←2010-12-24 2010-12-25 2010-12-26→ ↑2010 ↑all
00:00:04 <Vorpal> fizzie, I think my sb live card has S/PDIF
00:00:18 <Vorpal> but hm
00:00:22 <Vorpal> it is probably all dusty
00:00:25 <fizzie> The USB stick I have has one of those. There's a little piece of plastic that converts it into the usual optical S/PDIF connector: http://ecx.images-amazon.com/images/I/61JztQ1Y7eS._AA1500_.jpg
00:00:27 <Vorpal> since I never used that hole
00:00:46 <fizzie> Don't they tend to have some plastic covers in there?
00:01:13 <Vorpal> fizzie, there is S/PDIF listed in alsamixer, but I don't think I have any connector that isn't either 3.5 mm style or gameport
00:01:16 * Vorpal checks
00:01:30 <fizzie> Which hardware is this?
00:02:19 <Vorpal> fizzie, sb live 5.1
00:02:26 <Vorpal> fizzie, PCI card
00:02:38 <Vorpal> and indeed, only the connector styles I mentioned
00:02:40 <fizzie> My motherboard-integrated Intel HD Audio thing does (coaxial) S/PDIF via a header on the motherboard; had to get the coax connector back-panel plate separately. I guess something similar could be done on some sound cards too.
00:02:41 -!- cheater99 has quit (Read error: Connection reset by peer).
00:03:05 <Vorpal> fizzie, I'm pretty sure the card does /some/ digital output however
00:03:19 <fizzie> http://upload.wikimedia.org/wikipedia/commons/d/d0/Sound_Blaster_Live!_5.1.jpg -- there's something that says "SPDIF" up there.
00:03:43 <fizzie> I'm not entirely sure what the topmost orange connector is either, from this angle.
00:04:09 <Vorpal> fizzie, I could check the back...
00:04:15 <fizzie> Looks like another stereo plug in another image.
00:04:43 <Vorpal> fizzie, can't tell the symbol, it is under the case metal
00:04:54 <fizzie> "ANALOG (center and subwoofer)/DIGITAL OUT (front and rear SPDIF digital outputs) via 4-pole 3.5 mm minijack on rear bracket" says a spec sheet.
00:05:01 <fizzie> So it's actually another hybrid thing.
00:05:10 -!- Slereah has quit (Ping timeout: 272 seconds).
00:05:12 <Vorpal> fizzie, but an electrical sort?
00:05:20 <fizzie> "4-pole" sounds like that, yes.
00:05:32 <Vorpal> fizzie, I only every used green and pink (and gameport)
00:05:35 <Vorpal> ever*
00:08:13 <Vorpal> hm. fizzie you should write a midi->speaker string thingy
00:08:23 <Vorpal> so we can get the whole TOS on there
00:09:14 <fizzie> Right, with some strong-AI modules to understand the semantics and reconstruct the best possible rendition out of a polyphonic midi file.
00:10:43 <Vorpal> fizzie, well, I don't demand that
00:11:10 <fizzie> I don't know the speaker string syntax. Is there a spec for that?
00:11:18 <Vorpal> fizzie, sec
00:11:25 <Vorpal> fizzie, http://www.freebsd.org/cgi/man.cgi?query=speaker&apropos=0&sektion=0&manpath=FreeBSD+8.1-RELEASE&format=html
00:11:29 <Vorpal> fizzie, scroll down a bit
00:11:41 <Vorpal> fizzie, until you see a two column bit
00:11:58 <Vorpal> (though some of the text before and after is relevant too)
00:12:38 <fizzie> It did look quite BASICy indeed.
00:13:24 <Vorpal> fizzie, yeah I think it is similar, with some minor changes
00:13:47 <Vorpal> " The `octave-tracking' feature and the slur mark are new."
00:14:57 <Vorpal> "The MB, MF, and X primitives of PLAY are not useful in a timesharing environment and are omitted."
00:15:01 <Vorpal> I wonder what those did
00:15:45 <fizzie> MF/MB are probably "play in foreground/play in background" switches.
00:15:56 <Vorpal> and X?
00:16:11 <fizzie> And X defererences a pointer, basically. :p
00:16:15 <fizzie> http://zem.fi/~fis/qbc.html#QEw4MDNl
00:16:19 <Vorpal> fizzie, heh?
00:16:43 <fizzie> It's followed by a string representation of a variable's address, and it goes and executes that.
00:16:56 <fizzie> I guess you could also consider it a subroutine call.
00:17:01 <Vorpal> fizzie, why...
00:17:16 <fizzie> You don't need to construct long strings that way.
00:17:25 <Vorpal> fizzie, you could call PLAY several times?
00:17:49 <fizzie> Yes, but that way you can do a hierarchical string that includes other parts several times.
00:17:54 <Vorpal> hm
00:17:55 <Vorpal> okay
00:18:07 <Vorpal> fizzie, well not strange they don't want that, since it was done in the kernel
00:18:08 <fizzie> Note: I don't really know what it's for. The example given is pretty contrived.
00:18:22 <Vorpal> haha
00:19:19 <Vorpal> fizzie, now I want to play monkey island on the pc speaker. I know I seen that on youtube somewhere (it is probably in ~/tmp or maybe /mnt/old-system/home/anmaster/tmp or such
00:20:12 <Vorpal> ah found the MT-32 version of it
00:20:23 <Vorpal> which is awesome too
00:20:59 <Vorpal> wait, no, that is a different version of monkey island
00:21:15 <Vorpal> still, awesome music
00:23:36 <Vorpal> ah found the mt-32 one of the same game
00:23:47 <Vorpal> (as I had the pc-speaker one of)
00:25:46 -!- j-invariant has joined.
00:28:54 <Deewiant> http://www.youtube.com/watch?v=a324ykKV-7Y has a bunch of versions of the monkey island tune for the interested
00:30:47 <fizzie> When I was walking to a friend's graduation party thing, there was a dude at the local train station underpass playing the Monkey Island tune on a melodica.
00:31:50 <fizzie> (Actually it turned out the "dude" (which we only noticed really briefly, as we were in a hurry to get that party) was a quasi-friend of ours, who was also at the same graduation event thing too.)
00:32:23 <fizzie> (Also, we hadn't realized that and started to tell him an interesting story about someone playing the Monkey Island tune under the railway station while we passed by.)
00:32:25 <Vorpal> Deewiant, the one I thought of was http://www.youtube.com/watch?v=7DLoSAb1-bc
00:32:53 <Vorpal> fizzie, XD
00:33:18 <Vorpal> fizzie, you thought it was some beggar or such?
00:33:28 <fizzie> Well, I think they like to be called "street musicians".
00:33:33 <Vorpal> fizzie, ah.
00:33:40 <fizzie> But sure, though the choice of song did make us wonder.
00:33:51 <Vorpal> fizzie, yeah, those poor comp sci students
00:33:52 <fizzie> The "uh, it was actually I" realization was a bit awkwardly entertaining.
00:36:13 <fizzie> The song's a perennial favourite of everybody. I think I've seen a recording of Press Play On Tape performing it somewhere too.
00:36:28 <Vorpal> <Deewiant> http://www.youtube.com/watch?v=a324ykKV-7Y has a bunch of versions of the monkey island tune for the interested
00:36:29 <Vorpal> hm
00:36:41 <Vorpal> I thought MT-32 was a standalone box?
00:36:49 <Vorpal> and that card is FUCKING huge
00:37:14 <Deewiant> The LAPC-I is an MT-32 compatible card
00:37:17 <fizzie> They put it on an ISA card too, I think.
00:37:18 <fizzie> Right.
00:37:21 <Deewiant> That is an ISA card.
00:37:29 <fizzie> I didn't take a look.
00:37:30 <Sgeo> Is it really that acceptable for a language to make no distinction between x(5) and x=5?
00:37:41 <Vorpal> also... are they all from the same version of monkey island?
00:37:56 <Vorpal> I thought it was made for MT32
00:37:58 <Vorpal> -23*
00:37:59 <Vorpal> gah
00:38:01 <Vorpal> -32*
00:38:15 <Vorpal> so it shouldn't sound better on more recent hardware
00:38:20 <Deewiant> They are from the same
00:38:30 <Vorpal> Deewiant, then how can it sound better than MT-32?
00:38:31 <Deewiant> And, as typical for games of the time, it has different versions for different cards.
00:38:45 <Vorpal> Deewiant, yes quite. But some of them were made /after/ monkey island iirc
00:38:48 <Sgeo> http://www.youtube.com/watch?v=h6A0CsU3fh8
00:39:03 <Deewiant> And if you noticed the last annotation, the final one played ("CD audio quality" or whatever it was called) is from a different version
00:39:09 <Deewiant> But the ones before that are from the same.
00:39:15 <Vorpal> Deewiant, yes quite
00:39:21 <Vorpal> ah
00:39:24 <Vorpal> Deewiant, didn't noticed that
00:40:15 <Vorpal> Deewiant, and adlib was quite decent (unlike gameblaster) though nowhere near MT-32 of course
00:40:35 <fizzie> First two versions sound curiously unlike.
00:40:40 <Vorpal> fizzie, yeah...
00:41:00 <Vorpal> the ultrasound one is good, but I think the MT-32 is even better.
00:41:00 <fizzie> Not that there's a very good standard for the PC beeper, of course.
00:41:13 <fizzie> Card looks like a full-size ISA one.
00:41:25 <fizzie> I have a 256K add-on memory card of that size somewhere.
00:41:34 <Vorpal> and SB16 sounded bad
00:42:06 <Vorpal> fizzie, dude I believe in my computer case it would be inside the harddisk bay at that size!!
00:43:38 <fizzie> I don't think there's supposed to be *that* much difference between Adlib's OPL2 FM synthesis and SB16's OPL3; the latter just adds features. (Of course many things that play on SB16 are actually sampled stuff, I guess.)
00:43:50 <fizzie> (If that's in the video, I'm not that far yet.)
00:44:06 <Vorpal> fizzie, the MT-32 sounds better than the SB16
00:44:09 <Vorpal> by far
00:44:18 <fizzie> Well, *that*'s not surprising.
00:44:22 <Vorpal> fizzie, true
00:44:33 <Vorpal> the SCC-1 is nice too
00:44:54 <Vorpal> but the ultrasound is really bad considering it comes after the MT-32
00:45:19 <Vorpal> the SCC-1 is mainly crisper
00:45:37 <Vorpal> which I don't really think fits
00:45:47 <fizzie> I also have a GUS MAX somewhere; GUS is a rather PC-demoscene-famous thing. It's not a bad piece of hardware, compared to the similar-cost SB16.
00:45:54 <fizzie> Has some on-board memory for wavetabley stuff.
00:46:08 <fizzie> Tracker music players could do hardware-accelerated mixing on that.
00:46:17 <Vorpal> fizzie, right, but I don't really care about anything that sounds /worse/ than an MT-32 :P
00:46:31 <fizzie> That just depends what you play with it.
00:46:58 <Vorpal> fizzie, anything designed to make use of MT-32's capabilities :P
00:47:21 <Vorpal> (pc speaker is an exception)
00:47:40 <fizzie> I'm not entirely sure that's a fair benchmark.
00:47:48 -!- cheater99 has joined.
00:47:57 <fizzie> In unrelated news, the new Special Edition re-orchestrated re-played soundtrack doesn't sound half bad.
00:47:59 <Vorpal> (but the segment between pc speaker and MT-32 I find throughoutly boring)
00:48:11 <Vorpal> fizzie, oh? link?
00:48:23 <fizzie> It's that last thing in the video.
00:48:46 <Vorpal> fizzie, didn't it have the issue of not doing the proper transition-and-blend thingy?
00:48:52 <Vorpal> or was that monkey island 2 only?
00:49:10 <Vorpal> "iMUSE" I think it was called
00:50:07 <fizzie> Well, that sounds possible.
00:50:16 <fizzie> Since it's recorded-on-real-instruments stuff.
00:50:17 <Vorpal> fizzie, iirc elliott played it or something
00:50:31 <Vorpal> fizzie, basically you got lots of CD seeking even
00:51:19 <fizzie> I haven't actually played it. (Though I did try out that episodic what-was-it-called modern thing, which was I-guess-nice but nothing that much to, as they say, write home about.)
00:51:31 <Vorpal> fizzie, though if recoreded in *perfect* sync (note: impossible with real instruments) you could mix the channels
00:51:48 <Vorpal> fizzie, what episodic thing?
00:51:55 <fizzie> Tales of Monkey Island.
00:51:59 <fizzie> Telltale Games' thing.
00:52:24 <fizzie> "Free wovels! Get them while they're wovel!" (Heard from a wovel salesman in the game.)
00:53:44 <fizzie> Also there were some chuckle-worthy jokes about "U Tubes".
00:53:55 <Vorpal> fizzie, ... there is no excuse for that pun which doesn't even really work
00:54:09 <Vorpal> wait, misread it
00:54:14 <Vorpal> as "free wolves"
00:54:16 <Vorpal> somehow
00:54:22 <Vorpal> okay, now it makes more sense XD
00:54:48 <Vorpal> <fizzie> Telltale Games' thing. <-- never heard of that
00:55:10 -!- MigoMipo has quit (Read error: Connection reset by peer).
00:56:04 <fizzie> The company, or the game?
00:56:21 <Vorpal> fizzie, the phrase in the line I highlighted
00:56:30 <Vorpal> fizzie, I have no idea IF it is a company or a game
00:56:51 <fizzie> Telltale Games is the company, and Tales of Monkey Island is the game.
00:56:57 <Vorpal> ah
00:57:04 <Vorpal> fizzie, well neither then
00:57:49 <fizzie> They've done episodic-style modern adventure games into existing "franchises"; that Monkey Island game, and some Sam & Max stories, and I don't know what else.
00:59:42 <Vorpal> hm should I use the FreeBSD name in the BSD license
00:59:44 <Vorpal> or my own
00:59:49 <Vorpal> since it is based on freebsd code
00:59:58 <Vorpal> I mean in the line "The views and conclusions contained in the software and documentation are those of the authors and should not be interpreted as representing official policies, either expressed or implied, of the FreeBSD Project."
01:00:06 <Vorpal> (I do not plan to re-license it under GPL)
01:00:41 <Vorpal> fizzie, any idea?
01:02:03 <Vorpal> hm
01:02:59 <Vorpal> meh, I'll settle for something like:
01:03:01 <Vorpal> (In part) Copyright (c) 2010 Arvid Norlander. All rights reserved.
01:03:01 <Vorpal> (In part) Copyright (c) 1992-2010 The FreeBSD Project. All rights reserved.
01:03:19 <Vorpal> (the "all rights reserved" amuse me)
01:06:47 <Vorpal> elliott, fizzie: http://sporksirc.net/~anmaster/software/pcspeak.tar.xz
01:06:55 <Vorpal> if you want to play with the software
01:08:50 <elliott> <fizzie> The song's a perennial favourite of everybody. I think I've seen a recording of Press Play On Tape performing it somewhere too.
01:09:02 <elliott> They've done an *awesome* version of it segued into LeChuck's theme.
01:09:28 <elliott> <Deewiant> The LAPC-I is an MT-32 compatible card
01:09:37 <elliott> Deewiant: Proper compat or just one of those ones with its default soundfont?
01:10:13 <elliott> Can confirm that the PC speaker and Tandy ones are the same game.
01:10:32 <Deewiant> The Roland LAPC-I contains an MT-32 compatible synthesizer
01:10:49 <elliott> Also, you can't compare cards with the MT-32 based on one track, since the MT-32 is far more reprogrammable than any of the others.
01:10:53 <elliott> <Vorpal> fizzie, didn't it have the issue of not doing the proper transition-and-blend thingy?
01:11:06 <elliott> That's the CD version of MI1, but it didn't even have iMUSE; it was just slightly less smooth.
01:11:18 <elliott> MI2 has never been CD-soundtracked and was the first to use iMUSE.
01:11:23 <elliott> Special Edition != CD rerelease.
01:11:23 <Vorpal> elliott, what did you do to that line, it only highlighted /part/ of it
01:11:29 <Vorpal> right, it ended just after >
01:11:36 <elliott> <Vorpal> fizzie, basically you got lots of CD seeking even
01:11:43 <Vorpal> elliott, you did it again
01:11:44 <elliott> Was MI2 even released on CD when it came out?
01:11:47 <elliott> <Vorpal> fizzie, though if recoreded in *perfect* sync (note: impossible with real instruments) you could mix the channels
01:11:50 <elliott> iMUSE didn't mix any channels.
01:11:58 <Vorpal> but no longer
01:11:58 <elliott> It waited until an appropriate transition point and jumped to the appropriate transition.
