -!- dbc has quit (brown.freenode.net irc.freenode.net).
09:43:40 -!- MizardX has quit (brown.freenode.net irc.freenode.net).
09:43:40 -!- oklopol has quit (brown.freenode.net irc.freenode.net).
09:43:41 -!- ski__ has quit (brown.freenode.net irc.freenode.net).
09:43:44 -!- mtve has quit (brown.freenode.net irc.freenode.net).
09:43:44 -!- ineiros has quit (brown.freenode.net irc.freenode.net).
09:43:45 -!- SimonRC has quit (brown.freenode.net irc.freenode.net).
09:43:45 -!- kerlo has quit (brown.freenode.net irc.freenode.net).
09:43:46 -!- evenant has quit (brown.freenode.net irc.freenode.net).
09:43:46 -!- pikhq has quit (brown.freenode.net irc.freenode.net).
09:43:46 -!- Ilari has quit (brown.freenode.net irc.freenode.net).
09:54:39 -!- Corun has joined.
09:54:39 -!- AnMaster has joined.
09:54:39 -!- Leonidas has joined.
09:54:41 -!- Slereah_ has joined.
09:54:41 -!- rodgort has joined.
09:55:00 -!- Deewiant has joined.
09:55:00 -!- Dewi has joined.
09:55:00 -!- oklopol has joined.
09:55:00 -!- fizzie has joined.
09:55:00 -!- chuck has joined.
09:55:00 -!- comex has joined.
09:55:00 -!- dbc has joined.
09:55:00 -!- MizardX has joined.
09:55:00 -!- ski__ has joined.
09:55:06 -!- mtve has joined.
09:55:06 -!- kerlo has joined.
09:55:06 -!- ineiros has joined.
09:55:06 -!- SimonRC has joined.
09:55:17 -!- evenant has joined.
09:55:17 -!- pikhq has joined.
09:55:17 -!- Ilari has joined.
09:55:21 -!- ais523 has joined.
09:55:21 -!- sebbu2 has joined.
09:55:21 -!- psygnisfive has joined.
09:55:31 -!- ehird has joined.
09:55:31 -!- GregorR has joined.
09:55:31 -!- lifthrasiir has joined.
09:55:41 -!- bsmntbombdood has joined.
09:55:41 -!- Asztal has joined.
09:55:41 -!- lament has joined.
09:56:00 -!- Corun has quit ("Leaving").
10:06:13 -!- ais523_ has joined.
10:06:22 -!- ais523 has quit (Remote closed the connection).
10:06:27 -!- ais523_ has changed nick to ais523.
10:17:48 -!- ais523_ has joined.
10:18:14 -!- ais523 has quit (Nick collision from services.).
10:18:17 -!- ais523_ has changed nick to ais523.
10:24:32 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
10:34:08 -!- Corun has joined.
10:45:24 -!- Slereah has joined.
10:47:28 -!- Slereah_ has quit (Read error: 60 (Operation timed out)).
10:48:39 -!- Corun has quit ("This computer has gone to sleep").
10:51:07 -!- Mony has joined.
10:53:44 -!- ais523 has joined.
11:17:18 -!- Corun has joined.
11:19:08 -!- Corun has quit (Client Quit).
11:21:14 -!- oerjan has joined.
12:09:54 -!- jix has joined.
12:54:14 -!- M0ny has joined.
12:58:56 -!- jix has quit (Read error: 60 (Operation timed out)).
12:59:05 -!- jix has joined.
13:01:13 -!- Mony has quit (Read error: 60 (Operation timed out)).
13:35:45 -!- Judofyr has joined.
13:50:53 -!- oerjan has quit ("leaving").
13:54:10 -!- bsmntbombdood has quit (Read error: 113 (No route to host)).
13:54:26 -!- bsmntbombdood has joined.
14:38:35 -!- ehird has left (?).
14:38:37 -!- ehird has joined.
14:39:37 <ehird> bsmntbombdood: wherza lisp machine
14:39:41 <ehird> 21:01:15 <GregorR> ehird: I see you posted something about WSGIScriptAlias in .htaccess ... did you ever find a workaround?
14:39:43 <ehird> 21:01:24 <GregorR> (Unless this is a different ehird :) )
14:39:47 <ehird> just put it in an httpd.conf :P
14:39:49 <ehird> less elegant? yep. meh.
14:40:23 <ehird> ELLIOTT, bitch ass.
14:40:40 <ehird> ais523: I've been playing with massively overcomitting mmaps
14:40:47 <ehird> I can mmap 2.5 GB but not (size_t)-1
14:42:54 <ais523> ehird: no, I ran into a vampire in minetown and wasn't watching my hitpoints
14:43:01 <ais523> s/minetown/mine's end/
14:43:32 -!- sebbu2 has quit (Read error: 113 (No route to host)).
14:43:52 * ehird decides to write a program to search for the maximum I can mmap
14:47:10 <ehird> which is greater than the memory I actually have..
14:47:42 <ais523> how much swap do you have?
14:48:04 <ehird> ehm... I think OS X has infinite swap
14:48:13 <ehird> I certainly can't configure it
14:49:36 <ehird> ais523: /private/var/vm
14:49:41 <ehird> -rw------T 1 root wheel 64M 11 Feb 21:18 swapfile0
14:49:41 <ehird> -rw------T 1 root wheel 64M 12 Feb 23:26 swapfile1
14:49:41 <ehird> -rw------T 1 root wheel 128M 12 Feb 23:26 swapfile2
14:49:41 <ehird> -rw------T 1 root wheel 256M 16 Feb 00:16 swapfile3
14:49:42 <ehird> -rw------T 1 root wheel 512M 16 Feb 00:27 swapfile4
14:49:42 <ehird> -rw------T 1 root wheel 1.0G 17 Feb 16:31 swapfile5
14:49:59 <ais523> heh, I like the filesizes
14:50:02 <ehird> an odd arrangement indeed
14:50:09 <ais523> it's clearly been using an exponential allocation strategy
14:50:27 <ehird> so I have 2GB of swap
14:50:32 <ehird> ... so, I should try and allocate 4gb
14:50:34 <ehird> maybe a bit less than 4gb
14:51:09 <ehird> http://answers.yahoo.com/question/index?qid=20080926085342AA29rTU "Would a 4GB flashdrive hold up 726 bytes?"
14:51:32 <ehird> ais523: ok, I can mmap 4gb
14:51:37 <ehird> I bet it fails if I +1
14:52:11 <ehird> well, most apps are 32-bit. but Factor and this have been -m64'd.
