00:12:59 -!- augur has joined. 00:25:30 yessir 00:25:38 -!- cal153 has joined. 00:26:40 quintopia: in a freetext calculator thing, variable assignment has to come after the expression :) 00:26:43 -!- iamcal has quit (Ping timeout: 240 seconds). 00:26:47 so you can do [complex workings out]->var 00:26:51 like STO on a calculator 00:27:28 oh goody 00:27:41 i always thought that made more sense than left assignment 00:28:06 why would one put the operation that comes last on the left? we read left to right damnit 00:31:12 -!- catseye has joined. 00:32:27 In my opinion, which you will all immediately disavow while calling me an idiot, having the target on the left of assignments in an imperative languages improves code skimmability. 00:32:38 *in imperative languages 00:32:46 Gregor: Duh, I agree. 00:32:50 I'm saying for a calculator. 00:32:56 Where everything is evaluated immediately. 00:33:07 2+2 shows 4 somewhere else, then + shows 4+, then 3 shows 7 00:33:14 then ->x or whatever stores it into x 00:33:21 or rather counts as a definition of x, but the point is 00:33:27 you can use it like a pocket calculator pushing buttons 00:33:29 except sleeker 00:33:37 I was responding to quintopia, whose statements seem to have suggested wider applicability (maybe :P ) 00:33:44 2+2(cursor here)=4(this is updated as you type) 00:33:48 an esolang has that iirc 00:33:56 catseye: perhaps, yeah 00:34:24 have it draw the ast as you type! silly. 00:34:31 http://peterstuifzand.nl/2010/05/30/calculators-on-computers.html, http://stuifzand.eu/abacus/ is what inspired me 00:34:33 although it's imperfect 00:34:39 instructive, for a prog lang env for beginners, though, maybe 00:34:55 "I created a new calculator. This short video shows how it works. I use it more often than calculator that's included on my computer." ;; i dislike this fashionable style of writing 00:34:58 have it draw the ast as you type! silly. // .. omg. 00:35:00 Best idea ever. 00:35:06 where talking in really simple retard-statements is considered the mark of a minimalist 00:35:16 Gregor: Yes, yes it is. 00:35:16 -!- FireFly has quit (Quit: swatted to death). 00:35:39 elliott: It is the fashionable style. It works well. I use it. 00:35:45 Gregor: Codekana, a Visual Studio extension for Windows, has a partial-AST-parser thing that it uses to highlight invalid code (missing a } somewhere and recovering after and stuff) 00:35:52 The blog post had diagrams. 00:35:58 That would be very cool to see in realtime. 00:36:26 elliott: ... Codekana? 00:36:30 catseye: I wrote a program to do arithmetic. You type in formulas with the keyboard and it tells you the results. I find it a very useful piece of software. 00:36:35 pikhq: http://www.codekana.com/ It's just some random VS extension. 00:36:52 elliott: Or as I'd write it, コデカナ? 00:36:52 http://www.codekana.com/blog/2009/04/02/on-the-speed-of-light-innovation-and-the-future-of-parsing/ Here's the "blah blah blah" blog post about it, with diagrams. 00:36:53 Skimmable. 00:37:08 Does feature phi in a drawn example of C code to denote "any integer", though. 00:37:11 Which is something you don't see every day. 00:37:20 elliott: with the keyboard! 00:37:21 pikhq: It's developed by the guy who wrote ViEmu. 00:37:31 (which was a surprisingly complete vi-esque interface to Visual Studio) 00:37:32 elliott: Yes, I'm just going "what the hell is with the name". :P 00:37:34 and, uh 00:37:38 "Word, Outlook and SQL Server" too 00:37:56 pikhq: so it can use ninja silliness and, uh, 00:37:59 [[Inspired on “kana”, the Japanese name of reading and writing systems, I decided to call the product Codekana. And I set out to write it.]] 00:38:07 it's, like, meant to make you look at your code in a WHOLE NEW WAY or something 00:38:07 anyway 00:38:14 i'm only talking about the partial ast stuff 00:38:25 which gives a reasonable parse even if you type in partially valid code in the middle ({ without }) 00:38:32 by using history 00:38:50 elliott: So, it's actually a reference to kana. Without even the faintest resemblance. 00:38:56 pikhq: Oh shut up :P 00:38:58 elliott: I could respect it if it used kana. :P 00:39:10 -!- catseye has changed nick to DrNinja. 00:39:19 忍者医師! 00:39:24 -!- DrNinja has changed nick to catseye. 00:39:29 No, it's too early for THAT. 00:39:30 猫目! 00:39:50 Anyway yes calculators could be so much better. 00:43:03 "But the intermediate version, when you have typed the new code only partially, is incorrect. In this incorrect version, braces dont pair properly." 00:43:19 I know I'm weird but... you could just prevent an unmatched brace from ever appearing. 00:43:33 -!- new-lisper has joined. 00:43:44 The solution is to not use braced languages. 00:43:45 If you type {, you get {}. If you delete }, you delete {*}. 00:43:46 :P 00:44:08 catseye: ITT: paredit 00:44:12 pikhq: i'm sure new-lisper disagrees ;D 00:44:15 it's actually quite irritating. 00:44:16 Funny how you start talking about non-braced languages right as new-lisper comes in :P 00:44:21 pikhq: Even Haskell pairs 'let' and 'in'... 00:44:55 catseye: not to mention, you know, _actual_ brackets 00:44:56 elliott: I realize it can be irritating, but I don't think we've explored the space very much either 00:46:07 The ultimate expression of this idea is probably ZX81 BASIC program entry :) 00:46:16 at least, for the time 00:46:56 Bad idea of the day: A platformer in which the world you're playing again is the web. 00:47:12 s/again/in/ wtf 00:47:29 Gregor: Totally preempted that one in 2006. (Long story.) 00:47:32 Gregor: very popular in china, that 00:47:44 elliott: ??? 00:47:46 oerjan: ??? 00:47:58 Gregor: Turns out that in reality nobody actually wants to wade through levels of shock pictures and therefore it should never be made. 00:48:06 note: okay, so that isn't what you meant 00:48:39 No. No it is not :P 00:49:26 You have been rickrolled! Return to level 4. 