01:12:01 <elliott> They were handmade.
01:12:05 <Vorpal> elliott, right
01:12:19 <elliott> I don't like the Special Edition theme; the main lead isn't pronounced enough.
01:12:25 <elliott> There's a Telltale Games guy on the esowiki, btw.
01:12:41 <elliott> Vorpal: just xchat copying colour codes for some reason
01:12:54 <Vorpal> elliott, but it only did it 2 out of 3 times
01:12:58 <elliott> Deewiant: Vorpal doesn't see annotations, re: <Deewiant> And if you noticed the last annotation, the final one played ("CD audio quality" or whatever it was called) is from a different version
01:13:05 <elliott> since he uses some youtube-dl thing, not flashl.
01:13:08 <elliott> *flash.
01:13:16 <Vorpal> elliott, and the modes of this channel should filter colour codes
01:13:19 <elliott> Vorpal: I think it's based on the precise moon-phase.
01:13:25 <elliott> Also, no, annoying colours are GREAT!
01:13:26 <Vorpal> elliott, aaah
01:13:32 <elliott> Vorpal: Aaaaah?
01:13:40 <Vorpal> elliott, (moon phase)
01:13:47 <fizzie> Irssi shows "< elliott> H<HVorpalH>H fizzie, ..." with inverse-video Hs.
01:13:54 <Vorpal> fizzie, hah
01:14:18 <elliott> Vorpal: Have you ever actually played the Monkey Island games?
01:14:31 <hagb4rd> i luv them
01:14:35 <Vorpal> elliott, no, it is on my todo list however
01:14:51 <Vorpal> elliott, but I never really liked point and click
01:14:58 <Vorpal> elliott, I prefer more free form in general
01:15:14 <Vorpal> elliott, open world I love (ev override is very open world)
01:15:16 <hagb4rd> for 5: where is the fuel for the chainsaw?
01:15:44 <elliott> 1 is great (but short; once you've beaten it once it only takes like 2-3 hours to do), 2 is probably the best adventure game ever written, and 3 (Curse of) is excellent too (although basically just ignore-retconned 2's ending and didn't involve Ron Gilbert; it is still good, however; note that it's 640x480x8. ScummVM can play it.) 4 (Escape from) is utter terrible keyboard-controlled 3D dreck and should be avoided at all costs.
01:15:51 <Vorpal> and there is the craziness of the puzzles too
01:15:52 <elliott> Tales I haven't played yet.
01:15:56 <Vorpal> I prefer RPG in general
01:16:05 <elliott> Vorpal: Of course you use the rubber chicken with a pulley in the middle to cross over the chasm ...
01:16:07 <elliott> What are you, an idiot?
01:16:19 <elliott> (Note: Actual spoiler.)
01:16:28 <Vorpal> elliott, I heard about that example before
01:16:33 <Vorpal> so I'm not spoiled by you
01:17:10 <Vorpal> elliott, but yeah I don't really enjoy exploring the combinatorial explosion of all items in all places at all times
01:17:16 <Vorpal> which is what it seems like pretty much
01:17:20 <elliott> Vorpal: Not really.
01:17:23 <Vorpal> unless you use spoilers
01:17:26 <elliott> You just have to sense a pun when one comes along.
01:17:37 <elliott> The combine-X-with-Y approach only happens when you're really stuck.
01:17:51 <Vorpal> elliott, so is "<elliott> Vorpal: Of course you use the rubber chicken with a pulley in the middle to cross over the chasm ..." from a pun?
01:18:03 <elliott> Vorpal: To be fair, it comes with the pulley.
01:18:09 <elliott> And there's a line conveniently crossing the chasm.
01:18:21 <elliott> But, okay, so that one isn't terribly intuitive, but it /is/ lampshaded in the actual game. :p
01:18:26 <Vorpal> elliott, what about the rubber chicken?
01:18:37 <Vorpal> I mean, the pulley, sure
01:18:40 <elliott> It comes with the pulley.
01:18:49 <Vorpal> elliott, which seems rather weird
01:18:55 <elliott> Yes, well.
01:19:00 <elliott> The Voodoo Lady is weird.
01:19:00 <Vorpal> I never seen a rubber chicken with a pulley built in before
01:19:16 <Vorpal> come to think of it...
01:19:21 <Vorpal> I never seen a rubber chicken
01:19:24 <Vorpal> rubber ducks sure
01:19:26 <Vorpal> but not chicken
01:19:41 <fizzie> They serve different roles.
01:19:54 <elliott> http://media.photobucket.com/image/rubber%20chicken%20with%20a%20pulley%20in%20the%20middle/ErDracu/Pollo.jpg
01:19:56 <elliott> WHAT NOW
01:19:58 <elliott> SCIENCE
01:20:03 <fizzie> For one thing, rubber chickens are far more floppy and less bath-toyey.
01:20:10 <elliott> Perfect for crossing chasms.
01:20:30 <Vorpal> elliott, not science. Just... technology
01:20:37 <elliott> It disproves science.
01:20:43 <Vorpal> elliott, so how much load can it take?
01:20:57 <elliott> Vorpal: Approximately one (1) Guybrush Threepwood./
01:20:59 <elliott> s/\/$//
01:21:12 <Vorpal> elliott, I meant the real-world one
01:21:25 <elliott> Vorpal: Approximately one (1) Guybrush Threepwood.
01:21:34 <elliott> (Turns out they live in a world where everything is tiny.)
01:21:49 <Vorpal> har
01:22:57 <elliott> Was Tales really released in 2009? jeez.
01:23:37 <hagb4rd> painted with deluxe paint.. btw ;)
01:23:47 <fizzie> Tales wasn't relatively speaking too far on the "horrible annoying puzzlery" scale, but it did have its share. (And personally I am most annoyed by those cases where I clearly know what to do, it has a reasonable probability of being the right thing, but I just can't seem to figure out how to communicate that to the game through the UI.)
01:24:15 <hagb4rd> brushguy
01:24:45 <Vorpal> fizzie, yeah that sucks
01:24:46 <elliott> indeed
01:24:49 <elliott> named after guy.brush
01:24:59 <elliott> since deluxe paint used .brush
01:25:10 <hagb4rd> it did
01:25:57 <fizzie> Guybrush Threepwood, Mighty Pirate(TM).
01:26:04 <hagb4rd> one of the first apps using thi terminology, as far as i know
01:26:09 <Mathnerd314> elliott: do you still think I wasted your time?
01:26:16 <Vorpal> why on earth is niceness value so backwards. In other words: why does *nix use an inverse of the more straight-forward concept "priority"?
01:26:20 <hagb4rd> wanna buy a map? *g
01:26:23 <Vorpal> elliott, ^
01:26:24 <elliott> Mathnerd314: Everyone always wastes my time!
01:26:36 <elliott> Vorpal: So they could use C's default initialisation to 0? :)
01:26:44 <elliott> OK, so that would work with priority too I guess.
01:26:50 <elliott> Vorpal: I suspect positive nice values came much before negative.
01:26:57 <Vorpal> hm
01:27:04 <Mathnerd314> elliott: ok...
01:27:05 <fizzie> Also it's nicer.
01:27:11 <Vorpal> elliott, so they started with lower-valued priority?
01:27:25 <elliott> probably, yes
01:27:35 <Vorpal> fizzie, you are trespassing on oerjan's territory!
01:28:40 <Mathnerd314> elliott: so is that a yes?
01:28:53 <elliott> Mathnerd314: it's file_not_found
01:29:02 <Vorpal> elliott, I think I might want to write 4'33" for the pc speaker
01:30:11 <Mathnerd314> Vorpal: it's called "mute button"
01:30:16 <elliott> Vorpal: That's 273 seconds. So, if ...
01:30:21 <elliott> How many Hz does one 386 cycle take?
01:30:30 <elliott> Rather.
01:30:37 <Vorpal> elliott, actually the time tempo to give it
01:30:41 <fizzie> Non-audible-frequency beeps were a common-ish way of doing sub-second resolution delays in some system.
01:30:44 <elliott> How many one-cycle instructions can you fit into one second of 386 execution?
01:30:45 <Vorpal> I just need to figure out*
01:30:50 <Vorpal> got dropped from it
01:30:50 <elliott> At 12 MHz.
01:31:05 <Vorpal> elliott, presumablly 1/12 MHz?
01:31:11 <Vorpal> that is
01:31:18 <Vorpal> 1/(12 MHz) btw
01:31:34 <elliott> Vorpal: Right. So 83 ns for one instruction.
01:31:44 <Vorpal> elliott, assuming it takes one cycle
01:32:15 <Vorpal> elliott, anyway, the tool doesn't work like that
01:32:38 <elliott> Vorpal: OK, 2^23 nops on a 12 MHz 386.
01:32:44 <elliott> Assuming a 1-cycle nop.
01:32:52 <elliott> That's close to 4'33", except as a power of two.
01:33:08 <Vorpal> elliott, ... but how does that help me?
01:33:10 <elliott> (Man, even 386s are /fast/.)
01:33:17 <elliott> Vorpal: Well, you can do it without even touching the PC speaker!
01:33:25 <Vorpal> elliott, but I don't have a 386
01:33:32 <elliott> That way you can listen to 4'33" even if you don't have a PC speaker to listen to it with.
01:33:50 <fizzie> "Early in production, Intel discovered a bug that could cause a system to unexpectedly halt when running 32-bit software. Not all of the processors already manufactured were affected, so Intel tested its inventory.
01:33:56 <fizzie> Processors that were found to be bug-free were marked with a double-sigma (ΣΣ), and affected processors were marked "16 BIT S/W ONLY". These latter processors were sold as good parts, since at the time 32 bit capability was not relevant for most users. Such chips are now extremely rare."
01:34:04 <fizzie> That's so very nice.
01:34:12 <elliott> That's where 3-core AMDs come from.
01:34:19 <elliott> 4-core AMDs born with Down's syndrome.
01:34:22 <Vorpal> elliott, here is the easy version: "". Note: program exits right away and it continues playing in background
01:34:36 <elliott> Vorpal: But how will I know when to stop listening?
01:35:01 <fizzie> It's a bit like the "only connections to IP addresses inside Finland" cheaper-rate thing our ISP used to offer.
01:35:07 <Vorpal> elliott, use a wall clock
01:35:20 <elliott> http://en.wikipedia.org/wiki/Wall_clock_time
01:35:23 <Vorpal> fizzie, so no google for example
01:35:25 <elliott> Vorpal: Yes, but I need a process that runs for 4'33" to do that.
01:35:30 <elliott> Vorpal: Which my example gave.
01:35:37 <elliott> <Vorpal> fizzie, so no google for example ;; I doubt Google existed then
01:35:41 <Vorpal> ah
01:35:43 <Vorpal> true
01:35:44 <fizzie> Vorpal: It was around 1991, so no Google anyway.
01:35:53 <Vorpal> fizzie, aaah THAT far back
01:35:59 <fizzie> (Disclaimer: Year is purely guesswork.)
01:36:06 <elliott> 1991, before the dinosaurs got wiped out.
01:36:09 <elliott> When Jesus roamed the Earth
01:36:13 <Vorpal> elliott, oh?
01:36:22 <Vorpal> elliott, he was long lived then
01:36:31 <Vorpal> of we have exponential year length
01:36:39 <elliott> http://farm1.static.flickr.com/172/431306643_528c65a6b3.jpg
01:36:55 <Vorpal> elliott, ...
01:37:34 <hagb4rd> elliot: "Brushes can be cut from the background by using the box, freehand, or polygon selection tools. They can then be used in the same manner as any other brush or pen. This functionality is simpler to use than the "stamp" tool of Photoshop or Alpha Channels as provided in later programs. Brushes can also be rotated and scaled, even in 3D. After a brush is selected, it appears attached to the mouse cursor,
01:37:34 <hagb4rd> providing an exact preview of what will be drawn. This allows precise pixel positioning of brushes, unlike brushes in Photoshop CS3 and lower, which only show an outline." --wikipedia, on the amiga version of deluxe paint (1985!) [appendix]
01:37:45 <Vorpal> ooh found a quickcam in a box
01:37:49 <Vorpal> I wonder if it works with linux
01:37:56 <Vorpal> (logitech quickcam express)
01:38:05 <Vorpal> (one of those with a pyramid base)
01:38:15 <fizzie> It's popular enough to probably do.
01:38:22 <Vorpal> hm
01:38:28 <hagb4rd> i loved it *sigh
01:38:36 <Vorpal> hagb4rd, loved what?
01:38:43 <Vorpal> fizzie, some googling suggests a out of kernel driver, which seems unmaintained
01:38:50 <hagb4rd> deluxe paint
01:38:56 <Vorpal> hagb4rd, never heard of it
01:38:57 <fizzie> Quickcam-in-a-box, sounds like a trademark.
01:39:27 <fizzie> Vorpal: Well, you were a mac guy, weren't you? (I ay misremember.)
01:39:30 <Vorpal> "This functionality is simpler to use than the "stamp" tool of Photoshop or Alpha Channels as provided in later programs."
01:39:35 <Vorpal> hagb4rd, I see an issue here
01:39:58 <Vorpal> hagb4rd, mostly in that it won't 1) do blending very well 2) follow where you drag the mouse
01:40:02 <elliott> <Vorpal> ooh found a quickcam in a box
01:40:03 <elliott> <Vorpal> I wonder if it works with linux
01:40:03 <elliott> <Vorpal> (logitech quickcam express)
01:40:03 <elliott> <Vorpal> (one of those with a pyramid base)
01:40:07 <elliott> I HAD ONE OF THOSE OMG.
01:40:12 <elliott> Such shitty quality.
01:40:16 <Vorpal> elliott, it's usb
01:40:18 <hagb4rd> vorpal: a painting app, i remember from the amiga, with quite innovative, and impressing func.
01:40:20 <Vorpal> and yes shitty quality
01:40:33 <Vorpal> hagb4rd, ah, so not for retouching photos then
01:41:03 <hagb4rd> ithttp://en.wikipedia.org/wiki/Deluxe_Paint#Functionality
01:41:04 <elliott> Vorpal: Also, careful with what you say about Deluxe Paint.
01:41:08 <hagb4rd> sry
01:41:08 <elliott> http://ui28.gamespot.com/475/full20040105092056_2.gif <-- It produced things like this.
01:41:19 -!- cheater99 has quit (Read error: Connection reset by peer).
01:41:21 <hagb4rd> http://en.wikipedia.org/wiki/Deluxe_Paint vorpal
01:41:26 <Vorpal> elliott, okay, good for drawing
01:42:01 <elliott> "the character's name derived from the file used to store his image data. Contrary to popular belief however, the original source sprite was not named "guy.brush" as the file extension used was ".bbm" and not ".brush". The file was in fact named "guybrush.bbm", the "brush" portion of the file name being included by the artist behind the character, Steve Purcell." <-- aww
01:42:03 <elliott> OH WELL
01:43:29 <hagb4rd> aye :)
01:44:00 <elliott> http://www.effectgames.com/demos/canvascycle/?sound=1 ;; these are beautiful btw, colour cycling
01:44:04 <elliott> (turn sound on)
01:44:23 <fizzie> Vorpal: gspca (which is in mainline kernel nowadays) supports at least some hw revisions of quickcam express.
01:44:37 <Vorpal> fizzie, oh nice
01:44:44 <Vorpal> fizzie, will try that when I recompile kernel
01:44:46 <Sgeo> Who is Pull?
01:44:48 <Vorpal> (going to anyway)
01:44:54 <elliott> Sgeo: ?
01:44:55 <Vorpal> Sgeo, what?
01:45:11 * Sgeo is being silly. Look at the image of the game
01:45:15 <Vorpal> or wait
01:45:21 <Vorpal> *tries on ubuntu box*
01:45:23 <Vorpal> ah
01:45:30 <Vorpal> fizzie, so... what program does one use then?
01:45:52 <fizzie> /* QuickCam Express */
01:45:52 <fizzie> {USB_DEVICE(0x046d, 0x0840), .driver_info = BRIDGE_STV600 },
01:46:05 <elliott> Vorpal: Anything? Cheese is popular for photo/video-taking, I gathe.r
01:46:07 <elliott> *gather.
01:46:09 <fizzie> You can lsusb to see if it has those ids.
01:46:22 <j-invariant> elliott: are there any games iwth graphics like that throughout?
01:46:26 <fizzie> Just dmesg for v4l messages about finding a new device. :p
01:46:33 <Vorpal> fizzie, fizzie nop, 0x0870
01:46:34 <elliott> j-invariant: On the Amiga, I would expect so.