14:52:29 <ehird> ais523: I'm basically trying to find a O(1) way to find out how much I can mmap without mmap realising I'm being tricksy on BSD/Linux
14:52:32 <ais523> ehird: I'm worried by the "best answer" on that, which implies 726 MB > 4 GB
14:52:36 <ehird> but it seems to be semi-arbitrary
14:52:58 <ehird> I'm worried by Yahoo answers i ngeneral!
14:53:07 <ehird> "The 4 GB Flash drive will work, but I would recommend a pocket media drive. They start from 100 GB and can go up to 500 GB. They are very small (fit in a pocket)."
14:53:13 <ais523> because pretty much every question seems to have stupidity in that order in it
14:53:19 <ehird> BECAUSE WHAT IF YOU HAVE TO STORE _MANY_ 726 BYTE FILES
14:53:24 <ais523> also, the first answer is correct, if not properly explained
14:53:36 <ais523> and it's pretty clear the person asking the question dropped the word "mega" from it
14:53:40 <ais523> well, the word-segment
14:53:49 <ehird> It reads size as 726 bytes and size on disc 4.00KB (4,096bytes).
14:53:59 <ehird> that was clearly copied out from the windows property box
14:54:02 <ais523> well in that case, wtf?
14:54:17 <ehird> times like these make me wish computers required licenses.
14:54:23 <ehird> ais523: they could just memorize the file :P
14:54:47 * ais523 picks the most recent question in the electronics section
14:54:51 <ais523> to get a sort of random sample
14:54:52 <ais523> http://answers.yahoo.com/question/index;_ylt=Apngm_fViZNk8wvQ7TC8QjGzxQt.;_ylv=3?qid=20090218064820AAcLxie
14:54:57 <ehird> http://gyu.que.jp/jscloth/touch.html <- holy feck (enable JS)
14:55:15 <ehird> ais523: haunted computer
14:55:26 <ais523> I was actually wondering if he'd got his computer confused with a TV
14:55:36 <ais523> although it seems to have task manager, so maybe not
14:56:36 <ais523> http://answers.yahoo.com/question/index;_ylt=AtBwU5Fdpb8IgZySgDYGvDCzxQt.;_ylv=3?qid=20090218015557AA7g5TU is another pretty bad one too
14:56:47 <ehird> factor, by the way, is an excellent language made by someone who knows what they're doing and it's also fast, and has great unicode support and tons of libraries/
14:56:51 <ehird> everyone should try it.
14:57:15 <ehird> (also, the workspace UI is like Smalltalk and Emacs put into one, except it integrates with the OS and you can use your own editor)
14:57:34 <ais523> does it work without the UI too?
14:57:44 <ehird> in which case it's just a regular REPL
14:57:47 <ehird> and you can run programs like that
14:57:50 <ehird> including hash-bang style
14:57:59 <ehird> but the UI is really great for development
14:58:24 <ais523> language tied to IDE = bad
14:58:28 <ais523> language which comes with an IDE = good
14:58:39 <ehird> well, the actual editing is outsourced to whatever editor you wish
14:58:46 <ehird> but it provides tons of hooks and documentation and a REPL etc
14:58:58 <ehird> ais523: the UI itself is rather crazy; it's drawn manually with OpenGL...
14:59:12 <ehird> i'd just use a gui toolkit :D
14:59:18 <ehird> hmm, I forgot to mention that factor's a functional, concatenative lang
14:59:30 <ais523> I guessed, I think I've heard of it vaguely
14:59:33 <ehird> basically joy except usable.
15:00:00 <ehird> It's made by the guy who wrote jEdit, if you've heard of that
15:01:08 <ehird> ais523: it's crazy how much random stuff it has
15:02:30 <ehird> it comes with a library for graphs, continuations, a full objective-c bridge with helper functions for making cocoa UIs, a bloody web framework, cellular automata stuff, coroutines, a pong game using the UI, a tetris game using the UI,...
15:02:46 <ehird> oh, and a sudoku solver
15:02:50 <ehird> (the latter ones are just demos, though)
15:03:15 <ehird> it also has proper compiler macros, which is always nice
15:05:03 <ehird> it reminds me of smalltalk quite a lot
15:05:13 <ehird> and it pops up the word's definition in your editor
15:07:42 <ehird> apparently the optimizing native-code compiler is really good, not as good as ocaml though
15:11:33 <ehird> <eiz> ESR wrote "The Cathedral and the Bazaar" and, uh... maintains fetchmail
15:11:39 <ehird> -- 2004-10-12, but not #esoteric
15:11:52 <ais523> how could he forget about C-INTERCAL?
15:12:07 <ehird> he gave up on C-INTERCAL.
15:12:11 <ehird> it was too cool. not crap enough.
15:12:15 <ais523> well, yes, but he still invented it
15:12:26 <ais523> and 2004 is well before I started maintaining it
15:12:35 <ehird> ais523: he thought it was a serious language, and was going to rewrite fetchmail in it
15:12:44 <ehird> when he realised it was a joke, he hastily amended the README to reflect this and abandoned it
15:13:00 <ehird> that's what he wants you to think
15:13:05 <ehird> "Your IP address is: 0:0:0:0:0:0:0:1" -- factor example app
15:13:09 <ehird> oh so fancy, with yer ipv6.
15:13:54 <ehird> so when is everyone going to start using ipv6 again?
15:14:56 <ehird> <eiz> Oh yeah, and he also single-handedly ruined the Jargon File. I hate him =(
15:14:58 <ehird> this guy talks sense.
15:15:10 <ehird> the sad face does it.
15:17:44 <ehird> ais523: 16:15:56 <eiz> Oh, there was that INTERCAL compiler too
15:17:58 <ehird> an eizaphant never forgets
15:23:11 <ehird> haha wow, you can use ^H as a variable name in SBCL
15:23:40 <ehird> you can use ^X though
15:39:01 <ehird> I Have an important request that made me to contact you; I am Mr.Wolak Rakan, I found Your profile very interesting and decided to reach you directly to solicit your assistance and Guidelines in making a business investment and transfer of (£12.5M GBP) to your country within the Next few days.
15:39:13 <ehird> gmail has failed to filter this, so I think I'll play along
15:39:45 <oklopol> will you give me a bit too?
15:40:06 <ais523> some people have even died as the result of scams or scambaiting attempts
15:40:24 <oklopol> why haven't i heard, links
15:40:24 <ehird> I'm kind of doubting sending a few silly emails will result in my death...
15:40:31 <ais523> basically, they got involved with criminals as a result and the criminals didn't like what happened
15:40:57 <ehird> i was going to go for the sillier tactics. like 'what about my fluffy bunny pal'.
15:41:15 <ehird> it's from a "Wolak Rakan"
15:41:18 <ehird> and the sig is in chinese
15:41:22 <ehird> it's an exotic 419!