00:49:40 The playing field is not FIGURATIVELY the web. 00:49:46 The playing field is LITERALLY web pages. 00:49:46 Gregor: Ohhhh. 00:49:55 oerjan: go back three spaces. 00:50:05 Gregor: I think something like that has maybe been done. Don't recall. 00:50:12 Gregor: So, what, you shimmy down the jagged cliff of /b/? 00:50:23 Wade through the gaps in long pages of styleless text? 00:50:29 Yes! Links are warps! You want to find certain things? 00:50:33 Jump into Wikipedia links? :D 00:50:36 There could be powerups for certain elements. 00:50:37 Yes! 00:50:43 Please tell me it can POST. 00:50:51 fuk 00:50:53 Achievement Unlocked: Vandalise Wikipedia! 00:50:57 X-D 00:51:14 Gregor: You have to collect letters from pages on the internet, travel back to Google, and put them in the search box. 00:51:18 This is the only form of key entry supported. 00:51:41 That ... is awesome. 00:51:43 letters are too easy. use whole words. 00:51:48 Note to self: Telling all my problems to the wrong channel is a bad idea. Prevents me from having the energy to repeat my problems to the right channel 00:51:49 "I need to optimize my slacking -- hey, I can surf the web *and* play a platformer *at the same time*!" 00:51:57 elliott: You carry them like monsters in Mario 2. 00:52:11 oerjan: Yes 00:52:13 Yes yes yes 00:52:27 Aardvark ... municipal ... barely legal ... artichoke 00:52:34 [SEARCH] 00:52:47 Hottest - porn - ever. 00:52:48 Gregor: Entering a googlewhack gets you points. 00:52:56 also, you have only "I'm feeling lucky" 00:53:10 Man, landing on Last Measure would be the worst thing ever. 00:53:12 Game popups! 00:53:12 OMG 00:53:16 You can jump on game popups 00:53:17 And they like 00:53:19 bob down 00:53:21 Because of your weight 00:53:25 And you can jump on top of them 00:53:26 Then climb in 00:53:32 When you climb in they zoom to full screen 00:53:40 Casino ads will never be the same 00:53:49 Note to self: Telling all my problems to the wrong channel is a bad idea. Prevents me from having the energy to repeat my problems to the right channel 00:53:51 ctrl+c ctrl+v 00:54:01 Gregor: ...now implement it in javascript 00:54:11 Gregor: and if you visit the game in the game, all your points in the metagame count to the rael game 00:54:11 I mean emotionally 00:54:12 *real 00:54:13 oerjan: go back three spaces. <-- or maybe like three years, come to think of it. 00:54:23 Sgeo: ctrl+c ctrl+v requires no emotional energy. 00:54:29 oerjan: burn :D 00:55:16 elliott: The problem with implementing it in JavaScript is that the only way you can inject yourself onto an arbitrary page is with a bookmarklet ... 00:55:34 Gregor: Nonono, just use Web Sockets or whatever (when that gets implemented :P) 00:55:37 Gregor: And render it manually. 00:55:43 ...come to think of it, how about you do it in NOT JAVASCRIPT 00:55:47 (rendering engine in javascript = lol) 00:55:56 firefox plugin FIREFOX PLUGIN yeah 00:56:14 you could do it as java with http://www.interactivepulp.com/pulpcore/? 00:56:19 the only thing that makes java applets not sucK! 00:56:20 *suck! 00:56:28 Writing it as a Firefox plugin has the advantages of discriminating against Chrome users AND being JavaScripts! 00:56:42 or 00:56:48 how about writing it in C or something like that 00:56:50 CRAZY IDEA I KNOW 00:56:54 Gregor: it's SYSTEM JAVASCRIPTS now 00:57:12 I like how Gregor is now disparaging Javascript. 00:57:15 elliott> *real <-- and here i was imagining some game with ufos and aliens 00:57:22 Unless he's beign serious. 00:57:23 *being 00:57:34 elliott: I'm not sure whether I was being serious or not. 00:57:34 oerjan: i approve 00:57:36 ruby on raels 00:57:38 elliott: I haven't yet decided. 00:57:42 "Opponents of the mosque, who have sued the planning commission and other county officials, have argued that it shouldn't have been granted a religious use permit because, according to them, Islam isn't really a religion." 00:57:49 On the building of a mosque in Tennessee. 00:57:51 http://railsblob.blogspot.com/2006/08/ruby-on-raels.html 00:57:53 (RIP why) 00:58:07 WHY CAN PEOPLE BE THAT DUMB. 00:58:08 WHY. 00:58:18 WHY DOES PHYSICS ALLOW FOR THIS. 00:58:30 elliott, why probably isn't dead 00:58:37 WHAT SORT OF KIND, LOVING, RANDOM EXPLOSION THAT CREATED EVERYTHING ALLOWS FOR THIS. 00:58:40 Sgeo: whoever-posted-as-why isn't 00:58:41 why is 00:58:41 :P 00:59:26 Who gets to decide what is and what isn't really a religion? 00:59:42 Sgeo: Jesus. 00:59:53 ITT: Sgeo manages to uncover the point with blind fumbling 01:00:20 Sgeo: In the US legal system? The courts. 01:00:30 Your MO-- 01:00:34 pikhq's mom. 01:00:36 gets to decide 01:00:49 Yes. The courts act in her name. 01:00:57 Much like the Queen, except nobody knows. 01:00:58 :P 01:01:04 Except elliott. 01:01:11 Now I'll have to remove his brain. 01:01:11 uh oh 01:01:17 And now, all of #esoteric 01:01:20 * elliott flees; helplessly 01:01:28 * elliott misuses; semicoli 01:01:45 Gregor: You'll just get surgically induced amnesia. 01:01:51 * Gregor misuses the fourteeth letter of the alphabet 01:02:32 Also the ninth and fifteenth :P 01:02:54 Also, it's quite amusing that people are going full retard on this but praise Bush. 01:03:13 Bush ran around saying "Islam is a religion of peace". Honest. I am not making this up. 01:03:30 (I presume he meant "as opposed to Christianity. We'll kill them all!") 01:03:47 Are there current politicians saying that Islam is dangerous and needs to be destroyed? 01:03:55 I think no politician would be that stupid 01:03:56 I hope 01:03:58 Sgeo: Tea Party. 01:03:59 HAHAHAHA 01:04:04 HAHAHAHA 01:04:06 ROFLMAO 01:04:15 gjsdiogjsdiogjaahahahahahahahahahahahahahahahahahahahahahaha 01:04:16 AHAHAHAHAHAGoodnight. Bye. 01:04:17 -!- elliott has quit (Quit: Connection reset by peer review). 01:04:32 Tea is for emacs users. 01:04:48 Did you just call me a Tea Partier? 01:04:48 Emacs is for tea drinkers. 