01:46:41 <Vorpal> fizzie, same for the first id
01:46:48 <j-invariant> I want to play them
01:46:49 <elliott> j-invariant: Perhaps none quite so beautiful,t hough; they look quite high-res.
01:47:04 <fizzie> Well, it could still be compatible. Or handled by something else.
01:47:11 <elliott> j-invariant: btw do you know about IΞ?
01:47:21 <Vorpal> fizzie, hm
01:47:25 <j-invariant> no what is that
01:47:37 <fizzie> /* Dexxa WebCam USB */
01:47:38 <fizzie> {USB_DEVICE(0x046d, 0x0870), .driver_info = BRIDGE_STV602 },
01:47:48 <fizzie> That's in the same list.
01:47:59 <Vorpal> fizzie, aah
01:48:04 <elliott> j-invariant: a really awesome dependently typed lambda calculus extension that does it in a totally different way to everything else
01:48:10 <elliott> j-invariant: see these posts (I haven't found a better source):
01:48:14 <fizzie> Marketing names are of course always pretty random.
01:48:19 <elliott> http://lukepalmer.wordpress.com/2009/02/18/dependent-types-are-ridiculously-easy/
01:48:22 <elliott> http://lukepalmer.wordpress.com/2009/03/25/system-ig-semantics/
01:48:28 <elliott> http://lukepalmer.wordpress.com/2009/04/12/some-constructions-in-ixi/
01:48:33 <elliott> http://lukepalmer.wordpress.com/2009/04/29/dana-actual-progress/
01:48:35 <elliott> http://lukepalmer.wordpress.com/2009/06/06/recursive-types-in-ixi/
01:48:41 <elliott> wish someone other than luke talked about it though :P
01:48:46 <elliott> j-invariant: oh you can ignore the system-ig-semantics one...
01:48:51 <elliott> that's ig not iXi
01:49:06 <j-invariant> im really in love with this color cycling stuff :(
01:49:24 <j-invariant> someone should make a game with these graphics
01:49:35 <elliott> it's really cool because the type system is sort of, to the side
01:49:38 <elliott> j-invariant: that would be awesome
01:49:45 <elliott> i'm nowhere near a good enough artist to though :P
01:49:48 <Vorpal> the image quality: aieee
01:50:21 <elliott> j-invariant: the problem is that palette cycling only works for some things
01:50:24 <elliott> so you can't do all animation withi t
01:50:31 <elliott> *with it
01:50:52 <elliott> j-invariant: and also the way some of these are done you'd have very little palette left ... that is, on "authentic" hardware
01:51:08 <fizzie> I have some el-cheapo webcam which for some reason worked a lot better in Windows than in Linux for non-daylight levels. I suspect some sort of amplifier AGC thing not being handled properly.
01:51:33 <elliott> j-invariant: anyway you should read those posts about i\Xi, it's a really cool system
01:51:48 <elliott> the posts get better as you go down :p
01:52:04 <elliott> j-invariant: it has things like a universal set ... but is still consistent-looking
01:52:11 <elliott> *the universal set, I suppose
01:54:23 <Vorpal> <j-invariant> im really in love with this color cycling stuff :( <--- what colour cycling?
01:54:52 <Vorpal> fizzie, ah yes mine seems to have some issues in here too
01:55:01 <j-invariant> Vorpal: http://www.effectgames.com/demos/canvascycle/?sound=1
01:55:03 <elliott> Vorpal: the one i linked.
01:55:07 <elliott> note: requires js, best with sound turned on
01:56:00 <Vorpal> elliott, heh, cool
01:56:04 <elliott> "But there's also a case to be made that we all *really really suck* at
01:56:04 <elliott> a game we joined under a horridly stupid misunderstanding of what it
01:56:05 <elliott> was all about, and that this is actually still the first era." -- teucer, talking at the end of B's supposed 7th (I think) era
01:56:05 <Vorpal> but where does the cycling come into it
01:56:13 <elliott> Vorpal: it's done entirely by changing the entries in a palette table
01:56:17 <elliott> the actual pixel values stay the same
01:56:22 <elliott> they're just mapped to different RGBs with the palette table
01:56:30 <elliott> (this is how you did it on actual hardware, very common)
01:56:44 <Vorpal> elliott, they change it midway through or such?
01:56:52 <elliott> Vorpal: it changes every frame
01:56:55 <Vorpal> ah
01:56:56 <elliott> that's what makes it look like it's animated
01:57:00 <elliott> they never touch the actual image
01:57:06 <elliott> just the palette
01:57:13 <Vorpal> aha
01:57:29 <elliott> "
01:57:29 <elliott> Pragmatism is Agora's way of ensuring stability. B's is emergencies -
01:57:29 <elliott> if things are breaking down, we have an emergency."
01:57:36 <Vorpal> elliott, if doing it through the middle of a refresh you could get more colours maybe
01:57:48 <elliott> Vorpal: Possibly. That would be a bitch though. :p
01:57:54 -!- cheater99 has joined.
01:58:07 <elliott> "As for the present situation with the 2E issue, I'm gonna do what we
01:58:07 <elliott> always do: enumerate the Woobleverses and try to recombine and/or
01:58:07 <elliott> destroy as many as possible."
01:58:11 <elliott> Gee, B, what do you wanna do tonight?
01:58:26 <elliott> The same thing we do every night, players -- try to figure out a consistent gamestate!
01:58:44 <elliott> ALTERNATIVELY
01:58:51 <elliott> The same thing we do every night, players -- have an Emergency!
02:00:11 <j-invariant> elliott: what's good about that XI thing?
02:00:12 -!- cheater99 has quit (Read error: Connection reset by peer).
02:00:20 <elliott> j-invariant: it's just cool :P
02:00:25 <elliott> j-invariant: i think it's more elegant than regular type systems
02:00:29 <elliott> j-invariant: since you don't need the concept of a type
02:00:31 <elliott> you just have predicates
02:00:37 <elliott> it's sort of "more integrated"
02:01:51 <Vorpal> elliott, do you watch youtube in the browser
02:01:52 <Vorpal> ?
02:02:03 <Vorpal> elliott, if so, does sound and video sync properly for you?
02:02:34 <elliott> Vorpal: Yes, and yes. It didn't use to because of no native x86-64 player, but it works now.
02:02:47 <elliott> (I'm using the native x86-64 Flash.)
02:02:53 <Vorpal> elliott, but I thought there wasn't a native one currently?
02:03:12 <elliott> Vorpal: there is, I don't think it's totally "final" though, but Debian's package uses it... at least on squeeze
02:03:20 <Vorpal> heh
02:03:34 <Vorpal> elliott, I get perfect sync in mplayer btw
02:03:42 <elliott> Yes, well, that's hardly surprising.
02:03:46 <Vorpal> elliott, :P
02:03:52 <Vorpal> elliott, and I have for the past years
02:04:04 <elliott> Vorpal: I didn't use Linux the past years.
02:04:12 <Vorpal> true
02:04:13 <elliott> In OS X, btw, everything syncs up perfectly, but Flash likes to take 100% of the CPU.
02:04:24 <elliott> I am, incidentally, not planning to run Flash on my new laptop. :p
02:04:27 <Vorpal> elliott, does it under linux (take 100%?)
02:04:33 <elliott> Vorpal: No.
02:04:38 <elliott> Adobe just really hate Steve Jobs.
02:04:49 <Vorpal> elliott, I assume you will use linux on that air?
02:05:11 <elliott> I'm planning to dual-boot it. I can't throw away OS X because of EFI updates and, well, it is nice in some ways.
02:05:28 <elliott> But I'm gonna stick Ubuntu on it. :p
02:05:48 <elliott> Not sure how much Kitten will like such foreign hardware.
02:05:51 <Vorpal> elliott, without or with flash?
02:06:03 <elliott> Who knows :p
02:06:03 <Vorpal> elliott, uh, couldn't you make kitten support it?
02:06:18 <elliott> Vorpal: Yes, but even Ubuntu's support was flaky as of a year or two ago.
02:06:21 <elliott> (For Macs.)
02:06:24 <Vorpal> elliott, elilo?
02:06:30 <elliott> Vorpal: Nobody uses elilo :p
02:06:30 <Vorpal> or grub2?
02:06:36 <elliott> GRUB 2 wasn't stable then.
02:06:50 <Vorpal> elliott, well, people used elilo when grub2 didn't yet exist iirc
02:08:57 <elliott> The BIOS emulation is the most common thing to do, I think
02:08:59 <elliott> *think.
02:09:02 <elliott> Since that's automatic.
02:09:33 <Vorpal> elliott, bootcamp?
02:09:38 <Vorpal> or what?
02:09:58 <Vorpal> elliott, also iirc it works well with freebsd
02:10:00 <Vorpal> (EFI that is)
02:10:09 <elliott> Vorpal: Boot Camp is just the marketing name for (1) an EFI update that added BIOS emulation, and (2) a tool that partitions your drive for you then reboots.
02:10:15 <elliott> (1) is the important thing here.
02:10:45 <Vorpal> elliott, ah, so how does EFI know if you want BIOS emulation for a given partition?
02:11:15 <elliott> Vorpal: It ... doesn't; it just starts emulatin' that thar BIOS if it looks like the bootloader wants a BIOS. I'm not sure how it works. Maybe EFI has different boot sectors.
02:12:04 <Vorpal> elliott, I presume you still have to deal with GUID partition tables?
02:12:15 <elliott> Vorpal: Well, "deal"; (GNU) fdisk supports them.
02:12:19 <elliott> (Or was it GNU?)
02:12:23 <Vorpal> elliott, the kernel needs to support it
02:12:36 <Vorpal> elliott, I think parted handles it too btw
02:12:43 <elliott> Right, I meant parted.
02:13:02 <Vorpal> is parted gnu?
02:13:53 <elliott> I think so.
02:13:56 <elliott> Yes.
02:14:01 <Vorpal> elliott, but... it is good?
02:14:11 <elliott> Well, I wouldn't go that far!
02:14:21 <Vorpal> elliott, it actually is
02:14:34 <Vorpal> elliott, with gparted it is a bliss compared to fdisk
02:14:43 <elliott> I like fdisk.
02:14:48 <Vorpal> wait, it is tricky to edit partition type manually from gparted
02:14:49 <elliott> The commands are shorter. :p
02:14:51 <Vorpal> so that lacks
02:14:57 <Vorpal> not sure about from parted
02:15:24 <Vorpal> elliott, I ended up having to use fdisk to touch up the partition types afterwards a few times
02:16:39 -!- cheater99 has joined.
02:17:29 <elliott> fizzie: Vorpal: http://www.youtube.com/watch?v=mYZXNVHVfhc The excellent Press Play On Tape Monkey Island theme / LeChuck's theme thing.
02:19:08 <Vorpal> elliott, decent performance
02:19:27 <elliott> The LeChuck part is better IMO.
02:19:35 <Vorpal> elliott, but not same as the intro melody iirc?
02:19:36 <elliott> But then, LeChuck's theme is /really/ catchy.
02:19:40 <Vorpal> elliott, some minor variations
02:19:40 -!- cheater99 has quit (Read error: Connection reset by peer).
02:19:43 <elliott> Vorpal: Indeed not; it changes at 1 minutes something.
02:20:01 <Vorpal> elliott, well even before that
02:20:13 <elliott> Vorpal: No, all that's the actual theme.
02:20:19 <Vorpal> elliott, hm
02:20:35 <Vorpal> elliott, I prefer monkey island 2 theme personally
02:20:52 <elliott> The one I linked gets rather metal at the end. :p
02:21:02 <Vorpal> elliott, yeah....
02:21:08 <Vorpal> elliott, I doubt it does that in the game?
02:21:15 <elliott> No ... but it totally should.
02:21:21 <elliott> All the game songs loop cleanly. :p
02:21:32 <Vorpal> elliott, wouldn't it be anachronistic?
02:22:18 <elliott> Vorpal: Almost everything in MI2 is anachronistic. Observe: http://www.mrbillsadventureland.com/reviews/m-n/monkeyR/stans1.jpg
02:22:24 <Vorpal> elliott, I have to say I prefer the game theme song (even discarding the metal bit)
02:22:27 <elliott> (Also "Ask me about LOOM(tm).")
02:23:03 <Vorpal> hah
02:23:06 <Vorpal> elliott, LOOM?
02:23:10 <elliott> The game.
02:23:20 <elliott> There's a guy who gives you a sales pitch for it in the SCUMM Bar.
02:23:33 <elliott> (SCUMM would also count as an anachronism, except it can sort of pass as not being one.)
02:25:59 <Vorpal> SCUMM not being the VM?
02:28:22 <elliott> Vorpal: SCUMM = Script Creation Utility for Maniac Mansion, the engine used for Maniac Mansion and then the Monkey Island games.
02:28:26 <elliott> Along with every other LucasArts adventure game.
02:28:37 <elliott> ScummVM is just a program that runs SCUMM games... along with several other engines, but :P
02:30:36 <Vorpal> <elliott> Along with every other LucasArts adventure game. <-- http://en.wikipedia.org/wiki/GrimE
02:31:04 <elliott> Vorpal: Yes, well, up to then.
02:31:12 <Vorpal> elliott, quite so
02:31:12 <elliott> Vorpal: To be fair, only one good GrimE game was ever released.
02:31:19 <Vorpal> elliott, oh?
02:31:24 <Vorpal> elliott, which one?
02:31:25 <elliott> (Grim Fandango; the only other was Escape from Monkey Island, which was beyond terrible.)
02:31:46 <elliott> Grim Fandango is probably the #1 or #2 best adventure game ever, though -- and that's *despite* the keyboard-arrow-button controls.
02:31:53 <Vorpal> elliott, then what came after GrimE?
02:31:54 <elliott> Which are *terrible*, and yet it still comes out on top for being just that good.
02:32:00 <elliott> Vorpal: Nothing; LucasArts stopped making adventure games.
02:32:07 <elliott> Telltale have their own thing, but that's years later.
02:32:10 <Vorpal> ah
02:32:25 <elliott> LucasArts pretty much just do Star Wars games... well, until the two Special Edition rereleases of Monkey Island.
02:33:41 <elliott> Seriously, how do you move XChat tabs. Eurgh.
02:33:48 <elliott> I can't get #esoteric back in its rightful place.
02:34:29 <Vorpal> elliott, when I use xchat I use the tree list view
02:34:33 <Vorpal> and sorted by name
02:34:59 <elliott> Vorpal: Yes, well, I'm using tabs.
02:36:46 -!- cheater99 has joined.
02:38:17 <j-invariant> elliott: http://coq.pastebin.com/sjjga8Uj
02:38:59 <j-invariant> this looks like a nice way to bootstrap, just set up enough machinary so that you can automatically prove equations - then start developing the theory
02:39:17 -!- cheater99 has quit (Read error: Connection reset by peer).
02:39:35 <elliott> #
02:39:35 <elliott> Section PublicService.
02:39:40 <elliott> j-invariant: make this "Section Socialism."
02:39:42 <elliott> :D
02:39:45 <j-invariant> hehe
02:39:50 <elliott> yeah that looks much nicer
02:39:59 <j-invariant> it would be a lot more effective if it took equations like fg = id into account (rather than just solving associativity) HINT HINT elliott
02:40:27 <j-invariant> im just saying that because I don't know how to implement it :/
02:41:10 * Sgeo may switch to Pigdin from Digsby
02:41:12 <elliott> j-invariant: you think *I* know? i'm an amateur :D
02:41:13 <j-invariant> elliott: look at line 314 though. I need to prove a complicated equation, so I name the symbolic category - ask it for the proof, then functor map it into MY category
02:41:44 <j-invariant> makes the proof so much easier
02:41:50 <elliott> yeah it looks awesome
02:41:54 <elliott> i should learn category theory properly
02:41:59 <Sgeo> Features I need: To know when a particular person goes online
02:42:09 <Sgeo> And more importantly right now, a log of when everyone goes on and offline
02:42:16 <elliott> <Sgeo> Features I need: To know when a particular person goes online
02:42:18 <elliott> Um, Pidgin does that ...
02:42:20 <Sgeo> A bit late for that now
02:42:28 <elliott> <Sgeo> And more importantly right now, a log of when everyone goes on and offline
02:42:29 <j-invariant> elliott: this is whatt I am using mostly http://www.scss.tcd.ie/Edsko.de.Vries/ct/catsters/linear.php
02:42:30 <Sgeo> elliott, I know. Reread what I wrote
02:42:33 <elliott> Uhh, it might do th-- why do you want that.
02:42:39 <j-invariant> elliott: but programming it helps a LOT
02:43:09 <Sgeo> My friend's gf is worried about him, he was supposed to meet up with her, asked when I last spoke with him
02:43:30 <Vorpal> Sgeo, do you mean on irc?
