00:00:02 pikhq: I no longer have a volume control, but I do have my freedom. 00:01:01 gregor: needs crouching badly :( 00:02:38 also: bottom of page=platform and top of page=no boundary 00:02:51 pikhq: WHY IS ALSA CONFIG IN /USR/SHARE/ALSA 00:02:58 WHAT THE FUCK HAPPENED TO /ETC 00:04:03 pikhq: I think that Linux is just too shitty to run bsnes well on this hardware. 00:04:11 elliott: :/ 00:04:15 pikhq: Sorry dude, I think I might install ZSNES until I get Kitten working :< 00:04:18 I AM BETRAYING THE CLAN. 00:04:36 zsnes is p cool tho 00:04:45 quintopia: I thought that then I learned how it worked. 00:04:45 if only it could do sound right 00:04:49 ... 00:04:51 HAHAHAHA 00:04:57 Sound being my only problem with bsnes. 00:05:18 i thought it rather ironic myself 00:05:30 It's like rain on your wedding day. 00:05:34 elliott: SNES9x at *least*. 00:05:45 pikhq: Okay fine. 00:05:47 elliott: ZSNES is absolutely *horrid*. 00:05:56 pikhq: I just, Linux hates me. 00:05:59 elliott: SNES9x at least has niceties like "correct sound emulation". 00:06:08 Yes, ZSNES's sound emulation is broken beyond belief. 00:06:15 So bsnes is out of the question until my OS is something more than a piece of wool balanced on top of a stick. 00:06:17 And the stick is cracking. 00:06:22 That's basically my impression of Linux. 00:06:24 Works just fine for me. :P 00:06:26 pikhq: But it doesn't have a water effect for the mouse cursor! 00:06:27 (NetBSD is something akin to a very heavy stone.) 00:07:27 I seem to recall some DOS NES emulator (Nesticle?) having this really "kewl" dripping-blood theme. 00:07:43 Yes. 00:07:49 Cursor almost looked like a testicle even. 00:07:51 Almost. 00:07:53 It was actually a hand. 00:07:57 But if you looked away... 00:08:03 pikhq: Snes9x goes at perfect speed. :/ 00:08:07 Yup. 00:08:14 Snes9x is quite a bit faster. 00:08:18 Well, hands are very related to testicles anyway. 00:08:18 does MAME work in linucks? 00:08:18 And no audio glitches. 00:08:20 Blearh. 00:08:26 quintopia: Yes. 00:08:33 sweet 00:08:36 pikhq: Would tweaking with latency help? 00:08:38 It works in all the Linucks of the world. 00:08:40 Higher is better, right? :P 00:08:44 elliott: XD 00:08:53 pikhq: IT'S A SETTING TWEAKING IT DOES THINGS HELP HELP 00:09:14 hmm 00:09:19 * quintopia tries the 9x thing 00:09:28 * olsner tries the sleeping thing 00:12:09 elliott: For the sake of curiosity, I think I'm going to encode a lossless video with x264 and see how much space it takes. 00:12:52 x264 is lossless? 00:12:58 Also, "lossless video"? 00:13:08 You mean, encode with x264 losslessly? 00:13:28 Sgeo: x264 can do lossless encoding, yes. 00:15:35 Just set the quantiser to 0 and voila. 00:16:13 why don't these horrible things just get deleted 00:16:34 cpressey: ? 00:16:51 ZSNES. PulseAudio. why can't i rm them 00:17:16 why do others insist of keeping their reference counts above zero 00:17:34 cpressey: drunk again i see :P 00:18:51 ... damn it. 00:18:54 pikhq: Things I don't like: Scaling filters 00:18:57 I think I just totally broke my good keyboard. 00:19:18 Gregor: What model? 00:19:34 * elliott utter keyboard nerd who has no good keyboards due to apathy and money! 00:19:44 so, snes9x does sound emulation excellently, but sucks balls at mouse emulation. maybe if i ran it in fullscreen mode it'd be less annoying. 00:20:22 i only have one keyboard, and it gets replaced regularly due to spills mostly 00:20:28 mouse emulation?? 00:20:29 why 00:20:35 elliott: SNES mouse. 00:20:40 ...why 00:21:00 Doom! 00:21:23 Mario Paint! 00:21:29 pikhq: I hereby approve of Snes9x for those of us who use terrible OSes. 00:21:48 elliott: It's imperfect but a *hell* of a lot better than ZSNES. 00:21:51 pikhq: Even if Chrono Trigger's opening sequence makes no sense at all. 00:21:52 elliott: not YET 00:22:00 Things are happening! I HAVE NO IDEA WHAT THEY ARE 00:22:05 elliott: Which, BTW, has been unmaintained for 6 years. 00:22:10 cpressey: Oh, incomprehensible spoilery opening sequences? 00:22:15 TV shows do that a lot. 00:22:19 Oh, sorry, 3 years. 00:22:21 "First episode, here's what's gonna happen in the next ten!" 00:22:25 elliott: i am setting the MOOD 00:22:33 And it's written in x86 assembly. 00:22:34 elliott: You just asked me what keyboard I use, then BEGGED me not to tell you. 00:22:36 Honestly. 00:22:47 Gregor: By... mentioning I'm a keyboard nerd? 00:22:50 pikhq: OMG 3D 00:22:54 I totally saw those bikes on the 3D PLANE 00:22:56 elliott: Yes. 00:23:00 elliott: Mode 7! 00:23:03 Gregor: Do tell. 00:23:04 Gregor: Oh just tell me, I'm using a terrible one now. 00:23:19 I'll just make a snide remark about it not classing as "good" and leave it at that. 00:23:29 pikhq: Seriously, does this opening sequence make any fucking sense at all? 00:23:33 dear people who use snes9x: is there a tag to make it automatically lower my resolution and fullscreen instead of windowing? 00:23:34 No. No it doesn't. 00:23:37 ...IT GOES BACK TO THE START 00:23:39 Start is... Enter 00:23:44 So that's what I had to do. 00:23:51 ...Battle mode? 00:23:56 pikhq: WHAT. 00:24:07 elliott: I've not played Chrono Trigger. :) 00:25:12 fizzie: What do I push. 00:26:58 Wait is more turn-based than active. 00:27:42 fizzie: Oh so those are the two modes for the actual game? 00:27:43 Oh, it's ATB? 00:27:43 And you press start in some name-entering thing and in the start screens, but after that it's mostly just abxy. 00:27:55 fizzie: Which is the "proper" one? 00:27:58 elliott: Active. 00:28:01 Please don't say "there's no proper one". 00:28:06 pikhq: Okay. 00:28:13 Active is for "real men", yes. 00:28:24 Enter a name woop woop 00:28:28 Am I super-boring if I just use "Crono" 00:28:55 (I sleeps now.) 00:29:06 elliott: Nah. 00:29:42 Gotta say, those graphics are amazing. 00:29:51 I... this is totally not my mental image of "SNES". 00:29:59 What *is*? 00:30:10 pikhq: I, whatever the first Super Mario Bros. game for it was numbered. 00:30:13 3? Whatever. 00:30:19 elliott: Super Mario Bros. 3 is NES. 00:30:27 The one after that? :P 00:30:30 Super Mario World 00:30:33 That, then. 00:30:37 #World 00:30:39 Dear God the colours are nice. 00:30:47 Dear GOD the colours are nice. 00:30:55 elliott: http://www.mariowiki.com/images/d/d0/SMWMvsB.png 00:30:56 No mom, YOUR bell makes such beautiful music. Wait, what? 00:31:11 pikhq: Yes. That. 00:31:21 elliott: And that was a launch title. 00:31:26 I wonder if everyone is the same height in this game. 00:31:36 I wonder why the bottom few pixels are unused. 00:32:03 elliott: The SNES had a lot more power in it. And even more if you stuck a RISC chip onto the cartridge! 00:32:41 Uhh, can I actually do anything before I go downstairs? To anyone who knows. 00:32:45 I haven't mastered these controls yet. 00:34:54 -!- FireFly has quit (Quit: swatted to death). 00:36:39 snes9x (this version) is unplayable with mouse and keyboard afaik 00:36:43 *afaict 00:37:36 Y'know, I'm quite sad that sticking CD-quality audio in games wasn't practical until recently. I would love to play FF6 without synthesised music. 00:38:19 She called 00:38:22 Had no idea who I was 00:38:29 And I could barely understand her accent 00:38:50 * Sgeo cries 00:39:04 -!- BeholdMyGlory has quit (Remote host closed the connection). 00:39:30 * elliott wonders if Sgeo actually cried there 00:39:45 * elliott closes curtains 00:39:46 FUCK YOU LIGHT 00:39:49 (in Chrono Trigger) 00:40:31 No I didn't 00:40:32 omg i want the cat 00:40:42 But I think I hung up on her on accident 00:40:52 pikhq: wow, she doesn't give you your allowance unless you talk to her twice 00:40:53 bitch :| 00:41:18 pikhq: I wonder who the hell the Japanese think they're portraying with their families. 00:41:37 Always single mother, always says the exact same thing to the kid 00:41:42 Always an almost-identical house 00:42:18 pikhq: I... a guy is walking on the spot. Emulation issue? 00:42:20 You think? 00:43:55 Keyboard has nearly restored itself. 00:44:16 Gregor: WHICH 00:44:49 Vorpal: ping 00:47:07 liquid metal self-restoring TERMINATEROR keyboard 00:47:36 i use a BRAIN keyboard. 00:47:41 it plugs into my BRAIN socket 00:48:12 -!- cpressey has quit (Quit: actually, i'm NOT thinking what you are thinking). 00:51:12 elliott: japanese fathers are never home, you know. 00:51:12 Cool, gold is worthless in this game. 00:51:22 oerjan: And Japanese houses are TINY 00:51:30 work during day, drink during night 00:51:32 You can get arbitrary gold here just by pressing a button at the right time ten times. 00:51:33 That's one gold piece. 00:51:35 Also trivial. 00:51:41 (and the drinking is also work) 00:52:00 oerjan: psht that's clearly the Norwegians 00:53:31 yep. baseless, that is. 00:54:24 pikhq: I saved to a save slot on the cart. Am I a bad person? 00:54:37 elliott: No. 00:55:28 elliott: Awesome. I appear to be getting 15,000 kbit/s for this LOSSLESS x264 encode. 00:55:44 (note: 420p24) 00:55:48 s/420/480/ 00:55:50 FREUD! 00:56:16 elliott: For comparison, a DVD is generally 9,000 kbit/s or so. 00:56:52 Approximately doubling the storage space would let us have lossless 480p. 00:57:08 pikhq: 1080p 00:57:25 ... I'll try that later. Quite a bit later. 00:57:38 Wow, this random girl I just met sure is dedicated to following a few pixels behind ,e/ 00:57:39 *me. 00:57:41 pikhq: with FLAC soundtrack 00:57:43 (I'd need to download the uncompressed video) 00:57:46 elliott: Well, of course. 00:58:07 "They're still setting up! Why not go and figure out what you're supposed to do before we let you in." 00:58:22 Come, random girl who follows a few pixels behind me. 00:58:28 Let us find our quest as you get in my way annoyingly. 01:01:15 elliott: This is actually a low enough bitrate that it would be ENTIRELY FEASIBLE to stream lossless 720x480 video over ATSC if the spec allowed. 01:01:23 pikhq: I approve. 01:01:32 Wait, ATSC? 01:01:36 Some zany American thing? 01:01:48 The American digital terrestrial television standard. 01:02:21 The only real reason we use it instead of DVB is that ATSC broadcasts started before DVB was invented. 01:02:43 The only time America has ever been ahead of the curve. 01:02:49 Ever. 01:03:03 ATSC's at least not a terrible standard. 01:03:26 "This Kingdom's been through a lot, like the war against Magus, 400 years ago." 01:03:36 Oh, you fought a guy 400 years ago. Such hardship you peaceful motherfuckers face. 01:03:40 Not even Switzerland has that kind of record. 01:05:25 pikhq: Oh, so the one I selected was basically "pseudo-realtime battles" 01:05:56 Yup. That's the Active Time Battle system for you. 01:06:05 pikhq: Am I... meant to like that? 01:06:28 It'll either grow on you or make you very glad that it fell out of style. 01:06:29 :P 01:06:37 I NEED TIME TO THINK ;_; 01:06:45 My Elephant's Dream encode is 1.2G 01:06:51 And 100% lossless. 01:06:53 Victory! 01:06:56 My elephant's dream encode is 1.2G. 01:07:42 "Jurassic (music symbol) rhythm (music symbol)" 01:07:45 Them some prehistoric lyrics. 01:08:00 This is, in fact, ENTIRELY PRACTICAL. 01:10:10 Now, then. 1080p, you say? 01:10:16 46GB uncompressed. 01:10:34 And it doesn't exist in such a format. 01:10:45 pikhq: It doesn't? :( 01:10:57 Lemme grab the PNGs, then. 01:10:59 pikhq: Anyway a dual-layer Blu-Ray can store 50 gigs. 01:11:04 y4m is just much more *convenient*. 01:11:05 pikhq: Are dual-layer Blu-Rays okay? 01:11:13 elliott: Yuh. 01:11:23 pikhq: I remember with DVDs it kinda sucked. 01:11:26 Lag between scenes. 01:11:47 elliott: That's called a "shitty encode". 01:12:25 pikhq: Or do I not mean multiple layers... 01:12:34 pikhq: I may mean double-sided. 01:14:39 elliott: There's two ways to do a dual-layer DVD. 01:14:55 The second layer can start from the inside or the outside of the disc. 01:15:08 With it starting at the outside of the disc, the transition time is essentially nil. 01:16:01 Anyways, hooray, 21G of PNGs. 01:16:16 I anticipate the x264 will be similar in size. 01:16:22 BARELY PRACTICAL! 01:22:18 goodnight 01:22:18 bye 01:22:20 -!- elliott has quit (Quit: Connection reset by peer review). 01:28:54 -!- augur has quit (Remote host closed the connection). 01:29:07 -!- augur has joined. 01:32:19 -!- catseye has joined. 01:37:48 -!- augur has quit (Ping timeout: 276 seconds). 01:43:45 someone ought to computerize Spawn of Fashan. on a really objectionable platform, like, MS-DOS w/CGA graphics. ALSO: it should crash a lot, for full effect. 01:52:48 OMGOMGOMG!!! 01:52:54 MythBusters is finally testing "When the shit hits the fan" 01:52:55 BEST 01:52:57 DAY 01:52:57 EVER 01:54:08 so, um 01:54:09 *boggles* *facepalm* *boggles* 01:54:12 culture 01:54:17 ....yeah. 01:56:21 Well, that's fun. It seems that mplayer seems to fail at a crazy, crazy Matroska file with lossless video and audio. 01:57:11 it's matryoskas all the way down 01:57:31 *matryoshkas 01:58:04 It seems to have trouble with the *FLAC* in there. 01:58:35 It handles being told to play a seperate AVI and WAV file just fine. 02:21:35 -!- SimonRC has quit (Ping timeout: 265 seconds). 02:24:17 http://cgi.ebay.co.uk/Commodore-64-Quintic-Warrior-Quicksilva-/110597345971 02:24:25 "you will never discover MY roots, challenger!" 02:31:38 -!- calamari has joined. 02:31:45 hi 02:31:52 hi calamari 02:31:58 Gregor: The requested URL /webplat/webplat.js was not found on this server. 02:32:30 how's it going catseye 02:32:41 calamari: back-up-y. 02:33:21 oh I see, he changed it to loader.py 02:33:24 calamari: Name changed, load method changed, recreate your bookmarklet from http://codu.org/websplat/ 02:33:31 loader.js rather 02:33:37 Errr? 02:33:40 loader.js should work ... 02:33:45 (As a symlink) 02:33:45 i was wondering what .py was involved for 02:34:16 Ohoho 02:34:21 * Gregor doesn't read properly :P 02:34:35 Anyway, yeah, remake your bookmarklet :P 02:34:40 cool 02:34:45 it's a mini gregor! 02:34:58 You haven't web*lat'ted in a while :P 02:35:07 weird, I can't seem to double-jump through "Some pages I recommend platforming on: " 02:35:13 You can thank quintopia for the images. 02:36:12 oh, yes I can, guess I just needed more height 02:40:47 falling off then clicking it again to load another copy produces very interesting results lol 02:41:06 pretty cool, I like it! 02:41:33 http://i41.tinypic.com/a4ymva.png This is an actual frame from a Bluray. HOW DO YOU ENCODE THAT POORLY. 02:41:58 MPEG2 at those bitrates can produce better. 02:44:24 Gregor: if there are 2 lines of text, I can't jump through.. is that by design?> 02:44:50 calamari: You can only jump through one (z-ordered stack of) element(s). 02:45:25 oh so it's the one above it that is stopping me 02:45:26 calamari: try increasing the font size! or disabling css! 02:45:34 i have no idea what will happen if you try those, btw. 02:46:34 without css it doesn't work 02:52:17 interesting.. after you fall all the way off, you still get one final jump 02:53:58 "tar: TAP: file changed as we read it" 02:54:06 is that the "royal we"? 02:54:25 or is tar actually a gang of little gnomes? 03:02:51 -!- calamari has set topic: Mandelbrot stabbed to death; on autopsy, authorities found smaller version of Mandelbrot inside | TOO SOON | http://is.gd/g4ullllID. 03:07:39 ppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppoooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooopppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp 03:07:39 pppppppppppppppppppppppppppppppppppppppppppppppppppppppoooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooookkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu 03:07:39 uuuuuuuuuu;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm 03:07:45 mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmlllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll///////////////////////''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''/ 03:07:58 -!