01:04:51 Take that back 01:04:53 WHAT SORT OF KIND, LOVING, RANDOM EXPLOSION THAT CREATED EVERYTHING ALLOWS FOR THIS. <-- i was going to say kind, loving, random explosions don't exist, but then i realized i would inevitably get a response involving roughly "in my pants". 01:05:11 No, I didn't 01:05:20 oerjan: That was a hybrid of "kind loving God" and "Big Bang", of course. :) 01:05:22 Emacs Party 01:05:24 -!- Behold has quit (Remote host closed the connection). 01:05:44 So, I am from Ed Party 01:05:54 Notepad Party 01:05:57 me ducks 01:06:05 Dear key. Please work 01:06:23 Or better, Cat Party 01:06:34 SciTE-nano Coalition 01:06:42 Can Has Cat Party 01:06:46 nothing beats "cat - > myprogram.c" 01:07:09 CAN HAS CHEEZBURGUER? Party 01:09:15 Sgeo: it is the UNKEY 01:14:14 -!- augur has quit (Ping timeout: 264 seconds). 01:15:43 it's unkey-dorey 01:20:25 bbl 01:22:16 -!- wareya has quit (Read error: Connection reset by peer). 01:22:55 -!- wareya has joined. 01:29:16 bah nosebleed 01:33:50 -!- ChanServ has set channel mode: +o oerjan. 01:34:24 -!- oerjan has set channel mode: -o oerjan. 01:48:31 -!- augur has joined. 01:49:46 -!- augur has quit (Read error: Connection reset by peer). 01:51:10 -!- Sgeo has quit (Ping timeout: 252 seconds). 01:54:38 -!- Sgeo has joined. 02:14:05 beautiful 02:14:24 ntfs-3g created some files on my NTFS partition that contain backslashes in their names 02:14:46 WINDOWS IS CONFUSE 02:33:02 enum 4 0 0 doif cage lt 4 setv va00 4 subv va00 cage ages va00 endi next 02:33:06 Hyuk 02:36:52 Idea: create tarball of 1,024 files, each named "abracadabra" but with different capitalization. Distribute to friends who use case-insensitive file systems. 02:37:03 -!- wareya has quit (Read error: Connection reset by peer). 02:37:22 hm no. 2,048 files, I guess 02:37:49 what hocuspocus 02:37:53 -!- wareya has joined. 02:38:20 OH i just made 2,047 files disappear!! 02:38:55 -!- new-lisper has quit (Read error: Connection reset by peer). 02:46:10 Wait what? 02:46:17 catseye is now a Creatures person? 02:46:54 Or, at least, is mocking the "Age all creatures to adult" code 02:47:06 But... catseye provided absolutely no context. 02:47:15 Sgeo: No context for you! 02:47:34 CONFUSION REIGNS. 02:47:51 I know exactly what the code you pasted does 02:47:58 So will elliott, probably 02:48:06 I just copied it from a wiki. 02:48:12 I have never played this game myself. 02:49:13 Sgeo: I also saw an SHA1 hash of your name. 02:49:36 I was more paranoid about my name back then 02:50:43 Right now, given that hash, and what #esoteric knows about my name, and this sentence, you could bruteforce my middle initial. 02:51:35 How did you stumble upon that wiki? 02:52:17 S. Hubert G. 02:52:34 Sgeo: tbh, I googled "Sgeo" :) 02:53:22 "Seirei Gakuen Educational Organization", says the first hit 02:54:17 If you googled for my real name, you'd find a gay DJ 02:54:22 * Sgeo is not a gay DJ. 02:54:55 SAYS YOU 02:56:40 Sgeo: Are you a Windows 7 user now? er, part time anyway? Or did you install something else... 02:56:52 Windows 7 02:57:04 At some point, I'll get around to installing some Linux distro 02:57:07 Maybe 02:59:16 Sgeo: Use the PowerShell! Or, y'know, don't. 02:59:43 I'm trying to figure out how to install it on Vista... PowerShell 2.0 seems to come with some intimidating "framework" thing... 03:00:08 "WinRM is the Microsoft implementation of WS-Management Protocol, a standard Simple Object Access Protocol (SOAP)based, firewall-friendly protocol that allows for hardware and operating systems from different vendors to interoperate." 03:00:12 ewwwwwwwwwwww 03:00:28 That's like a triple dose of evil all in one package. 03:01:14 Mmmmaybe I will stick with 1.0 03:03:44 This is actually pretty nice music 03:10:34 http://djsethgold.com/demo/DjsethGoldLive.mp3 03:12:36 Windows' handling of case in filenames is so confusing nowadays. 03:12:46 Now, it *stores* case, but is still case insensitive. 03:14:13 -!- augur has joined. 03:14:28 Dear Steve Jobs: open/closed app development and integrated/fragmented are orthogonal issues 03:17:12 Sure, openness to manufactuters can result in fragmentation, but there are other kinds of openness, that Android is and that iOS is not 03:19:10 -!- augur has quit (Ping timeout: 252 seconds). 03:25:54 -!- catseye has quit (Quit: leaving). 03:26:27 -!- catseye has joined. 03:26:31 POWERSHELL!!! 03:27:48 POWERTHIRST 03:28:01 400 babies. 03:30:11 enum 0 0 0 kill targ next 03:30:35 That one kills everything everywhere 03:30:52 I was going to ask if it was your intention to kill 400 babies. 03:31:04 (Destroys, really. Kill all life is enum 4 0 0 dead next ) 03:31:29 Well, unless someone stuffed a creature into a non-4 family 03:31:34 Which I've done 03:35:42 \o/ 10.6G free, so I can actually install stuff now 03:35:42 | 03:35:43 |\ 03:36:28 And I wonder if I can get Cygwin bash running inside a Powershell window. I mean, I can get cygwin's irssi running here, so... it would be nice anyway. 03:36:31 bbiab 03:36:33 -!- catseye has quit (Quit: leaving). 03:41:30 I love this music! 03:41:48 -!- Zuu_ has joined. 03:41:50 -!- Zuu has quit (Read error: Connection reset by peer). 03:42:41 -!- Zuu_ has changed nick to Zuu. 03:42:49 -!- Zuu has quit (Changing host). 03:42:49 -!- Zuu has joined. 03:43:38 -!- augur has joined. 03:48:15 -!- catseye has joined. 03:48:31 I wonder if his friends called him Benny 03:50:39 ahhhh i bet i have to start a login shell for it to work 03:53:49 -!- augur has quit (Ping timeout: 252 seconds). 03:55:10 ha HAAAAAAAAAA 03:55:11 %SystemRoot%\system32\WindowsPowerShell\v1.0\powershell.exe -noexit C:\cygwin\cygwin.bat 03:55:53 FINALLY, decent copy-and-pasting functionality while within Cygwin. 03:59:48 And forget '-noexit'; it actually works better without. 04:09:53 -!- jcp has joined. 