02:43:37 <Sgeo> Vorpal, MSN
02:43:38 <elliott> Vorpal: Nobody uses IRC.
02:43:40 <Vorpal> oh
02:43:55 <Vorpal> elliott, ... your statement is absurd given the medium of it's transmission
02:44:12 <elliott> Vorpal: You know what I mean, though.
02:44:26 <Vorpal> elliott, no one non-tech savey
02:44:37 <Vorpal> (% typos)
02:44:46 <Sgeo> Eh, some tech morons manage to get on sometimes
02:45:21 <Vorpal> Sgeo, rare though
02:45:31 <Vorpal> at least in channels like this
02:45:34 <Vorpal> #ubuntu is another thing
02:46:11 <j-invariant> elliott: right now I can automatically prove all equations of the form f(g(1h)1)(i(j1)1)k = (f1g)((1(h1))(il))k (for example) because they both get reduced to f(g(h(i(j(k1)))))
02:46:39 <elliott> j-invariant: that is really awesome ... can coq even do that normally?
02:46:45 <j-invariant> elliott: but it would be useful to be able to add reductions like pq --> 1 so that it could prove stuff like pp1qq = 1
02:47:13 <j-invariant> elliott: well you can do it without category theory, but it's the "same thing" really
02:47:22 <j-invariant> you have to program it of course
02:47:37 <elliott> <j-invariant> elliott: but it would be useful to be able to add reductions like pq --> 1 so that it could prove stuff like pp1qq = 1
02:47:44 <elliott> can't you just like, add that as an equality theorem
02:47:49 <elliott> and eliminate for equality in the tactics?
02:47:50 <elliott> generalising
02:48:39 <j-invariant> that's what I do now: have to use tactics for each rewrite (that includes working your way into the middle of a deep expression you want to rewrite it), what I mean is the automatic equation prover should be able to take these sort of equations into account too
02:49:02 <j-invariant> a proof like (fg)(hk) = f((gh)k) needs like 40 rewrites or something
02:49:54 -!- FireFly has quit (Quit: swatted to death).
02:50:17 <j-invariant> (N.B. this is only difficult because "=" is an arbitrary equivalence relation, proving this stuff is really easy if it's actual equality)
02:50:31 <Sgeo> elliott, did you reread my line?
02:50:35 <elliott> Sgeo: yes
02:50:46 <elliott> j-invariant: put a "do_common_elimination" thing at the start and end of every tactic
02:50:50 <elliott> so it's "automatic" most of the time
02:51:10 <j-invariant> elliott: I can't implement do_common_elimination! that's too difficult
02:51:20 <elliott> j-invariant: well you can eliminate equality as a tactic, right>
02:51:21 <elliott> ?
02:51:24 <j-invariant> no
02:51:25 <elliott> just make that do_common_elimination for now
02:51:26 <elliott> and expand it later
02:51:27 <elliott> *right?
02:51:30 <elliott> j-invariant: oh. what can you do?
02:51:33 <elliott> **can*
02:53:12 <j-invariant> elliott: see http://coq.pastebin.com/sjjga8Uj line 72. That says that if f = f' then fg = f'g.. you have to apply it manually if you want to prove something like (jh)k = 1k given jk = 1
02:53:19 <j-invariant> I mean jh = 1
02:56:12 -!- cheater99 has joined.
02:57:26 <j-invariant> elliott: which is a huge hassle... but now I can just state an equation in the symbolic category, which then gets proved automatically.. and I just map the equation back into my category!
02:57:42 -!- cheater99 has quit (Read error: Connection reset by peer).
02:59:16 <j-invariant> I would guess the genenral theory of typed equations would be decidible though?
02:59:39 <j-invariant> actually no you can encode the word problem can't you...
02:59:44 <elliott> j-invariant: that is awesome
03:00:00 <elliott> j-invariant: hmm can you do language syntax extensions in coq?
03:00:03 <elliott> like, not as expressions
03:00:08 <elliott> but actually defining things like "Record x := y"
03:00:21 <elliott> j-invariant: if not, write a preprocessor for using coq with categories :)
03:00:24 <j-invariant> elliott: you need to hack into the ocaml stuff to do that
03:00:28 <elliott> things like "Category blah ..."
03:00:33 <elliott> just do a preprocessor, way easier than hacking ocaml
03:00:44 <j-invariant> elliott: well it would be easier to just make a new programming language that checks with Coq to see that everything checks out
03:00:58 <elliott> j-invariant: are you sure? this way you get tactics for free
03:01:11 <elliott> j-invariant: a preprocessor just has to transform some half-assedly parsed text into Coq boilerplate :P
03:01:14 <elliott> for defining a category or whatever
03:01:20 <j-invariant> heh yeah thats true
03:01:34 <elliott> j-invariant: you can even shout it CATEGORY to show that it's not going to be parsed decently :P
03:02:31 <elliott> j-invariant: this is really cool though
03:02:42 <elliott> j-invariant: hey is Lam a category? for lambda calculus
03:02:43 <elliott> Hask is
03:02:48 <elliott> but LC doesn't really have types
03:02:49 <elliott> so i guess not
03:02:54 <j-invariant> elliott: yeah I feel like with this I might be able to speed up a bit
03:02:59 -!- wareya_ has joined.
03:03:40 <Vorpal> night →
03:06:01 <j-invariant> elliott: one category I really want to get defined is one where objects are prolog terms and maps are substitutions
03:06:01 -!- wareya has quit (Ping timeout: 250 seconds).
03:06:38 <j-invariant> elliott: so e.g. {X|-->e(Y,k)} : f(X,g(X)) ---> f(e(Y,k),g(e(Y,k)))
03:07:59 -!- wareya has joined.
03:08:12 -!- wareya_ has quit (Ping timeout: 272 seconds).
03:08:15 <j-invariant> there's a book that shows how to implement unification on it
03:09:04 <elliott> j-invariant: that is awesome. do that :D
03:10:03 <elliott> j-invariant: what about CHR?
03:11:27 <j-invariant> well that is jsut syntax stuff
03:11:32 <j-invariant> no actual execution
03:14:39 <elliott> ah
03:14:42 -!- cheater99 has joined.
03:14:45 <elliott> i don't know much about chr
03:15:11 <j-invariant> http://www.cs.man.ac.uk/~david/categories/programs/
03:15:20 <j-invariant> stealing ths stuff once I get the scene set
03:15:55 <elliott> http://www.cs.man.ac.uk/~david/categories/programs/x.cat huh what is this, category theory in ML?
03:16:22 -!- cheater99 has quit (Read error: Connection reset by peer).
03:19:19 <elliott> j-invariant: do the Hask category next :D
03:20:15 <j-invariant> no way lol
03:20:19 <j-invariant> you do it
03:20:29 <j-invariant> I don't even know wha the Hask is
03:20:47 <j-invariant> can define simple data types using initiality, though - that needs to be tested out
03:21:11 <elliott> j-invariant: Hask is the category of Haskell types
03:21:16 <elliott> category-extras uses it :)
03:23:51 <elliott> j-invariant: so to implement Hask you need to implement Haskell's type system... and I think probably most of the values too :P
03:24:08 <elliott> j-invariant: http://hackage.haskell.org/packages/archive/category-extras/0.53.5/doc/html/Control-Category-Hask.html
03:24:11 <elliott> j-invariant: Make it clearer when we are dealing with the category (->) that we mean the category of haskell types via its Hom bifunctor (->)
03:27:28 <elliott> j-invariant: any opinions on lambda Prolog?
03:29:14 <j-invariant> what do you mean
03:29:21 <j-invariant> I don't use it
03:30:35 <elliott> j-invariant: any interesting insights, i guess
03:33:10 -!- cheater99 has joined.
03:34:53 -!- cheater99 has quit (Read error: Connection reset by peer).
03:39:05 <Mathnerd314> elliott: I'm going off to tell children that Santa is coming to a town near them, but will only visit their house if they're asleep. Anything else I should tell them?
03:39:53 <Sgeo> "Punching trees gives me wood"
03:40:40 <Sgeo> http://www.levelupstudios.com/punching-trees-gives-me-wood
03:42:06 <elliott> Mathnerd314: "I killed your parents."
03:42:44 <Mathnerd314> elliott: *young, impressionable children*
03:42:58 <Sgeo> REMEMBER ME
03:43:24 <elliott> Mathnerd314: "I killed your parents... while they slept. If you sleep... well, who knows what could happen?"
03:43:29 <j-invariant> Tell them that Santa only exists because people beleive in it, and that they should be very careful not to beleive in anything BAD incase it becomes real. And then make up a scary story about this happening in the past with a terrible monster
03:43:40 <Sgeo> j-invariant, I hate you.
03:43:44 <elliott> "Of course... Santa only brings presents to those who sleep."
03:43:50 <elliott> Mathnerd314: "So, you know... it's your decision..."
03:44:41 <Mathnerd314> elliott: you realize that their parents are standing right next to them, and the call is probably on speakerphone?
03:44:53 <j-invariant> Sgeo: whuy
03:45:09 <elliott> Mathnerd314: "The reason why your parents seem to be next to you is ... they're replaced with duplicates."
03:45:14 <elliott> Mathnerd314: "And they want your lungs..."
03:45:33 <Mathnerd314> elliott: I don't think I can get this all in before they hang up
03:45:43 <elliott> Mathnerd314: practice talking REALLY QUICKLY
03:45:49 <Sgeo> j-invariant, when I was a little kid, I read a book about the power of wishes. Being a kid, I believed it.
03:46:03 <Sgeo> I really wish I never read that book
03:46:08 <elliott> j-invariant: note -- Sgeo is crazy
03:46:08 <j-invariant> hehe
03:46:53 <Mathnerd314> elliott: also, the people sitting next to *me* will overhear and pull we away from the phone
03:47:02 <Mathnerd314> s/we/me/
03:47:13 <elliott> Mathnerd314: wrestle with bears to practice your self-defence skills
03:47:32 <Sgeo> How long is this phone conversation that you can talk about it right now?
03:47:37 <Mathnerd314> elliott: this is just getting more and more insane.
03:47:43 <elliott> Mathnerd314: "Santa loves you... but he has a secret. If you swap around two letters in his name... it becomes Satan."
03:47:47 <elliott> Mathnerd314: [whisper] "Behind you."
03:48:08 <Mathnerd314> Sgeo: it's Norad tracks Santa, and my shift is in an hour
03:48:18 <Mathnerd314> elliott: that's more like it
03:48:24 <Sgeo> Ah
03:48:51 <Sgeo> Why propagate lying to little kids?
03:48:59 <elliott> lawl
03:49:05 <elliott> Santa is EVIL CHRISTIAN LIES
03:50:06 * Sgeo wonders if there's been any studies on how being taught that Santa is real or not as a kid affects the kid
03:50:46 <elliott> It doesn't.
03:51:43 -!- cheater99 has joined.
03:51:44 <Mathnerd314> elliott: you know from experience?
03:52:33 <elliott> Mathnerd314: From what I know of Sgeo finding out Santa wasn't real traumatised him and caused him to dedicate his life to overcoming the evil lies of Santa or something, so I hardly find the topic worth discussing *shrug*
03:53:10 <Mathnerd314> anyone seen the new Tron?
03:53:11 -!- cheater99 has quit (Read error: Connection reset by peer).
03:53:27 <elliott> not I
03:53:28 <Sgeo> elliott, uh, no
03:53:42 <elliott> Sgeo: I may have exaggerated slightly.
03:53:52 <Sgeo> elliott, I was raised Jewish
03:54:39 <elliott> Precisely.
03:54:41 <elliott> Jewish Santa.
04:00:44 <Mathnerd314> elliott: how did you get to be so insane?
04:01:56 <j-invariant> Is the new TRON good?
04:02:57 <elliott> Mathnerd314: ENDLESS AMOUNTS OF DRUGS*
04:02:58 <elliott> *lies
04:03:16 <elliott> j-invariant: The soundtrack is good (Daft Punk), dunno about the film :-P
04:04:16 <Mathnerd314> random google/twitter/reddit link: http://mimeti.ca/journal/?p=1481
04:05:57 <elliott> "# Tagged Christianity, Gnu, Jeff Bridges as Richard Stallman, root access, Sun Microsystems, Tron: Legacy"
04:05:59 <elliott> lol @ first tag
04:06:04 * elliott tries to figure that out :D
04:10:14 -!- cheater99 has joined.
04:11:05 -!- zzo38 has joined.
04:11:39 <Mathnerd314> yeah, a post carefully designed to be so obscure that elliott can't understand it
04:11:56 <elliott> ok it's christianity because of sacrifice son i think
04:11:58 <elliott> I THINK
04:11:58 -!- cheater99 has quit (Read error: Connection reset by peer).
04:13:10 -!- zzo38 has set topic: Count the seconds it takes to stop thinking about this sentence. | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
04:13:31 <Mathnerd314> -1
04:13:34 <elliott> argh :D
04:13:43 -!- elliott has set topic: IT IS CHRISTMAS AND SO WE WILL NOT HAVE ANY CONFUSING TOPICS | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
04:13:48 <zzo38> elliott: It's Christianity because it is Christmas, I think.
04:14:03 <elliott> but that has nothing to do with tron
04:14:06 -!- Mathnerd314 has set topic: IT IS SECULAR CHRISTMAS AND SO WE WILL NOT HAVE ANY CONFUSING TOPICS | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
04:15:12 <zzo38> Is "The sillier you are to the batsman, the closer you are." a confusing topic? (Probably it is because I switched a few words around)
04:16:09 <elliott> Yes :P
04:16:24 <elliott> google's right sidebar has baubles
04:17:13 -!- elliott has set topic: IT IS PAGAN YULE AND SO WE WILL NOT HAVE ANY CONFUSING TOPICS | HAPPY TWO DAYS AFTER FESTIVUS | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
04:17:42 <zzo38> The "proper" way is "The closer you are to the batsman, the sillier you are." (but of course it is not completely proper either)
04:17:47 <zzo38> O, now the new topic message is OK.
04:18:00 <zzo38> You can make Yule as well as Christmas, now!!
04:18:01 <j-invariant> is that cricket?
04:18:07 <elliott> i think so
04:18:09 <zzo38> j-invariant: What do *you* think?
04:18:23 <Mathnerd314> elliott: I don't see baubles
04:18:31 <j-invariant> I can't decide because I find it odd that "sillier" is a cricket concept
04:19:06 <zzo38> j-invariant: "Sillier" isn't really a cricket concept. But the fielders standing very close to the striking batsman are said to be in the "silly" position.
04:19:28 <j-invariant> oh
04:19:44 <elliott> aha
04:19:46 <elliott> http://www.google.com/search?q=christmas&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-GB:unofficial&client=iceweasel-a
04:19:49 <j-invariant> I will try watching cricket next time it's on TV
04:19:52 <elliott> it's only for "christmas"
04:19:59 <j-invariant> I ifeel like I am mssing out
04:20:06 <zzo38> (I don't think they say that the ones closer are "sillier", though.)
04:20:31 <elliott> j-invariant: don't worry, IIRC zzo38 says he has never watched a game of cricket
04:20:42 <elliott> j-invariant: so obviously you can reach this level of enjoyment without ever playing it or seeing it played
04:21:02 <j-invariant> lol
04:21:09 <zzo38> elliott: You are correct, I don't even know what channel it is on. (Or if it is on any channel in where I am, at all.)
04:21:22 <j-invariant> so you are a fan of cricket without having seen any games?
04:21:22 <elliott> j-invariant: so watching it is completely superfluous!
04:21:27 <elliott> yes, isn't it great :D
04:21:30 <j-invariant> I like this idea
04:21:43 <elliott> i think i'm going to be a fan of, uhh
04:21:48 <elliott> american football
04:21:53 <elliott> gonna buy ALL the rulebooks
04:22:00 <elliott> mark the superbowl in my calender
04:22:03 <elliott> not gonna watch it though
04:22:14 <elliott> *calendar
04:22:28 <zzo38> If I had time, and I know what channel, I might watch a Test match.
04:22:43 <elliott> but what about when they go for the real thing?!
04:22:56 <zzo38> elliott: What real thing?
04:23:04 <elliott> the non-Test match!
04:23:12 <Sgeo> Insane asylums: Where the sane go insane
04:23:14 <Sgeo> </tasteless>
04:23:18 <zzo38> elliott: That isn't what Test match means.
04:23:26 <elliott> bah!
04:23:38 <elliott> Sgeo: Don't worry, your stay will be peaceful.
04:24:05 <zzo38> Test match means the long game, with simpler rules but more strategy.