15:41:39 <ehird> ... the sig links to an ecard service.
15:54:27 -!- psygnisfive has quit (Read error: 110 (Connection timed out)).
16:07:02 -!- BeholdMyGlory has joined.
16:31:16 -!- sebbu has joined.
16:32:00 <ais523> <liquidpele> And no, smilies cannot ever face left just like you can never go back in time.
16:34:33 * ais523 is reading the source to netcat
16:37:05 <ehird> ais523: Hobbit netcat
16:37:16 <ehird> (I prefer the former, as you know...)
16:37:26 <ehird> ais523: it is typical for ~1996 unix hacker sort of culture.
16:37:55 <ais523> nc: invalid option -- '-'
16:38:00 <ais523> I'm guessing it isn't the GNU version, then
16:38:06 <ehird> nc: illegal option -- -
16:38:06 <ehird> usage: nc [-46DdhklnrtUuvz] [-i interval] [-p source_port]
16:38:06 <ehird> [-s source_ip_address] [-w timeout] [-X proxy_version]
16:38:06 <ehird> [-x proxy_address[:port]] [hostname] [port[s]]
16:38:21 <ehird> ais523: does your usage look like that?
16:38:34 <ais523> it just said "nc -h for help"
16:38:47 <ehird> ais523: you have the source, surely you can check the comment header?!?!?!
16:39:10 <ais523> ah yes, it's the hobbit version
16:39:30 <ehird> yep, that's the latest
16:39:53 <ehird> it's solid code, what's not typical about it, ais523?
16:40:05 <ehird> well, okay, it uses poop as a variable name, that is a bit strange.
16:40:17 <ehird> register HINF * poop = NULL;
16:40:23 <ehird> from function HINF * gethostpoop (name, numeric)
16:40:27 <ehird> I assume poop actually means something
16:40:36 <ehird> ais523: they're entertaining
16:40:46 <ais523> how often do you see that in source nowadays?
16:41:16 <ehird> #ifdef GAPING_SECURITY_HOLE
16:41:16 <ehird> char * pr00gie = NULL;/* global ptr to -e arg */
16:41:25 <ehird> i bet -e was a last-minute afterthought
16:41:40 <ehird> oh? did you ask him? :P
16:41:42 <ais523> according to the readme if you turn that on in a suid root version of netcat, you're in trouble
16:42:00 <ehird> I just mean 'pr00gie'
16:42:14 <ehird> incidentally, this guy fell off the face of the earth after releasing a few netcat releases
16:42:16 <ais523> the readme implies that he was preempting stupid packaging by giving the option a scary name
16:42:29 <ehird> to a sufficient degree that bitrot has made google return no relevant results
16:42:49 <ehird> and the only page you can get is the netcat page via the internet archive, IIRC
16:43:02 <ehird> I wonder if this guy is a perl coder
16:43:32 <ais523> no, I think this is the sort of person who invented Carp
16:43:38 <ais523> it's the other way round
16:43:54 <ehird> Debug (("ipoptions ret len %d", x))
16:43:57 <ehird> damn, it's old-school debug macros.
16:44:49 * ais523 suddenly wonders how many commercial programs look like netcat
16:45:06 <ehird> /* can *you* say "cc -yaddayadda netcat.c -lresolv -l44bsd" on SunLOSs? */
16:45:11 <ehird> I don't know, can I?
16:45:40 <ehird> more cursing, though.
16:45:44 <ehird> and less.. working
16:45:53 <ehird> /* If your shitbox doesn't have getopt, step into the nineties already. */
16:45:57 <ehird> wow, a time when systems didn't have getopt
16:46:13 <ais523> at least, not without cygwin or some other library
16:46:16 <ehird> does windows count as a system, relaly
16:46:32 <ehird> helpme();/* exits by itself */
16:46:32 <ehird> bail ("no help available, dork -- RTFS");
16:46:38 <ehird> now why would you not define HAVE_HELP?
16:46:46 <ais523> to make the executable smaller, obviously
16:47:36 <ehird> ais523: nope - smugness
16:47:36 <ehird> #ifdef HAVE_HELP/* unless we wanna be *really* cryptic */
16:48:09 <ais523> tbh, options are one thing that can normally be deduced from the source, unless it's intentionally obfuscated
16:48:39 <ehird> /* None genuine without this seal! _H*/
16:48:48 <ehird> Unforgable electronic signature
16:49:03 <ais523> hey, it didn't say it guaranteed the source was genuine
16:49:08 <ais523> just that the source wasn't genuine without it
16:50:55 <ehird> ais523: when you wanna make an FPGA-type thing you use vhdl or verilog right?
16:51:16 -!- sebbu has quit (Read error: 110 (Connection timed out)).
16:51:30 <ehird> ais523: see i was thinking about bsmntbombdood's yesterday idea of a modern lisp machine in fpga
16:51:40 <ehird> and it wouldnt' feel right to make a lisp machine in anything but something using lisp, y'know?
16:51:46 <ais523> I have no idea how to code up a lisp machine, really
16:51:52 <ais523> and you have to start somewhere
16:52:03 <ehird> well, if you're in the 80s and a company, you make yer own damn chip
16:52:06 <ais523> I mean, I can sort-of understand how to implement Lisp in an imperative manner
16:52:13 <ais523> but not in a behavioural manner
16:52:15 <ehird> oh, you don't implement lisp on the chip directly
16:52:37 <ehird> ais523: what machine code is to C, lisp machine code is to lisp
16:52:55 <ais523> I guessed it would be something like that
16:52:58 <ehird> i.e., a relatively simple translation, with features specificalyl designed to make parts of the latter easier
16:53:16 <ehird> although, lisp machines are closer to Lisp than x86 is to C
16:53:24 <ehird> since they were bsaically designed to do lisp and nothing else
16:58:04 <ehird> ais523: say, that solving a maze with random walk thing in ocaml you said was so fast
16:58:16 <ehird> care to paste your code? I'd be interested in seeing how well factor does, speed-wise
16:58:21 -!- Hiato has joined.
16:58:41 <ais523> ehird: legally can't, I don't have the copyright on it
16:58:50 <ehird> Ouch; copyright assignment?
16:58:53 <ais523> working for a university is ridiculous
16:59:24 <ehird> Just imagine, people do that willingly. With the FSF.
16:59:28 <ehird> They think it's a _good thing_...
16:59:57 <ais523> well, for small submissions, when you don't care about the licence anyway, it makes it easier for the FSF to sue people infringing the copyright
17:00:37 <ehird> if you don't care there's no point suing.