- ChanServ has set channel mode: +o oerjan. 03:08:15 * oerjan taps his fingers 03:09:14 -!- oerjan has set channel mode: -o oerjan. 03:10:15 bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbboops sorry .; b 03:10:15 xx x 03:10:21 sorry 03:10:35 bbl :) 03:10:37 -!- calamari has quit (Quit: Leaving). 03:10:37 SOMEONE IS NOT TAKING A HINT HERE 03:12:23 w 03:13:32 the um,link in the topic goes nowhere now 03:13:55 also, was he drawing a mandelbrot set? i can only hope, but i don' think so 03:14:33 hm 03:15:18 -!- catseye has set topic: Mandelbrot stabbed to death; on autopsy, authorities found smaller version of Mandelbrot inside | TOO SOON | http://tunes.org/~nef/logs/esoteric/. 03:16:04 it doesn't look much like a mandelbrot set 03:16:42 maybe in 300 columns and really large.....mmmstill no. 03:17:20 only the first two lines even resemble each other, but a part of the mandelbrot set still wouldn't look like that 03:17:46 conclusion: he was drawing it *poorly*. 03:17:55 ...that may be. 03:19:08 -!- catseye has set topic: Mandelbrot stabbed to death; on autopsy, authorities find smaller version of Mandelbrot inside | TOO SOON | http://tunes.org/~nef/logs/esoteric/. 03:19:15 newspaper headline grammar 03:20:29 -!- oerjan has set topic: Mandelbrot 'stabbed to death'; on autopsy, authorities 'find smaller version of Mandelbrot inside' | TOO SOON | http://tunes.org/~nef/logs/esoteric/. 03:20:38 no _this_ is newspaper headline grammar 03:21:26 yes 03:32:26 -!- SimonRC has joined. 03:33:28 dive into cohomology 04:05:53 -!- SimonRC has quit (Ping timeout: 255 seconds). 04:07:33 -!- wareya has quit (Read error: Connection reset by peer). 04:08:28 -!- wareya has joined. 04:26:55 fiber bundles for dummies 04:50:43 Things I wish I had known about: document.elementFromPoint 04:51:04 Why does it have a name inconsistent with all the other getElement* functions? Because Microsoft made it. 04:56:19 -!- catseye has quit (Ping timeout: 252 seconds). 05:00:55 -!- SimonRC has joined. 05:17:51 -!- catseye has joined. 05:18:11 hi fungot 05:18:12 catseye: so in pi calculus, join calculus, etc. they can also sell copies of software ( the gnu project has written things like gcc, mozilla, and probably breaks horribly in lots of hitchcock films, the fnord 05:18:30 ... 05:18:33 i love you. 05:22:08 pikhq: http://www.fantasyscotus.net/ 05:22:42 -!- oerjan has quit (Quit: Good night). 05:37:54 coppro: ... 05:40:43 -!- augur has joined. 05:49:03 -!- antivigilante has joined. 05:56:11 -!- wareya has quit (Read error: Connection reset by peer). 05:57:00 -!- wareya has joined. 06:13:30 netcat, on cygwin, doesn't do the -e thing it seems -- at least not for MY bot. 06:13:33 :( 06:13:36 good night 06:14:16 -!- catseye has quit (Read error: Connection reset by peer). 06:18:32 -!- augur has quit (Remote host closed the connection). 06:18:43 -!- augur has joined. 06:18:52 -!- Sgeo has quit (Ping timeout: 245 seconds). 06:20:03 Re the topic, a Finnish parody newspaper web thing announced Mandelbrot's death with something like (badly paraphrased) "The decades-long project to measure the boundary of Benoît Mandelbrot had to be discontinued on Thursday. 06:21:11 The project, started in 1967, took significantly longer than expected, as the perimeter of Mandelbrot grew longer every time the researchers switched to more accurate methods of measurement. 06:21:58 According to the latest measurement, made with an electron microscope, Mandelbrot was longer than the British coastline." 06:22:02 And so on, and so forth. 06:25:42 Well, since I went that far, I might do the last sentence too: "The experts described the challenged faced by the project also with exact mathematical expressions, but of that the reporters, having received a social science education, did understand nary a thing." 06:25:53 s/ged/ges/ and meh. 06:31:26 nice 06:31:36 linky? 06:36:19 Uh, well, http://lehti.samizdat.info/2010/10/mandelbrotin-aariviivan-pituuden-mittaus-keskeytettiin/ to that particular piece. 06:46:22 -!- Mathnerd314 has quit (Remote host closed the connection). 07:23:29 -!- Zuu has quit (Read error: Operation timed out). 07:23:50 -!- Zuu has joined. 07:41:36 Is there any fractal that does not have boundary length go to infinity as the scale appoaches zero? 07:52:10 -!- augur has quit (Remote host closed the connection). 07:52:20 -!- augur has joined. 07:59:55 -!- FireFly has joined. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:16:50 -!- jix has joined. 08:16:51 -!- jix has quit (Client Quit). 08:17:03 -!- jix has joined. 08:27:15 -!- augur has quit (Remote host closed the connection). 08:27:24 -!- augur has joined. 08:32:26 -!- atrapado has joined. 08:38:00 -!- ais523 has joined. 08:41:49 -!- jcp has quit (Ping timeout: 250 seconds). 08:46:35 -!- pikhq_ has joined. 08:49:59 -!- fizzie` has joined. 08:51:01 -!- jcp has joined. 08:54:06 -!- EgoBot has quit (*.net *.split). 08:54:06 -!- pikhq has quit (*.net *.split). 08:54:06 -!- bsmntbombdood has quit (*.net *.split). 08:54:06 -!- fizzie has quit (*.net *.split). 08:54:06 -!- cal153 has quit (*.net *.split). 08:54:06 -!- myndzi has quit (*.net *.split). 08:56:21 -!- cal153 has joined. 08:56:21 -!- myndzi has joined. 08:56:45 -!- EgoBot has joined. 08:56:45 -!- bsmntbombdood has joined. 09:11:53 -!- jcp has quit (Ping timeout: 255 seconds). 09:16:32 -!- jcp has joined. 09:44:18 -!- ais523 has quit (Read error: Operation timed out). 09:52:20 -!- ais523 has joined. 10:03:58 -!- Zuu has quit (Read error: Connection reset by peer). 10:07:28 Vorpal: ping <-- pong 10:09:03 elliott: for log reading purposes: I realised that trying to prove anything useful about much of libc is useless, for example, take strlen(), you can't even say anything about it really because there might not be any \0 byte. 10:10:10 it is easier to prove useful properties for stuff like memcpy and so on that takes a max length parameter 10:26:22 -!- Ilari_antrcomp has quit (Read error: Operation timed out). 10:26:22 -!- Ilari has quit (Read error: Operation timed out). 10:27:27 -!- Ilari has joined. 10:27:47 -!- Ilari_antrcomp has joined. 10:49:49 -!- antivigilante has quit (Ping timeout: 252 seconds). 10:53:35 Vorpal: you can prove that strlen returns the length correctly if given a null-terminated string 10:54:55 meanwhile, on the C module I'm trying to TA: you can't use scanf("%s") because it leads to a buffer overflow if the user gives more input than the size of the buffer can't you just use sizeof() 10:55:02 I do not have high hopes for this module... 11:01:00 -!- fizzie` has changed nick to fizzie. 11:01:04 That is very impressive. 11:01:30 ais523, yes indeed. Hm I wonder how to express that to frama-c, I'm not completely sure it is possible 11:01:51 ais523, TA? 11:02:29 -!- ais523 has quit (Read error: Operation timed out). 11:05:48 -!- ais523 has joined. 11:06:27 "Teaching assistant", I'd guess. 11:08:48 -!- GreaseMonkey has quit (Quit: Welcome honored guest. I got the key you want! would you like onderves. of Yourself). 11:42:48 -!- ais523 has quit (Remote host closed the connection). 11:43:28 fizzie, got any good suggestion for safe prefix for identifiers in library code? _ is for the standard and __ is for libc. ___ looks a bit silly... And so does prefixing parameter names in the headers with the library prefix 11:44:08 (and I can't just skip them, since there will be formal verification comments referring to those in the headers) 11:46:47 Not triple-underscores, at least. 11:46:56 fizzie, hm, why not? 11:47:06 fizzie, it looks so silly that no one else uses it 11:47:15 Yes, but it still looks so silly. 11:47:23 good point 11:48:37 I'm not sure what would be sensible, though. Any non-odious things I can think of are more or less conflict-prone. 11:49:58 fizzie, error prone is something I want to avoid everywhere in this case. 11:52:13 As a quick poll: the SDL/OpenGL/glib headers at least include parameter names but don't seem to use any sort of prefix. 11:53:10 fizzie, true, but I don't want defines from something like ncurses messing up, which is infamous for doing just that. 11:54:35 Add libpng to that list of no-prefix names, though the libpng parameter names are so silly ("void png_destroy_write_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr);") that maybe they don't need any. 11:54:44 It's like reverse hungarian notation. 11:54:56 fizzie, :D 11:55:21 fizzie, amusing define from deep in glibc: 11:55:25 /* This is not a typedef so `const __ptr_t' does the right thing. */ 11:55:25 #define __ptr_t void * 11:55:37 the comment makes sense certainly but... why define that at all 11:56:36 It's a nice tradeoff: you get "const __ptr_t" working, but on the other hand multi-declarations like "__ptr_t a, b" then won't. 11:57:33 fizzie, well yeah, but why do they define that thing at all 11:58:09 I don't really know. Maybe they don't like the * character? 12:00:01 Maybe it's for when they try to port glibc to a compiler that doesn't support void pointers? 12:01:56 fizzie, har har :P 12:02:23 Hey, void's a new feature, you can't expect everyone to support it yet. 12:03:43 Seminar session time real soon now, I wonder what it's going to be about. 12:04:03 "Multi-task learning with kernel and nonparametric methods" 12:43:16 -!- nooga has joined. 12:47:28 -!- BeholdMyGlory has joined. 12:52:13 -!- oerjan has joined. 13:00:44 -!- ais523 has joined. 13:06:22 -!- antivigilante has joined. 13:24:45 -!- antivigilante has quit (Read error: Connection timed out). 13:29:10 -!- atrapado has quit (Read error: Connection reset by peer). 13:30:07 -!- BeholdMyGlory has quit (Ping timeout: 245 seconds). 13:30:26 -!- nooga has quit (Ping timeout: 264 seconds). 13:31:38 -!- BeholdMyGlory has joined. 13:32:46 -!- oerjan has quit (Quit: leaving). 13:34:37 -!- Mathnerd314 has joined. 13:44:17 -!- elliott has joined. 13:44:31 -!- elliott has set topic: Mandelbrot 'stabbed to death'; on autopsy, authorities 'find smaller version of Mandelbrot inside' | TOO SOON | http://tunes.org/~nef/logs/esoteric/?C=M;O=D. 13:45:31 17:14:39 elliott: There's two ways to do a dual-layer DVD. 13:45:31 17:14:55 The second layer can start from the inside or the outside of the disc. 13:45:31 17:15:08 With it starting at the outside of the disc, the transition time is essentially nil. 13:45:43 pikhq_: would it work to transition to another layer, say, in the middle of fast action? 13:47:47 18:41:33 http://i41.tinypic.com/a4ymva.png This is an actual frame from a Bluray. HOW DO YOU ENCODE THAT POORLY. 13:47:47 18:41:58 MPEG2 at those bitrates can produce better. 13:47:47 Wow. 13:47:50 I'd prefer a DVD. 13:49:03 `addquote 21:18:12 catseye: so in pi calculus, join calculus, etc. they can also sell copies of software ( the gnu project has written things like gcc, mozilla, and probably breaks horribly in lots of hitchcock films, the fnord 13:49:04 elliott: also more fat burns better. there's no finer-grained editing than page-level atm. you may want to just do 13:50:19 02:09:03 elliott: for log reading purposes: I realised that trying to prove anything useful about much of libc is useless, for example, take strlen(), you can't even say anything about it really because there might not be any \0 byte. 13:50:20 well 13:50:58 (exists n. byteAtLocEq (p+n) 0) -> strlen n = n 13:50:59 or something 13:51:02 erm 13:51:05 (exists n. byteAtLocEq (p+n) 0) -> strlen p = n 13:51:10 hm 13:51:14 or whatever 13:51:27 Vorpal: of course you could package that up as nullTerminated p 13:51:32 nullTerminated s -> strlen s = n 13:51:32 hah 13:51:38 indeed 13:51:54 of course you'd probably not be able to have them as actual coq functions without some zaniness 13:51:55 but eh 13:52:03 or at all really 13:52:05 meh 13:52:09 elliott, anyway, if you really care that much about stuff being correct that you formally prove it you probably don't want to use C strings anyway 13:52:17 03:43:28 fizzie, got any good suggestion for safe prefix for identifiers in library code? _ is for the standard and __ is for libc. ___ looks a bit silly... And so does prefixing parameter names in the headers with the library prefix 13:52:20 surely ___ is reserved for libc 13:52:23 since it's __X for X=_ 13:52:37 elliott, no __ is for libc, and _ is for the standard 13:52:45 Vorpal: if you care about it being formally proven and you want to write it in C you're either controlling a nuclear reactor or stupid. 13:52:53 elliott, indeed 13:52:56 Vorpal: i mean 13:52:57 ___ 13:52:59 has prefix __ 13:53:01 which is the libc prefix 13:53:02 no? 13:53:19 elliott, but then __ is for C standard since _ is, and __ has prefix _ 13:53:38 I think it says "followed by alphanumeric letter" or such, I need to check that 13:53:47 Vorpal: ok. 13:54:19 things I wish I could do: tell apt "I want ubuntu-desktop, but not this one dependency (*cough* pulseaudio) and you just have to accept that." 13:55:22 elliott, I wish the same. 13:55:39 there are a few more ones that just pulseaudio I wish to avoid 13:55:42 Vorpal: instead, I'm just going to try to install ubuntu-desktop every now and then, note the new packages, and install them :P 13:55:48 hmm, like what? 13:55:58 Cough, Ubuntu One, cough, but that isn't depended on by ubuntu-desktop. 13:56:19 elliott, hm forgot, I think it used to depend on openoffice or something in older releases. 13:56:26 oh. i think it still does 13:56:26 seems that is just recommends nowdays 13:56:31 ah 13:56:38 fuck openoffice with a rusty chainsaw 13:56:56 it's perhaps the only office suite worse than Office 13:57:07 elliott, anyway, there was some other thing too, evolution. 13:57:22 what's wrong with having evolution installed, even if you don't use it? 13:57:28 it's not like ubuntu isn't full of such stuff anyway 13:57:32 hah true 13:57:48 and you can hide it from the memories and try and forget about the disk space it's using up 13:57:51 *menus 13:58:10 elliott, you could just disable pulseaudio and not uninstall it? 13:58:13 Vorpal: interesting, I have libpulse but no pulseaudio. 13:58:14 Wonder how that works. 13:58:19 Vorpal: you *can* do that, but it's a bitch 13:58:28 elliott, not just an init script? 13:58:31 also I think to hide the indicator applet icon for the pulseaudio audio stuff it's uh 13:58:37 you have to uninstall it maybe? or just fuck with the files 13:58:43 and that would probably break ubuntu-desktop if the former 13:58:50 Vorpal: no, because it also redirects ALSA to PulseAudio 13:59:04 ouch 13:59:15 Vorpal: well it's to be expected, it's reasonable to do if you actually want it 13:59:18 i just don't 13:59:18 ever 13:59:59 elliott, exactly do you expect anything relating to sound more advanced than the build in pc speaker to be sane under linux ;P 14:00:07 wait, that one isn't sane either 14:00:13 *built-in 14:00:16 it shows up in /dev/input 14:00:17 that sentence took a while to parse... 14:00:24 elliott, simple typi 14:00:25 argh 14:00:28 typo* 14:00:29 typi, the plural of typo 14:00:38 typoing typo is kind of awkward 14:00:44 Vorpal: yeah, the only way i've managed to get a sane interface to linux audio that also works well is OSSv4 14:01:00 decent API (from what I gather), proper /dev files that work reasonably and mix properly 14:01:05 low latency 14:01:15 and it actually makes Flash audio sync up 14:01:28 Vorpal: but OSS on Ubuntu is just ... lol 14:01:37 elliott, there are a few alsa drivers that actually have low latency and so on. But they are the exception rather than the norm 14:01:42 the repository version is old. the non-repository version is a pain to set up. 14:01:43 and stuff. 