04:25:13 <elliott> i didn't even realise it was festivus :(
04:25:27 <j-invariant> what's festivus?
04:25:56 <elliott> j-invariant: http://en.wikipedia.org/wiki/Festivus
04:25:58 <elliott> j-invariant: For the rest of us.
04:26:04 <elliott> [["Festivus" was a term used by the 2000 Baltimore Ravens of the National Football League (NFL) and their fans to denote the NFL Playoffs. During the season, Ravens head coach Brian Billick, wanting his players to focus on every game, banned the word "playoffs." Players substituted the term "festivus" for playoffs and "Festivus Maximus" for the Super Bowl. The Ravens eventually won the 2001 Festivus Ma
04:26:04 <elliott> ximus, Super Bowl XXXV.]]
04:26:11 <elliott> I love it when Wikipedia humours other people's jokes for a sentence
04:26:18 <elliott> "The Ravens eventually won the 2001 Festivus Maximus"
04:26:39 <elliott> [[# In 2007, in a commercialization of the holiday, the first Festivus Pole Lot opened [20] in downtown Milwaukee.]] oh the irony
04:27:26 <j-invariant> I wish they would play Seinnfeld on TV
04:27:46 <elliott> i think they do ... on one of Sky's 7 billion channels
04:28:19 <elliott> at least here.
04:28:45 -!- cheater99 has joined.
04:30:33 -!- cheater99 has quit (Read error: Connection reset by peer).
04:42:00 -!- j-invariant has quit (Quit: leaving).
04:42:32 -!- pikhq has quit (Ping timeout: 264 seconds).
04:43:25 -!- elliott has quit (Quit: Leaving).
04:44:21 -!- pikhq has joined.
04:47:12 -!- cheater99 has joined.
04:48:38 * Sgeo wonders if elliott realizes I often rail against things that haven't affected me personally. I am utterly against faith healing, although no one I know has fallen for that BS. I am against thinking that vaccinations cause autism.
04:48:40 <Sgeo> etc. etc.
04:49:20 -!- cheater99 has quit (Read error: Connection reset by peer).
04:55:50 <zzo38> I am trying to think of what kind of algorithms and so on to use in TeXnicard for plurals and other word forms.
04:57:44 <zzo38> Maybe a list of patterns such as "]:1:s" and so on, but what do with other word forms?
04:59:31 <Sgeo> I am against drunk driving, texting while driving, and tired driving, but that might not be the best example -- I haven't learned to drive yet, and I have fears
05:00:07 <zzo38> Sgeo: I haven't learn to drive and never plan to. I don't want to have driving license, please.
05:01:34 -!- SgeoN1 has quit (Quit: Bye).
05:05:47 -!- cheater99 has joined.
05:08:17 -!- cheater99 has quit (Read error: Connection reset by peer).
05:25:13 -!- cheater99 has joined.
05:26:52 -!- cheater99 has quit (Read error: Connection reset by peer).
05:43:42 -!- cheater99 has joined.
05:44:51 -!- cheater99 has quit (Read error: Connection reset by peer).
05:46:15 * Sgeo vaguely gets offended at "Rides the short bus"
06:01:48 -!- cheater99 has joined.
06:03:25 -!- cheater99 has quit (Read error: Connection reset by peer).
06:05:30 -!- oerjan has joined.
06:06:10 <zzo38>
06:08:48 -!- zzo38 has quit (Remote host closed the connection).
06:09:56 <oerjan> <elliott> I love it when Wikipedia humours other people's jokes for a sentence
06:11:52 <oerjan> like in http://en.wikipedia.org/wiki/Will_Smith#Family_and_early_life. actually it looks sillier now than when i last saw it.
06:13:29 <oerjan> just an edit from yesterday though, so will probably be reverted
06:13:44 <quintopia> they should leave it :P
06:14:15 <oerjan> oh it wasn't _entirely_ serious before, it just that someone yesterday made it ungrammatical
06:17:54 <quintopia> filebin.ca is down
06:18:03 <quintopia> what's a non-suck file upload site these days?
06:20:12 -!- cheater99 has joined.
06:21:32 -!- cheater99 has quit (Read error: Connection reset by peer).
06:23:12 <oerjan> *it's
06:23:49 <quintopia> well yes. it does suck that it's down
06:23:52 <quintopia> make a recommendation
06:38:11 -!- cheater99 has joined.
06:40:55 -!- cheater99 has quit (Read error: Connection reset by peer).
06:56:43 -!- cheater99 has joined.
06:59:31 -!- cheater99 has quit (Read error: Connection reset by peer).
07:02:15 <pikhq> Merry Christmas!
07:02:30 <pikhq> (mêri kurisumasu!)
07:02:45 <pikhq> s/mêri/merî/
07:04:20 <oerjan> merry christmas
07:04:42 * oerjan chews som dried fish
07:04:44 <oerjan> *some
07:06:35 <pikhq> How Norwegian of you.
07:07:37 <oerjan> it's a norwegian specialty, although really i think it's mostly sold to tourists these days :D
07:09:35 <oerjan> stockfish it's called in english, according to the bag
07:10:26 <oerjan> in this case, dried haddock
07:10:43 <pikhq> Never heard of it. It just really seemed to me that dried fish would be something exceptionally Norwegian for some reason.
07:10:56 <oerjan> oh wait
07:11:24 <oerjan> it's also exported in large amounts to southern europe and (the worst quality) africa
07:13:40 <oerjan> although i don't think they usually eat it without further preparation
07:14:37 <oerjan> "Beside oil, gas, and income from the merchant fleet, stockfish is Norway's longest sustained export commodity, and the socioeconomically most profitable export over the centuries.
07:14:47 -!- cheater99 has joined.
07:16:44 <oerjan> oh ... and it's of course the pre-stage to lutefisk!
07:17:18 <pikhq> Aaaah, so it's a traditional part of the traditional penance!
07:17:18 -!- cheater99 has quit (Read error: Connection reset by peer).
07:17:19 <pikhq> :P
07:17:27 <oerjan> heh :D
07:18:38 * oerjan considered buying lutefisk, but apparently it cannot be prepared in microwave </bachelor frog>
07:20:11 <oerjan> "After sorting by quality, most of the stockfish is exported to Portugal, Italy and Croatia. In Norway and Iceland, the stockfish is mostly used as a snack and for lutefisk production. In Italy and Portugal, the fish (called stoccafisso) is soaked and used in various courses, and is viewed as a delicacy.
07:21:04 <pikhq> oerjan: According to Wikipedia, "Lutefisk sold in North America may also be cooked in a microwave oven."
07:21:14 <pikhq> Because here in AMERICA we disbelieve in proper cooking!
07:21:18 <oerjan> ah.
07:21:56 <oerjan> well the package had no instructions for microwaving, unlike basically all the other ready-meals
07:22:59 <pikhq> It amuses me that lutefisk is more commonly eaten in the US than in Norway...
07:23:22 <pikhq> (largely courtesy of Scandinavian immigrants up north)
07:23:25 <oerjan> heh. well iirc there are more people of norwegian ancestry in the US than in norway
07:23:41 <pikhq> Probably.
07:24:55 <pikhq> Hmm. Perhaps I should actually try lutefisk.
07:25:02 <coppro> pikhq: If you guys disbelieve in proper cooking, what do the Texans do?
07:25:22 * oerjan assumes they put everything on the barbecue
07:25:36 <pikhq> oerjan: Not only. But stereotypically, yes.
07:25:58 <coppro> ah, no, there you would be mistaken
07:26:06 <coppro> the Texans are actually aware of another cooking method
07:26:12 <pikhq> Deep frying!
07:26:16 <pikhq> Oh how they love it.
07:26:30 <coppro> a little too much, if you know what I mean
07:26:30 <oerjan> ah.
07:27:12 <pikhq> Allow me to introduce you to "chicken-fried steak".
07:27:20 * oerjan has vaguely <strike>heard</strike>read on reddit about deep fried bacon
07:27:20 <pikhq> It is a *deep fried steak*.
07:27:50 <pikhq> And it's actually fairly typical in the South.
07:28:16 <pikhq> Oh, and served with gravy.
07:28:27 <pikhq> As is mandatory.
07:29:04 <oerjan> consider a spherical texan in vacuum...
07:29:25 <pikhq> (the name comes because it's made very similarly to fried chicken. Another typical food.)
07:29:30 <pikhq> (also delicious.)
07:34:12 -!- cheater99 has joined.
07:36:38 -!- cheater99 has quit (Read error: Connection reset by peer).
07:37:07 -!- Sgeo has quit (Read error: Connection reset by peer).
07:38:14 -!- hagb4rd has quit (Ping timeout: 240 seconds).
07:53:18 -!- cheater99 has joined.
07:55:19 -!- cheater99 has quit (Read error: Connection reset by peer).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:11:48 -!- cheater99 has joined.
08:12:49 -!- poiuy_qwert has joined.
08:13:47 -!- cheater99 has quit (Read error: Connection reset by peer).
08:30:49 -!- cheater99 has joined.
08:32:47 -!- cheater99 has quit (Read error: Connection reset by peer).
08:49:12 -!- cheater99 has joined.
08:50:14 -!- oerjan has quit (Quit: leaving).
08:50:46 -!- sebbu2 has joined.
08:51:11 -!- cheater99 has quit (Read error: Connection reset by peer).
08:53:36 -!- sebbu has quit (Ping timeout: 260 seconds).
09:07:47 -!- cheater99 has joined.
09:09:11 -!- cheater99 has quit (Read error: Connection reset by peer).
09:25:49 -!- cheater99 has joined.
09:27:36 -!- cheater99 has quit (Read error: Connection reset by peer).
09:44:12 -!- cheater99 has joined.
09:47:05 -!- cheater99 has quit (Read error: Connection reset by peer).
10:03:44 -!- cheater99 has joined.
10:05:32 -!- cheater99 has quit (Read error: Connection reset by peer).
10:14:59 -!- hagb4rd has joined.
10:22:10 -!- cheater99 has joined.
10:25:31 -!- cheater99 has quit (Read error: Connection reset by peer).
10:40:50 -!- cheater99 has joined.
10:43:53 -!- cheater99 has quit (Read error: Connection reset by peer).
10:44:35 <Vorpal> morning
10:47:32 -!- Slereah has joined.
10:47:42 <Quadrescence> Vorpal: hi
10:48:18 <Vorpal> merry <day after xmas> I guess
10:48:27 <Quadrescence> it is xmas here tho
10:49:19 <Vorpal> Quadrescence, well, it's the 25th here, but only in Scandinavia (as far as I know) does the celebration take place on the proper day, the 24th
10:49:33 <Vorpal> ;P
10:50:23 <Quadrescence> i see
10:50:30 <Quadrescence> Vorpal: i pmed u
11:00:51 -!- cheater99 has joined.
11:03:18 -!- cheater99 has quit (Read error: Connection reset by peer).
11:04:38 -!- hagb4rd has quit (Ping timeout: 240 seconds).
11:20:12 -!- cheater99 has joined.
11:21:26 <Vorpal> anyone happens to know if handlers registered with atexit() are called on ctrl-c?
11:21:40 -!- cheater99 has quit (Read error: Connection reset by peer).
11:21:51 <Vorpal> or if I need a signal handler
11:23:21 <Vorpal> seems like the answer is no
11:24:01 <Vorpal> signal handlers it is then
11:38:16 -!- cheater99 has joined.
11:39:47 -!- cheater99 has quit (Read error: Connection reset by peer).
11:56:25 -!- cheater99 has joined.
12:03:02 -!- cheater99 has quit (Ping timeout: 240 seconds).
12:03:54 <fizzie> Yes, they are only called on normal exit.
12:04:09 <fizzie> Not for signal-caused termination.
12:15:51 -!- cheater99 has joined.
12:16:38 <Vorpal> fizzie, hm
12:16:58 <Vorpal> fizzie, do you know if calling complex stuff in a signal handler is safe if you exit at the end?
12:17:22 <Vorpal> fizzie, basically I just want to stop any sound generation in the beeper program if it is ctrl-ced
12:17:22 -!- cheater99 has quit (Read error: Connection reset by peer).
12:17:36 <Vorpal> fizzie, so I need a ioctl, then exit()
12:17:39 <Vorpal> an*
12:17:47 <Vorpal> and I'm not sure that is allowed in signal handlers
12:18:56 <fizzie> I'm not entirely sure. But perhaps there were some relaxations of the rules if the signal handler doesn't exist.
12:19:36 <Vorpal> fizzie, I thought so, but I can't find any mention of it
12:19:54 <fizzie> ncurses' signal handlers do manage the terminal cleanup (incl. raw → cooked input flip) on signals, so it must be possible.
12:20:02 <Vorpal> hm
12:24:03 <fizzie> Hm.
12:24:14 <fizzie> tcsetattr is in fact in the list of signal-safe functions, so they can call that.
12:24:16 <Vorpal> I do mask off all signals, so that should not be an issue at least
12:24:18 <fizzie> Unfortunately, ioctl is not.
12:24:22 <Vorpal> fizzie, gah
12:25:24 <fizzie> I assume it is because ioctl seems to be the sort of "catch-all" function of doing everything that doesn't fit anywhere else.
12:25:47 <fizzie> The list does include all kinds of IO stuff.
12:25:53 <Vorpal> hm
12:26:13 <Vorpal> fizzie, so I guess I need a volatile sig_atomic_t then
12:26:37 <fizzie> Well, there is that solution. Your sleeps should be interrupted by the signal anyway, I guess.
12:27:17 <Vorpal> fizzie, yes quite. But what if someone hits ctrl-c when it isn't sleeping, sure not likely, but it could happen
12:27:17 <fizzie> It might be "safe enough" to risk, though. And of course there's not much you can do if you get SIGKILL'd during a beep.
12:27:27 <Vorpal> which means I can't keep it off the "fast path"
12:27:33 <Vorpal> fizzie, hm
12:27:34 <Vorpal> true
12:28:21 <fizzie> You could sig-mask everything during the call to ioctl(); I think the only thing that's likely to break is if you call your beep-ioctl() while in the middle of another beep-ioctl() call.
12:28:24 <Vorpal> fizzie, ooh wait, init can (and afaik does) ignore kill. You could ptrace init (hm does that actually work?) and inject the beeping code into it.
12:28:25 <Vorpal> XD
12:28:51 <Vorpal> fizzie, you mean sigmask while calling it in the signal handler?
12:29:03 <fizzie> No, in the app side.
12:29:35 <Vorpal> fizzie, how would that help, the ioctl is just "start playing note at this frequency", then I have to sleep and do another ioctl to stop it playing
12:29:36 <fizzie> Or does sigblock just ignore it, instead of keeping it pending? I haven't really done much messing with signals.
12:30:09 <Vorpal> oh wait, you mean like that
12:30:10 <Vorpal> hm
12:30:31 <fizzie> I guess blocking does mean actual ignoring.
12:30:40 <Vorpal> hm
12:30:44 <fizzie> It would be nice if there was some way of keeping the signal in the pending state until it's safe to handle.
12:31:25 <fizzie> No, it actually does work the way I think it would, I think.
12:31:28 <fizzie> "If the action associated with a blocked signal is anything other than to ignore the signal, and if that signal is generated for the thread, the signal shall remain pending until it is unblocked, --"
12:31:50 <Vorpal> hm
12:32:02 <fizzie> So you could hopefully that way make sure the signal is handled when you're safely (for some values of safe) outside the ioctl.
12:32:55 <fizzie> Of course if you want to be literal with the spec, just avoiding the in-ioctl call-ioctl case doesn't make it fully proper.
12:33:30 <Vorpal> hah
12:33:33 <fizzie> Or you could fork(), let the child handle the beeping (and ^C-from-terminal reception), and have the parent issue an extra "stop with the beep" ioctl whenever the child terminates.
12:33:48 -!- cheater99 has joined.
12:34:15 <fizzie> Doesn't help if someone deliberately sends a SIGTERM (or some other) first to the parent, then to the child.
12:35:39 <fizzie> (Well, except if you add in the parent a signal handler to kill+wait the child first, and only then stop the beeping.)
12:36:11 -!- cheater99 has quit (Read error: Connection reset by peer).
12:36:40 <Vorpal> fizzie, this seems a lot more complex :P
12:37:02 <Vorpal> did the sig_atomic_t now
12:37:41 <Vorpal> fizzie, also debugging this /does/ make me nervous. sudo valgrind feels so.... scary
12:40:20 <fizzie> For debugging, could you just temporarily sudo chown the device node? (Or does it test against something else?)
12:43:38 <Vorpal> fizzie, do you know why kernel code seems to shun floating point?