17:00:40 <ehird> I <3 centralization. ... not
17:00:46 <ais523> ehird: if you don't care and they do
17:00:58 <ais523> then you might agree to the requirement so as to get your code in their projects
17:01:02 <ais523> I mean, you don't care but the FSF does
17:01:09 <ehird> I'm saying that the whole concept is ridiculous
17:04:02 -!- Judofyr has quit (Remote closed the connection).
17:04:28 <ehird> the factor CLI reminds me of running the mathematica kernel directly and I don't know why
17:04:28 <ehird> [ehird:/Applications/Factor] % ./factor
17:04:29 <ehird> ( scratchpad ) 1 0 /
17:04:29 <ehird> Type :help for debugging help.
17:04:46 <ais523> well, that's reverse polish, which looks concatenative right away
17:04:58 <ais523> also, I've never tried to run the mathematica kernel directly
17:04:59 <ehird> I just mean the general feel of it
17:05:29 <ehird> [ehird:/Applications/Mathematica.app/Contents/MacOS] % ./MathKernel
17:05:29 <ehird> Mathematica 7.0 for Mac OS X x86 (64-bit)
17:05:29 <ehird> Copyright 1988-2008 Wolfram Research, Inc.
17:05:31 <ehird> Power::infy: Infinite expression - encountered.
17:05:35 <ehird> Out[1]= ComplexInfinity
17:05:40 <ehird> er, sorry for the flood
17:05:53 <ehird> I guess it's just having used a similar GUI interface,then seeing it emulated at a console
17:06:13 <ais523> hmm... they don't seem all that similar
17:06:22 <ais523> they're both REPLs, but I don't see more of a similarity than that
17:06:54 <ais523> in fact the Mathematica kernel reminds me more of the Perl debug REPLs on CPAN than of the Factor one
17:07:08 <ehird> ais523: they're both mainly used via UI interfaces that look essentially the same, but are richer in the UI environment
17:07:25 <ehird> giving a sort of detached feeling. But I'm just odd.
17:09:39 <ehird> also, there's a slight problem with factor's pong demo
17:09:44 <ehird> specifically, the AI is unbeatable.
17:09:56 <ais523> that's not a problem, it just makes the game different
17:10:05 <ais523> as in, changing it from "can you win" to "how long will you last"?
17:10:22 <ehird> ais523: true. but you can't score at all
17:12:20 <ehird> ais523: how much do you reckon my system will let me mmap?
17:12:52 <ais523> it would be nice if it was exactly your free memory + your free disk space
17:12:58 <ais523> I'd be really impressed with Apple if it was that
17:13:02 <ehird> naw, it's a bit less unfortunately
17:13:07 <ehird> I'm thinking it's 3.5GB
17:13:15 <ehird> as I have 2.5GB and 2GB of swap
17:13:37 <ehird> [problem is, determining this in a portable way across linux/bsd
17:14:07 <ehird> hmm nope, more than 4.5GB
17:14:28 <ehird> i can allocate 5gb
17:14:45 * ehird puts sleep(10), to see if Activity Monitor puts it at 'Virtual memory: 5gb'
17:14:55 -!- sebbu has joined.
17:15:24 <ehird> ais523: it pputs it at virtual memory = 7.5gb...
17:16:21 <ehird> i guess it realises it's playing dirty tricks
17:16:40 <ais523> can you quickly write to all that 4.5GB in ten seconds
17:16:47 <ais523> maybe one byte per megabyte for speed?
17:16:58 <ehird> heh, I'll make it write to it
17:17:09 -!- sebbu2 has joined.
17:17:13 <ehird> I'll write to it at 5368709110
17:17:17 <ehird> which is about 10 bytes from the end
17:17:36 <ehird> zsh: segmentation fault ./overcommit
17:18:36 <ais523> well, I'm slightly surprised it's a segfault not some other sort of error
17:18:54 <ehird> ais523: the OS just lies to the program
17:18:58 <ehird> and gives it back a smaller space
17:19:04 <ehird> (that can expand at will)
17:19:55 <ehird> writing to it puts it at 7.5GB real m emory then grows to 1.19GB real memory, then my system lags like fuck and I have to terminate it
17:20:04 <ehird> somewhere in the vicinity of 323000000 in
17:20:11 <ehird> (out of 5368709120)
17:20:17 <ehird> 7.5GB virtual memory
17:20:58 <ehird> ais523: I can write to [323000000] then read it fine
17:21:00 <ehird> (if I do nothing else)
17:21:10 <ehird> presumably, it breaks down if I go further than my free memory
17:21:16 <ehird> which OS X sez is 1.21GB
17:22:55 <ehird> ais523: writing to 5000000000 out of 5368709120 segfaults
17:23:00 <ehird> i can use up to my total memory
17:23:18 <ais523> ah, I can see how that would happen
17:23:26 <ais523> or maybe up to your total memory - the amount the OS absolutely needs
17:23:39 <ehird> yep, [1610612736] segfaults
17:23:47 <ehird> so I'll try 1610000000
17:23:59 <ehird> that also segfaults
17:24:03 -!- sebbu3 has joined.
17:24:10 <ehird> so does 1600000000...
17:25:07 <ehird> ais523: heh, if I assign to [1000000000] activity monitor says 7.56GB, as if I was using all the previous data
17:25:12 <ehird> overcomitting is _weird_
17:25:49 <ehird> ais523: http://answers.yahoo.com/question/index?qid=20080513171119AAYgfzE <- wtf @ best answer
17:26:39 <ais523> I reckon there are people who go around yahoo answers upvoting obviously stupid answers for fun
17:26:56 <ehird> ais523: that was chosen by the asker
17:27:17 -!- sebbu has quit (Connection reset by peer).
17:28:54 <ehird> anyway, here's my current code:
17:29:00 -!- FireFly has joined.
17:29:23 -!- sebbu has joined.
17:29:30 <ehird> ais523: http://pastie.org/393011.txt?key=hgvohjpomjrqkeebq3smg
17:29:45 <ehird> could you try it on your system? I assume it'll fail due to the large amounts, but I'm curious as to what point
17:29:49 <ehird> the mmap? the assignment?
17:30:02 <ais523> ehird: I know from experience not to mess with overcommitting on here
17:30:17 <ehird> I don't think my program can break your computer ...