14:01:48 luckily my sb live card is one of the exceptions 14:02:26 elliott, as for flash audio sync, how comes video syncs fine with audio when watched in, say, mplayer or vlc? 14:02:42 Vorpal: well if you're me it doesn't, not when using pulseaudio or alsa :) 14:02:44 Vorpal: but uh 14:02:47 Vorpal: (1) flash sucks 14:02:52 right 14:02:53 Vorpal: (2) flash uses OSS all the time, I think 14:02:59 (so obviously it's faster without a translation layer) 14:03:00 and 14:03:01 hm that would explain parts of it 14:03:10 Vorpal: (3) nspluginwrapper is insane and god knows what it does to desync everything up. 14:03:16 elliott, it even syncs up on my thinkpad, with intel hd audio 14:03:25 elliott, right 14:03:30 and i hate nspluginwrapper and HURRY THE FUCK UP ADOBE RELEASE THE 64-BIT VERSION. 14:03:37 Vorpal: i have intel hd audio but... yeah 14:03:41 alsa, not a fan. pulse, not a fan. 14:03:43 that is 14:03:45 is-not 14:03:51 alsa is not a fan of my card 14:03:52 not 14:03:54 i'm not a fan of alsa 14:04:04 elliott, well thinkpad uses alsa with pulse. Because so far I haven't had sync issues I never bothered replacing it 14:04:18 guess I'm just lucky 14:04:41 Vorpal: In this case I decided to just get rid of PulseAudio when it was too shitty to handle bsnes giving it 32khz audio :P 14:05:33 pikhq_: re: that screenshot 14:05:35 pikhq_: "The project team was able to produce a x264 video compliant Blu-ray. The first and most important stone from which to build authoring programs. 14:05:35 The content already present on the network and released under a permissive license, were compressed at a rate sufficient to be burned on a standard DVD disc." 14:05:45 pikhq_: i.e. "We're innovators! Also, pointless goal that makes quality useless!" 14:06:19 note: google translate is *great* at italian -> english 14:07:55 It's official; Empathy is so bad that I'm installing Pidgin. 14:07:59 WHYYYYYY 14:08:38 You must be some sort of a robot to so completely lack any empathy. 14:08:47 (Or maybe a replicant.) 14:09:33 Or a replicator! http://www.yehppael.com/images/uploaded/0000/0006/img-21.jpg Although that makse little sense. 14:12:56 Just to add one more data point: I don't have any AV sync issues when it comes to (regular) video playback through pulse/ALSA (except for one bluetooth headset which is pretty flaky everywhere), and haven't bothered to replace it because (a) Ubuntu and (b) I actually use Pulse's "use pavucontrol to route audio from different apps to different devices in one easy, centralized way" feature; I guess that's not something OSSv4 does? (Though perhaps if I just did pe 14:12:56 r-application symlinks to /dev files and some sort of custom kludge to manipulate those.) 14:13:07 -!- Slereah has quit (Ping timeout: 240 seconds). 14:14:25 http://www.opensound.com/wiki/index.php/Tips_And_Tricks -- that mess re "add a "vmixctl attach" command to the $OSSLIBDIR/soundon.user file" makes it sound uncomfortably close to fiddling with asoundrc's. :p 14:16:01 fizzie: It isn't actually /dev-based. 14:16:06 OSS is nothing like old OSS :P 14:16:32 Well, again, I was just going by http://www.opensound.com/wiki/index.php/Tips_And_Tricks "Changing the default sound output" -- "The typical method is to relink /dev/dsp to the desired /dev/oss/.../ device." 14:16:46 That sounds pretty old-fashioned. 14:17:11 fizzie: Files; very old-fashioned. 14:17:26 What is that if not "/dev-based", though? 14:17:47 fizzie: Well, I'm pretty sure there's an API that doesn't actually touch any /dev files. 14:17:56 Presumably it treats /dev/dsp as configuration. 14:18:34 -!- Slereah has joined. 14:19:21 fizzie: I actually don't know, but I do know that it works perfectly. 14:19:28 And doesn't feel hacky or anything. 14:21:44 Well, I'm just looking at this http://manuals.opensound.com/usersguide/ and from what I can tell, you're certainly supposed to configure "where does this sound go" in the application doing the playback, which is something I don't like, because so many apps tend to hide that thing in all kinds of uncomfortable places. (I don't mind "hacky", though.) 14:22:20 I don't do crazy shit like that, but, uh, I'd trust the wiki more than the manual. 14:22:26 Documentation is one thing I do not believe it is so strong on. 14:23:11 The whole discussion is a bit academic, since my laziness is so endemic I'm sure I couldn't be arsed to switch from "whatever Ubby does by default" unless (until?) things actually break. 14:27:47 fizzie: OSSv4 is a pain to get working on Ubuntu because Ubuntu sucks, so yeah. 14:34:35 Uhh, you can't resize Pidgin's text box thing to be only one line by default, can you? 14:35:02 Wait. no, one works. 14:40:21 "National Flags 14:40:21 Due to political reasons, they are not distributed with GNOME unfortunately." 14:40:22 lol 14:41:34 -!- Slereah has quit (Ping timeout: 240 seconds). 14:42:24 Haha... 14:45:00 Certain flags could be used to annoy certain people... 14:45:43 http://developer.pidgin.im/raw-attachment/ticket/2367/pidgin201gui.jpg "I will now point out your UI's flaws using badly-spelled red on black, great big red borders around things I already have arrows pointing to, and Fixedsys as one of my main fonts." 14:45:51 "Also, I will spell any word greater than six letters long incorrectly." 14:46:02 (Okay, so there's only two typos, but come on.) 14:46:28 -!- Slereah has joined. 14:52:42 -!- pikhq_ has changed nick to pikhq. 14:52:56 pikhq_: re: that screenshot 14:52:56 pikhq_: "The project team was able to produce a x264 video compliant Blu-ray. The first and most important stone from which to build authoring programs. 14:52:56 The content already present on the network and released under a permissive license, were compressed at a rate sufficient to be burned on a standard DVD disc." 14:52:56 pikhq_: i.e. "We're innovators! Also, pointless goal that makes quality useless!" 14:53:12 also 14:53:12 17:14:39 elliott: There's two ways to do a dual-layer DVD. 14:53:12 17:14:55 The second layer can start from the inside or the outside of the disc. 14:53:12 17:15:08 With it starting at the outside of the disc, the transition time is essentially nil. 14:53:12 pikhq_: would it work to transition to another layer, say, in the middle of fast action? 14:53:14 :P 14:55:35 Yeah, it's small enough that the buffer takes up the slack. 14:55:57 Also, some DVD players can do the switch to the other layer start effectively instantly; hooray, buffering. 15:01:41 Wow, Chrono Trigger's graphisc are awesome. 15:01:42 *graphics 15:04:07 Oh look at that, I've gone back just in time for the war. 15:11:31 -!- MigoMipo has joined. 15:16:47 -!- cpressey has joined. 15:17:22 -!- BeholdMyGlory has quit (Remote host closed the connection). 15:17:53 cpressey: NETBSDNETBSD 15:18:04 elliott: Tonight. It shall happen. 15:18:14 cpressey: How many hours is "tonight"? 15:18:18 I'm a little irritated by Maverick, tbh. 15:18:20 I cannot miss the achtung. 15:18:23 achtung 15:18:28 achtungachtungachtung 15:18:56 eh. yeah, you usually log off less than an hour before i get back online at home. 15:19:36 like 8 hours from now, bahah. 15:19:56 cpressey: I TOTALLY SUGGEST THE WEEKEND 15:20:05 (like it will be SO ENTERTAINING from your end to see me disappear, then reappear 20 minutes later saying "I can't get the network up"_ 15:21:21 but yes, weekend is probably a better plan 15:21:55 * cpressey can't wait to see the NetBSD One Online Music Store plugin for madplay though 15:22:03 (see? it doesn't even make any sense, right?) 15:26:16 cpressey: You are quite thoroughly insane. 15:26:22 (like it will be SO ENTERTAINING from your end to see me disappear, then reappear 20 minutes later saying "I can't get the network up"_ 15:26:24 yes it will 15:26:29 also it takes longer than that :P 15:28:18 DISCOVERY: document.elementFromPoint 15:28:25 FURTHER DISCOVERY: document.elementFromPoint is effectively unusable. 15:30:46 -!- Harpyon has joined. 15:31:50 Gregor: You must make moving elements work. Wanna know why? 15:32:15 Without elementFromPoint being tolerable, I /can't/. Period. But why? 15:32:17 Gregor: Marquees as moving platforms. 15:32:26 Gregor: Jumping at the right time to avoid falling from blinking ones. 15:32:29 I rest my fucking case. 15:32:35 The movement of marquees isn't detectable from within JS anyway. 15:32:45 All you know is the box. 15:33:03 Gregor: JS is unsuitable for this? 15:33:07 Y'DON'T SAY :P 15:33:24 Doing it directly in WebKit would be hell on Earth. 15:33:36 Doable, but I'd sooner castrate myself with a rusty soup spoon. 15:35:36 Gregor: Use WebKit, write the bulk of it in JS, but for special bits have WebKit-using C++ that exposes marquee positions and other useful stuff to JS? 15:35:57 That's exactly what I'm referring to. 15:36:08 Hacking out the part of WebKit that draws that, detecting whereTF it's putting things, and barfing that back to JS = hell on Earth. 15:36:19 Gregor: Yeah because this is an easy project :P 15:36:29 Up until now it has been. 15:36:38 Gregor: Yay another dormant Gregor project 15:37:33 Yes, "I'm not adding a pointless feature that would take me a week of work" = "I'm abandoning this project" 15:37:33 Totally the same. 15:39:30 Oh, and which would significantly reduce the userbase. From "whoever wants to" (tens) to "whoever downloads this and compiles a hacked WebKit into a browser" (zero) 15:40:10 Gregor: Can you not write your own elementFromPoint? Sounds "easy". 15:40:51 cpressey: Uhh, how am I writing elementFromPoint not in terms of elementFromPoint? Iterating over every element and checking if it's there? 15:43:39 I suppose I could have an "enhanced" hacked-WebKit branch and a "normal" unhacked-WebKit branch ... 15:43:59 And if I implemented a non-shitty getElementsByRect into the enhanced one ... 15:45:36 Oh, and which would significantly reduce the userbase. From "whoever wants to" (tens) to "whoever downloads this and compiles a hacked WebKit into a browser" (zero) 15:45:39 Into a browser?? 15:45:44 I meant making it a standalone application. 15:45:55 Just embed a webkit widget in a window. Done. 15:46:07 This also gets rid of having to make it a bookmarklet. 15:46:35 Gregor: So how is elementFromPoint useless? 15:47:02 It's too damned slow because: 15:47:25 1) It only returns the highest-zIndex element at the point, so I need to CSS-modify that and recurse to get every element. 15:47:41 2) It's only by-point, not by-box, but I'm doing box-box collision detection here. 15:48:50 Gregor: Yeah, I'm totally gonna say "Write ultra-trivial application that just starts up a window with WebKit on some page (put a URL box there or something, the "start level") that automatically injects the JS into any non-starting page. (That would be, what, 200 lines at the most? Surely.) 15:49:01 Then you can add little enhancements for the JS code when you want. 15:49:21 That is not the hard part. At all. That's the already-done part. 15:49:39 Gregor: Oh, I wasn't thinking about the "realtime" version, if things are changing. Just, survey every element, record its BB (in a quadtree?), test points in that structure. 15:49:43 Gregor: You already have an application? 15:50:02 cpressey: That's more-or-less exactly what I'm already doing. 15:50:05 Gregor: I'm just saying, do that and ditch the bookmarklet model. Then you can add stuff when you want without making shit change. 15:50:17 cpressey: I was hoping elementFromPoint would help me make it dynamic :P 15:50:20 Gregor: Oh, and you could load the JS from the web, so that updates would be automatic unless you added some WebKit stuff... 15:50:25 I see. 15:50:25 (Which could easily just tell you "yo upgrade") 15:50:26 elliott: Well OBVIOUSLY you ditch the bookmarklet if you're doing it that way. 15:50:44 Gregor: Sheesh, I know it gives equivalent functionality, I'm just saying: Do this now and the amount of work later is decreased :P 15:50:51 Also it'll be less silly. 15:51:22 Out of curiosity, have you ever compiled WebKit? Have you ever hacked its build system at all? 15:51:34 Gregor: ...you wouldn't need to? 15:51:39 Just link to it. 15:51:51 I would need to give its library a name unambiguous from "real" WebKit. 15:51:51 I'm assuming that its API is good enough to add a proper elementFromPoint replacement. 15:51:57 Gregor: No you wouldn't. 15:52:01 You could *link to actual WebKit*. 15:52:09 All you're doing is binding some C++ stuff to JS. 15:52:10 OHOHOHOH 15:52:20 Suddenly I have 100% more understanding of what you're suggesting X-D 15:52:23 Gregor: Is that laughter or "oh, right, do the *obvious* thing" :P 15:52:26 The latter! Yay 15:52:50 OK, so that idea has just gone from "wtf, hacking WebKit = I want to kill myself" 15:52:59 To "ahh, simply providing the necessary interfaces" 15:53:01 To "elliott is a design GOD" 15:53:14 Gregor: Note: I just filed for a patent on "Linking to WebKit to make a game" 15:53:41 DAMN! 15:53:57 Gregor: Darn, it just came back. Apparently there's already a patent for that 15:54:01 SO YOU SHOULD BE FINE 15:54:06 Only I'm litigious enough. 15:54:20 "There's a patent for that. From Apple." 15:57:04 pikhq: Does libavcodec do replaygain? 15:57:35 Gregor: I wonder if you could actually get the "real" bounding box of a marquee... 15:57:55 I guess internally, it actually has "100%" width, and just scrolls like that. 15:58:03 But you could get the bounding box of all elements inside. 15:58:10 Gregor: Of course it'd be easier with JS that actually moves stuff. 15:58:15 Now to find a page that actually does that! 15:58:21 -!- nooga has joined. 16:00:09 elliott: Right now I don't even support JS that /adds/ stuff, e.g. Facebook's dynamic content. 16:00:14 Gregor: Right. 16:00:24 Gregor: I bet WebKit's API lets you hook into newly-created elemeeeents :P 16:00:47 I wouldn't need to if I just exposed a fast getElementsByRect. 16:00:54 Or that, right. 16:01:01 haha, you know Yvette's Bridal Formal? 16:01:08 the worst designed website in the world? 16:09:18 I am aware of the work of Yvette, yes 16:09:31 -!- sshc has quit (Ping timeout: 240 seconds). 16:13:29 cpressey: http://yvettesbridalformal.com/Fastings1.html -> "please click here to see what you came to see" -> http://yvettesbridalformal.com/Psychological_Thriller_Horror_Success.html 16:13:36 [[There used to be alot more conspiracies, death letters, FEMA cover ups, but it seems like theyve been adding new material every so often. I'll edit this post later to make sure I link the yahoo answers community to other odd pages...feel free to use anything you find as a source. A bunch of us have been working on this site for weeks now trying to come up with something, but we keep coming to dead ends.]] 16:13:39 [[Leads: 16:13:40 Apparently she has a son who's either Schizophrenic or mentally ill 16:13:40 Yvette no longer works there 16:13:40 When company is called, there seems to be a snooty or suspicious person who picks up. Any questions pertaining to the website ends in a quick "Have a happy 2009"]] 16:13:47 cpressey: tl;dr more than just a terrible web designer 16:15:06 Gregor: Does installing plash on a system do anything weird to it? should I use a chroot? 16:15:14 anyone played with Open MPI ? 16:15:27 nooga: iirc cfunge has some openmpi code. lawl 16:15:34 wait no 16:15:39 openmp 16:17:44 holy shit "/x/ discovered Yvette's a while ago, and they are 100% serious alien conspiracy theorists." 