12:44:11 <Vorpal> fizzie, well, the device node is /dev/console, which is as far as I know a bit special, in that it depends on the current vt
12:44:21 <Vorpal> I don't know what messing with chown there would do
12:44:34 <Vorpal> fizzie, possibly it would mess something up
12:45:44 <Vorpal> I could however run it from a non-X vt
12:45:52 <Vorpal> that should change permissions.
12:46:15 <Vorpal> hm actually it doesn't
12:46:25 -!- oerjan has joined.
12:47:07 <fizzie> Floating point doesn't exist everywhere, which might be a good reason to shun it.
12:47:18 <Vorpal> fizzie, I use the same interface as beep(1) but I can't even get beep to work on a vt as a normal user, though I remember that it worked in the past, and the man page states it should work
12:47:26 <Vorpal> (see section "ioctl wackiness"
12:47:27 <Vorpal> )
12:52:27 -!- cheater99 has joined.
12:54:40 <Vorpal> fizzie, fun: http://en.wikipedia.org/wiki/PC_speaker#Pulse-width_modulation
12:54:54 -!- cheater99 has quit (Read error: Connection reset by peer).
12:57:24 <Vorpal> fizzie, btw I think I found out why that tandy sounded so different from the ibm in that video
12:57:45 <Vorpal> according to wikipedia tandy had 3 channels
12:57:51 <Vorpal> unlike the ibm which had one
13:00:37 <Vorpal> it was mentioned in http://en.wikipedia.org/wiki/GW-BASIC not sure how reliable it is
13:11:28 -!- cheater99 has joined.
13:14:55 -!- cheater99 has quit (Read error: Connection reset by peer).
13:21:07 <fizzie> Ooh, three-channel beeping.
13:26:19 -!- TLUL has joined.
13:26:51 <TLUL> Message on all protocols: Merry Christmas!!!!!
13:29:28 -!- cheater99 has joined.
13:32:14 -!- cheater99 has quit (Read error: Connection reset by peer).
13:48:41 -!- cheater99 has joined.
13:50:58 -!- cheater99 has quit (Read error: Connection reset by peer).
13:56:41 <Vorpal> fizzie, quite
13:57:10 <Vorpal> fizzie, btw... tried MT-32 emulation in scummvm. But... ALSA lib pcm.c:7245:(snd_pcm_recover) underrun occured
13:57:20 <Vorpal> I think it may be time to upgrade this computer
13:58:03 <Vorpal> besides it didn't sound like the videos
13:58:12 <Vorpal> (why I wonder)
13:58:26 <Vorpal> (it sounded like even more advanced synthing)
13:58:33 <Vorpal> could be sample rate or something I guess
13:58:38 -!- oerjan has quit (Quit: leaving).
13:59:53 <Vorpal> huh, my thinkpad has two cards in /proc/asound... card0 and card29
14:00:13 <Vorpal> card29/id says ThinkPadEC
14:00:17 <Vorpal> card0/id says Intel
14:00:22 <Vorpal> card0 is what I would expect
14:01:28 <Vorpal> ThinkPadEC just has one control in alsamixer, named Console. And it is just mute/unmute, no volume control
14:03:10 -!- Mathnerd314 has quit (Disconnected by services).
14:03:31 -!- Mathnerd314_ has joined.
14:04:14 -!- Mathnerd314_ has changed nick to Mathnerd314.
14:07:30 -!- cheater99 has joined.
14:08:58 -!- cheater99 has quit (Read error: Connection reset by peer).
14:10:43 -!- Mathnerd314 has quit (Disconnected by services).
14:13:14 -!- Mathnerd314 has joined.
14:21:34 -!- FireFly has joined.
14:25:30 -!- cheater99 has joined.
14:27:35 -!- cheater99 has quit (Read error: Connection reset by peer).
14:30:21 <fizzie> Sounds like a ThinkPad console beep.
14:31:21 <fizzie> The motherboard-integrated Radeon card has an Alsa entry too, because there's a HDMI port that can be used for audio output.
14:31:36 <fizzie> There's a single alsamixer control called "S/PDIF", and it's also just mute/unmute.
14:32:26 -!- ais523 has joined.
14:43:42 -!- cheater99 has joined.
14:45:37 -!- cheater99 has quit (Read error: Connection reset by peer).
14:48:01 -!- TLUL has changed nick to TLUL|afk.
14:57:28 -!- TLUL|afk has changed nick to TLUL.
15:02:16 -!- cheater99 has joined.
15:03:31 -!- cheater99 has quit (Read error: Connection reset by peer).
15:07:03 -!- TLUL has changed nick to TLUL2.
15:07:33 -!- TLUL2 has changed nick to TLUL.
15:13:02 -!- ais523 has quit (Ping timeout: 240 seconds).
15:13:17 -!- TLUL has changed nick to TLULbot_.
15:13:18 -!- TLULbot_ has quit (Changing host).
15:13:18 -!- TLULbot_ has joined.
15:14:14 -!- TLULbot_ has changed nick to TLUL2.
15:14:35 -!- TLUL2 has changed nick to TLUL3.
15:14:57 -!- TLUL3 has changed nick to TLUL4.
15:15:22 -!- TLUL4 has changed nick to TLUL5.
15:16:25 -!- TLUL5 has changed nick to TLUL.
15:16:26 -!- TLUL has quit (Changing host).
15:16:26 -!- TLUL has joined.
15:20:14 -!- cheater99 has joined.
15:21:48 -!- cheater99 has quit (Read error: Connection reset by peer).
15:32:24 -!- hagb4rd has joined.
15:37:44 <Vorpal> <fizzie> Sounds like a ThinkPad console beep. <-- ah perhaps
15:38:16 -!- cheater99 has joined.
15:38:38 <Vorpal> fizzie, the thinkpad just beeps using the normal speakers. Normal volume control affects it
15:40:28 -!- cheater99 has quit (Read error: Connection reset by peer).
15:57:16 -!- cheater99 has joined.
15:58:07 -!- ais523 has joined.
15:59:02 -!- cheater99 has quit (Read error: Connection reset by peer).
16:04:12 -!- oerjan has joined.
16:15:25 -!- cheater99 has joined.
16:17:30 -!- cheater99 has quit (Read error: Connection reset by peer).
16:17:47 -!- Slereah has quit (Ping timeout: 240 seconds).
16:21:25 -!- Slereah has joined.
16:33:49 -!- cheater99 has joined.
16:36:34 -!- cheater99 has quit (Read error: Connection reset by peer).
16:46:22 -!- jcp has quit (Ping timeout: 265 seconds).
16:53:16 -!- jcp has joined.
16:53:19 -!- cheater99 has joined.
16:54:55 -!- cheater99 has quit (Read error: Connection reset by peer).
17:11:19 -!- cheater99 has joined.
17:13:19 -!- cheater99 has quit (Read error: Connection reset by peer).
17:21:49 -!- Wamanuz5 has joined.
17:23:09 -!- Wamanuz4 has quit (Read error: Operation timed out).
17:29:51 -!- cheater99 has joined.
17:32:13 -!- cheater99 has quit (Read error: Connection reset by peer).
17:32:59 <nooga> hoooh
17:33:26 <oerjan> haaah
17:34:12 <ais523> hieeee
17:34:42 <ais523> I've been meaning to ask; what /is/ the other meaning of esoteric?
17:34:45 <ais523> the non-programming one, that is
17:35:00 <oerjan> hidden mystic knowledge?
17:35:10 <ais523> hmm, perhaps
17:35:27 <oerjan> "confined to and understandable by only an enlightened inner circle; "a compilation of esoteric philosophical theories"
17:35:41 <Slereah> Well, that is basically also the meaning we use!
17:35:45 <Slereah> Except less magical
17:35:47 <oerjan> `define esoteric
17:35:58 <Slereah> `swedish esoteric
17:36:01 <Slereah> !swedish esoteric
17:36:27 <HackEgo> isutereec
17:36:35 <HackEgo> * confined to and understandable by only an enlightened inner circle; "a compilation of esoteric philosophical theories" \ [23]wordnetweb.princeton.edu/perl/webwn \ * Esotericism or Esoterism is a term with two basic meanings. In the dictionary sense of the term, "esoterism" signifies the holding of opinions
17:36:37 <EgoBot> esutereec
17:36:43 <oerjan> ooh it's working
17:37:11 <oerjan> huh
17:37:18 <oerjan> `swedish esoteric
17:37:19 <HackEgo> isutereec
17:37:22 <oerjan> !swedish esoteric
17:37:23 <EgoBot> esutereec
17:38:14 <Slereah> Swedish and finno-swedish
17:42:15 <nooga> russe-norsk
17:42:42 <nooga> ( http://en.wikipedia.org/wiki/Russenorsk )
17:43:18 <oerjan> i don't think that survived the soviet union. maybe it'll redevelop now...
17:44:30 <oerjan> although nowadays it's probably replaced by english
17:46:32 <nooga> uhuh
17:48:26 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner]).
17:48:49 -!- cheater99 has joined.
17:52:13 -!- cheater99 has quit (Read error: Connection reset by peer).
18:19:17 <nooga> uhuh
18:19:33 <oerjan> ^echo uhuh
18:19:33 <fungot> uhuh uhuh
18:22:30 <nooga> anyone tried Scala?
18:22:32 <nooga> or Io?
18:42:14 -!- TLUL has left (?).
18:42:27 -!- TLUL has joined.
18:49:59 -!- KingOfKarlsruhe has joined.
19:27:13 -!- elliott has joined.
19:28:21 <ais523> oh, hi elliott btw
19:28:27 <elliott> hi ais523!
19:28:29 <elliott> happy christmas
19:28:36 <elliott> and/or other holiday of various kinds
19:28:36 <ais523> (saying "btw" at the end of that makes it look like the channel's been active...)
19:28:44 <ais523> and indeed, insert typical christmas greeting here
19:28:58 <elliott> i got SICP :)
19:29:06 <ais523> have you read it today?
19:29:18 <elliott> like ten pages! finally i am a real programmer
19:30:29 <elliott> `addquote * oerjan considered buying lutefisk, but apparently it cannot be prepared in microwave </bachelor frog>
19:30:31 <HackEgo> 255) * oerjan considered buying lutefisk, but apparently it cannot be prepared in microwave </bachelor frog>
19:30:45 <elliott> only a Norwegian could decide not to bother with the delights of lutefisk because it would be too inconvenient
19:33:11 <elliott> 04:47:07 <fizzie> Floating point doesn't exist everywhere, which might be a good reason to shun it.
19:33:14 <elliott> And also it is scary.
19:33:34 <elliott> 05:57:10 <Vorpal> fizzie, btw... tried MT-32 emulation in scummvm. But... ALSA lib pcm.c:7245:(snd_pcm_recover) underrun occured
19:33:39 <elliott> Vorpal: You did pirate the ROMs, right?
19:33:53 <elliott> Doesn't work without them.
19:33:58 <elliott> 09:34:42 <ais523> I've been meaning to ask; what /is/ the other meaning of esoteric?
19:34:07 <elliott> ais523: Aleister Crowley magick, except more general
19:34:23 <elliott> ais523: that's a hideous overgeneralisation, but
19:34:48 <elliott> anybody stupid enough to think #esoteric on freenode would be about esoterica probably hasn't got a better definition themselves
19:36:04 <elliott> i am, incidentally, very tired
19:40:38 <ais523> so am I
19:40:49 <ais523> I haven't been online recently because I've been pretty ill
19:40:58 <ais523> still am, actually, but Christmas takes precedence for some reason
19:41:38 <elliott> sorry to hear it
19:41:43 <oerjan> heck, some people put off dying until after christmas
19:41:44 <elliott> did those keyloggers ever get written?
19:41:51 <ais523> oh right, I have to mark those somehow
19:41:59 <ais523> they've been submitted, and I have a bunch of them to mark
19:42:03 <ais523> but I haven't dared to actually look at them yet
19:42:09 <ais523> and now may not be a good time
19:43:19 <elliott> heh
19:43:29 <ais523> I need to find a suitable VM to mark them in
19:43:35 <ais523> and it'll be a pain rebooting it all the time
19:43:56 <ais523> a) because the kernel modules in question probably won't unload properly, b) because the very nature of the problem means that pretty much any error will just crash the system
19:46:44 <elliott> ais523: qemu? that can boot without assembling a hard disk image
19:47:02 <elliott> ais523: -kernel bzImage -initrd foo -append root=blah
19:47:08 <elliott> well, root= just beind an example
19:47:49 <elliott> http://syntensity.com/static/python.html <-- CPython, compiled to LLVM bitcode, compiled to JavaScript.
19:50:23 <elliott> ais523: here's a riddle/thought experiment for you
19:50:30 <elliott> ais523: (car '()) is an invalid Scheme program.
19:50:37 <elliott> ais523: is (eval '(car '())) a valid Scheme program?
20:01:41 <elliott> Gregor: http://syntensity.com/static/python.js
20:15:51 -!- Wamanuz5 has quit (Remote host closed the connection).
20:18:17 -!- Wamanuz has joined.
20:24:08 -!- zzo38 has joined.
20:25:01 <zzo38> Although I have never watched any cricket game on television or in a stadium, I have heard the game on the radio, I have seen photographs, and I have seen scorecards.
20:30:29 <elliott> > ((evil '(lambda (x) x) '())
20:30:29 <elliott> '(hello))
20:30:29 <elliott> Error: attempt to call a non-procedure
20:30:29 <elliott> ('hello '())
20:30:30 <elliott> huh what
20:33:58 <elliott> [[ Interesting, but please, the wiki is really not a discussion board. We were using it that way before, but it sucked. Tell the forum about this matter. --Graue 02:02, 24 May 2006 (UTC)]], in reply to ais523
20:34:05 <elliott> Graue must hate us so much :)
20:34:20 -!- poiuy_qwert has quit (Quit: This computer has gone to sleep).
20:34:33 <zzo38> elliott: I disagree. The wiki works better as the discussion board than the forum we have. IRC can also be used for discussion.
20:37:09 -!- Sgeo has joined.
20:37:25 * Sgeo needs to take a chill pill sometimes
20:38:00 <oerjan> liquid nitrogen with raspberry flavor
20:38:37 <Sgeo> =P lol
20:38:52 * Sgeo wonders what it would take to have solid nitrogen
20:42:47 <oerjan> 63.153 K, -210.00 °C, -346.00 °F
20:43:52 <oerjan> hm that's only about 15 °C below boiling
20:48:02 <ais523> elliott: I think everyone disagrees with em on that
20:48:15 <ais523> wikis make very good forums
20:48:25 <ais523> I find them much more convenient than most actual forums
20:48:32 <elliott> ais523: I agree completely -- I was remarking on the fact that /none/ of Graue's policies seem to be popular
20:48:41 <zzo38> ais523: I think you are correct. I agree with you about that
20:48:54 <elliott> The "no categories without discussing" thing is just weird, the keeping "User:" in links is weird too...
20:49:00 <ais523> zzo38: I have watched cricket games, actually
20:49:16 <ais523> it's rarely worth going to a stadium to actually watch the game, the field's so big you can hardly see what's going on, on TV works a lot better
20:49:20 <ais523> (I've watched them both ways)
20:49:23 <elliott> so, anyway, I'm trying to beat cpressey at his own game
20:49:26 <ais523> also, I agree with User: in links
20:49:38 <ais523> if you know me from Wikipedia, I'm a stickler for namespace separation
20:49:57 <elliott> specifically, I'm trying to beat him at the "Scheme subset with a short self-interpreter" game
20:50:14 <elliott> since, well, even reading the first few pages of SICP does things like this to you
20:50:42 <zzo38> elliott: You can ignore any policies that don't work that are bad except for the one about public domain. (Even wikipedia has one rule "ignore all rules" and it is basically of a similar idea; if the rule doesn't work, do something else. But also fix things to meet the rule if they should. There are rules for a reason!)
20:50:55 <elliott> zzo38: that's not such a wise idea, at least it wasn't when Graue was around
20:51:00 <elliott> his reaction to people breaking policies was to yell at them a lot
20:51:03 <elliott> :)
20:51:09 <ais523> only once, but it was pretty famous
20:51:15 <ais523> combined with a mass revert of changes and a short ban
20:51:22 <elliott> ais523: more than once
20:51:25 <ais523> really?
20:51:31 <ais523> the year categories thing is the only thing I can think of
20:51:33 <elliott> yes, I found another Graue-rage recently, lemme grep the logs
20:51:38 <zzo38> ais523: You have watched at a stadium and television. What match was it and where was the stadium?
20:51:38 <elliott> ais523: wasn't he against them?