04:11:04 eww 04:11:18 w3m requires libgc 04:11:25 that's vaguely disturbing 04:11:46 (at least the cygwin port does) 04:11:50 *insert your mom joke* 04:12:41 your mom is so fat she had to install cygwin to install bc to run bash to compute her weight LOL 04:14:22 ...needs work. 04:17:01 -!- calamari has joined. 04:17:24 Your mum is so fat, she gives RAMSAN's buffer overflows 04:17:46 Your mum is so ugly, she has to be rendered in parallel to avoid the OS from rebooting 04:18:11 Your mum is so slow, she lost against Java in a garbage collecting contest 04:18:31 i wonder if i really wanted a your face joke, actually. 04:18:44 Your mum is so poor, google won't let her search because showing her ads is pointless 04:20:13 RAMSENs? 04:20:19 *RAMSANs? 04:21:17 http://www.ramsan.com/ 04:22:04 Gregor: i think having written the code yourself AND having it be in a language you are familiar with improves code skimmability... 04:22:18 Yes. Yes they do. Wooh. 04:25:18 Having written the code yourself in a language you are unfamiliar with, however. 04:25:39 Improves only "skimbility". 04:26:24 yes 04:26:33 -!- awilcox has joined. 04:26:34 I would just like to take this moment to declare that Daft Punk is awesome. 04:26:41 elliot, python is awesome 04:26:48 -!- awilcox has left (?). 04:26:54 o.O 04:26:58 wut 04:27:01 but i was meaning that AND capitalized as an "and/or" without taking up the space that takes 04:27:40 (a ^ b ) v (a v b) = a v b 04:27:43 So just say or 04:31:50 :( 04:31:57 configure: error: 'i686-pc-cygwin' is not (yet) supported by pcc. 04:53:05 sgeo: in english, "or" = ^, and "and/or" = || 04:53:38 I thought "or" = xor? 04:53:56 Wait, that's what you said, isn't it? 04:55:17 Sgeo: In English, "or" = xor *only* if talking to someone who is actually familiar with xor. :P 05:03:58 Ok, what's the difference between ^ and ... wait 05:04:04 Who uses "or" to mean AND? 05:04:10 * Sgeo headaches 05:04:25 I think I'm failing to understand some notation or another 05:06:09 Um yes, I hadn't noticed until now, but... 05:06:26 In C, ^ is xor, while in logic, ^ is and 05:07:32 ^ should be banned from everything. It's used too much. 05:08:32 No, ^ should be the only unbanned ASCII character. 05:08:46 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 05:11:01 in Spiral ^ means peek/copy 05:22:04 -!- catseye has quit (Ping timeout: 241 seconds). 05:24:03 -!- catseye has joined. 05:33:44 -!- bsmntbombdood_ has changed nick to bsmntbombdood. 05:39:10 -!- augur has joined. 05:46:32 -!- oerjan has quit (Quit: Good night). 06:10:15 -!- Harpyon has joined. 06:12:01 http://codu.org/webplat/ SO MUCH STUPID 06:12:07 (It's also quite broken right now, but *eh*) 06:14:15 -!- Harpyon has quit (Client Quit). 06:15:58 what's the point of this? 06:16:11 epic win 06:16:17 Epic wiiiiin! 06:16:25 Run that bookmarklet on a page OTHER than that one :P 06:16:31 yeah I did 06:16:37 Use WASD 06:16:51 it shows a happy face and scrolls it down past the bottom of the page forever 06:16:58 -!- sftp has quit (Remote host closed the connection). 06:17:09 You loaded it on a page with no platforms it could use :P 06:17:31 (Also, I've only tested this on Firefox) 06:17:47 he needs a hot air balloon 06:17:51 lol 06:18:20 This is just a proof of concept. There are lots of bugs to iron out, then I intend to add powerups when you find certain elements, etc. 06:18:24 And, y'know, a goal >_> 06:18:52 so I need a page with a lot of images 06:18:55 I guess? 06:19:41 Text works too. Wikipedia articles work. 06:20:03 google doesn't 06:20:44 lol 06:20:47 The Google homepage doesn't, but search results do. 06:20:48 wikipedia is good 06:21:17 search results didn't work for me, weird 06:21:21 Browser? 06:21:34 It does not work nicely in Chrome on Wikipedia 06:21:44 It does not work nicely in Chrome /anywhere/ 06:21:46 firefox 3.6.10 06:22:10 Gregor, I'm resuming the war against you 06:22:23 Sweet. 06:22:47 it's awkward that you can't move upwards out of a box 06:23:06 need to be able to break boxes somehow so you can go back up :) 06:23:20 This is a proof of concept, people :P 06:23:29 The biggest issue right now is that you can clip through some boxes for no obvious reason. 06:23:31 Gregor: amstan suggests jquery 06:23:31 I have to fix that first. 06:23:31 -!- augur has quit (Remote host closed the connection). 06:23:49 Bleh@jquery 06:23:59 I guess this is EXACTLY the kind of thing you use jquery for X-P 06:24:06 (I never use libraries) 06:24:16 01:23 < ebering> aspergers, the game 06:24:18 lol 06:24:31 What channel is this that you're advertising my stuff on X-P 06:25:02 the UW computer science club 06:25:34 You're at UW? 06:25:42 pink ties forever! 06:25:51 Tell Brian Burg "I don't know you!" Unless you know him, in which case just say hi or something. 06:26:03 OH 06:26:04 I don't know him 06:26:05 That UW X-D 06:26:11 Waterloo, not Washington :P 06:26:39 I broke I phone .. can't get nandroid to work 06:31:28 calamari: Android probably works better than Nandroid 06:31:51 Nandroid is a thingy for Android 06:31:55 Backups, I think 06:32:20 yeah 06:32:35 trying to recover via fastboot instead of the recovery image, hopefully that'll work 06:32:35 -!- pikhq has quit (Ping timeout: 255 seconds). 06:33:51 -!- pikhq has joined. 06:37:23 Gregor: cool btw :) 06:38:01 AND PIKHQ MISSED IT 06:38:03 HA HA HA 06:38:06 let me know when you've replaced the smiley face with a paypal donation link, then I'll know it's done :) 06:38:16 Bahaha 06:39:43 can it tell that it's gotten to the bottom of the page? 06:39:51 Vorpal: pikhq: you also (but for different reasons) <-- morning 06:40:14 I mean of the logical page not the visible page 06:42:22 yay, phone is restored :) 06:42:25 bbl. university 06:42:58 calamari: Not at present, no :P 06:43:07 You just fall for all eternity. 