16:17:56 elliott: Ah. I thought you were going to tell us of the super fun of WebPlat on that page 16:18:02 cpressey: tried that earlier 16:18:06 it was super fun, but this is even funner 16:18:17 dear god http://yvettesbridalformal.com/FineArtPanamaCity19.html 16:18:20 the jumping things 16:18:35 [[A couple of guys from /x/ actually drove out there and talked with them and took some pictures in front of the place. The Yvette's site blogged about it later, but I'll be damned if I can find anything on that site. 16:18:36 Apparently they told them that they were from an online magazine looking to investigate the link between the US government and aliens, and the Yvette's people talked with them (warily) for a good length of time.]] 16:18:41 http://www.yvettesbridalformal.com/interview1.html 16:19:03 "...........I didn't even know there was a 16:19:03 magazine named UFO Magazine..........when I got 16:19:03 to my PC I searched and sure enough, there is a 16:19:03 magazine/Ezine of the title UFO 16:19:03 Magazine..........." 16:19:16 "Frank seemed like an honest and decent sort of 16:19:16 fellow......just an ordinary, regular guy.........like 16:19:16 maybe someone who would ask you if he could 16:19:16 borrow your lawnmower and actually bring it 16:19:16 back to you after he mowed his lawn.............." 16:19:34 this is just 16:19:36 cpressey: this is the greatest thing ever 16:20:03 i'm trying to setup a small ompi cluster in a laboratory on my university 16:20:08 like, uh 16:20:19 "additionally ~~ 16:20:19 20 machines 16:20:19 the most famous painting in the world..........The 16:20:19 Last Supper by Leonardo da Vinci.......the two 16:20:19 central figures leaning away from each other to 16:20:19 form a perfect V and on either side of the 16:20:20 refectory hall, 4 tapestries for a total of 8....... 16:20:22 yes, V8 encoded right there in broad daylight for 16:20:24 all the world to see......... 16:20:26 and why is everyone out of printer ink for their 16:20:28 PC printers !?" 16:20:32 cpressey: "The Last Supper has a hidden encoded message, now what IS it with printer ink?" 16:20:51 it's way too early for this level of intense wisdom 16:21:14 http://www.yvettesbridalformal.com/Lyra_V8_Sean_Terrence_Best.html 16:21:41 http://www.yvettesbridalformal.com/sitebuilder/images/test196_028-450x600.jpg i think this is sean 16:22:01 http://www.yvettesbridalformal.com/Alphabet_Angels_Alpha.html yeah another pic of him here 16:25:58 http://yvettesbridalformal.com/aboutYvettesPortraitPainter1.html 16:27:29 "I have been to yvette's, it is a lot like this: very colorful friendly place with lots of things moving. you can get a tuxedo and also get your portrait painted. they are very positive and believe in the power of dreams." 16:28:51 -!- Zuu has joined. 16:28:51 -!- Zuu has quit (Changing host). 16:28:51 -!- Zuu has joined. 16:29:06 wtf is this encoding: ./share/libc/__fc_string_axiomatic.h: Non-ISO extended-ASCII C program text 16:29:16 it isn't UTF-8 either 16:29:23 or anything I tried in my editor 16:29:41 less manages to display the chars in it 16:29:48 but both emacs and kate are totally lost 16:30:16 I have UTF-8 locale set so not sure what less is trying 16:30:33 does it have any weidr characters in it with less? 16:30:34 *weidr 16:30:36 *weird 16:30:52 Gregor: wait, why does plash development seem dormant? 16:30:58 elliott, well yes: 𝔹 ℤ but they are intended 16:30:59 news posts from 2007 and 2008, edgy eft as last mentioned ubuntu version 16:31:08 wait hm 16:31:10 Vorpal: copy it all from less and pastie it 16:31:14 /* CEA (Commissariat l'nergie atomique et aux nergies */ 16:31:23 elliott, I think it might be mixed encodings 16:31:25 lets see... 16:31:27 Vorpal: copy it all from less and pastie it 16:31:29 :P 16:31:54 -!- Slereah has quit (Ping timeout: 240 seconds). 16:32:22 elliott, anyway yeah it was mixed encoding, the copyright header was in some ISO-ish thingy while the rest was UTF-8 16:32:25 problem solved 16:32:43 Vorpal: please tell me the blackboard bold isn't just in comments or strings 16:32:44 somehow 16:33:27 elliott, it is in C comments, but they are used for the proof. 16:33:39 elliott, as in, they are used in the syntax of proof specification 16:34:00 elliott, the file declares the concept of a 0-terminated string, looks quite scary 16:34:08 (Moo) 16:34:34 Gregor: ooM 16:34:41 Vorpal: i wanna seeeee and am lazy pastie 16:34:43 or sprunge 16:34:43 :P 16:35:18 cpressey: oh INTERESTING, apparently those recipe pages 16:35:23 used to be conspiracy ones 16:35:32 well 16:35:37 no web archive though 16:35:44 and google cache has updated 16:35:45 elliott, http://sprunge.us/LFBe 16:35:46 elliott, this is raw 16:35:52 Vorpal: my browser handles it! 16:35:57 Vorpal: wow at prefixing every line with @ 16:36:07 Vorpal: ...wow at that syntax :D 16:36:09 elliott, yeah mine screws up the comment at the top 16:36:15 "(char *s1, char *s2, ℤ n)" 16:36:16 there are no words 16:36:20 oh, mine does too 16:36:20 -!- Slereah has joined. 16:36:38 elliott, is that good or bad? ;P 16:36:43 that, uhh 16:36:50 i'm not sure what was wrong with "nat" 16:36:51 er 16:36:51 "int" 16:37:17 Vorpal: have you ever used plash :| 16:37:18 elliott, indeed. Except it might conflict with a C variable, typedef or function called that 16:37:22 elliott, no I haven't 16:37:26 elliott, indeed. Except it might conflict with a C variable, typedef or function called that 16:37:28 only as much as "char" might 16:37:31 char, int 16:37:33 same thing 16:37:39 elliott, well, they *are* the C types there 16:37:43 Vorpal: no 16:37:47 "ℤ n" 16:37:50 elliott, unlike ℤ 16:37:54 ℤ is integers 16:37:55 elliott, yes ℤ is the logic type 16:37:59 Vorpal: i... see 16:38:01 char * is the C type 16:38:07 i, uh 16:38:10 that distinction makes little sense 16:38:13 elliott, it is proving stuff about variables of that C type 16:38:25 elliott, well ℤ would be bignum I think 16:38:31 unlike, say, an int or such 16:38:40 for which it would model the overflow behaviour 16:39:16 ...but the last argument to memcp could overflow 16:39:24 elliott, anyway I saw somewhere in the docs that C typedefs took priority in case of a name conflict, but you could always reach the logic types using the unicode equivalents. 16:40:09 Vorpal: did they never think of just using an invalid char like $? 16:40:12 $Z 16:40:22 elliott, I don't know :P 16:41:20 elliott, anyway, I don't understand all in there, but memcmp isn't used to describe a C function called that. It is used to represent the concept of comparing memory in order to be able to describe the concept of a null terminated string in the end.... I think. 16:41:47 i will have to take your word for it 16:42:26 -!- Slereah has quit (Ping timeout: 265 seconds). 16:43:22 Some program called ReportCrash is taking 100% CPU. 16:43:26 ... thanks, Apple. 16:43:46 it's reporting a crash in ReportCrash 16:43:46 Gregor, wait, are you using a mac!? 16:44:04 Vorpal: For my research I have stuff on EVERY MAJOR OS! Yes, that's right. Victory. 16:44:06 elliott, or hm you could learn how it works :P 16:44:25 Gregor, Solaris? HP-UX? 16:44:31 wait, HP-UX isn't major 16:44:31 MAJOR 16:44:36 Vorpal: Solaris isn't major either. 16:44:36 [negates what you said] 16:44:41 Gregor, oh? 16:44:47 Gregor, what are the major ones then? 16:44:53 Vorpal: Windows, Linux, OS X 16:44:55 OMG THAT WAS HARD 16:44:56 Windows, Mac OS X, Linux. 16:45:09 hm, sad it is only those that count as major these days 16:45:16 "These days"/ 16:45:17 *? 16:45:19 It's mostly sad that Windows is on the list :P 16:45:19 You mean "all days". 16:45:25 Ever since OS X came out that's been it. 16:45:29 elliott: Solaris was big once. It was contender! 16:45:31 elliott, well, go back to the 80s and it wasn't like that 16:45:34 Before that, what, Mac OS was on the list too. And Windows was a lot bigger. 16:45:38 (Rather than X) 16:45:40 (Mac OS X that is) 16:45:48 Before that, WINDOWS WINDOWS WINDOWS, Mac OS. 16:45:53 Before that, WINDOWS WINDOWS WINDOWS, OS/2. 16:46:00 Before that, ... nothing. 16:46:10 elliott, DOS 16:46:14 Well, yeah. 16:46:18 If you can call that an OS. 16:46:24 elliott, and before that, SunOS and so on 16:46:28 Rather than a few API functions, a filesystem and a shell :P 16:46:29 and several other ones 16:46:29 I've got an iMac on my desk now 16:46:31 "shell" 16:46:32 elliott: It calls itself an OS! 16:46:41 Gregor: So does my feces. 16:46:47 (Note: Lies) 16:46:48 elliott: That's weird. 16:46:49 DOS is so not an OS 16:46:55 DNOS 16:47:00 elliott: Also, I'm surprised you didn't use the Britlish spelling. 16:47:21 Gregor, "faces"? ;P 16:47:41 I think it should be parenthesized like (Disk Operating) System. It operates your disk. 16:47:44 Gregor: Thanks, I just viewed the Wikipedia article "feces" to see if "faeces" was a hypercorrection (my usual justification for not using the British spelling). 16:47:46 Gregor: I regret this. 16:47:57 Gregor: I really don't care though :P 16:48:19 -!- Slereah has joined. 16:48:21 Gregor: I do hate "foetus", though, as it's just something that was invented because lots of Latin words have "oe" in them, right? Must be "foetus" really. 16:48:28 I raeally doen't caere. 16:48:36 Actually it's "fetus" in Latin and the British medical literature even calls it that. 16:48:47 Gregor: Nosrsly Plash, is it actually developed? 16:48:55 elliott: Yes. 16:49:02 elliott: I had a chat with the main Plash dev a while ago.\ 16:49:13 Gregor: Okay, so their website is just ancient. 16:49:15 there's no goddamn irc channel for open-mpi 16:49:18 elliott: It's still up to date in the repo, and it works on the latest glibc, there just hasn't been a /release/. 16:49:29 Gregor: And their nightly repo appears to only go up to Ubuntu Edgy Eft. 16:49:35 Circa 2006. 16:49:45 elliott: That's just their auto-builds. 16:49:52 Gregor: I really don't want to build it. 16:50:00 Neither do I :P 16:50:06 Hence why I just use lenny. 16:50:27 Gregor: What, so I should make a Debian chroot to install it in? 16:50:48 Not ideally, but yeah >_> 16:50:59 Gregor: Did *you* make a Debian chroot to install it in? :P 16:51:10 Am I installing debootstrap right now? (Yes) 16:51:19 elliott: Yes. 16:51:25 elliott: Hackiki and HackBot are in chroots. 16:51:28 Gregor: Are *you* insane? (Yes.) 16:51:36 debootstrap is SO EASY 16:51:41 E: debootstrap can only run as root 16:51:44 That.... why... no. 16:51:52 how am i supposed to ask few simple questions 16:51:56 nooga: what 16:51:57 elliott: It needs to make dev files. 16:52:01 Gregor: Your mom needs to-- 16:52:10 elliott: Then it needs to chroot for the final step of installation. 16:52:11 Gregor: OK, can I chown -R after that though? :P 16:52:21 nooga: open-mpi is way too SERIOUS BUSINESS for IRC :) 16:52:36 elliott: I ... suppose? But you can't chroot into it as non-root anyway ... 16:52:56 Gregor: But having other people own stuff in my ~ feels disturbingly kinky. Wait, what? I mean disturbingly weird. 16:53:23 Dear debootstrap: "debootstrap lenny --help" does NOT mean "create a directory called --help". 16:53:48 X-D 16:53:52 Hokay, totally debootstrapping lenny. 16:54:41 Gregor: How long is this gonna take? :P 16:54:47 Well, I've debootstrapped on an ARM machine. But still. 16:54:51 elliott: Not as long as you think, longer than you'd like. 16:55:08 (That's what HE said) 16:55:25 ...THAT MAKES NO SENSE 16:55:29 (SORT OF) 16:55:53 I: Retrieving ed 16:55:53 I: Validating ed 16:55:56 Yayyyyy 16:56:00 *Yaaaaay 16:56:21 Thank you for correcting the spelling of "Yaaaaay" ... 16:57:58 elliott: I was expecting an "I: ed is ded" after that. 16:58:08 Gregor: It... it's important! 16:58:15 I also use * to reword sentences to be nicer after-the-fact. 16:58:17 I am a bit strange. 16:59:15 Rhythmbox's successor should be called Melodycube. 16:59:19 Then Harmonyprism. 16:59:24 Or 16:59:29 Bluescube 16:59:33 Rhythmbox and Bluescube 17:00:36 -!- tombom has joined. 17:01:29 I'll install the Timbreoctahedron. 17:01:38 -!- augur has quit (Ping timeout: 252 seconds). 17:02:12 fizzie: Metretesseract 17:03:17 Gregor: So, uh, don't run Plash as root, yeah? :P 17:04:21 Spectromandelbulb, for the mathematically inclined listener. 17:04:41 Perhaps even the mathematically reclined. 17:06:59 Gregor: So I want the nightly build things, right? 17:08:51 Gregor: RITE? :P 17:08:55 elliott: I use the release *shrugs* 17:08:59 Gregor: Alright then 17:09:45 Gregor: LAWL it wants to install X11 libs. 17:09:49 Go ahead, young Debian 17:09:56 WARNING: untrusted versions of the following packages will be installed! 17:09:56 Untrusted packages could compromise your system's security. 17:09:56 You should only proceed with the installation if you are certain that 17:09:56 this is what you want to do. 17:09:56 plash 17:09:58 Gregor: Now that *is* ironic. 17:10:08 Yeah, Plash has an utterly-useless "powerbox" feature that uses X ... 17:10:16 Gregor: Hey, I'm installing it to test that out. 17:10:28 lawl 17:10:32 It's a good idea fundamentally, I'm just not sure they can pull it off with the current state of graphical applications. 17:10:34 OK, utterly useless /to me/ 17:10:36 *Especially* X11. 17:10:43 So I basically want to laugh at it. 17:11:08 Gregor: http://www.combex.com/tech/edesk.html This is how it should actually be done :P 17:11:13 (And one of the inspirations for the powerbox.) 17:12:19 Gregor: It does seem awfully silly just to handle files. 17:12:27 Because who will ever have resources on a network? 17:12:33 What silliness! 17:12:38 The problem is there technique /can't/ block network access. 17:12:43 *their 17:12:49 Since you can always use the raw syscalls. 17:12:58 Gregor: lawl, indeed 17:13:12 Gregor: wait couldn't they disable them? 17:13:15 and provide their own bsd sockets 17:13:19 of course bsd sockets is so hilariously low-level 17:13:22 as to make such a thing pointless 17:13:34 SYSTEMS CONTINUE TO SUCK BECAUSE OF HOW SYSTEMS SUCK; NEWS AT 11 17:13:34 Hence why I disable them (with firewall) and provide an HTTP proxy. 17:14:06 Gregor: Can you make debootstrap use hardlinks? 17:14:07 That'd be really nice. 17:14:20 That WOULD be really nice ... but I doubt it. 17:14:21 A bunch of chroots taking up effectively 0 additional space than if there was no chroot. 17:14:31 Gregor: Probably you can make one and then just hardlink-copy it. 17:14:33 And go from there. 17:14:46 I mean, I doubt it depends on what directory name you use. So it should work fine. 17:14:58 Gregor: Of course, creating new hardlinks on updates and the like would be the issue. 17:15:01 elliott: Except you wouldn't want it to be ALL hardlinks, since you'd want an independent /etc. 17:15:02 But yeah. 17:15:03 But *eh* you could probably automate it. 17:15:06 Gregor: So? 17:15:11 Gregor: The default /etc files can be hardlinked. 17:15:16 If you modify them, it'll no longer be hardlinked. 17:15:17 Obviously. 17:15:19 or, wait 17:15:28 Depends on how who modifies them :) 17:15:28 modifying a hardlinked file doesn't modify other links right? 17:15:33 fairly sure 17:15:36 Depends on the editor. 17:15:39 Gregor: ...lawl 17:15:45 Gregor: What editors do the stupid thing? 17:15:50 (modify the other linked files) 17:16:06 echo 'foo' >> /etc/options 17:16:07 Gregor: ooh, that probably breaks file-change notification daemons 17:16:15 since the other linked files change 17:16:19 but this isn't notified about 17:16:27 (presumably, since that requires traversing the entire FS) 17:16:50 Gregor: What about x=$(cat /etc/options); echo $x'foo' >/etc/options? 17:16:56 Please don't tell me that behaves differently. 17:16:57 I will cry. 17:17:21 I'm actually not sure off the top of my head :P. I think that truncating it might cause recreation. 