20:51:51 <ais523> zzo38: it was Edgbaston stadium in Birmingham, more than once
20:52:04 <ais523> elliott: not in principle, e was against the fact they were implemented without asking for permission first
20:52:24 <ais523> e doesn't mind the categories /themselves/
20:52:40 <Sgeo> What was wrong with the year categories?
20:52:48 <elliott> <ais523> elliott: not in principle, e was against the fact they were implemented without asking for permission first
20:52:54 <elliott> ais523: grr, ever said something and then been unable to find it in the logs?
20:53:06 <zzo38> elliott: You are correct partially I think, of course there are rules for a reason. But you have to know the purpose of the rules too.
20:53:09 <ais523> elliott: I can't remember
20:53:18 <zzo38> ais523: I mean, what was the match format?
20:53:20 <ais523> it seems plausible that I've done that, but I can't think of an occasion on which it's happened
20:53:32 <ais523> zzo38: five-day, either international or county
20:53:40 <ais523> but not seeing all five days
20:53:40 <elliott> 10.12.22:19:37:47 <elliott> haha, and some classic Graue rage:
20:53:40 <elliott> 10.12.22:19:37:49 <elliott> [[I reverted your purge of Talk:Udage because that isn't the way wikis work. You do not own that page, nor do you own the Udage article. Do not delete valid information from this site again. --Graue 19:16, 10 Oct 2005 (GMT)]]
20:53:47 <Sgeo> Is Graue even still active?
20:53:47 <ais523> more commonly 1 or 2
20:53:51 <ais523> Sgeo: nope
20:53:53 <elliott> (he wiped the talk page after it got big, IIRC)
20:53:55 <ais523> e only wakes up in emergencies nowadays
20:54:13 <Sgeo> That's more active than I
20:54:15 <Sgeo> >.>
20:54:24 <elliott> (I agree with not blanking the pages, but the reaction is a bit over-the-top)
20:54:29 <ais523> me, Keymaker and cpressey deal with most of the spam problem, that's pretty much the only thing that normally needs admin intervention
20:54:55 <zzo38> ais523: Well, that is OK. If there is enough of wrong thing that is emergency, it ought to be corrected like that.
20:54:57 <elliott> ais523: I like how cpressey didn't even realise he was a sysop until I told him
20:55:01 <elliott> ais523: despite him deleting pages before that
20:55:49 <elliott> grr, I just realised I'm basically cheating at the self-interpreter game
20:55:53 <elliott> accidentally
20:56:16 <elliott> because I'm using various non-trivial procedures, and then just including them in the self-interpreter with
20:56:18 <elliott> (foo . ,foo)
20:56:20 <elliott> in the default environment
20:56:29 <elliott> i.e., abusing metacircular privileges to vastly inflate the language
20:57:17 * Sgeo is excited for the upcoming release of Newspeak, despite not having enough time to submit my trivial IDE changes
20:57:32 <Sgeo> Well, change. And it would probably be easier to describe the change than to package it up in any way
20:57:43 <elliott> uh oh
20:57:45 <elliott> my mouse is ... leaking
20:58:40 -!- elliott has left (?).
20:58:44 -!- elliott has joined.
21:00:22 <elliott> ais523: is there an idealised algol self-interp? :)
21:02:13 <ais523> it doesn't have I/O, so no
21:02:39 <ais523> also, the interesting subsets are incapable of self-interpreting due to being FSMs
21:02:52 <Sgeo> elliott, I assume that's a metaphor for trying to avoid me?
21:03:00 <elliott> Sgeo: what
21:03:11 <Sgeo> What could a mouse possibly leak?
21:03:11 <elliott> ais523: you don't need IO, just have a program that is a function
21:03:14 <elliott> taking a program and returning the result
21:03:17 <Sgeo> *non-biological mouse
21:03:21 <elliott> Sgeo: i wish i knew
21:04:16 <ais523> that reminds me of my /quit I think my computer is on fire
21:04:35 <ais523> elliott: it doesn't have any sort of string type either
21:04:48 <ais523> not even an easy equivalent to lists
21:04:57 <elliott> ais523: sounds useful!
21:05:11 <ais523> idealised, see?
21:05:28 <Sgeo> Have I actually talked about Alluded To Female enough in here for her to be called Alluded To Female?
21:05:38 <elliott> Sgeo: No, and please don't make it that way.
21:05:45 <ais523> PSOX!
21:05:49 <elliott> ais523: assume a perfectly spherical programming language in a vacuum
21:06:09 <Sgeo> So why did you mention Alluded To Female the other day?
21:06:21 <elliott> I did?
21:06:51 <ais523> elliott: hmm, you know, say, Iterable<bool[8]> from Java, or your equivalent in insert-favourite-programming-language-here?
21:07:01 <elliott> ais523: sure
21:07:06 <ais523> that's pretty much as close to a string as you can get in Idealised Algol
21:07:10 <Sgeo> When I said something about sending a message to Fidelity, and you thought Fidelity was her name
21:07:29 <ais523> it's probably close enough to work from
21:07:29 <elliott> ais523: so all strings are 8 bytes? :)
21:07:37 <ais523> no
21:07:37 <elliott> (say yes)
21:07:39 <elliott> darn
21:07:52 <ais523> they're things that support the same methods as a list of 8-bit characters would
21:08:00 <elliott> ais523: oh, bool
21:08:02 <elliott> not byte
21:08:05 <ais523> yep
21:08:19 <ais523> who needs data types other than booleans?
21:08:35 <ais523> (actually, IA's type system is hilarious; it's based around 1-bit integers and 0-bit integers)
21:08:40 <elliott> ais523: idealised algol has functions, does it not?
21:08:43 <ais523> (and the 0-bit integers are used more often)
21:08:46 <elliott> then why does it need booleans?!
21:08:49 <ais523> it does, but only in the typed lambda calculus sense
21:08:55 <Sgeo> 0-bit... integers..
21:09:07 <ais523> I suppose it doesn't actually need booleans
21:09:07 <elliott> ais523: you can do booleans in typed lambda calculus :P
21:09:15 <elliott> *the typed
21:09:20 <ais523> on the other hand, I think they end up equivalent in the final circuitry
21:09:58 <ais523> (that is, \a\b.a and \a\b.b are wired the same way as true and false)
21:09:59 <elliott> your language is bloated! I'm switching to Idealised Concurrent Lambda Calculus
21:10:32 <elliott> ais523: how generic is idealised algol? e.g. could there be a VHDL compiler outputting idealised algol?
21:10:39 <ais523> you have three or possibly four base types: command (0-bit int), integer (1-bit int), variable (like bool* in C), and perhaps Semaphor
21:10:43 <ais523> *semaphore
21:10:51 <elliott> 0-bit int is just data () = (), surely?
21:10:54 <ais523> there's some debate as to whether the semaphores are intrinsically needed
21:10:56 <ais523> elliott: yep
21:11:06 <ais523> but it can still be evaluated for its side effects
21:11:07 <elliott> ais523: also, like (bool *) sans pointer arithmetic, one presumes
21:11:13 <ais523> and yes
21:11:21 <ais523> more like (bool &) from C++, except everyone hates C++
21:11:25 <elliott> ais523: I would call 0-bit int "the unit type"
21:11:31 <elliott> 0-bit integer just makes it sound strange for no real reason
21:11:31 <ais523> or ref bool from ML
21:11:36 <elliott> ais523: *bool ref
21:11:36 <ais523> elliott: oh, we normally call it "command"
21:11:39 <ais523> but that sounds even weirder
21:11:41 <ais523> elliott: err, yes
21:11:50 <elliott> in ML, type parameters come first, because inconsistency is a virtue
21:12:28 <ais523> IA/ICA is pretty heavily inspired by ML, actually
21:12:42 <ais523> except it's call-by-name
21:13:11 <ais523> (this makes it difficult to determine whether it's lazy or eager, it has properties of both; mostly lazy, I think)
21:13:21 <Sgeo> 4:30 I quit IRC
21:13:23 <Sgeo> (local time)
21:13:38 <Sgeo> No, elliott. Not forever.
21:14:01 <elliott> you have 16 minutes
21:14:03 <elliott> oerjan!
21:14:13 * oerjan commends tswett on counting the number of sand particles in Nubia
21:14:55 <oerjan> no wonder he was away from the channel so long
21:15:06 <Sgeo> oerjan, hmm?
21:16:49 <elliott> wow, beating cpressey is really hard
21:17:41 <elliott> oerjan: can you get a rubber hose to beat cpressey with?
21:18:04 <ais523> elliott: just steal something from a shop, kill the Kops that come after you, and take theirs
21:18:12 <ais523> (note: this may only work in NetHack, not in Real Life)
21:18:43 -!- poiuy_qwert has joined.
21:18:46 <elliott> ais523: killing the Kops is not exactly /easy/
21:18:53 <ais523> it is, Kops are pretty rubbish
21:18:55 <oerjan> elliott: here ===================================
21:18:58 <ais523> it's the shopkeeper that's the issue
21:19:03 <ais523> the Kops are just there to get in the way
21:19:10 <elliott> ais523: oh, the Kops are fine, I forgot
21:19:14 <elliott> ais523: I was thinking of the Watch
21:19:22 <elliott> who, er, bad memories
21:19:36 <elliott> ineiros: YO HMOD UPDATE OUT CUZ IT'S LIKE THEI T
21:19:38 <elliott> *THE IT
21:19:47 <Sgeo> elliott, you escaped the watch IIRC
21:20:09 <elliott> Sgeo: Correction -- I, pressing the keys that #nethack told me to, escaped the Watch.
21:20:12 <elliott> #nethack is a very good player.
21:30:12 <Sgeo> Well, bye all
21:30:24 <Sgeo> Bye all.
21:30:35 <zzo38> I am sorry I am late.
21:30:44 <zzo38> ais523: What was the result of the match?
21:30:45 -!- Sgeo has quit (Quit: Leaving).
21:33:10 <ais523> zzo38: I can't remember
21:33:12 <ais523> it was ages ago
21:33:19 <ais523> I don't think I was supporting either team
21:34:11 <elliott> emacs appears to rearrange itself in my task bar if i move it
21:34:15 <elliott> go next to firefox, stupid thing!
21:34:22 <elliott> no, don't go next to two firefoxes
21:34:22 <zzo38> I found a description of algorithm for plurals, I might use something similar with TeXnicard.
21:34:27 <elliott> I dragged you next to /one/
21:35:24 <zzo38> But first I should add conditional processing commands @< ... @>
21:35:43 <zzo38> Or perhaps @[ ... @]
21:36:02 <zzo38> ais523: Do you need to support either team?
21:36:52 <zzo38> ais523: Do you remember if there were any ties or draws?
21:37:04 <elliott> wow
21:37:07 <elliott> my program has a cadadr in it
21:37:09 <elliott> (cadadr e)
21:37:20 <ais523> zzo38: draws are quite common, ties are pretty rare
21:37:25 <ais523> (draws happen when you run out of time)
21:39:32 <zzo38> ais523: How common are draws? (I have only seen a few scorecards, so I don't know.)
21:43:49 <ais523> pretty common
21:44:03 <ais523> sometimes sides get greedy and get too many runs, then can't win in time
21:44:17 <ais523> and sometimes sides stall for a draw because they know that declaring would probably make them lose
21:44:40 <ais523> sometimes, about half the games are draws
21:44:45 <elliott> grr, I think I might have to convert my defines into lets
21:44:55 <ais523> (to the extent that county cricket introduced a tiebreak rule for draws, to stop people doing it so much)
21:45:33 <coppro> declare?
21:45:46 <ais523> concede the rest of your innings
21:45:49 <zzo38> coppro: Declare an innings closed is you end your current innings early.
21:46:08 <ais523> you only do it when you think you'll win anyway, and fear that if you continue you'll end up running out of time and only getting a draw
21:48:50 <elliott> OK, I /think/ I have BEGIN and DEFINE support now.
21:52:49 <elliott> ugh, my default environment is going to be ugly though
21:52:51 <elliott> since I have no LIST
21:53:24 <elliott> (cons (cons 'cons cons) ...)
21:53:30 <elliott> Cons cons cons cons cons cons cons.
21:57:09 <zzo38> Here is one possible rule (I don't know how well it would work): If one side has wasted more time than the other side in total, and the game is running out of time while the side wasting more time is the batting side, the batting side loses. (I don't know how well such a rule would work.)
21:59:37 <ais523> how can you tell if a side is wasting time?
21:59:44 <ais523> the rule actually used is based on runs per over
21:59:48 <ais523> which is similar, in a way
21:59:57 <ais523> in that timewasting tends to be quite low on runrate
22:00:09 <ais523> whereas trying to hurry up so you can declare tends to be quite hich
22:00:10 <ais523> *high
22:03:16 <elliott> "I need help with another scheme question. I have four different functions that check four conditions to determine validity of the input. Each function checks one condition and returs a boolean. Normally, I would just check ((func1) && (func2) && .....)and print out the boolean. But appearantly Scheme48 doesn't allow logic operators. So How do I check the four functions without using logic operators?"
22:03:56 <ais523> it'd be something like (lazy-and (func1) (func2) (func3) (func4)), although I don't know what the function's actually called in Scheme
22:04:11 <elliott> ais523: "and", and it's a special form; but I know this, I was just quoting it for amusement.
22:04:12 -!- wareya has quit (Read error: Connection reset by peer).
22:04:16 <zzo38> ais523: I don't know if you can tell if a side is wasting time much, except possibly in a computer game. Or maybe with TV recording. I don't know.
22:04:20 <elliott> [[I tried doing cond( (func1) (cond( (func2) (cond (func3) (cond (func4)#T))))) but it returns
22:04:20 <elliott> Warning: invalid variable reference
22:04:20 <elliott> cond
22:04:20 <elliott> #{Package 173 user}
22:04:20 <elliott> ; no values returned]]
22:04:23 <elliott> ^ lol
22:04:54 -!- wareya has joined.
22:05:37 <zzo38> Maybe runs per over might work. Either way, if you think you could tell which side wasted more time in the match in total, do you think my rule would work? (I don't know for sure?)
22:09:37 <elliott> In any case, I strongly think that the primitive for assigment
22:09:37 <elliott> should be SET and not SET!. In fact, since no one likes assignment
22:09:37 <elliott> anyway, I don't see any reason why anyone should object to just
22:09:37 <elliott> leaving this undefined in the standard.
22:09:51 <elliott> -- Kent M Pitman, the RRRS standardisation list, 1984
22:09:54 <elliott> If only.
22:13:39 <zzo38> elliott: What is it that you are making?
22:14:52 <elliott> zzo38: I am attempting to implement an interpreter of a restricted subset of Scheme in that subset, such that the self-interpreter is shorter than that of Chris Pressey's Pixley effort (http://catseye.tc/projects/pixley/) without the language being much bigger at all.
22:17:51 <zzo38> OK. Try to make that.
22:23:51 -!- j-invariant has joined.
22:27:01 <elliott> wow I think this self-interpreter might actually work right now
22:27:04 -!- TLUL has changed nick to TLUL5.
22:27:09 -!- TLUL5 has changed nick to Cook_Me_Flax.
22:27:19 <elliott> well, if I fix the ONE OR TWO errors :P
22:28:27 <elliott> hmm, where _is_ that unterminated list
22:30:55 <elliott> ais523: you never tried that riddle, did you?
22:31:07 <elliott> well
22:31:08 <elliott> "riddle"
22:31:12 <elliott> in a very lose sense of the word
22:41:28 <elliott> ugh, more bugs
22:42:11 <zzo38> This is the document I found about English plural: http://www.csse.monash.edu.au/~damian/papers/HTML/Plurals.html (I plan to use a somewhat more generic algorithm in TeXnicard)
22:42:14 -!- zzo38 has quit (Remote host closed the connection).
22:43:38 <j-invariant> What is TeXnicard?
22:45:05 <elliott> yay damian conway
22:48:05 <j-invariant> im not sure ym little finger can take much more proving
22:48:49 <elliott> j-invariant: you prove things with your little finger?
22:48:55 <elliott> i tend to use my other fingers, too
22:49:47 -!- TLUL_ has joined.
22:50:14 -!- KingOfKarlsruhe has quit (Remote host closed the connection).
22:50:25 <j-invariant> even with Ctrl moved to the caps-lock key, it's taking a beating with all these lemmas
22:50:51 <elliott> j-invariant: the best place for ctrl is where alt is (on pc keyboards) IMO, that's where it originally was
22:50:54 <elliott> and alt was at the extremities
22:51:00 <elliott> originally = on ibm pc keyboards
22:51:01 <j-invariant> hmm
22:51:12 <elliott> although on os x of course this is done by default
22:51:14 <elliott> well, with apple keyboards
22:51:20 <j-invariant> any idea how to tell emacs this ? :D
22:51:26 <elliott> not in emacs, though, which still uses ctrl...