17:30:25 <ais523> it's taken me half an hour to unthrash it after accidentally using up too much memory when compiling C-INTERCAL
17:30:37 <ehird> ais523: ah, well mine won't use more than 1 integer's worth of memory
17:30:37 <ais523> so it doesn't break the computer, but it makes it unusable for a while
17:30:48 <ehird> since it doesn't use any more than one integer's worth
17:30:49 <ais523> well, I don't have much memory
17:30:58 <ehird> do you have sizeof(int) memory? :P
17:31:06 <ais523> considerably less IIRC
17:31:29 <ehird> what kind of computer is that
17:31:32 <ehird> that cannot hold an int
17:31:41 <ais523> apparently I have 1000.2 MiB of memory
17:32:11 <ais523> ehird: I wasn't sure of the units, I assumed you were referring to an int-range of memory
17:32:17 <ehird> no, I meant one integer :P
17:32:20 <ais523> but that max memory value is pretty suspicious by itself
17:32:30 <ehird> well, technically it stores an integer pointer too.
17:34:45 <ehird> ais523: I predict the mmap will just fail
17:34:55 <ehird> as overcommitting seems to be careful to a degree
17:34:58 <oklopol> for a second there i thought ais523 has a bignum computer.
17:35:05 <oklopol> well i didn't, but i hoped.
17:35:08 <ais523> meh, maybe I'll turn overcommit off and then run it
17:35:14 <ehird> that will definitely fail...
17:35:22 <ehird> oklopol: omg that'd be amazing.
17:35:28 <ehird> i bet the lispms had that
17:35:29 <ais523> or set it to always overcommit even in ridiculous circumstances mode
17:35:44 <ehird> who came up with overcomitting
17:35:50 <oklopol> yessssssss everything should be implemented hardware level, it may be stupid, but damn it's cool.
17:38:08 <ehird> On the structure of the cohomology of nilpotent Lie algebras (2007) - my masters thesis, written under the supervision of Barry Jessup and Paul-Eugene Parent.
17:38:13 <ehird> I think this is the only reason I actually like factor
17:38:28 <ehird> the author wrote that :P
17:38:37 <ehird> and it makes absolutely no sense
17:38:40 <ais523> have you read it, or do you just like the na,e?
17:38:53 <ehird> it reads like it was generated
17:38:54 <ehird> http://factorcode.org/result.pdf
17:40:02 <ais523> ehird: you clearly don't read many academic papers, then
17:40:11 <ehird> I just don't understand them
17:40:21 -!- sebbu2 has quit (Connection timed out).
17:41:05 <ais523> have you read mine about the 2,3 machine, by the way?
17:41:13 <ais523> admittedly, it's rather boring
17:41:17 <ehird> wasn't it typeset in arial?
17:41:26 <ais523> no idea, quite possibly
17:41:30 <ais523> it was just an openoffice file
17:41:48 <ais523> ah, I remember what happened
17:42:00 <ehird> http://www.wolframscience.com/prizes/tm23/TM23Proof.pdf
17:42:04 <ais523> I think I was using a linux-specific font, they wanted to put some mathematica code at the end so I sent them the .odt
17:42:07 <ehird> looks like Univers
17:42:15 <ehird> the actual text is times
17:42:16 <ais523> and they somehow managed to mess up the conversion of the edited version to pdf
17:42:21 <ehird> in fact, I think it's times new roma
17:43:05 <ais523> wow, that Mathematica at the end is ugly
17:43:12 <ehird> it's really long too
17:43:31 <ais523> I hadn't really mastered the habit of concise Mathematica, that requires memorising the entire stdlib really
17:44:18 <ais523> ironically, the Perl does the same thing as the Mathematica and runs a lot faster
17:44:28 <ais523> so if they wanted to show off their prize programming language, they failed
17:45:04 -!- sebbu3 has quit (Connection timed out).
17:46:51 <ehird> It is irritating that SBCL's only UI is emacs.
17:47:14 <ehird> (The non-editing-supporting, dumb-terminal sbcl(1) does not count)
17:47:41 <ais523> obviously, readline should be a wrapper program, not a library
17:47:59 <ais523> that's a lot more UNIXy, and would probably work just as well
17:48:19 <ehird> but sbcl is a mainly closed-world system, like most lisps.
17:48:24 <ehird> you shouldn't have to do that
17:48:38 <ehird> also, it doesn't work as well
17:48:41 <ehird> e.g., tab completion
17:48:53 <ais523> does readline do tab-completion?
17:50:08 <ehird> ais523: if readline was a wrapper, rms couldn't harass the clisp author with false accusations to make him use the gpl.
17:50:17 <ehird> think of the childr^Wfsf
17:50:21 <ais523> yes, I was just thinking the same thing
17:55:23 <ehird> ais523: i wonder if my cynicism rubs off on everyone
17:56:27 -!- sebbu2 has joined.
17:57:26 <ehird> Wonder how much an old lisp machine costs.
17:57:32 <ehird> (A: A lot. Probably.)
17:59:33 <ehird> lol, the scammer replied
17:59:46 <ehird> Thank you for your mail, however I want to bring it to your notice that this business we are trying to do must Need at least 7 working days with full concentration, no matter how engage you may be in order for it to be concluded successfully.
17:59:48 <ehird> [blah blah blah blah]
17:59:59 <ehird> Finally, I want to remind you of the importance for you to keep this business very secret and confidential until this fund is transferred into your account, bearing in mind the nature of what we are doing. If you think we should proceed and agree with the terms then reconfirm your name and address ,telephone and fax number for me to prepare the agreement.I want you to bear it in mind that this is 100% risk free and legal.
18:00:22 <ais523> 7 working days with full concentration? I wouldn't be able to concentrate for that long...
18:00:35 <ehird> i'm tempted to ask some sort of question about child labour
18:05:38 <ehird> what's the emacs way to hook into when an autoload triggers?
18:05:48 <ais523> I can't remember offhand
18:05:51 <ais523> I think it involves add-hook
18:05:57 <ais523> but I'm not sure what hook to hook into
18:08:02 -!- sebbu has quit (No route to host).
18:31:12 <bsmntbombdood> if you have an fpga you can have a piece that's constantly garbage collecting in the background
18:34:02 <ehird> bsmntbombdood: mark-and-sweep is hideously stupid.
18:35:15 <ehird> sure is, if you're doing it for a whole freaking machine
18:35:26 <ehird> also, you can't really do much more than a conservative gc
18:35:43 <ehird> lisp machines run other code too
18:35:48 <ais523> doesn't refcounting work in lisp?
18:35:49 <ehird> and you can access memory directly with them
18:35:56 <ehird> ais523: no. circular data structures.
18:35:56 <ais523> because there's no way to express a cycle of references?
18:36:05 <bsmntbombdood> a lisp machine knows what you can do with pointers in lisp
18:36:09 <ais523> ah, how do you get a circular data structure in lisp?
18:36:28 <ehird> (defvar butts (cons 1 nil)) (replacd butts butts)
18:36:32 <oklopol> in some lisps there's even syntax for naming parts of a structure and referring from within
18:36:37 <ehird> butts ;=> #1#=(1 . #1#)
18:36:42 <ehird> (Printing syntax may be wrong, it's from memory)
18:36:56 <ais523> hmm... I didn't even realise Lisp allowed that sort of thing
18:37:15 <oklopol> hmm right that's scheme's output syntax wasn't it?