17:17:41 Now I get to make $DISPLAY work from inside a chroot! 17:18:39 ...how DO You do that? 17:18:41 *you 17:19:11 Copy .Xauthority from the host, make sure you have TCP enabled, use DISPLAY=127.0.0.1:whatever instead of DISPLAY=:whatever 17:19:16 Huh, localhost.nl isn't registered. Isn't that against the lawz of DNSzzz? 17:19:37 Gregor: I have no .Xauthority :-D 17:19:46 TCP enabled in my X server, you mean? 17:19:51 Yeah 17:19:53 Can I enable that without restarting X? 17:19:56 Say yes or I will kill you. 17:20:11 I have no idea. 17:20:52 Alternatively, punt and start Xephyr with whatever option tells it to not do any permissions checks :P 17:21:11 Gregor: Or... ssh in to the chroot and use X forwarding. 17:21:12 Alternatively alternatively, install a VNC server in the chroot. 17:21:21 Alternatively alternatively alternatively, do that. 17:21:29 Gregor has invented unary 17:21:46 "How many pigs do I have? Alternatively, alternatively alternatively, alternatively alternatively alternatively..." 17:22:21 How many bigs do I have? One metric lot. 17:22:41 -retro start with classic stipple and cursor 17:22:42 dear god no 17:23:02 AWESOME 17:23:22 hey, that's useful! 17:23:22 if you do 17:23:24 x & y 17:23:27 then ^C only kills y 17:23:28 but if you do 17:23:29 (x & y) 17:23:32 then ^C kills them both 17:24:32 Gregor: Haha wow, metacity thinks compositing is a good idea in Xephyr. 17:25:11 WOW metacity has a lot of dependencies. 17:28:02 -I ignore all remaining arguments 17:30:46 powerbox.Powerbox(user_namespace = state.caller_root, 17:30:46 AttributeError: 'module' object has no attribute 'Powerbox' 17:31:42 Gregor: Running your WM with pola-run SO DOESN'T WORK 17:31:58 X-D 17:32:04 sh-3.2$ pola-run -B --prog /usr/bin/icewm-lite --env DISPLAY=127.0.0.1:1 --x11 --pet-name IceWM 17:32:04 /usr/bin/icewm-lite: error while loading shared libraries: libgcc_s.so.1: cannot open shared object file: No such file or directory 17:32:08 I... uh... 17:32:12 Well that's a weird one. 17:32:24 ls -l /usr/lib/libgcc_s.so.1 17:32:28 It's in /lib instead. 17:32:37 -fl /lib doesn't help though 17:32:40 ls -l /lib/libgcc_s.so.1 then :P 17:32:51 If that's a symlink to somewhere weird, that would be the issue. 17:33:00 Don't know why it would be though. 17:33:06 l Follow symbolic links (the "l" is for "foLLow"). If sym- 17:33:16 Ohheh :P 17:33:17 sh-3.2$ ls -l /lib/libgcc_s.so.1 17:33:17 -rw-r--r-- 1 root root 93016 Jan 2 2009 /lib/libgcc_s.so.1 17:34:11 Welp, E_WTFBBQ 17:34:21 elliott, from example output in the manual: "Warning : entering loop for the first time", it states this is completely normal. 17:34:26 XD 17:34:28 Vorpal: heh 17:34:39 Gregor: I guess I oughta trust my WM :P 17:35:02 elliott: Well, it also oughtn't to have that problem >_> 17:35:05 elliott, while the software is indeed screwy, it actually seems to do the job damn well. 17:35:36 Gregor: I LOVE THE PART WHERE GTK STUFF DOESN'T LISTEN TO $DISPLAY. 17:36:26 Gregor: Plash is such a gigantic hack :P 17:36:44 elliott: But it works. 17:36:49 "Works" 17:36:55 Well, maybe powerbox doesn't :P 17:37:12 leafpad: can't save config file - /home/u/.config/leafpad/leafpadrc 17:37:13 MWAHAHAHA 17:37:25 * Gregor is morbidly amused by his icon-character's death. 17:37:43 Gregor: OMG 17:37:47 The smallest enemies, right 17:37:52 Little tiny critters that are only a mild annoyance 17:37:55 Gregor: Make them the favicon. 17:38:02 The website fights back! 17:38:06 It doesn't want you to steal its images! 17:38:14 ... YES. YES, YES. 17:38:29 Favicon goombas! 17:38:32 YESSS 17:38:34 elliott, frama-c normalises the source code as an early step of the analysis, but I wonder why it turns *ptr++ = val; into: (where "unsigned char val, *ptr;") into: 17:38:35 unsigned char tmp = ptr; 17:38:35 ptr ++; 17:38:36 *tmp = (unsigned char) val; 17:38:41 Vorpal: Because. 17:38:41 that seems a backward way to normalise it 17:39:14 elliott, a more logical way to do it would be: 17:39:16 *ptr = val; 17:39:16 ptr ++; 17:39:18 IMO 17:39:41 Vorpal: Maybe it relies on a certain evaluation order for some reason or something. 17:39:49 -!- wareya has quit (Read error: Operation timed out). 17:39:49 elliott, perhaps 17:39:59 no that's silly 17:41:41 elliott, awesome thing I just noticed: frama-c documentation uses spivak in at least one place 17:42:23 quintopia: ? 17:42:56 hi gregor 17:43:04 Gregor: Ahh, that marshalling error is when a program doesn't have access to itself :P 17:43:18 elliott: ... laaaaaaaaaaaaaaaawl 17:44:20 elliott, what language is plash coded in? 17:44:35 Vorpal: Probably mostly C and then Python for the tools 17:44:47 actually, favicon enemies would be cool...if they dropped out of the address bar entirely, grew a face and legs, and could charge at you really fast...and could respawn at the address bar! 17:44:52 that sounds complicated... 17:44:53 elliott, considering it's security-criticallity it is imperative that it should be formally proved! Go do it! 17:44:55 ;) 17:45:36 quintopia, that sounds like "no the browser won't let you do that" 17:46:02 Vorpal: that's what i was thinking 17:46:07 but it would be damn cool 17:47:10 They could just drop from above the visible page. 17:47:21 They can't actually drop from the browser chrome, but they can drop from "nowhere" 17:47:50 Gregor: Especially since you're totally making it an application and therefore it won't have any chrome >_> 17:47:57 * elliott *subtle* 17:48:07 wait, youtube changed logo? 17:48:09 it is yellow now 17:48:12 wtf happened 17:48:28 Vorpal: Yeah it's a total rebrand. 17:48:32 elliott: At this point my plan is to make that a branch. There's really only a few things the browser can help me with, so having those as optional is still a nice option. 17:48:43 elliott, but why... 17:48:58 elliott, people knew the old logo 17:48:59 Vorpal has never seen occasion-specific logos before. 17:49:08 elliott, oh is that what it is 17:49:09 right 17:49:22 Gregor: But bookmarklets are fugly :P 17:49:35 Vorpal: try clicking the thing to the right of Tube. 17:49:43 elliott: Apps for web-based things are also fugly. Having both options is less fugly than either option. 17:49:53 Gregor: It's hardly "web-based" :P 17:49:59 It's a game that uses web pages. 17:50:10 It's not a web agme in the sense of, say, a Flash game. 17:50:21 elliott, yes but it doesn't work without flash whatever it is 17:50:26 ... 17:50:29 ... 17:50:35 Gregor: I WANT TO CRYYYY BECAUSE OF VORPAL 17:50:40 Gregor, I normally use youtube-dl... 17:50:40 elliott: Me toooooooooooooo 17:50:48 Oh, he's still on YouTube :P 17:50:56 today is celebration of mind day 17:50:58 Gregor: I have a knife. 17:51:02 And I know Vorpal's... in Sweden. 17:51:05 Gregor: Let's go. 17:51:09 elliott: I have a mullet. 17:51:10 Meet you in Stockholm in a few hours. 17:51:11 Bring weapons. 17:51:15 elliott: I'm offering you a piece of bread. 17:51:16 Also a laptop. 17:51:23 elliott: How can you possibly refuse a man with a mullet. Piece o' bread. 17:51:42 Mullet. 17:52:03 eh. #6 was better. 17:52:16 Gregor: Are you suuuuuure that powerbox is in the default thing. 17:52:19 You know, the thing with a thing. 17:52:49 * Gregor tries desperately to interpret that as words. 17:52:54 You do have to enable it with an option. 17:52:57 Gregor: The. Plash package. 17:52:59 Are you su. 17:53:00 Gregor: Yes. 17:53:05 yes it says the powerbox module has no powerbox in it. 17:53:12 which is sort of non-reassuring from a reassuring-things standpoint of view 17:53:19 No, I'm not sure it's in the package. 17:53:19 (standpoint of view) 17:53:24 Gregor: lulz 17:53:25 Maybe it's plash-powerbox or something. 17:53:27 aptitude search plash 17:53:31 yeah there's nothing else although 17:53:34 the repo *does* have two packages 17:53:37 but i dunno what the other is because i'm lazy 17:53:43 sh-3.2$ aptitude search powerbox 17:53:43 sh-3.2$ 17:53:57 What's the other package? 17:54:18 Oh, /me reads :P 17:55:42 PLASH SITE DOWN LOL 17:55:46 wth is a powerbox? 17:55:53 quintopia: a box with power inside 17:55:57 FWZOOOOOM 17:56:00 -!- nooga has quit (Ping timeout: 272 seconds). 17:56:11 i was asking gregor 17:56:14 who tends to be more helpful 17:56:17 Gregor: The other package is just glib-source. 17:56:30 quintopia: It's what happens when you stop believing. 17:56:34 elliott: Lamesauce ... maybe you need to -f in the path for powerbox ;) 17:56:36 elliott, hm C totally needs mode unicode syntax 17:56:41 Gregor: :-D 17:56:51 gregor: what does powerbox add? 17:56:57 It adds meaning to life. 17:57:11 quintopia: http://plash.beasts.org/ <-- step one, know what Plash is. I don't want to explain Plash :P 17:57:18 we've discussed plash 17:57:22 what is powerbox 17:57:25 Ah, good. 17:57:28 Powerbox is... all you've ever wanted. 17:57:29 quintopia, it's a plash thing 17:57:43 Powerbox makes it so when you try to access something that you don't have permission for, rather than just failing it pops up a box. 17:57:46 It's a thingy plash thing thing. 17:57:47 Asking whether the access is OK. 17:57:53 Gregor: Technically not really. 17:57:57 It's more a replacement file chooser. 17:58:11 for...when you have actual users running under plash? 17:58:12 Dear god, someone's ported PowerShell to other operating systems. 17:58:16 Nobody is safe. 17:58:25 elliott, but isn't it closed source? 17:58:29 Yes 17:58:31 *Yes. 17:58:37 elliott, so... just wine or what? 17:58:41 http://pash.sourceforge.net/ 17:58:43 Think Mono. 17:58:51 Sort of. 17:58:53 With less specification. 17:59:10 oh, reimplementation. Still very bad 17:59:41 THIS IS INEXPLICABLEDDDDDDDDDH 18:00:06 true 18:00:10 no need to cry about it tho 18:00:16 * quintopia hands elliott a gummy worm 18:00:19 I blame Gregor. 18:00:24 Also capitalism. 18:00:41 elliott, also they seem to like the word "rich" 18:00:51 I'm not even sure what the heck it is supposed to mean there 18:01:39 Vorpal: You're just not understanding-rich enough to understand it. 18:02:57 Gregor, :P 18:03:01 WHY THIS NO THE WORK IN THE BIT WHAT THE IM FFFF 18:03:42 "Grant access to a file or directory (source-pathname), but attach it to a different pathname in the file namespace(dest-pathname). " 18:03:43 I approve 18:03:44 *." 18:03:56 Expose ~/.foo as ~/config/foo :P 18:03:58 Erm. 18:03:59 other way around 18:06:32 elliott: I use that for /tmp 18:06:53 Gregor: hmm, like what? 18:07:00 /tmp inside maps to /tmp/FOO/ outside? 18:07:08 or /foo inside maps to /tmp outside 18:07:12 /tmp/foo that is 18:07:17 elliott: /tmp maps to /tmp/foo.$$, yeah. 18:07:32 OMG WEBSPLAT WORKS ON MY PHONE 18:07:48 OMG LEMME TRY ON MY IPOD 18:08:15 Unfortunately bookmarklets don't, so it only works directly on the page. 18:08:38 quintopia: Without a keyboard? Good luck :P 18:09:09 well i can at least see if it loads 18:09:23 -!- wareya has joined. 18:10:20 Gregor: Lame, -fl,w=foo doesn't work if foo doesn't exist 18:10:21 e.g. 18:10:24 Luckily my phone has a computer :P 18:10:26 Errr 18:10:27 Wow 18:10:30 -fl,w="$HOME/.config/leafpad" 18:10:31 Luckily my phone has a keyboard :P 18:10:34 Gregor: XD 18:10:37 alright, it may take forever to load...... 18:10:39 My phone has a computer too! 18:10:43 HELP COMPUTER 18:10:55 let's say it works if you can wait for an infinite amount of time 18:11:38 Gregor: Things that suck about Plash for desktop use: You have to trust your menu with complete privileges :P 18:11:51 "EXPLOIT IN GNOME-PLASH-MENU FOUND" 18:11:54 "AAAAAAAAH OH GOD HOLY SHIT" 18:12:14 [ten bajillion hax0rs exploit it overnight] 18:12:24 I guess you don't really. 18:12:30 It could just tell its parent, "yo, start this program". 18:12:38 And its parent would only allow it to give a name. 18:12:51 It would then look up, e.g. /etc/plash/apps/[name] and use those options. 18:12:53 But still :P 18:12:59 elliott, gnome-plash-menu? what, really? 18:13:10 Vorpal continues to demonstrate his lack of one-line scrollback 18:13:25 Gregor: Things that suck about Plash for desktop use: You have to trust your menu with complete privileges :P 18:13:38 yes indeed, and why 18:13:46 elliott, it is meant to run untrusted code 18:13:50 Vorpal is expressing incredulity that such a thing would be created 18:14:03 quintopia, yes I was doing that 18:14:08 Vorpal: Yees... 18:14:08 wasn't it obvious? 18:14:11 -!- Zuu has quit (Ping timeout: 276 seconds). 18:14:20 no, it sounded like you were wondering if gnome-plash-menu actually existed. 18:14:20 it's rather more obvious that such a thing would be created 18:14:21 elliott, it needs to get privs to chroot, then drop them 18:14:27 than that it should be incredible 18:14:39 Gregor: Please transport me to a world where people can actually understand what I say X_X 18:15:23 elliott, I don't see why it need to be very risky to have a small wrapper that takes a path and executes it securely inside a plash chroot 18:15:28 elliott: why wouldn't you want to go to a world where you can understand what other people say? that seems infinitely more useful 18:15:40 quintopia: because even when i understand what Vorpal says, it's still stupid 18:15:44 :p 18:16:10 elliott: yes, but if i were in such a world, i could understand what you say when you're being particularly unclear 18:16:27 yes, but it's not where you'd go 18:16:28 it's where i'd go 18:16:52 yes, so you could receive that same advantage when i'm being unclear 18:17:09 no, because the world where i can understand others perfectly 18:17:11 we could both go there, and we'd both understand each other perfectly 18:17:15 isn't also the world where others can understand me 18:17:17 quintopia, elliott: why not go to a world with perfect understanding in both directions? 18:17:19 quintopia: who said you can go anywhere? 18:17:28 you get to stay in this shithole :D 18:17:35 Vorpal: that is implied 18:17:51 a world where "x can understand everything anyone says when x goes there" 18:18:12 quintopia, no I meant for communication back to here 18:18:22 who wants to communicate back to here? 18:18:30 this other world would be so much better... 18:18:40 Vorpal can never go to the other world, because it is literally impossible to understand anything he says 18:18:49 wonder how much tickets cost 18:18:53 :P 18:19:15 elliott: approximately two lisp machines 18:19:48 hell, people will perfectly understand why i need lisp machines to continue my wonderful work there 18:19:51 so i'll get them for free 18:20:11 indeed, but you have to pay two lisp machines to get tix there in the first place 18:20:32 quintopia: i'm saying 18:20:36 the money for two lisp machines 18:20:39 can either get me 18:20:42 1) two lisp machines 18:20:42 or 18:20:45 1) paradise + two lisp machines 18:22:34 shit\ 18:22:56 what 18:23:28 i just forgot i was supposed to be giving a presentation today in 15 minutes. i didn't make a presentation 18:23:42 it's been rescheduled for tuesday, but i still feel like a failure 18:24:00 dialog = gtk.FileChooserDialog() 18:24:00 Segmentation fault 18:24:28 Gregor: Powerbox is BROKEN BROKEN BROKEN 18:26:40 elliott, know anything about formal verification for haskell code? Such as what tools are available and how good they are? 18:26:58 Vorpal: Here's how you do that: 18:27:03 Vorpal: - Write your library in Coq, prove it there. 18:27:08 Vorpal: - Extract it to Haskell. 18:27:15 elliott, hah, really? 18:27:20 Vorpal: - If you want, write a Haskell module wrapping your functions to give some nicer interface stuff. 18:27:32 (That you have trivially wrapped the functions correctly is up to you to verify.) 