22:51:36 <elliott> j-invariant: dunno, what OS?
22:51:41 <j-invariant> ubuntu
22:52:04 <coppro> elliott: I have to say I like them where they are
22:52:38 -!- Cook_Me_Flax has quit (Ping timeout: 240 seconds).
22:52:55 <elliott> j-invariant: system → preferences → keyboard → layouts → options... → ctrl key position → ...never mind, it doesn't have an option for it
22:53:04 <elliott> j-invariant: you could mess around with xkb. but that way madness lies.
22:54:53 <elliott> j-invariant: telling emacs, though... I /think/ you can tell it to consider another key to be C-
22:55:07 <elliott> but i don't know how.
23:00:40 <elliott> wait /what/
23:01:41 -!- ais523 has quit (Remote host closed the connection).
23:04:08 <elliott> ok this /almost/ works
23:05:10 <Vorpal> <elliott> Vorpal: You did pirate the ROMs, right? <-- I found some copies on an old floppy. It said "this is totally legal" on the sticker
23:05:23 <elliott> Vorpal: I meant as opposed to not having the ROMs. :p
23:05:33 <Vorpal> elliott, I do have the roms yes
23:05:40 <Vorpal> elliott, it failed in a different way before that
23:05:46 <Vorpal> elliott, which was to exit
23:05:55 <Vorpal> elliott, and the music /plays/, it just stutters
23:06:00 <Vorpal> every now and then
23:06:00 <elliott> Vorpal: Note: The MT-32 has a superior sound to the emulator, even when it works properly.
23:06:06 <elliott> And, also, uses less CPU.
23:06:12 <elliott> Did I mention ScummVM can interface with a real MT-32?
23:06:17 <Vorpal> elliott, I don't have one
23:06:18 <elliott> What I'm saying is: BUY IT FROM ME GOD DAMMIT.
23:06:21 <Vorpal> so that doesn't help
23:06:26 <Vorpal> elliott, don't you want to use it
23:06:29 <Vorpal> elliott, free shipping?
23:06:46 <Vorpal> also I have nowhere to put it anyway
23:06:49 <elliott> Vorpal: Free shipping if the price is at least \epsilon more than the cost of shipping :p
23:06:50 <elliott> *:P
23:06:52 <elliott> Also, it's not big.
23:07:05 <Vorpal> elliott, approx dimensions?
23:07:22 <elliott> Umm... it's wider and about as long as this 13" laptop, I'd say. Pretty flat; just enough for a display and some knobs.
23:07:25 <elliott> Light, too.
23:07:32 <elliott> Vorpal: http://upload.wikimedia.org/wikipedia/en/0/05/MT_32.jpg
23:07:34 <elliott> Not big.
23:07:46 <Vorpal> elliott, that doesn't really show the scale
23:07:48 <elliott> Vorpal: http://www.chrisguitars.com/rol-mt32.jpg
23:07:51 <elliott> Vorpal: Comparison with big power brick.
23:08:04 <Vorpal> ah
23:08:19 <Vorpal> elliott, won't fit. Only place I have is on top of the tower case really
23:08:37 <elliott> Just put it on the floor :P
23:09:05 <elliott> Only $1e9999999999999999999999.
23:09:19 <Vorpal> elliott, besides, the actual copy of the game for scummvm I have includes some *.ogg renderings that sound like the mt-32
23:09:22 <Vorpal> and that don't stutter
23:09:29 <Vorpal> and that scummvm seems able to use
23:09:33 <elliott> Vorpal: Is that Monkey Island 1?
23:09:40 <elliott> Vorpal: If so, those are rips of the CD tracks.
23:09:42 <Vorpal> elliott, 1 and 2. Have only tried 1 so far
23:09:51 <Vorpal> both have *.ogg
23:09:54 <elliott> Vorpal: Right, 2 you don't need MT-32 for, just tell it to use Adlib.
23:10:01 <elliott> Vorpal: Oh, and BTW, you want to turn on aspect ratio correction in the ScummVM settings.
23:10:02 <Vorpal> elliott, but adlib sounds horrible
23:10:04 <Vorpal> I heard that
23:10:08 <elliott> Vorpal: Not for MI2. It's designed for Adlib.
23:10:11 <elliott> (More or less.)
23:10:11 <Vorpal> elliott, oh
23:10:24 <Vorpal> elliott, aspect ratio correction?
23:10:33 <Vorpal> elliott, won't it just resize the window?
23:10:34 <elliott> Vorpal: Yes. global scummvm settings -> graphics.
23:10:47 <elliott> Vorpal: tl;dr the games are 320x200. This was shown on a 4:3 monitor.
23:10:49 <elliott> i.e. non-square pixels.
23:11:00 <elliott> i.e. everything will look subtly wrong if you don't turn on aspect ratio correction.
23:11:01 <Vorpal> yeargh
23:11:11 <elliott> Vorpal: You probably want 2x or 3x scaling (with no filters) too.
23:11:16 <elliott> Higher is better, since there's less error in the aspect ratio correction then.
23:11:18 <Vorpal> elliott, btw xdpyinfo or whatever thinks I have (very slightly) non-square pixels
23:11:31 <Vorpal> the dpi differs in like 1-2. I think it is wrong
23:11:36 <elliott> Vorpal: I think that's just because you have non-square pixel distance, possibly.
23:11:40 <elliott> With the actual pixels still being square.
23:12:01 <Vorpal> elliott, well measuring and dividing by pixels gives it as square
23:12:10 <elliott> Fair enough then.
23:12:12 <Vorpal> but the difference is too small to actually be certain about
23:12:26 <elliott> Now to figure out how my self-interpreter is broken.
23:12:41 <elliott> Ahhhhhh-ha!
23:12:43 <elliott> It isn't.
23:12:48 <elliott> I was just foolishly using an INVALID PROGRAM.
23:13:16 <Vorpal> elliott, I mean, I /could/ use the steel calipers I have. But the risk of scratching screen is high
23:13:27 <Vorpal> elliott, also I don't think it would be enough for the width
23:14:42 <Vorpal> elliott, it could be some integer rounding error for the dpi I guess
23:15:03 <elliott> http://i.imgur.com/WhkPl.gif <-- Oh man, time to get some wood, I'm all out. La la la... la la la... WHAT AAAAAAAA
23:15:12 <elliott> (on SMP, apparently)
23:15:21 <elliott> Better with TNT though :P
23:15:40 <fizzie> Two of these three monitors report non-equal X/Y DPI numbers too.
23:15:42 <fizzie> $ xdpyinfo | grep -i resolution:
23:15:42 <fizzie> resolution: 93x95 dots per inch
23:15:42 <fizzie> resolution: 99x98 dots per inch
23:15:42 <fizzie> resolution: 96x96 dots per inch
23:15:58 <Vorpal> elliott, wait, water on the wood?
23:16:04 <Vorpal> elliott, oh right, you did that?
23:16:25 <elliott> Vorpal: No.
23:16:29 <Vorpal> elliott, we have 2 or 3 "water trees" on our server
23:16:34 <elliott> We do?
23:16:54 <Vorpal> elliott, two near 4000,4000, I made one (4x4x2 pool) and PH made one (1x1x1 pool)
23:17:00 <Vorpal> elliott, and a third one elsewhere iirc
23:17:06 <Vorpal> somewhere northeast of spawn
23:17:07 <elliott> Oh, so a tree in water. Right.
23:17:38 <elliott> Feck, it do not wurk.
23:17:39 <Vorpal> elliott, that happens too
23:17:47 <elliott> This a bads!
23:18:03 <elliott> $ mzscheme -f diuerse.scm
23:18:03 <elliott> (begin (define x (quote x)) x)
23:18:03 <elliott> (begin x)
23:18:03 <elliott> x
23:18:03 <elliott> car: expects argument of type <pair>; given ()
23:18:04 <Vorpal> elliott, or rather, way east from spawn there is a waterfall with multiple partially submerged trees
23:18:06 <elliott> WHAT HAVE I EVER DONE TO YOU
23:18:18 <Vorpal> elliott, this was natural
23:19:13 <elliott> If I fix this bug, though, I think I've beaten cpressey.
23:19:27 <elliott> Although I'll want to implement a compiler to C, so we can have a wonderful Diuerse interpreter in C.
23:19:36 <elliott> Well. With a hardcoded program.
23:20:08 <Vorpal> elliott, beaten him at what?
23:21:18 <fizzie> Beaten him with what?
23:21:51 <fizzie> "A ray of sunlight is made up many atoms." -- Plan 9 from Outer Space.
23:22:20 <fizzie> Supposedly with "of" in there, that might be a typo of whoever made this script transcription thing.
23:23:14 <Vorpal> fizzie, well, not of sunlight, But what about a ray of CME? (Coronal Mass Ejection iirc)
23:23:41 <elliott> Vorpal: Beaten him at "a Scheme subset with a short self-interpreter".
23:24:05 <elliott> My language is slightly bigger than his (but smaller in some other aspects), but the self-interpreter is shorter and simpler.
23:24:08 <Vorpal> elliott, oh, but didn't he hate the begin keyword?
23:25:04 <fizzie> Vorpal: "Take a can of your gasoline. Say this can of gasoline is the sun. Now you spread a thin a line of it to a ball, representing the Earth. Now, the gasoline represents the sunlight, the sun particles. Here we saturate the ball with the gasoline, the sunlight. Then we put a flame to the ball. The flame will speedily travel around the Earth, back along the line of gasoline to can, or the sun itself. It will explode this source, and spread to every place tha
23:25:04 <fizzie> t gasoline, or sunlight, touches. Explode the sunlight here, Gentlemen, and you explode the universe."
23:25:20 <elliott> Vorpal: I could do without begin, it'd just be slightly uglier. I don't know that he hates it, anyway, and Scheme has no keywords; "begin" is a perfectly valid symbol, for instance. But, yes, the BEGIN special form.
23:25:39 <elliott> fizzie: That is my new favourite science.
23:25:57 <Vorpal> fizzie, ... this is so.... absurd
23:26:01 <fizzie> It's about a solaronite bomb.
23:26:44 -!- oerjan has quit (Quit: BOOOM).
23:26:58 <elliott> Explode The Sunlight Here, Gentlemen, And You Explode The Universe sounds like a post-post-indie post-post-post-pre-'pata-post-hipster album title.
23:27:02 <elliott> Okay, so not really.
23:27:05 <elliott> BUT IT WOULD, IN A BETTER UNIVERSE.
23:27:10 <fizzie> Oh no, oerjan must've stumbled across some solaronite. Bye bye, universe.
23:27:19 <elliott> RIP universe a long time ago -- recently.
23:27:40 <Vorpal> elliott, I wonder what would happen in MC if the admin was to /give 64 blocks of 0 to a player
23:27:44 <Vorpal> (that is air iirc)
23:27:50 <elliott> Vorpal: I don't think you can hold air.
23:28:15 <fizzie> Also the "space soldier" bad guy is called Eros.
23:28:45 <Vorpal> fizzie, .... ....
23:28:57 <fizzie> I wonder if pre-beta you could have placed air (with the item-placement hack) as opposed to mining.
23:30:07 <fizzie> At least with /give you can (or could) give someone water/lava source blocks, and they could hold those just fine. (They were rendered as thin, flat squares of water/lava when held in the hand.)
23:30:56 <fizzie> And protocol-wise the "no block" code tends to be -1, not 0. So it could do something.
23:31:47 <elliott> fizzie: Minepedia is very adamant that you can't do any funny stuff with air, at least: http://www.minecraftwiki.net/wiki/Air
23:32:40 <fizzie> "There's no item with id 0"
23:32:48 <fizzie> Apparently not, then.
23:33:25 <fizzie> Water is nice and animated in the inventory too.
23:33:55 <fizzie> Both spring and block forms have the same tooltip ("Water").
23:34:02 <elliott> $ mzscheme -f diuerse.scm
23:34:02 <elliott> (quote x)
23:34:06 <elliott> No, no, no ...
23:34:12 <elliott> fizzie: Portals even have a nice inventory icon. :p
23:34:19 <elliott> As in, actual portal block. It just dies after a few seconds if you place it.
23:35:08 <fizzie> The non-source water block is pretty silly. You can place a block of still, non-flowing water anywhere, apparently.
23:35:29 <elliott> Awesome.
23:36:42 -!- calamari has joined.
23:36:42 <elliott> It works!
23:36:45 -!- sshc has joined.
23:36:58 <calamari> hi
23:37:54 <elliott> hi
23:41:28 <elliott> hm I think that my procedures may not support recursion as-is
23:41:54 <elliott> indeed
23:43:04 <calamari> working on something fun?
23:44:01 <elliott> calamari: semi-fun
23:44:14 <elliott> calamari: a self-interpreter for a tiny subset of Scheme
23:44:28 <elliott> the goal is to beat cpressey by having a shorter self-interpreter with a language about as small :)
23:44:53 <calamari> cool
23:45:30 <elliott> currently I have a 66-line interpreter which does everything except recursion.
23:45:46 <elliott> and which I think can make self-interpret by changing (define (f ...) ...) to (define f (lambda (...) ...)).
23:45:49 <elliott> well, and adding recursion :P
23:46:15 <Deewiant> That's kind of like having a C compiler that doesn't support loops
23:46:43 <elliott> Deewiant: Technically, all I have to do is fix DEFINE.
23:47:04 <elliott> Deewiant: I could have it so that in (define name value), name is in value's scope. Except not really, because that'd be infinite recursion etc.
23:47:15 <elliott> But the way I do it, define is basically a let around the rest of the program.
23:47:23 <elliott> So I'm not quite sure what to do.
23:48:24 <Deewiant> For define, all you need to do is implement macros and convert (define (f ...) ...) to (define f (lambda (...) ...)) using them
23:48:35 <elliott> Deewiant: That's not the issue.
23:48:44 <Deewiant> I know, I was being snarky.
23:48:52 <elliott> Right.
23:49:05 <elliott> The real issue is that "(define x (lambda () (x)) ..." doesn't work, because it just adds x to the environment in which it evaluates "...".
23:49:10 <elliott> *x)))
23:49:16 <elliott> And it's not exactly clear how to fix this.
23:49:28 <elliott> I could add list mutation, but, uh, that's a lot of additional complexity and ugliness and horrible and ugh.
23:49:38 <elliott> Wait!
23:49:40 <elliott> Actually I can just fix define.
23:49:43 <elliott> Wait, no, I can't.
23:49:46 <elliott> Can I? Maybe.
23:49:55 <elliott> Maybe if I add another parameter to EVIL to denote ... no, no.
23:50:00 <elliott> But if ... no.
23:50:32 <elliott> Deewiant: Fix my program, yo!
23:50:43 <Deewiant> s/bugs//
23:50:46 <Deewiant> Done
23:50:46 <Vorpal> elliott, how goes the C game language thing?
23:50:58 <Vorpal> well, not quite c
23:51:12 <elliott> Vorpal: Waiting for Phantom_Hoover to return from his mandatory six-day relative-visiting-hell. (Note: Hell may be exaggerating. Slightly!) :p
23:51:35 <Vorpal> elliott, actually I had guests from US this evening. Very nice.
23:51:58 <elliott> Yes, but this is IRELAND.
23:52:21 <Vorpal> they are not old relatives however, but rather young friends of my mom.
23:52:41 <elliott> "mom"? You've turned into one of them already!
23:52:56 <Vorpal> elliott, oh right "mum" in UK
23:53:30 -!- TLUL_ has changed nick to TLUL.
23:54:25 <elliott> Grr, this is actually really subtle.
23:54:32 <elliott> Clearly I did not read sufficient amounts of SICP today.
23:56:05 * elliott notes:
23:56:06 <elliott> T3X.ORG : Goodbye
23:56:06 <elliott> Due to a revision of German law (the controversial JMStV), I will take this site off-line on 2010-12-31 until some degree of legal certainty has been established.
23:56:08 <elliott> and for posterity:
23:56:10 <elliott> Looking for Scheme 9 from Empty Space?
23:56:11 <elliott> It is now hosted at these locations:
23:56:11 <elliott> http://tx97.net/s9fes/
23:56:13 <elliott> http://telegraphics.com.au/s9fes/
23:56:15 <elliott> http://www.sacrideo.us/s9fes/
23:56:17 <elliott> http://xivilization.net/~marek/s9fes/
23:56:19 <elliott> Thanks for giving the code a new home!
←2010-12-24 2010-12-25 2010-12-26→ ↑2010 ↑all