06:43:28 well yeah, but is it possible to tell 06:45:19 * calamari steals Gregors secrets 06:47:06 Oh, sure, it's possible to tell. 06:47:07 I just don't. 06:47:17 Gimme a break, I wrote this in like half an hour X-D 06:47:58 no I was genuinely asking, I know nothing about js 06:48:05 i wrote jQuery in half an hour 06:48:15 js is easy 06:49:08 s/asy/vil/ 06:50:46 It chorks 06:54:57 it kinda works on the phone but typing forces open a search window so you can't move 06:55:22 Gregor: webplat is neat, but i don't like how it prevents me from jumping UP through bounding boxes unless i've dropped DOWN through them first, and then i can't land on them again! 06:55:48 quintopia: Proof of concept! ;) 06:55:57 better would be: can always jump up through them, can always land on them, can always drop through them 06:56:01 yes i know 06:56:07 perfect time for criticism! 06:56:23 inputs for future versions! 06:56:43 Gregor's secrets: http://codu.org/webplat/webplat.js 06:56:59 I wouldn't advise that you read that :P 06:57:07 i can find bookmarklet source myself kthx 06:58:18 this could be a game... 06:59:06 It's intended to be a game (eventually) 06:59:22 imagine two players hopping around on the same page....then creating portals to other pages....and following each other around.... 06:59:26 what is the goal in the game 07:00:08 quintopia: to skim credit card numbers and passwords 07:00:10 to be a proof of concept 07:00:26 ;P 07:00:28 success! 07:03:52 -!- evincar has joined. 07:04:26 Hello world. 07:05:17 world: Do I know you? 07:07:08 Only as a doll knows the paint of its eye. 07:08:04 Jeez, now I want to make some poetic language. 07:08:52 Don't do that, you'll get hives 07:08:57 Or worse, a beret 07:09:34 also, it's been done, several times. 07:10:07 yeah never do anything that has been done. breathing is RIGHT OUT 07:10:18 Actually, I was leaning toward something more along the lines of why the lucky stiff's language project... 07:10:26 ...but that's been done, or at least it was being done. 07:10:31 I forget the name. Potion? 07:10:46 -!- calamari- has joined. 07:10:48 Yeah, Potion. 07:10:55 originality is key, here 07:11:55 Originality is hard to come by in a field that's about originality. Working on mainstream languages, you're not expected to be particularly innovative. 07:12:14 -!- Sgeo has quit (Read error: Connection reset by peer). 07:12:19 true that 07:12:23 cool neighbor lost power.. too bad for the moon, would be a good time for the telescope 07:12:29 neighborhood rather 07:13:15 jQuery might be unacceptably slow. 07:13:40 Speed varies inversely as productivity? 07:14:01 -!- calamari has quit (Ping timeout: 272 seconds). 07:15:05 bbl.. Gregor tftf 07:15:21 I don't know what "tftf" means 07:15:28 -!- calamari- has quit (Client Quit). 07:15:47 gregorrrrrrrrrrrrrr 07:15:54 what is the goal of the planned game 07:17:31 I'm not sure yet X-D 07:17:57 * olsner has been reading the long cvs migration thread on the netbsd lists 07:18:00 unfortunately they seem to be getting nowhere 07:18:21 "tftf" means "true, false, true, false". he was trying to break your brain by toggling it too quickly. 07:18:24 good night. 07:19:40 jQuery does everything better than I did, but it's SO slow :( 07:19:46 Load time is now almost insufferable ... 07:22:42 Gregor: Using repeated lookups? Using plugins that are known to be performance-affecting? Using effects that create perceived lag rather than actual lag? 07:23:50 -!- catseye has quit (Ping timeout: 264 seconds). 07:23:53 evincar: No plugins, there's a repeated traversal over the entire DOM that's more-or-less unavoidable because I need to change the .display type of nearly every element in the DOM before I can start caching their locations. But all the traversals are the same as I was doing raw, just now with jQuery. Plus now they work better :P 07:24:46 so you've just migrated that POC to jQuery? 07:24:52 and it's slow? 07:25:24 Only the initial load. 07:25:52 so it would rule out a game which involves tunneling to other pages quickly and repeatedly 07:26:06 After that it's fast. 07:26:25 how do you decide which elements to make into platforms? 07:26:56 so far, only H1s, Ps, a subset of IMGs, and As have been platforms 07:27:30 It's a pretty weird heuristic >_> 07:28:05 I played around with it and found something I liked. 07:28:05 Just a sec ... 07:28:24 Right, if a node has non-whitespace text children, OR it has no non-inline children at all, then it's a platform. 07:28:52 ah 07:28:56 seems good 07:31:04 -!- GreaseMonkey has quit (Remote host closed the connection). 07:31:06 another interface suggestion: make the first jumps smaller, but allow for in-air double and triple jumps that double and double again jump height resp. 07:31:24 That would be a powerup. 07:31:30 Assuming I can decide how you find powerups. 07:35:06 jQuery version now online. 07:35:53 erm 07:35:56 :/ 07:36:02 ? 07:36:03 no worky? 07:36:15 trying to use here: http://eventactions.com/events-calendar/eastern-time/emory-college 07:36:15 Hmmm? 07:36:27 with same bookmark as before 07:37:05 Same bookmarklet as before won't work. 07:37:18 well, then 07:37:23 same link above at least? 07:37:26 I'm temporarily too lazy to actually put jQuery and my stuff in one file, so the bookmarklet now loads both :P 07:38:00 Yeah 07:38:19 Now, back to making it fast ... 07:38:35 wow that does take forever to load 07:39:43 What did you load it on? 07:42:35 ssame link above 07:42:38 -!- MigoMipo has joined. 07:43:27 Part of it is just Codu being slow, btw :P 07:47:17 By the way, if somebody wants to draw some transparent pngs in various relevant positions, I'd be ultra-thrilled to add them ;) 07:47:20 echo a loading... message to the status bar once the script has begun loading? 07:47:32 that ould give a hint how much is jquery and how much is codu 07:48:24 and what positions would you consider relevant? 07:49:36 Standing guy, running guy, jumping guy. 07:52:08 -!- MigoMipo has quit (Remote host closed the connection). 07:53:09 -!- tombom has joined. 