18:27:59 elliott, is the resulting code as slow as coq or? 18:28:13 Gregor: I have a WM running under plash \o/ 18:28:14 | 18:28:14 /< 18:28:19 Vorpal: It extracts it to normal Haskell code. 18:28:21 elliott: Schweet. 18:28:26 A bit more explicit recursion than you'd expect, but... 18:28:30 Gregor: Yes, now nobody can exploit IceWM! 18:28:33 Gregor: I am SAFE 18:28:36 elliott: why would you want to do that? 18:28:52 quintopia: Why not? 18:29:08 well, most folks get along without it... 18:29:10 elliott, okay, sounds awesome. Is that how everyone does it, or are there alternative ways, more like annotating haskell code and then proving it? 18:29:12 why? 18:29:33 Vorpal: There are no such ways as far as I know. 18:29:48 quintopia: Dude, this is #esoteric. We do shit for the hell of it. 18:29:49 elliott, because, doesn't coq require that your program terminates, which would make proving and extracting a main loop a bit tricky 18:30:14 Vorpal: It... there... You're asking way too general questions, bother someone else :p 18:30:22 I'm no expert, but, yeah, . 18:30:26 http://codu.org/projects/websplat/ has a TODO list by the way. 18:30:37 Your MO 18:30:41 -!- augur has joined. 18:30:59 elliott, right, I guess you could do a trivial mainloop wrapper around a function that just did the main loop once 18:31:08 not sure how to model state then but meh 18:31:24 Vorpal: Coq is more advanced than you think it is. 18:31:25 (state, as in finite state machine passed as a parameter or such) 18:31:36 elliott, I know it is a hell of a lot of advanced :P 18:32:57 Gregor: what about long term todos? 18:33:20 quintopia: I only have short-term todos right now :P 18:33:55 Gregor: By the way, you accelerate indefinitely while falling right? 18:33:57 If so, keep that forever. 18:33:58 If not, add that now. 18:34:04 you said once you planned to make it multiplayer? did you mean internetworked multiplayer or just "you use the arrows, i'll use wasd"? 18:34:16 Gregor: "slightly-enhacned" 18:34:22 Multiplayer will be easy when he makes it an application :P 18:34:36 quintopia: Internet multiplayer, and that would require a more-than-slightly-enhanced browser, right now I'm considering it hyper-long-term TODO. 18:34:45 elliot: the falling sprites include wind effects, so it would be inconsistent not to have a terminal velocity 18:34:46 elliott: Actually the biggest problem with that isn't solved even by having an application. 18:34:54 elliott: Eepending on the environment, you won't render the page the same. 18:35:04 Gregor: Eepending on it. 18:35:07 Gregor: That's true, but... 18:35:27 Gregor: Have multiplayer mode use a codu proxy for every page. 18:35:30 Maybe disable cookies too. 18:35:35 Gregor: And, of course, don't allow any sort of WebKit tweaking. 18:35:38 Job done. 18:35:51 And don't allow browser windows of different sizes or different fonts. 18:35:53 Gregor: Randomness doesn't affect it because you get the same page from the codu proxy. OK so javascript randomness would break it. 18:35:55 But who cares. 18:36:00 Gregor: Sure :P 18:36:13 Gregor: Or rather, 18:36:19 Gregor: If the page uses any fonts, check both of you have them. 18:36:26 i think all that is solved by having it be an application: if the application is the only one doing the rendering then it will always be the same, yes? 18:36:29 And make you two agree on a window size -- or just force 800x600. 18:36:33 quintopia: See everything above ......... 18:36:53 elliott: you're assuming the application uses an external renderer 18:36:59 Yes. 18:37:00 It does. 18:37:02 WebKit. 18:37:09 Yeah, I'm not writing a new rendering engine :P 18:37:13 lol 18:37:18 GregorKit 18:37:19 well, that would solve the problem wouldn't it? 18:37:27 elliott: "Both of you" -- so this is no longer MMO, just MP 18:37:35 Gregor: For MMO, just fix it at 800x600. 18:37:39 Or maybe slightly wider. 18:37:50 And assume the fonts are close enough? :P 18:37:54 quintopia: Other things that would "solve the problem": Make everyone go to a farm of identical computers where Gregor lives to play 18:38:04 BRILLIANT! 18:38:04 X-D 18:38:13 are the machines lisp machines? 18:38:16 Gregor: How about: Have a list of what fonts are sans, serif, monospaced; replace them with a single generic font for each of those types. 18:38:21 Gregor: Failing that... 18:38:34 Gregor: And, of course, don't allow any sort of WebKit tweaking. <-- ? 18:38:43 elliott, what tweaking are you referring to? 18:38:51 Gregor: Maybe the font used on the page at any given time is the font being used by the person who has the least fontso n it. 18:38:54 Gregor: that is, if it's 18:38:55 A,B,C,D 18:39:02 and someone hasn't got A or B but has C 18:39:04 then everyone sees C 18:39:06 even if that person leaves 18:39:09 (to avoid disrupting things) 18:39:11 until everyone leaves 18:39:21 Gregor: and if someone comes in with just D, it tells them to eff off 18:39:24 Gregor: ...actually fuck that 18:39:25 lol 18:39:28 if someone came in with Super Obscure Font A 18:39:32 then nobody else would get in 18:39:34 Gregor: Gregor: How about: Have a list of what fonts are sans, serif, monospaced; replace them with a single generic font for each of those types. 18:39:35 Do that. 18:39:40 Gregor: And, of course, don't allow any sort of WebKit tweaking. <-- ? 18:39:41 elliott, what tweaking are you referring to? 18:39:42 fix the prng? 18:39:43 Configuration of any sort 18:39:48 cpressey: Yes, agreed. 18:39:50 Gregor: Fix the PRNG :P 18:39:51 elliott, hm 18:40:05 HERE YOU GO EVERYONE. HERE'S YOUR SEED. 18:40:12 Gregor: Okay so it's not easy, but all this normalising stuff adds up to maybe 500-2,000 lines of code. 18:40:14 Gregor, anything happening with zee or is it still ENOSTORY? 18:40:15 Gregor: Plus a tiny codu.org HTTP proxy. 18:40:18 I honestly don't think randomness is soing to be an issue. 18:40:26 Vorpal: ENOSTORY. 18:40:28 Gregor: Well, yes, but in pathological cases... easy enough just to srand. 18:40:42 Gregor: And, I mean, it would be FRIGGIN' SWEET. 18:40:55 Gregor, but you had an awesome one about a hacker that you told me some months ago? 18:41:11 Vorpal: That's the baseline story, but that just gets you INTO the game, it's not the game :P 18:41:17 Gregor, ah 18:41:18 Also, E_NODATA 18:41:34 Gregor: Who needs story? 18:41:40 Gregor: OMG I KNOW WHAT THE HATS CAN BE USED FOR 18:41:42 Gregor: Multiplayer. 18:41:43 elliott: ZEE 18:41:46 Sure, it limits you to 30 people per page but. 18:41:49 elliott, there is no way zee can be done without a story 18:41:49 elliott: ... wowzers. 18:42:04 TOO MANY TOPICS AT ONCE *brain asplote* 18:42:06 Vorpal: why not? 18:42:12 Gregor: Actually that's lame, 18:42:14 elliott, how? 18:42:17 Gregor: Hats should totally be powerups. 18:42:21 Vorpal: ...what do you mean, "how"? 18:42:34 elliott: multiplay will require truly different sprites for every player... 18:42:41 guh 18:42:42 elliott, how would zee make sense without a story 18:42:44 quintopia: Not really. 18:42:48 elliott, it is inherently story-oriented 18:42:57 quintopia: Just change the shirt/pants colour and have a name tag above everyone. 18:43:06 elliott: color replacement is possible but still takes time 18:43:10 Vorpal: "Why can't you do ZEE without a story?" "Because you can't do ZEE without a story!" 18:43:40 The whole idea is to find clues within pictures. 18:43:53 Gregor: So make the clues abstract and meaningless :P 18:43:56 elliott, well, it wouldn't just make any sense, I mean, the gameplay become pointless without a goal induced by a story 18:43:59 For the clues to be interesting, they have to fit into a context. 18:44:03 Vorpal: Pinball is also pointless. 18:44:04 Gregor, indeed 18:44:28 I suppose without context, ZEE becomes ultra-where's-Waldo. 18:44:34 Which is a playable game, but not really what I'm going for. 18:44:41 elliott, so would you like a text adventure or table top RPG without any sort of story? 18:44:55 > look 18:44:58 There are things here. 18:45:00 > get thing 18:45:02 You get a thing. 18:45:05 > inv 18:45:06 Things. 18:45:11 Gregor: I approve. 18:45:19 -_- 18:45:35 how about a meta-text-adventure 18:46:18 "You are in a place describable by many adjectives, and containing many points of interest. You can make choices. Choose or don't choose?" 18:46:22 > choose 18:46:46 "You choose to go in a direction that leads to another place." 18:47:29 quintopia, too much of a story 18:47:38 You can go direction-1, direction-2, direction-3 and direction-4. 18:47:49 quintopia: NOWAIT 18:47:50 but that same sequence repeats in every place you go...how is that a story? 18:48:11 "You are in node 34 of a graph. There are arcs to nodes 12, 38, 382 and 3." 18:48:15 quintopia, what happens if you don't chose? 18:48:46 Vorpal: "You are in a place still. It is still like it was a moment ago. Choose or don't choose?" 18:49:00 hm 18:49:05 You are playing a text adventure game. Do you continue or quit? 18:49:07 > continue 18:49:08 You lose. 18:49:15 You are playing a text adventure game. Do you continue or quit? 18:49:16 > continue 18:49:19 You continue playing the game. 18:49:20 > continue 18:49:22 You continue playing the game. 18:49:23 > quit 18:49:28 You stop playing the game. 18:49:32 You are playing a text adventure game. Do you continue or quit? 18:49:33 I like Gregor's better 18:49:37 > I N C E P T I O N 18:49:40 I preferred Gregor's too 18:49:47 since it makes a subtle reference to War Games 18:49:59 -!- augur has quit (Ping timeout: 276 seconds). 18:49:59 i am not certain that reference was intentional... 18:50:08 What's WarGames? 18:50:11 I didn't intend it that way so much as just an insult to text-adventure games, but OK :P 18:50:19 elliott, ... do I look stupid? 18:50:25 elliott 18:50:26 Vorpal: Yes. 18:50:29 don't answer...damn 18:50:31 Vorpal: And you've asked that before. 18:50:37 ("What's WarGames?".) 18:51:04 Better reference: 18:51:29 You are playing tic-tac-toe. Do you continue playing, forfeit, or beat your opponent over the head with a rusty pipe? 18:51:32 elliott, well I know what they are nowdays. If I asked (can't say I remember asking that), I presumably learnt about it then 18:51:36 > beat opponent with pipe 18:51:40 You win! 18:51:44 Vorpal: ..."they"? 18:51:51 I think Vorpal thinks I just mean "wargames". 18:51:58 Because he's an idiot, you see. 18:52:53 You are playing tic-tac-toe. Do you continue, forfeit, or beat Col. Mustard in the Conservatory with a lead pipe? 18:53:26 elliott, well, yes I thought quintopia meant wargames. Since that is what e wrote. 18:53:31 He wrote War Games. 18:53:41 elliott, yes, that is different from WarGames 18:53:50 and different from wargames 18:53:54 * Gregor bashes his head into this conversation. 18:54:09 Gregor: I actually thought Vorpal might have demonstrated knowledge beyond what I would have expected of him for a second. 18:54:14 NOOOOOOOOOOOOOOOOOOOOPE 18:54:29 Gregor: do you need a hug? 18:54:32 quintopia, yes indeed, which meant either interpretation was a possible fuzzy match 18:54:37 < elliott> "You are in node 34 of a graph. There are arcs to nodes 12, 38, 382 and 3." 18:54:42 you realize what this is, right 18:54:50 I mean there should only be 3 arcs 18:54:56 a maze of twisty little passages all alike? 18:54:56 and 382 is too high 18:55:11 OK, so 38 is probably too high too 18:55:13 cpressey, wumpus? 18:55:18 cpressey: I was envisioning it as rooms 18:55:20 And the arcs are directions 18:55:22 N, E, S and W 18:55:34 Vorpal: way to even out the demonstrated knowledge score :) 18:56:37 I'm gonna go with cpressey: Way to win a "fails to understand abstraction humor" point. 18:56:46 cpressey, well, it is utterly obvious it has to be wumpus. Just add an attribute to each node that can have the value {normal,bats,pit} and the wumups and then you are done 18:57:02 Gregor: Wait, who wins that? 18:57:09 cpressey. 18:57:09 Gregor's adventures in javascript have given me a sick idea. 18:57:12 You are standing in a tesseract. Exits N,E,S,W,U,D,M,G. 18:57:17 cpressey: I like sick ideas! 18:57:25 Gregor: Let's rape a kitten! 18:58:03 quintopia, ;D 18:58:33 oh kay, well 18:58:38 not that sick. 18:58:42 TELL US 18:58:51 THE SUSPENSE IS U N B E A R A B L E 18:59:42 quintopia: combine jQuery and underload somehow 18:59:44 pikhq: I wonder how smallX is so small. 18:59:59 to loop, you create new DOM elements, then interpret those, etc 19:00:01 cpressey, did you write that wumpus in b93 or was it someone else who did? 19:00:13 Vorpal: someone else. Wim Rijnders, I believe was his name 19:00:16 cpressey wrote all befunge software ever 19:00:22 cpressey, hm okay 19:00:25 elliott IS TRUE 19:00:29 meeting, bbl 19:01:21 elliott, why is that supposed to be funny? 19:02:46 because it makes fun of you in a way? that's probably all there is to it 19:04:22 quintopia, it didn't really. I nowhere indicated that I thought cpressey wrote all. It was just that I got it from his website... 19:04:37 -!- nooga has joined. 19:04:44 it did though\ 19:04:51 it implied that it was something that you might think 19:05:33 elliott, see /msg. 19:06:15 quintopia, which was however utterly untrue. 19:07:18 do you guys know some examples of problems that can be easily parallelized? 19:07:23 like uh 19:07:33 raytracing, or seeking prime numbers 19:08:16 nooga: http://en.wikipedia.org/wiki/Embarrassingly_parallel 19:08:43 nooga, sorting, by merge sort (for example). Cache effects will probably make it a bit worse on most common computers 19:09:16 (as in, sub-embarrassingly) 19:09:45 elliott, did you see the /msg? 19:10:01 elliott: wtf 19:10:07 nooga: what about it 19:10:22 i wouldn;t think about word 'embarrassingly' 19:10:28 elliott, I guess not then, I shall however consider the /msg implicitly acked by your actions 19:10:44 nooga: it's a bit of a colourful term. 19:10:56 cool 19:11:05 i think i will take fractal stuff 19:11:17 i need simple problem to focus on open-mpi stuff 19:11:51 and also i will get some nice, colorful pictures that can be showed to students 19:11:57 "Date adopted: William Blake" 19:18:44 Last really easily parallelizable problem I worked with was raytracing an animation... 19:19:08 One could just render 8 frames in parallel... 19:25:08 Graphics is full of embarrassingly parallel problems. 19:25:18 With raytracing you can also render thousands of RAYS in parallel. 19:25:38 THOUSANDS OF RAYS YAAAAAAAAR 19:25:41 ...what. 19:25:59 Gregor: It's a bit intuitive: lots of pixels, usually don't depend on one another, kerching. 19:26:09 Or you can do photon-tracing instead. All the quality of ray-tracing with none of the benefits! 19:30:16 -!- Zuu has joined. 19:30:16 -!- Zuu has quit (Changing host). 19:30:16 -!- Zuu has joined. 19:31:34 http://upload.wikimedia.org/wikipedia/commons/6/63/Glas-1000-enery.jpg <-- photon mapping uncanny valley ahoy! 19:32:16 "Date adopted: William Blake" <-- where is that interesting failure? 19:33:17 gregor: damn that is so close. if it were people, i'd be distrubed 19:33:20 http://upload.wikimedia.org/wikipedia/commons/6/63/Glas-1000-enery.jpg <-- photon mapping uncanny valley ahoy! <-- looks great 19:33:33 Gregor, a bit too clean and perfect to be realistic of course 19:33:34 but a wine glass isn't disturbing for some reason 19:33:38 Gregor: Brr. 19:33:42 Vorpal: Exactly. Uncanny valley. 19:33:42 That hurts. 19:33:54 Gregor, you could add some dust or such perhaps? 19:34:04 Vorpal: I couldn't, but someone could :P 19:34:07 Fingerprints. 19:34:10 Gregor, indeed 19:34:14 NO 19:34:17 NO FUCKING FINGERPRINTS 19:34:19 Particulate ... 