18:37:20 <ais523> amusingly, Perl allows that sort of thing but explicitly says it creates a memory leak, the programmer has to break the cycle for the resulting object to be garbage-collected
18:37:32 <ehird> [18:37:15] <oklopol> hmm right that's scheme's output syntax wasn't it?
18:37:35 <ehird> well, it's an extension
18:38:06 <ais523> ehird: yes definitely, I've read the helpfiles about it
18:38:07 <ehird> lisp, in colloquial usage = common lisp and its kin/parents
18:38:12 <ehird> ais523: I was replying to bsmntbombdood
18:38:13 <ais523> oh, you're saying no to bsmntbombdood
18:38:25 <bsmntbombdood> lisp is a class of languages, containing LISP, common lisp, and scheme, among others
18:38:28 <ais523> also, Perl has weaken which I think is one of my new favourite useful yet esoteric keywords
18:38:39 <oklopol> bsmntbombdood: so {scheme} \subset lisp
18:38:49 <ais523> it causes a pointer to not count for the purposes of refcounting, and if the refcount goes down to 0, the pointer immediately becomes undef
18:39:40 <oklopol> or maybe scheme \subset (\union lisp), where \union unions the features of lisps in genera
18:39:49 <oklopol> but that might make a tiny bit less sense
18:40:12 <oklopol> not really, just a bit less.
18:40:54 <oklopol> ais523: python and c have that too, they're just not keywords
18:41:23 <ais523> oklopol: well, C++ needs all memory management to be done by the user, and explicit free
18:41:28 <ais523> so you could implement weaken by hand, I suppose
18:41:34 <ais523> how do you do it in python?
18:42:02 <oklopol> at least i assume that's what they do, could be something slightly different, since i haven't actually looked at them.
18:43:09 <ehird> del is completely unrelated
18:43:33 <oklopol> yeah it is pretty unrelated
18:44:56 -!- Corun has joined.
18:47:04 <oklopol> ais523: well if you just use pointers, they are automatically weak; i'm talking about boost's strong and weak pointers
18:47:36 <ais523> well, they aren't properly automatically weak
18:47:47 <ais523> because they don't become NULL when the thing they point to is freed
18:48:35 <oklopol> i don't see how that's something inherent to a weak pointer
18:48:49 <ais523> well, it's one of their more useful properties
18:49:35 <oklopol> i guess the issue is whether the weak/strong distinction is about technicalities, that is, getting refcounting to work, or whether it's something that's actually used for some purpose
18:49:40 <oklopol> for the latter, they need to be nulled
18:49:53 <oklopol> but i think boost's philosophy is the former
18:50:04 <ais523> ah, the latter is what I was using in my Perl program
18:51:18 <oklopol> i mean without nulling, there is no way to know whether the object has been removed, so the weak pointer will have to have died anyway at some point, or bugs will occur, so they could just as well have been strong, because from their perspective they're always pointing to a living object.
18:51:37 <oklopol> of course, this is kind of a triviality, dunno why i'm explaining it.
18:53:11 <ehird> Poor man's lisp comment: '
18:53:48 <ais523> that's like ()! in Underload/Underlambda
18:54:00 <ehird> it's useful because you don't have to comment out every line of the form
18:54:06 <ais523> or maybe ''' in Python
18:54:14 <ais523> is ''' or """ more common, by the way?
18:54:48 <ehird> the REPL prints out '.' unless the str contains ' in which case it prints "."
18:55:14 <ehird> ais523: common lisp has somethign specifically for it
18:55:23 <ehird> so if you just want something for sbcl you do
18:55:33 <ehird> #+nil obviously comments out the next form
19:06:13 -!- Judofyr has joined.
19:07:55 <ehird> this is rather difficult
19:10:32 -!- M0ny has changed nick to mony.
19:10:35 -!- mony has changed nick to Mony.
19:21:38 <ehird> http://www.isolani.co.uk/blog/standards/Ie8BlacklistForcingStandardsRenderingOptIn ;; Correct fix: write bot to mark all pages as that. distribute it. microsoft reverts change. profit.
19:22:36 <ais523> err... what? Microsoft /re-reverted/ that?
19:22:55 <ehird> nope, they just added an insidious feature that makes it actually neccessary
19:23:09 <ehird> microsoft 1 humans 0
19:23:10 <ais523> the more worrying thing there is the automatically updating list
19:23:19 <ais523> that implies that IE8 sends to Microsoft information on which websites you visit
19:23:22 <ais523> otherwise, how could it update?
19:23:31 <ehird> only for the sites you click the button on
19:24:37 <ais523> anyway, it's been discovered that IE8 has exponential performance on nested absolutely-positioned <div>s
19:24:54 <ais523> it seems that 25 nested absolutely-positioned <div>s is enough to crash most high-spec computers
19:25:26 <ais523> I wonder if Microsoft will fix that one?
19:25:42 <ais523> or if you can just put a website with that markup in, as a logo or something?
19:26:01 <ais523> if the website makes the computer, people will probably blame it on IE not the website
19:26:03 <ehird> i hope the IE team goes and does something more suited for their level of intelligence
19:26:06 <ehird> like, say, playschool
19:26:18 <ais523> actually, I think the IE team are very intelligent and given stupid orders by management
19:26:24 <ais523> such as what to prioritise, and what to implement or not
19:26:29 <ehird> surely they could just follow the orders on a technicality?
19:26:32 <ehird> how do they sleep at night?
19:28:35 <ais523> "Despite all the outreach to sites, we saw from the telemetry data that IE8 Beta 2 users still have to use Compatibility View a lot."
19:28:39 <ais523> why does that scare me a lot?
19:29:28 <ais523> also, they mention Opera does something similar, although it probably isn't at all similar
19:29:35 <ais523> that article looks like it's trying to pre-emptively avoid criticism
19:31:13 <ais523> hmm... http://www.gtalbot.org/BrowserBugsSection/MSIE8Bugs/ looks interesting
19:36:14 <ehird> can't figure out how to hook into slime to get "tell me when the repl buffer is ready"
19:36:18 <ehird> to resize it to be smaller
19:36:29 <ehird> superior lisp interaction mode for emacs
19:36:38 <ehird> it basically imitates the lisp machine editor
19:37:18 -!- olsner has joined.
19:38:36 -!- oerjan has joined.
19:42:08 <ehird> all the cool kids seem to use emacs fully maximized.
19:42:31 <ehird> also, does knuth really expect to be alive to write the 7th volume of taocp?