07:55:15 I still have some issues with bounds >_> 07:56:19 how about gifs so the running guy can be properly animated? 07:56:32 * quintopia curses the lack of mng support in the world 07:57:07 Bleh @ gifs *sobblecopter* 07:57:14 How about animated gifs only for the parts that need animation :P 07:57:20 Wooh, fixed bounds issue. 07:58:58 well, what would be best is if all animation happened within your code, since you're gonna be swapping out graphics anyway 07:59:31 I could do that, not sure how fast it'd be. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:00:10 that way, position changes can be timed with posture changes appropriately 08:00:20 (yeah games are complicated that way :/) 08:01:46 there was this really neat windows program i used once way back in the day for creating cursors. 08:02:26 it gave you a super-large grid to fill in the pixels on, like 500x zoom by default 08:02:53 what's the closest thing to that for drawing sprites in linux? 08:03:04 I cannot for the life of me figure out why it thinks it needs to scroll up here ... 08:03:09 Not a clue. 08:07:02 Fixed once again by jQuery :P 08:08:21 Damn it jQuery, stop solving all my problems! 08:09:42 -!- evincar has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]). 08:09:51 impossible 08:09:56 jQuery solves every problem 08:12:13 Hmmm, this KINDA works on Chrome. Better than it did before. 08:13:42 http://www.doxdesk.com/img/updates/20091116-so-large.gif 08:13:51 proof that jquery solves every problem 08:15:07 Seems like approximate rate IPv4 addresses are allocated is about 10 per second... 08:16:31 we still have enough to last for YEARS 08:16:36 ...right? 08:16:53 Well, two at most... 08:18:37 And IANA pool is likely going to run dry on early March to early June next year... 08:22:12 nothing like running out of numbers to accelerate the move to v6 08:23:07 -!- atrapado has joined. 08:25:39 Except that it could cause IPv6 migration plans that look like: http://www.codinghorror.com/blog/2009/12/our-disaster-recovery-plan.png 08:35:56 gregor: http://www.aseprite.org/ 08:36:56 ilari: y2k didn't look like that. it got fixed before it broke anything. why is that? 08:38:52 Self-dispelling prophecy... :-) 08:42:09 so if you're predicting ipv4 disaster...does that mean nothing interesting will happen? 08:44:28 quintopia: The reason I wanted somebody else to do it for me is that I'm sprite-incompetent :P 08:44:53 I was saying about Y2K. Don't know if the drumming about IPv4 address exhaustion is hard enough. 08:48:31 Similar counter of IPv6 would likely be fun to see. Much bigger numbers and much faster countdown. 08:55:23 Fun fact: Currently only 1/8th of IPv6 address space is for unicast use. Most of IPv6 addresses are of undefined kind. 09:01:31 -!- tombom has quit (Ping timeout: 240 seconds). 09:19:53 -!- catseye has joined. 09:20:44 -!- evincar has joined. 09:24:29 -!- catseye has quit (Ping timeout: 276 seconds). 09:29:38 -!- Rugxulo has joined. 09:40:04 anybody here mess with FALSE? 09:40:20 gregor: http://filebin.ca/qxuvnd 09:40:47 i'm not that competent either, but this'll do for now 09:41:07 (this is only left-to-right running) 09:41:40 mirror for right-to-left, stop animating for jump? 09:50:28 also, frames 0 and 2 should be 115ms, and frames 1 and 3 100ms for smooth running 10:10:36 -!- evincar has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]). 10:12:18 -!- dbc has quit (Ping timeout: 250 seconds). 10:28:45 -!- dbc has joined. 10:50:18 -!- pikhq has quit (*.net *.split). 10:51:09 -!- pikhq has joined. 10:56:16 -!- tombom has joined. 10:56:16 -!- tombom has quit (Changing host). 10:56:16 -!- tombom has joined. 11:03:36 -!- Rugxulo has left (?). 12:00:40 -!- ais523 has joined. 12:07:08 -!- ais523 has quit (Read error: Operation timed out). 12:15:38 argh, I hate cards with low quality magnetic strips. 12:27:42 I hate how our cluster has for some unfathomable reason stopped sending out the "job started" / "job ended" notification emails. They were useful in noticing when things were finished without extra polling, since I keep looking at emails anyway. 12:45:14 -!- tombom has quit (Quit: Leaving). 12:59:50 fizzie, heh 13:00:12 fizzie, write a shell script to poll it? 13:02:08 And send my own emails? Well, I guess I *could*, but since it already has the feature. 13:10:55 It's scheduled with SLURM, which has all kinds of nifty little tricks, like a dependency thing so that you can say "run D only after A, B and C have finished" (s/and/or/ or s/finished/started/ or other options also available). 13:11:04 Come to think of it, I wonder what it would do with circular dependencies. 13:12:41 Probably they'd just boringly stay in the queue until canceled. 13:17:17 fizzie, you could make the script notify you in some other way. Anyway if you make that poll using cron it would be trivial to send email: just echo what you want to send 13:17:22 and the cron daemon will send it 13:19:08 It's still a fact that the queue scheduler system should take care of that sort of stuff. I don't think we're supposed to be setting up all kinds of silly little cron jobs on the cluster front-end node, anyway. 13:21:27 Heh, decided to go check out the fancy Ganglia web-frontend of the cluster status; normally it shows all kinds of CPU/mem usage plots and other things like that you'd expect, but now it just says: "Cannot find any metrics for selected cluster "Triton", exiting. Check ganglia XML tree (telnet 127.0.0.1 8652)" 13:21:33 oh god, arch is switching /usr/bin/python to mean python3 13:21:50 this will cause havoc for sure 13:21:53 I'm pretty sure localhost:8652 won't help me here. 13:22:17 fizzie, maybe report it to whoever is the sysadmin? 13:23:22 If they don't notice it, sure. Though I'm not quite sure who's responsible. We have a department-internal issue tracker, but that thing is of wider scope. 13:38:58 -!- elliott has joined. 13:39:26 -!- oerjan has joined. 