19:34:21 Gregor, well it might be without fingerprints, if gloves were used 19:34:28 ;_; 19:34:33 Fine, but dust/particulate in the air. 19:34:37 i think it has more to do with the lighting on the floor 19:34:37 oh god 19:34:47 the glass is a bit too perfect too 19:34:52 especially near the extreme points 19:34:53 elliott, indeed 19:35:05 elliott, it actually looks quite okay at the top 19:35:48 just not where the "cup" joins up with the "non-cup" (presumably there are technical terms for these parts?) 19:35:52 The glass almost seems to float a bit in the air (probably really isn't)... 19:36:06 i think it's cool that we know enough optics to know how individual photons behave in situations like that tho. 19:36:10 Ilari, there is that too 19:36:12 it's nice to be alive today 19:36:34 Gregor, what tool was used to render it? 19:36:38 It's nice to be alive today, for today we have and no perspective. 19:36:50 indeed 19:36:53 Gregor, I can't trace it back since you gave me a link to the actual image 19:36:54 Vorpal: Idonno, just found it on the Wikipedia "photon mapping" page :P 19:37:02 There are other things too where the only way to distinguish natural and artifical is that artifical is "too perfect". 19:37:07 who needs perspective? 19:37:13 perspective is lame 19:37:16 Gregor, It is nice to be alive today, because we are still alive. 19:37:19 Gregor: It's nice to be alive today, for today we have specific places to go shit and no perspective. 19:37:23 Vorpal: *clap clap* 19:37:33 It's nice to be alive today, for today we have Vorpal and no perspective. 19:37:39 It is nice to be alive today, because the alternative is being dead. 19:37:42 (Paraphrasing) 19:37:46 Gregor, exactly :D 19:38:02 Gregor, (hey, wait, you did realise I was joking right?) 19:38:16 it's nice to be a conscious sentient and sapient entity today, for the alternative might be being a sponge 19:39:16 * quintopia imagines what it might be like to be spongebob squarepants 19:39:20 It's nice to be alive today, because otherwise it might be tomorrow. 19:39:31 Also, the lines where the segments of wood floor join look unnatural (no dark line due to gap). 19:39:50 Ilari: Might be fake wood flooring. 19:40:07 or really well done flooring 19:40:22 Gregor, wait, " Or you can do photon-tracing instead. All the quality of ray-tracing with none of the benefits!" <-- I thought there were stuff you could do with photon mapping that you couldn't do with ray tracing? Or is photon tracing different from photon mapping? 19:40:34 Even well done flooring has narrow dark lines. 19:40:43 Ilari: there are in fact such lines 19:40:47 look up and to the left 19:41:09 actually, i think this is a large wooden table 19:41:17 it has the feel of furniture more than flooring 19:41:20 Vorpal: Photon tracing is actually more powerful than photon mapping, and there is stuff you can do with either that you can't do with ray tracing. 19:41:21 Ilari, another thing wrong with the image: too sharp, perfect focus all the way 19:41:30 Vorpal: But both are so crazy-expensive that they're not worth it. 19:41:32 Gregor: Photon simulation 19:41:38 You create a universe in a box 19:41:43 And you put your objects in the box 19:41:45 Then you create photons 19:41:45 Gregor, ah 19:41:48 And fire them inside the box 19:41:54 They have little microchips in 19:41:56 To record their results. 19:42:03 (Hey, look, it's poetry.) 19:42:09 vorpal: that's not due to lack of realism. a camera set on infinite focus would do the same. or any orthographic projecting camera. 19:44:09 quintopia, a camera set on infinite would be blurred in the foreground, even with smallest shutter it isn't perfect. Also there would be other artifacts from the camera. Such as slight noise in the image. 19:44:12 and so on 19:45:24 quintopia, this image for example looks way more real (because it is): http://upload.wikimedia.org/wikipedia/commons/8/8e/Kaustik.jpg 19:45:50 vorpal: arguably it looks fake because it doesn't look like what we're used to (the view through a typical camera or our own eyes). but what we're seeing here may be more "real" than what our eyes see in the same situation. thus, it's in the uncanny valley because it is "too real" rather than because it is not real enough. 19:45:53 Vorpal: Totally 'shopped. 19:46:10 Gregor, har har 19:46:13 sort of the same as people thinking high frame rate movies look cheap, just because that is what camcorders produce. 19:46:27 Gregor: I can tell by the pixels. 19:46:30 *photons. 19:46:32 Gregor, alternative answer: yes, I guess someone bought that glass in a shop 19:46:43 Vorpal: Boo hiss X-P 19:46:47 Gregor, what 19:47:29 http://upload.wikimedia.org/wikipedia/commons/e/ec/Glasses_800_edit.png 19:47:34 If this was slightly more noisy, it would be very uncanny valley. 19:47:54 Dirty up the table a bit, add some imperfection in the glasses... 19:48:06 elliott, that is way more realistic than the other glass though 19:48:22 elliott, also I seem to remember that one was rendered with povray 19:48:23 It's not quite uncanny valley right now. 19:48:45 elliott, no or limited caustics in it 19:49:59 quintopia, if I look at stuff around me, they aren't perfect, there are small imperfections, stuff worn flat or chipped, grey dust on blank black surfaces. And so on 19:50:12 quintopia, the rendered images here are missing that 19:50:22 that is really only an issue if you see stuff close up in the image 19:50:43 if it is a more zoomed out view then it generally works better in 3D. 19:51:38 err, "in rendered 3D" 19:51:41 obviously 19:51:44 bbl 19:54:11 -!- Slereah has quit (Ping timeout: 240 seconds). 19:56:24 elliott, btw, wrt the "@ at every line", it isn't required it seems. Just the way everyone does it 19:58:02 -!- augur has joined. 20:00:15 Twitter? 20:00:20 -!- ais523 has quit (Remote host closed the connection). 20:01:16 Ilari, asking me? Then: No, the annotations for a formal verification system for C 20:01:19 and youtube 20:01:27 oh 20:01:33 which system is this 20:02:09 http://catseye.tc/projects/befunge93/eg/INDEX.html kind of sort of lists the author of every one of those Befunge-93 example sources 20:02:32 cpressey: I like to think that the intention of that page is to list every Befunge-93 prorgam ever. 20:02:38 cpressey: Also, yay Lahey 20:02:47 I like to imagine Lahey is a famous mathematician for inventing Lahey space. 20:02:54 Which REVOLUTIONARISED wrapping geometry. 20:03:02 *REVOLUTIONISED? Maybe not! 20:03:34 Hey, my HSV gar-collaborative-field thing went through: http://www.mezzacotta.net/garfield/?comic=519 20:03:50 That was fast; I was under the impression that the strip queue there is something like half a year long. 20:04:31 fizzie: Manyhills totally ruined that :P 20:04:44 quintopia, frama-c 20:05:00 fizzie: Did your Piet one get printed? 20:05:04 (TOTALLY PRINTED) 20:05:43 elliott: It got added to the queue, but since I sent it (a week or two) later than this one, it hasn't come out yet. 20:06:14 http://www.mezzacotta.net/garfield/?comic=514 this isn't really funny, it's just two strips! 20:06:21 wait 20:06:26 i guess the first one may not be the one referenced in the second one 20:06:30 if the second one actually references any real strip 20:07:23 fizzie: "Admin note: This submission was so timely and appropriate, I made a special exception and inserted it into the queue in place of another strip. (This is a *very* rare exception - it takes a lot of work to mess with the buffer queue.)" 20:07:28 fizzie: I doubt that that happened to you. 20:07:33 fizzie: So perhaps the queue isn't so long. 20:07:35 Hmm. 20:07:38 But maybe it is. 20:07:47 After all, that strip mentioned a massive backlog and it's recent. 20:09:04 fizzie: Oh jesus: "Garfield Pine Nuts Garfield" 20:09:28 worp 20:11:35 so this is like some spinoff of garfield minus garfield? how long has it been going? 20:15:12 quintopia: as long as mezzacotta 20:15:27 2008-11-15, it seems 20:16:32 fizzie: "Admin note: This submission was so timely and appropriate, I made a special exception and inserted it into the queue in place of another strip. (This is a *very* rare exception - it takes a lot of work to mess with the buffer queue.)" <-- on which one? 20:17:45 Vorpal: http://www.mezzacotta.net/garfield/?comic=513 20:19:45 ah 20:26:00 cpressey: OK, seriously, I'm sick to death of Python. Convince me that Lua is awesome. 20:27:31 it's one of a small handful of major projects published under the X11 license? 20:28:16 I... 20:28:18 Small handful? 20:28:38 quintopia: For one, X.org! 20:28:57 quintopia: Expat. PuTTY. 20:29:04 There are many more. 20:29:07 (Although none of the top of my head.) 20:31:00 elliott: surprisingly few worthy of mention tho. compared to GPL. 20:31:11 http://freshmeat.net/tags/mitx-consortium-license "1073 projects tagged 'MIT/X'" -- but of course "major". 20:31:23 There's curl, though. 20:32:12 quintopia: Well, y'know, I count a lot of MIT software as worthy of mention. 20:32:18 And a lot fewer pieces of GPL software than most people. 20:32:37 fizzie: Freshmeat is oddly patchworky at times. 20:32:51 I just made the collision detection really spanky :) 20:32:55 Also, MIT == ISC/BSD2/BSD3 in all countries that matter at all :P 20:33:00 Gregor: Oh yeah... so... spanky... 20:33:01 ?????? 20:33:26 Believe me, it's spanky :) 20:33:34 Gregor: It... spanks you? 20:33:43 Individual lines are separate platforms. 20:33:53 How... spanky... 20:33:56 So there's none of this "I'm standing on the magical edge of a box with text below me somewhere" 20:33:59 Gregor: Is it online now? 20:34:04 Yup 20:34:36 wooo 20:34:48 -!- BeholdMyGlory has joined. 20:35:00 It's also slower, but I think it's still fully usable. 20:35:32 Gregor: In the platform edition (let's call it that), you could do that initial loop in C++. 20:35:38 Wait, would you even need that loop with a getElementsByRect? 20:35:44 elliott: No 20:35:50 Gregor: Oh :P 20:36:00 Gregor: so falling damage is up next? 20:36:17 quintopia: The TODO list is not (necessarily) in priority order ;) 20:36:20 Falling damage might suck. 20:36:24 e.g. on looong pages with a lot of text. 20:36:28 It's useful to fall down the side. 20:36:31 Maybe just cap it at a maximum. 20:37:10 FALL = DIE MUAHAHAHAH 20:37:36 I just wish there was an element to complete sequence "Python, Ruby, Perl". 20:37:56 Something like Python but with proper lambdas and way less anal like Perl, except without Perl's complete and utter insanity and sigils. 20:37:59 You know, a scripting language. 20:38:06 elliott: nope, fall too far = death 20:38:14 this is why there will be a balloon... 20:38:16 Tcl! 20:38:16 no. 20:38:25 quintopia: Dude, your designing of Gregor's game for him is kinda creepy :P 20:38:40 No, it's not. 20:38:44 Yes it is. 20:38:46 I don't want to design it :P 20:39:00 Gregor: I think it should have sex robots in it that make you explode. 20:39:03 Also they shake the browser window. 20:39:05 Plz implement 20:39:16 Luckily I'm a filter :P 20:39:50 and luckily he leaks like a sieve 20:39:57 because splat=awesome 20:40:01 more death = more better 20:40:17 Basically I have a strange, morbid enjoyment of watching avatar-me die. 20:40:29 Funny, I have a strange, morbid enjoyment of watching you die. 20:40:32 wow me too! 20:40:35 Address please. 20:43:54 Dealing with strings in Lua is annoying... 20:44:16 And also integer division is annoying. 20:44:49 Ilari: The floats-only thing totally freaks me out, but what's wrong with its string handling? 20:46:09 The integer division wouldn't be so bad if there was operator that did integer division (even if it was really just division followed by floor). 20:46:41 No proper regexps for instance... 20:49:18 Ilari: Gotta be a binding for PCRE or something? 21:11:54 -!- augur has quit (Ping timeout: 252 seconds). 21:12:00 -!- oerjan has joined. 21:19:31 hi oerjan 21:19:45 HOW DARE YOU ADDRESS ME LIKE THAT 21:19:49 i mean hi 21:20:54 oerjan: what 21:21:16 oerjan: Just a typo? 21:21:40 clearly a typo. pesky keyboard. 21:22:43 i certainly do not have a paranoid split personality bent on world domination AT ALL 21:22:57 what a ridiculous thought 21:23:22 -!- GreaseMonkey has joined. 21:27:10 -!- webquint has joined. 21:27:27 so it seems google has forgotten what shape a G has 21:27:36 G looks nothing like dizzy gillespie 21:27:52 I... what? 21:28:06 Oh. 21:28:35 Uoogle. 21:29:40 * oerjan sees nothing, not even when selecting google in english 21:30:08 not that this is unheard of 21:30:44 oerjan: http://www.google.co.uk/logos/2010/gillespie10-instant.gif 21:30:59 cpressey: BetNSD. 21:31:39 * oerjan sees the top of the G behind gillespie 21:34:49 oerjan: I see he has a horn growing out of his head. He should get that looked at. 21:34:52 which combined with the rest looks more like an ø 21:35:41 an øoogle once bit my sister 21:36:04 did it get infected? 21:37:35 pøssibly 21:39:03 øoogle? 21:39:09 that's not even a word! 21:39:27 *WHOOSH* 21:40:08 lol 21:40:32 second-price actions are cool, agreed? 21:40:47 gah at ungoogable software names. 21:42:13 uhhhhhh 21:42:15 cpressey: pinginging. 21:42:32 s/actions/auctions/ 21:42:38 pinyin penguin 21:42:50 elliott: whatwhatwhat. 21:42:58 [[8====D "Surprise!", also represents a penis.]] 21:43:00 -- [[Emoticon]] 21:43:28 cpressey: cpressey: OK, seriously, I'm sick to death of Python. Convince me that Lua is awesome. 21:43:35 -!- sftp has quit (Remote host closed the connection). 21:43:39 I just wish there was an element to complete sequence "Python, Ruby, Perl". 21:43:39 Something like Python but with proper lambdas and way less anal like Perl, except without Perl's complete and utter insanity and sigils. 21:44:22 elliott: er... ok... well every time you quote someone using [[these things]], as you often do, you're using Lua quote syntax! OK, terrible argument. 21:44:26 but but but pythons are not venomous 21:44:29 elliott: does this mean you'll never finish vagrant? 21:44:40 cpressey: wait is that the string syntax? 21:44:41 elliott: Plof! 21:44:48 webquint: it's already platonically perfect! 21:44:56 cpressey: indeed (multi-line) also comments, heh 21:44:58 cpressey: no but srsly :P 21:45:14 elliott: it does have proper lambdas, and proper... something else. [[]] is the "omg all this" string syntax. (better than """ """ imo) 21:45:26 cpressey: yes but 21:45:26 --[[ omg all this comment ]]-- 21:45:32 function(x) return x+1 end 21:45:34 vs e.g. 21:45:36 \x. x+1 21:45:37 :P 21:45:48 even C++0x's lambdas are shorter 21:45:57 i speak english, i read english, i write english, i have no problem with code in english 21:46:02 [](int x) { return x+1; } 21:46:05 short is for golf 21:46:07 cpressey: nor I 21:46:09 but it's verbose 21:46:24 cpressey: you don't see an argument against verbosity? 21:46:26 i mean 21:46:27 i have no problem with verbose 21:46:29 it's like reading java in a way 21:46:33 -!- augur has joined. 21:46:34 * webquint likes smalltalk closures/lambdas best 21:46:39 when i stare at lua code, all the huge keywords stand out 21:46:45 and i can't see the actual functionality as well 21:46:49 that's my experience at least... 21:46:53 :/ 21:47:03 cpressey: but uh the lack of integers bothers me way more :p 21:47:10 ...also tables 21:47:10 I think lua is more sensible than python, but not significantly better... and lua's syntax is different from everything and there's almost no stdlib at all in lua 21:47:30 conflating arrays and dicts in tables is not cool, i agree 21:48:02 -!- Mathnerd314 has quit (Ping timeout: 245 seconds). 