19:43:31 <ais523> emacs is designed to be fully maximised, I think, as it's meant to be an entire UI not just an editor
19:43:48 -!- Judofyr has quit (Read error: 104 (Connection reset by peer)).
19:48:06 <ais523> as for that IE8 stuff, I'm wondering about submitting it to Slashdot to give kdawson something else to bash
19:48:09 <ais523> do you think that's a good idea?
19:48:40 <ais523> who posts all sorts of anti-Microsoft stories even if they make no sense
19:49:07 <ais523> when there are loads of legitimate reasons to bash Microsoft, why pick stupid ones?
19:50:08 <ais523> ehird: http://tech.slashdot.org/article.pl?sid=09/02/16/2259257&from=rss was one of the stupider ones recently
19:50:29 <ais523> "oh no, I cracked Photoshop by replacing a DLL, now it doesn't work, it must be Windows 7 DRM!"
19:51:48 <ehird> wow, Leopard makes dark windows actually look nice. now I can be a super-leet haxor.
19:52:13 <ais523> actually, dark windows are popular so as to hurt your eyes less
19:52:21 <ehird> i don't have crap eyes :D
19:52:22 -!- Judofyr has joined.
19:52:26 <ehird> i wonder where emacs's scroll-like-the-rest-of-my-system-not-a-jumpy-weird-ass-piece-of-crap setting is
20:00:40 <oerjan> <ehird> ais523: i wonder if my cynicism rubs off on everyone
20:05:38 <ehird> m-xc ustomize is so flaky
20:05:41 <ehird> why can't it write out regular variables
20:11:25 <ehird> ais523: oerjan: new vote: I switch to linux or bsd and use a tiling window manager so this isn't a problem.
20:11:54 <ais523> ehird: "vote"? for what?
20:12:03 <ehird> I am unable to make decisions :-D
20:12:19 * oerjan votes for xmonad, if only because it's in haskell
20:13:02 * oerjan goes back to IE in the other window
20:15:37 -!- ehird[erc] has joined.
20:16:55 -!- ehird[erc] has quit (Remote closed the connection).
20:21:11 -!- Slereah_ has joined.
20:21:30 <lament> I would use emacs if I were an idiot fuckhead moron.
20:21:48 <ehird> i'm a regular guy who wants to use lisp so I have to be an idiot fuckhead moron
20:22:02 <lament> good point, i suppose there's nothing better than slime
20:22:11 <lament> because nobody cares enough about lisp to write something better
20:22:26 <ehird> lament: lisp machine OS, duh.
20:23:13 <lament> the screenshots i've seen of them looked quite horrendous
20:23:26 <ehird> they weren't _pretty_, this was the 80s
20:23:33 <ehird> but they were efficient, usable and highly reflective
20:23:59 <lament> efficient, usable, highly reflective - pick any two
20:24:11 <ehird> on "modern" slum computers, sure. :)
20:24:39 <lament> the reason modern computers suck is not because they don't run lisp on bare metal
20:25:01 <ehird> the point is that you need hooks to allow the high-level features
20:25:05 <ehird> and modern machines, well, don't.
20:25:13 <ehird> so we get to use C! Joy!
20:29:40 <oerjan> some people also use Joy, C?
20:30:32 <ehird> mmph I wonder how to center the emacs frame
20:31:23 -!- Slereah has quit (Read error: 110 (Connection timed out)).
20:33:16 -!- Hiato has quit ("Leaving.").
20:38:33 <ehird> this should be trivial
20:38:33 -!- oklopol has quit (Read error: 104 (Connection reset by peer)).
20:39:54 -!- kar8nga has joined.
20:42:44 <ehird> ais523: do you know how/
20:42:53 <ais523> in what window manager?
20:43:03 <ehird> you can set window positions generically in emacs
20:43:09 <ehird> I've just forgotten how to get the frame's height in pixels
20:43:16 <ais523> I'm not sure what the elisp command for that is
20:43:24 -!- oklopol has joined.
20:43:28 <ais523> it's not the sort of thing that programs generally mess with, so I've never had to look it up
20:43:39 <oklopol> is there a shorter noun for "need to pee"?
20:43:56 <ais523> as far as I know, no to both, though
20:44:18 <oklopol> GregorR: i was wrong, it seems computers using vista can shut down due to hardware failures as well.
20:45:21 <oklopol> ais523: referring to a conversation GregorR probably doesn't even remember anymore.
20:45:51 <oklopol> ehird: is that a joke, or something urbandictionary just hasn't learned yet?
20:46:10 <ehird> oklopol: I invented it. go use it
20:46:16 <oklopol> ais523: well by noun i meant noun, but adjective works too.
20:46:23 <ehird> ais523: what does [[[...]]] around the modeline mean?
20:46:32 <ais523> it means you're in a triply recursive edit
20:46:41 <ais523> you can go back to editing in the middle of certain commands
20:46:49 <ais523> and then pop the edit back to the command when you're done
20:47:04 <ais523> useful in the middle of a long find/replace operation when you notice something else that needs fixing, for instance
20:47:22 <ehird> how can I exit the recedits?
20:47:41 <ais523> I can't remember offhand, maybe C-M-c
20:47:51 <ehird> Debugger entered--Lisp error: (error "Cannot return from the debugger in an error")
20:47:55 <ais523> I know ESC ESC ESC exits a recedit
20:48:03 <ehird> Debugger entered--Lisp error: (error "Cannot return from the debugger in an error")
20:48:12 <ais523> without resuming the command that started the recedit
20:50:36 <ehird> aaaaaargh I forgot how to center :<
20:50:50 <ehird> it's not -(/sh 2)fh
20:50:53 <ehird> (s=screen,f=frame)
20:50:59 * ehird 's brain is off today
20:51:08 <ehird> okay ehird let's think logically :|
20:51:20 <ehird> i wish you hadn't done that
20:51:28 <lament> you're not smart enough for lisp.
20:54:22 <ehird> okay, emacs environment all set up. umm, what was I going to write again?
20:54:52 <ehird> oh look, SBCL's running with 2GB virtual memory. so it definitely does the overcommitting trick
20:55:00 <ehird> ... wtf it's 32-bit :<
21:03:08 -!- kar8nga has quit (Read error: 60 (Operation timed out)).
21:14:54 -!- kar8nga has joined.
21:15:22 -!- ehird has changed nick to ehird|away.
21:38:35 -!- ehird|away has changed nick to ehird.
21:47:17 <oklopol> okokokokokokookokokokokokoo
21:47:26 <oklopol> okokokokokokokokokookokokokokokoko
21:47:46 <ais523> oklopol: you should do evaluations in s-k combinator calculus
21:47:49 <ais523> but with o rather than s
21:48:35 <oklopol> how would you indicate el structure?