13:42:43 elliott, arch linux went insane, /usr/bin/python is now python3... 13:42:50 Vorpal: Hahahaha 13:42:53 Vorpal: Is that a default? 13:42:59 Oh god it is 13:43:06 "any program requiring 2.x needs to point to /usr/bin/python2 instead" 13:43:07 tl;dr 13:43:13 any python program ever requires modification tor un 13:43:14 *to run 13:43:26 Vorpal: but don't you see?! Arch is for people on the EDGE, we don't care if shit works as long as it's NEW! 13:43:38 elliott, yeah, wtf. You go make kitten usable right now :P 13:43:57 Vorpal: Gimme a grant and it'll be 1.0 in weeks 13:43:59 elliott, I'm not going to switch to a non-rolling-release. And I'm not going to go gentoo 13:44:18 elliott, any suggestions? 13:44:33 Vorpal: I... well, no, that's sort of why I'm doing Kitten. pikhq also concluded that there's basically nothing. 13:44:38 Vorpal: There are very few rolling release distros. 13:46:55 elliott, requirements for me: rolling release, reasonably up-to-date (so nothing as outdated as, say, debian stable), allows easy tweaking of system without it trying to get in the way, not source-based. And hm probably linux kernel too. Well freebsd can support nvidia drivers iirc so... 13:47:21 Vorpal: Kitten is NetBSD-based, as I said. 13:47:54 elliott, yeah, well, I'll just port it to kitten/linux. I mean, it balances out against debian/bsd so that's okay 13:47:58 Vorpal: I, uh, try nouveau. Nvidia's driver *is* an OS-independent binary blob and a kernel interface, so you could port it, but nobody really has recently 13:48:24 elliott, nouveau does not yet 3D, which I um, need 13:48:41 at least it didn't when I checked about a month ago 13:48:48 "Any 3D functionality that might exist is still unsupported. Do not ask for instructions to try it. But you can read GalliumHowto in case you are brave enough." 13:48:56 sounds like it exists but they don't like to admit it 13:48:57 :P 13:49:01 maybe not though 13:49:35 elliott, well, I looked at that, and at best it was partial for some chipsets, nothing at all for my chipset 13:49:41 * Vorpal prepares to reboot for kernel upgrade 13:49:52 Vorpal: isn't there another nvidia driver? :/ 13:50:12 Traceback (most recent call last): 13:50:12 File "/usr/bin/denyhosts.py", line 5, in 13:50:12 import DenyHosts.python2_version 13:50:12 ImportError: No module named python2_version 13:50:14 right 13:50:20 they even fail to package stuff 13:50:57 lawl 13:52:10 removing the last bit on that line worked, as in, plain old "import DenyHosts" 13:52:17 will file a bug after I rebooted for kernel upgrade 13:53:53 Vorpal: I hereby file a bug in your system, you have proprietary components that refuse to work with software. I suggest replacing the broken element with an Intel card. :-P 13:54:08 Or at least ATI, since they have good open-source drivers. 13:54:20 (More gooderer, at least.) 13:54:32 elliott, well yeah, you find me one that is AGP and as fast 13:54:38 Vorpal: AGP? 13:54:41 I file a bug in your system. 13:54:42 It uses AGP. 13:54:42 elliott, yes my system is outdated 13:54:52 I suggest you apply a sledgehammer fix. 13:54:54 elliott, I do not have the money to upgrade currently. 13:55:11 elliott, besides: don't fix what isn't broken 13:55:16 as long as it works... 13:55:20 Totally is broken, it can't run NetBSD. 13:55:35 Well, quickly. 13:56:08 Vorpal: But fine, finding an AGP card now. >:) 13:56:30 elliott, old card was: geforce 7600 GS 13:56:34 need to be at least as good 13:57:22 Vorpal: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600007850&IsNodeId=1&name=AGP%204X/8X 13:57:27 Vorpal: Pick a Radeon :P 13:58:28 elliott, will recheck that when I get back after rebooting, I don't think it works in w3m and atm the nvidia userland/kernel module are out of sync, so can't start X. 13:58:30 brb 14:01:16 -!- ais523 has joined. 14:02:08 hi ais523 14:02:20 hi 14:02:40 gah, why do the seminars on Tuesday always have to be coincidentally relevant to what I was doing the week before? 14:03:29 -!- Vorpal has quit (Ping timeout: 255 seconds). 14:03:39 ais523: You realise now that if you work on a cure to cancer this week, next Tuesday... 14:03:47 Use your power for good! 14:04:02 elliott: well, I do multiple things in a week, it's only ever relevant to one of them, usually 14:04:12 ais523: Dedicate an entire week to curing cancer. 14:05:46 hmm, prediction for IPv4 exhaustion's around 9 or 10 months 14:05:54 that could be... interesting... when it happens 14:06:15 ais523: omg #esoteric totally needs to host an IPv4 Exhaustion party 14:06:17 i'll bring popcorn 14:06:28 someone bring lots of displays to put graphs and visualisations on 14:06:36 that would be great 14:07:21 -!- Vorpal has joined. 14:07:32 * elliott wonders what happened to IPv6 14:07:34 *IPv5 14:08:16 ah, apparently the protocol number was used by something else 14:08:19 IPv5 is an unrelated protocol, IIRC 14:08:20 or something 14:08:46 what i don't like about ipv6 is how willy-nilly people are giving out utterly gigantic ipv6 allocations 14:08:49 sure we have an awful lot of space 14:08:54 but didn't you learn anything from last time?! 14:09:12 I thought IPv6 was designed so that everyone, no matter how major or minor, got a /64? 14:09:16 "Version number 5 was used by the Internet Stream Protocol, an experimental streaming protocol." right 14:09:25 ais523: there are larger allocations 14:09:29 but i think /64 is the minimum or something 14:09:42 what plausible reason would you need for something larger than a /64? 14:09:51 ais523: allocations range from, like, ipv4 address space ("tiny!") to ... an utterly ridiculous number of ipv4 address spacse 14:09:52 /64 is so many IPs, I can't even visualise the number 14:09:54 *spaces 14:09:55 how big is /64? 14:10:01 i suck at /foo 14:10:04 same as a 64-bit integer 14:10:04 -!- ais523 has quit (Read error: Connection reset by peer). 14:10:10 RIP ais523 14:10:31 -!