21:48:35 cpressey: and objects 21:48:36 you can rebuild lua so you ONLY get integers -- i kind of prefer that :> 21:48:53 dicts as objects i don't mind nearly as much 21:48:57 cpressey: oh great, code configuration by process of recompiling the interpreter 21:48:59 praise be it 21:49:06 yes, PHP woo 21:49:19 i forgot PHP does that :D 21:49:24 php.ini fuck yeah 21:49:37 i swear i want to like lua. 21:49:40 it's just 21:49:41 eurgh 21:50:10 well, i'm not saying i like lua, just that, it doesn't exceed my annoy threshold 21:50:25 for its size, it's hard to beat 21:50:42 cpressey: you don't mind verbosity but mention size? :) 21:50:50 javascript doesn't exceed my annoy threshold either, really 21:51:04 elliott: yes. think about it 21:51:36 javascript doesn't exceed my annoy threshold either, really 21:51:39 ...have you ever used javascript 21:51:43 like really used it 21:52:53 --enable-broken-cxx=no Force configure to accept the compiler, 21:52:53 even if it thinks that it is broken. 21:52:56 elliott, ^ 21:53:03 just so wtf 21:53:15 elliott: I've used the living FUCK out of JS, and it doesn't even approach my annoy threshold. 21:53:26 Gregor: Yes, and I don't listen to your opinions on languages :P 21:54:32 elliott: whose opinion do you listen to on anything? 21:54:37 -!- Mathnerd314 has joined. 21:54:48 webquint: it de...pends 22:00:09 webquint, whoever agrees with his preconceptions... 22:01:18 the mystery trains with predetermined destinations... 22:06:55 the most annoying thing about JS is that (effectively) I have to run it in a web browser 22:08:03 all the good JS implementations have standalone interpreters afaik 22:08:38 oh, the other thing lua has proper is, if you are going to conflate objects and dictionaries, bloody well just conflate them -- I'm looking at Python's division between object attributes and dict entries here 22:08:54 olsner: with ~no libraries 22:09:31 elliott: yep, same as lua :) 22:10:55 cpressey: I just want (python - anal - weird Gudioness + additional flexibility a la Ruby or Perl (without their insanity) + nicer object system) 22:10:57 is that so much to ask? 22:11:03 *anality :D 22:11:36 elliott: i want that too. 22:11:39 elliott: yes it is. 22:11:52 cpressey: well... you could go about it almost patchwork 22:12:08 cpressey: i mean ruby is *surprisingly* close, but (1) THE FUCKING "COMMUNITY" and (2) it has a few really, really weird parts that just destroy it 22:12:44 elliott: i feel that way mostly about it too. actually i know not so much about the community, but, yes, from what i do know, probably. 22:12:45 cpressey: so... take Python, get rid of the indentation-based syntax (I don't inherently hate it but it contributes a lot to the anality). lax up function calls a bit, maybe make the parens optional 22:13:05 add one or two loop constructs, a bit of flexibility, y'know? 22:13:09 make the object system less crazy 22:13:12 GIVE IT PROPER FUCKING LAMBDAS 22:13:15 that are *nice* to write 22:13:37 accordingly: eliminate statement/expression distinction 22:13:45 give it block frickin structure at any rate 22:13:48 maybe replace some control structures with lambdas a la ruby 22:13:49 for instance 22:13:54 for x in y: ... 22:13:55 becomes 22:14:02 y.each {x => ...} 22:14:06 (or whatever lambda syntax you use) 22:14:13 cpressey: and, yes, give it sane scoping rules. 22:14:18 not that hard. 22:14:19 The end. 22:14:29 why not just take out the garbage parts of ruby? what are they in your opinion? 22:14:30 ^^ that is a *really* short change set for an *entire language* 22:14:40 cpressey: It's... hard to specify. 22:14:49 the perl legacy, i'd start with 22:15:00 cpressey: One of them is: Those little lambdas -- "y.each {|x| ...}", equivalently "y.each do |x| ... end" (this equivalence is *great*) -- they're not actually lambdas. 22:15:04 i like /^foo/ as a literal regexp but it ends there 22:15:05 Every function call has zero or one "blocks" attached. 22:15:12 You can't pass these as arguments. 22:15:13 You can do: 22:15:15 proc {|x| ...} 22:15:19 equivalently, s/proc/lambda/ 22:15:21 but it's Not The Same 22:15:25 there *is* some shorthand for proper lambdas now 22:15:29 yeah, the pseudo-functional-ness of ruby is awful 22:15:43 s/of/in/ 22:15:46 which is 22:15:48 * elliott looks it up 22:16:07 ah yes 22:16:11 cpressey: -> x,y,z {x+y+z} 22:16:13 which is a bit odd, but 22:16:16 also you have to call it as 22:16:18 f.(1,2,3) 22:16:20 or f[1,2,3] 22:16:26 -!- cpressey has quit (Quit: Lost terminal). 22:16:26 since functions/variables aren't the same for various complex reasons 22:16:30 ("f" is a zero-argument function call) 22:16:46 eugh, lambdas aren't functions? 22:17:37 olsner: no, because 22:17:38 you can say 22:17:43 array.sort! 22:17:46 to sort a list in place 22:17:50 whereas, if sort! was a lambda there 22:17:53 array.sort! would be a reference to it 22:18:04 obviously 22:18:17 obviously 22:18:18 olsner: so basically they're separate concepts. 22:18:22 well it is obvious 22:18:29 olsner: would you have just referring to a lambda by name call it?! 22:18:38 f = -> { puts "hi!" } 22:18:40 g(f) 22:18:43 WHOOPS F GOT EXECUTED THERE 22:19:48 coming from haskell, it's so natural that f would be IO () or something, and would have to be interpreted to have its effects performed 22:20:29 olsner: yeah, uh, not a functional language 22:21:12 olsner: anyway ruby isn't all that bad. and i gather the japanese ruby community, from experience and second-hand, is *really* nice 22:21:16 but the western ruby community 22:21:17 fucking 22:21:18 douchebag 22:21:19 shitheads 22:21:30 with no reason to exist 22:21:44 they are driving you to create something better! 22:21:59 just don't. shut up. i dealt with those guys for like a year 22:22:03 i hate them beyond death 22:22:44 -!- MigoMipo has quit (Quit: Quit). 22:22:58 olsner: and of course *because* of the horrible western community, 22:23:04 ruby is regularly mocked to death on all corners of the internet 22:23:05 for no technical reason 22:23:11 and also because of it, it attracts more of the same type of people. 22:23:45 olsner: and... why the lucky stiff used to moderate it. a lot. 22:23:46 now he's gone. 22:23:50 -!- nooga has quit (Ping timeout: 264 seconds). 22:24:54 -!- cpressey has joined. 22:25:00 cpressey: what was the last thing you saw? 22:25:03 -!- nooga has joined. 22:25:32 something about space cows and space milk and space cheese 22:25:38 although this probably does not help you 22:25:41 -!- Sgeo has joined. 22:25:42 cpressey: -> x,y,z {x+y+z} 22:25:44 did you see that? 22:25:49 yes, that was it 22:26:01 prefix -> is a lambda? 22:26:13 that's... i don't think i like it 22:26:45 { x,y: x+y } i like as a lambda 22:26:56 cpressey: lemme pastie what you missed 22:27:36 *paste 22:27:39 pastie is fucking up stuff 22:27:47 cpressey: http://sprunge.us/PeWh 22:30:26 ! could totally be a postfix operator that calls the lambda HAHA 22:30:37 cpressey: i... shut up :P 22:30:42 it sort of is, i guess 22:30:43 f.() 22:30:48 .() is the postscript operator! 22:30:53 it's part of the name, isn't it? 22:30:54 .call even more so 22:30:56 f.call a,b,c 22:30:58 cpressey: i mean 22:31:00 .() sort of is 22:31:01 oh yes there is a punctuation theme here 22:31:06 but yeah, ! is just valid at the end of names 22:31:08 as is ? 22:31:11 cue taken from scheme 22:31:32 yeah ruby is... yeah. 22:31:51 cpressey: it's... i actually recommend you try it sometime. compared to python, i mean... 22:32:06 it is not as bad as it sounds. just NEVER EVER look at ANY community for it if it's not in japanese. 22:32:28 in fact {learn japanese; use ruby} is probably way easier and nicer than {use python} if you have the patience :) 22:32:41 i've written a little ruby 22:32:55 cpressey: here's the interface of the blind gods: http://edbrowse.sourceforge.net/ 22:32:57 i think i got bored with it and stopped 22:33:02 it's ed + web browser 22:33:04 + mail client 22:33:11 akkhkkk 22:33:13 cpressey: http://www.eklhad.net/edbrowse/usersguide.html#guide take a look at these shortcuts 22:33:15 no it's awesome! 22:33:24 it is totally the best interface for blind people i'd say 22:33:57 "You will never be faster than your sighted colleague when traveling through unfamiliar territory, no matter what system you use." 22:33:59 *sniff* 22:34:12 cpressey: it's like this guy used ed as an editor and he was like 22:34:17 awesome this is totally how i want to work 22:34:19 and then he was like 22:34:30 you know... i wish i could g/re/p my web pages 22:34:33 or my emails 22:34:34 and he added it 22:34:36 and then he was like 22:34:40 would be nice if i could fill in forms... 22:34:42 and then that 22:34:45 that's cool 22:34:49 cpressey: oh file manager too 22:34:50 haha 22:35:09 cpressey: of course *ideally* it'd be, like, a bunch of programs 22:35:16 and the main ed thing would just be a "shell" using ed commands 22:35:17 but ... 22:35:21 even i wouldn't want to code *that* on unix 22:35:25 sounds like a pain 22:35:57 "Now sessions 2 3 and 4 are the subpages about plains trains and automobiels respectively. You can fill out forms or follow hyperlinks in any of them, or stay in session 1 and do something else." 22:36:00 cpressey: blind tabs 22:36:11 "If you are trying to listen to a speech synthesizer, the last thing you need is background music. Instead of playing the song, I make it available to you through a hyperlink." 22:36:12 how kindly 22:36:37 cpressey: ha it has macros 22:36:46 cpressey: this guy has managed to turn ed into emacs and *actually deemphasise* the editor part 22:37:03 wow it does frames. 22:37:07 and pdfs. (via pdftohtml) 22:37:21 cpressey: oh yeah and he wrote his own javascript compiler and engine to use with the web browser. 22:37:31 he's changed it to a mozilla one now, but STILL HOW HARDCORE IS THAT 22:38:53 i am now in awe at how comprehensive this thing's functionality is 22:39:23 cpressey: it has database functionality :D 22:39:24 that's kind of i'm in awe too 22:39:26 through odbc 22:40:07 -!- augur has quit (Ping timeout: 245 seconds). 22:40:54 cpressey: gahaha, he wrote his own speech system too 22:40:59 "Installation of the Jupiter speech package proceeds in two steps. First the Linux kernel is patched, so that it can accept adaptive modules. These modules intercept keyboard input and console output and modify it in some way, to make the computer more accessible. For instance, an adapter for the blind might divert console output into a speech synthesizer. Jupiter is such an adapter. This package includes the patch that "prepares" the Linux 22:41:00 kernel for adaptive modules, and the Jupiter module itself, which you load via insmod or modprobe." 22:42:30 Gregor: check your PMs! 22:43:05 cpressey: and he has an adopted kid with severe ADHD! question: where does his time come from? 22:46:51 brb 22:52:04 btw, I totally have an iMac now. 22:52:08 * cpressey instant mild dislike 22:56:39 -!- webquint has quit (Ping timeout: 265 seconds). 22:59:38 The OSX store doesn't allow system apps? 23:00:01 That's pretty much the sort of app that I'd actually _appreciate_ careful monitoring of 23:00:28 -!- augur has joined. 23:05:13 -!- tombom has quit (Quit: Leaving). 23:07:57 this keyboard is made of lead and chiclets 23:10:19 http://notalwaysright.com/try-explaining-that-to-your-insurance-agent/7819 ... yes. 23:19:04 eugh, "deja vu all over again" 23:20:39 cpressey: you bought it? 23:20:40 or 23:20:43 employer 23:21:10 employer 23:21:22 it has a key labelled "return" 23:21:45 because it totally want to return the texted I just typed in 23:21:51 s/it/I/ 23:21:56 *text 23:21:59 blargh 23:22:25 (I KNOW WHY IT'S CALLED THAT. DO NOT START.) 23:23:52 cpressey: i wouldn't start :P 23:24:12 and Vorpal's not here, so, ok 23:25:11 cpressey, am I not? 23:25:28 well, you weren't 23:26:26 I just like the irony of a company that prides itself on usability * key labelled with an obsolete unrelated function. 23:27:55 -!- sshc has joined. 23:27:55 cpressey: is that... meaning globbing? 23:28:02 OH LOOK new macbook air, i wonder how much it sucks 23:28:22 elliott: it means "too lazy to type this part; infer". alternately, it's a product of some kind 23:28:35 cpressey: * 23:29:35 ok this... actually... the new macbook air looks kinda nice. 23:29:59 wait, this marks Apple moving into the netbook industry 23:30:01 11.6" version 23:30:24 although it's... 23:30:27 a netbook for rich people 23:31:02 and mactards 23:31:07 -!- Mathnerd314 has quit (Ping timeout: 240 seconds). 23:31:09 ...there was a reddit picture about that yesterday 23:31:32 olsner: i.e. rich people. 23:31:35 :) 23:31:39 or just poor, drooling people. 23:35:14 -!- augur has quit (Ping timeout: 264 seconds). 23:36:37 in black clothes with slim cigarettes 23:36:53 and largish, purple-tinted glasses 23:37:55 theory: cpressey works for Microsoft 23:39:03 * Sgeo is tempted to try Opera 23:39:06 Don't know why 23:39:26 I don't know either 23:39:37 elliott: so M$FT got me a mac so I could... 23:39:39 Well, actually, I do know why 23:39:47 cpressey: precisely. 23:40:17 Maybe Opera will deal with Reddit nicely 23:41:47 cpressey: i hate... software 23:41:50 Why would Safari have privacy mode first? Isn't Jobs anti-porn? 23:42:08 Why would Safari have privacy mode first? Isn't Jobs anti-porn? Why would Safari have privacy mode first? Isn't Jobs anti-porn? Why would Safari have privacy mode first? Isn't Jobs anti-porn? Why would Safari have privacy mode first? Isn't Jobs anti-porn? Why would Safari have privacy mode first? Isn't Jobs anti-porn? Why would Safari have privacy mode first? Isn't Jobs anti-porn? 23:42:10 lolling 23:42:10 all 23:42:10 over 23:42:12 the 23:42:14 floor 23:42:32 I hate programming. Therefore, I will make my own programming language. 23:42:42 Sgeo: analogy 23:42:44 utterly 23:42:44 fails 23:42:48 What? 23:42:52 That wasn't an analogy 23:43:00 That was a completely disconnected thought 23:43:00 it wasn't? 23:43:05 ...okay. 23:43:08 * Sgeo was thinking about PHP 23:44:43 -!- Mathnerd314 has joined. 23:44:49 w(h^n) 23:45:30 cpressey: whhhhhhhhhhhhhhhhhh? 23:45:51 yesh 23:47:17 Opera actually works with my school's website! 23:47:53 Maximized Opera fails at that rule about clicky things near the top of the screen should go all the way 23:51:18 Now, how do I open new tabs 23:52:01 yesh, the mutt shell 23:52:50 cpressey: theory: maybe i should make my OWN SCRIPTING LANGno 23:53:01 oerjan: you have that comic strip up there in norway? 23:53:08 Sgeo: ctrl-t as every other browser with tabs? 23:53:23 I meant a link in a new ab 23:53:25 *tab 23:53:30 elliott: it would be the only one that isn't broken (except where it is) 23:53:33 well the library has it 23:54:16 oerjan: you have that comic strip up there in norway? 23:54:16 i had not actually seen it until being in contact with americans 23:54:18 which comic? 23:54:19 also here http://www.start.no/tegneserier/ 23:54:25 elliott: mutts 23:54:39 http://muttscomics.com/ 23:54:40 never heard of it 23:55:03 not exactly a great comic strip, but alright sometimes 23:55:26 um... the one on that page sucks 23:55:37 cpressey: question 23:55:42 cpressey: have you ever read the Perry Bible Fellowship 23:55:49 and are you too curmudgeonly to enjoy it or not 23:56:20 elliott: um... i don't *think* so... 23:56:27 cpressey: http://www.pbfcomics.com/ 23:56:34 pick one. click. laugh. failing that, start from step one. 23:57:01 confused by web page. oh. those are links. ok... 23:57:54 Opera works nicely with Reddit 23:57:58 And with my school's website 23:58:23 um 23:58:27 some of them are funny 23:58:38 cpressey: the ones near the bottom aren't so good 23:58:40 the earlier ones 23:59:19 'Miggs' is good 23:59:40 there's quite a few in that... "continuity" if you can call it that 23:59:44 pretty sure all of them have died at least once