21:48:51 <ais523> with o for * and k for
21:49:14 <oklopol> except you can't start or end with a k or have two adjacent k's
21:49:21 <oklopol> okokokokokokokokokookokookoko
21:49:50 <oklopol> why did i start okoing anyway
21:49:55 <ehird> have oko be binary numbers
21:49:57 <ehird> of a godel encoding
22:00:37 -!- fizzie has quit ("reconfigurationationary restart required").
22:07:47 -!- kar8nga has quit (Read error: 104 (Connection reset by peer)).
22:33:56 -!- chuck has changed nick to chuck|busy.
22:34:59 -!- jix has quit ("...").
22:41:25 <ehird> man, the sbcl documentation si refreshing
22:41:40 <ehird> it's comprehensive and informal and it has tons of places where it says, oh yeah, this sucks, or oh this is a bug
22:50:14 <ehird> heh #lisp are talking about lispms
22:57:59 <ehird> I'm downloading a lisp machine environment and emulator!
23:06:07 <Slereah_> What's lisp machine code, anyway?
23:06:14 <Slereah_> What are the differences with regular lisp?
23:06:49 <ehird> Slereah_: well, it's not lisp
23:06:56 <ehird> imagine x86 machine code, and imagine C
23:07:00 <ehird> -> lisp machine code / lisp
23:07:07 <ehird> lisp compiles down to llisp machine code
23:09:44 <Slereah_> Yes, but like, what are the basic functions?
23:10:03 <Slereah_> I hope it's at least RISC, 'cause otherwise, the answer could be long!
23:12:00 <ehird> Slereah_: you'd have to buy the manual to know exactly, probably
23:12:08 <ehird> also, there are multiple lisp machines
23:12:16 <ehird> from difffferent companies
23:13:02 -!- Mony has quit ("Quit").
23:13:07 <Slereah_> Is it something like theorical lisp + lambda + io or something?
23:13:19 <ehird> Slereah_: it's imperative
23:13:29 <ehird> it just has special things for bignums, dispatching, functions, gc
23:13:35 <ehird> Slereah_: umm, lisp is imperative
23:13:53 <Slereah_> I mean, it's not purely functional, but still.
23:13:53 <ehird> Scheme is functional, but lisp is imperative
23:13:59 <ehird> it has first-class functions
23:14:01 <ehird> that doesn't make it functional
23:14:24 <Slereah_> I just assume it's somehow similar to scheme
23:14:33 <ais523> lisp is imperative, but sufficiently functionallish that you can do functional programming in it in a pinch
23:16:30 <ehird> ais523: Slereah_: http://common-lisp.net/project/bknr/static/lmman/frontpage.html
23:16:34 -!- chuck|busy has changed nick to yourwiki-tech.
23:16:46 <ehird> mostly a lisp manual, it seems
23:17:03 <ehird> has assembly stuff
23:17:11 -!- yourwiki-tech has changed nick to chuck|busy.
23:17:18 <ehird> The first instruction here is a CAR instruction. It has the same format as MOVE: there is a destination and an address. The CAR instruction reads the datum addressed by the address, takes the car of it, and stores the result into the destination. In our example, the first instruction addresses the zeroth argument, and so it computes (car y); then it pushes the result onto the stack.
23:17:27 <ehird> Contents of address registeeeeeeeeeeeeeeeeeeeeeee
23:21:05 <ehird> ais523: how much does a working vhdl/verilog environment cost?
23:21:11 -!- chuck|busy has quit (Read error: 60 (Operation timed out)).
23:21:20 <ais523> ehird: GHDL you can get for free, but it's just a simulator
23:21:32 <ehird> does it support everything I'd use?
23:21:37 <ehird> also, what else would I need but a simulator...?
23:22:02 <ais523> I use gtkwave to view simulation output, though
23:22:06 <ais523> and I know how much you hate gtk
23:22:15 <ehird> ok, so ... free, is the answer?
23:22:18 <ehird> what other tools would I need?
23:22:30 <ais523> you need it, gcc I think, and something to view the output
23:22:34 -!- chuck|busy has joined.
23:22:41 <ais523> if you're just doing things with assertions and print statements, you need nothing else
23:22:57 <ais523> if you want to view the internal signals, which is useful when debugging, you need some program to actually show the simulation output
23:23:02 <ais523> and gtkwave's the only one I know of
23:23:31 <ais523> oh, ghdl's slightly buggy in that it sometimes accepts broken code, but you probably don't care about that
23:23:31 <ehird> i thought you said these tools were all highly expensive
23:23:38 <ais523> ehird: the synthesizers, yes
23:23:45 <ehird> oh, write to FPGA?
23:23:56 <ehird> FPGAs are only used for hobbies right?
23:24:00 <ais523> they all rely on the internal details of the chips, you see, which the manufacturers won't release
23:24:03 <ehird> as in no real-world chips us ethem
23:24:04 <ais523> and no, they're used for serious things too
23:24:13 <ais523> no real-world chips use them for the final product
23:24:20 <ais523> but things like Pentiums are simulated on them
23:25:11 <ehird> how fast can fpgas go?
23:25:22 <ais523> insanely fast, some of them
23:25:34 <ais523> even in my student project at University, I was measuring time in nanoseconds
23:26:27 <ehird> ais523: and these synths + the fpga chips cost..
23:26:36 -!- BeholdMyGlory has changed nick to Zetro.
23:26:49 <ais523> I think you can get one of the evaluation packs they market for hobbyists for not a ridiculous amount, although the synth is rather rubbish
23:27:07 <ehird> define not ridiculous
23:28:21 <ais523> $189 for an evaluation board and evaluation software for the bottom-of-the range version
23:28:52 <ehird> ais523: bottom range = crap, I assume
23:29:08 <ais523> the better evaluation boards, with the same software, cost hundreds to thousands of dollars
23:29:13 <ais523> and the better software is price on request
23:29:24 <ehird> yeah I think I'll stick to simulators
23:30:13 <ehird> i wonder how symbolics EVER made a profit
23:30:53 <ais523> you may want to look at opencores.org by the way
23:31:08 <ais523> for example VHDL/Verilog code
23:31:39 <ais523> most of the VHDL/Verilog code you can buy costs a fortune and requires an NDA before you can touch it, opencores.org is free
23:32:55 -!- Zetro has changed nick to BeholdMyGlory.
23:35:02 <ais523> ehird: by the way, GHDL is what happened when someone modified gcc to make it process VHDL rather than C
23:35:09 <ais523> it generates executables that run the simulations
23:36:20 -!- ais523 has quit.