- ais523 has joined. 14:10:36 wb ais523 14:10:37 [14:09] i suck at /foo 14:10:38 anyway 14:10:39 [14:09] same as a 64-bit integer 14:10:40 [14:09] around 16*10^18 14:10:41 right 14:10:44 IRC just disconnected for no obvious reason 14:10:46 saw the first one of those 14:10:54 ais523: you can definitely get /128 iirc 14:10:58 * coppro starts an election campaign 14:10:58 i think /32 is the smallest 14:11:00 ipv4 address space 14:11:08 no, /32 is the largest 14:11:11 the values work in reverse 14:11:14 oh right 14:11:15 fff 14:11:16 you know what i meant 14:11:17 *mean 14:11:20 an IPv4 address space would be a /96 14:11:21 32-bit is the smallest 14:11:33 ask Vorpal :P 14:11:42 which is still large; not quite incomprehensibly large, though, just "about half the population of the world" 14:11:44 there, back 14:12:10 ais523: there aren't 8 billion people alive! 14:12:22 elliott: I was approximating 14:12:36 ais523: strange approximation :) 14:12:44 closer to two-thirds 14:12:47 Isn't current IANA IPv6 unicast address pool at something like 99%? 14:12:56 Ilari: of course 14:13:00 Vorpal: what's the smallest/biggest ipv6 allocations? 14:13:03 and it will be for a very long time 14:13:04 smallest is /96 right? 14:13:05 [15:08:22] what i don't like about ipv6 is how willy-nilly people are giving out utterly gigantic ipv6 allocations <-- they are trying to make route aggregation more feasible 14:13:06 6,697,254,041 in 2008, according to Google 14:13:11 I think it's over 7 billion by now 14:13:21 elliott, well, any size is possible 14:13:21 but from whom? 14:13:28 Vorpal: you know what i meant. 14:13:29 * ais523 thinks that value is suspiciously precise 14:13:31 most common allocations 14:13:43 ais523: it's the sum of all the estimated populations of countries, i think 14:13:50 Alpha says 6.79 billion as of last year 14:13:59 ais523: which are usually census results or records or whatever + some fudge factor for everyone they missed 14:14:03 i think 14:14:17 by current trends we have about 2 more years to hit 7 billion 14:14:31 elliott, well, I have a /48 from sixxs, and I use a /64 of that for my LAN. You don't want less than /64 really, the stateless auto-configuration is based on filling in the mac for the last bit of the ip so... 14:14:38 err, less = smaller 14:14:44 ais523: /48. 14:14:57 elliott, on the other hand I heard of cases for /125 and other such 14:15:03 SO YEAH people are pretty much throwing away ipv6 address space because there's a lot of it 14:15:10 a /125 would contain just 6 addressible addresses? 14:15:23 For point-to-point tunnels I've seen /126's being used. 14:15:25 yeah, typically end user points get /64 iirc 14:15:26 ais523, yes, it isn't commonly used 14:15:44 /46. really. 14:15:49 elliott, yeah, a /62 or would be reasonable for end-user. That would allow a few subnets if you really wanted it. 14:16:14 IPv6 design is fundamentally "make it bloody impossible to screw this up 14:16:18 (and they will screw this up) 14:16:32 coppro: giving anyone a /46 is just unbelievable... 14:16:38 elliott, /48 I said 14:16:40 this is a *home* user or a *single* corporation 14:16:42 48, right 14:16:42 typo 14:16:49 there is *no way* they will use all those up! 14:16:55 if they do 14:16:56 elliott, yes indeed it is 14:17:00 give them some more! 14:17:00 a large corporation might want a /48 14:17:08 but most would not 14:17:09 so why not just make it designed for small allocations 14:17:10 and stack them? 14:17:13 coppro, only because of routing 14:17:16 coppro: you do realise /48 is bigger than ipv4? 14:17:20 elliott: yes 14:17:28 but the ability to conveniently divide it up is key 14:17:29 elliott, a /64 is bigger than ipv4 too 14:17:42 a corporation might allocate a /40 to each country it operates in, a /32 to each region, etc. 14:17:51 but we're talking multinationals here 14:17:53 coppro, indeed, please explain about /64 and ipv6 stateless autoconfigure to elliott. 14:18:01 "All Global Unicast addresses other than those that start with binary 000 have a 64-bit interface ID field (i.e., n + m = 64) [read: it shouldn't really be subnetted to anything below /64], formatted as described in Section 2.5.1." 14:18:03 coppro, a /32? what? 14:18:09 coppro, you are counting from the wrong end 14:18:13 Vorpal: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600007850%20600030349&IsNodeId=1&name=ATI AGP ATI cards 14:18:20 *AGP ATI cards (just highlighting it :P) 14:18:26 Vorpal: err, sorry 14:18:30 yeah, I am 14:18:38 coppro, indeed, please explain about /64 and ipv6 stateless autoconfigure to elliott. 14:18:46 how do you know i'm not criticising the designs that lead to such large alloactions too? 14:18:48 esr is, incidentally, working on an RFC for encoding INTERCAL programs in IPv6 addresses 14:18:50 *allocations 14:18:52 which is arguably even more wasteful 14:18:58 ais523: i suggest we shoot him to death. 14:19:00 esr? really? 14:19:08 coppro: esr is the authro of C-INTERCAL... 14:19:09 *author 14:19:11 oh 14:19:14 elliott, anyway for a LAN you want a /64. Due to stateless autoconfig (which is nice, don't deny it, dhcp is a sucky mess) that is reasonable. 14:19:19 elliott: I don't 14:19:22 demonstrating the height of his coding skill 14:19:31 sorry, wait, no, that's FETCHMAIL 14:19:34 ais523: but... but 14:19:39 for release roughly 5.5 months from now 14:20:08 *roughly* 5.5 months? 14:20:26 "Roughly 4 hours, 57 minutes and 29.32 seconds from now, I will..." 14:20:26 For the record, even the "not really a very big" ISP I use at home has a /32: 2001:1bc8::/32, FI-NBLNETWORKS-20040525. 14:20:26 elliott, no, roughly 5.5018375 months 14:20:49 elliott, actually "5 and a half month" isn't that extreme 14:20:52 waiting for April Fool's would be ridiculous, C-INTERCAL's pretty much ready for release in its current state 14:20:58 "5.5" though :P 14:21:07 elliott, it is shorter to write :P 14:21:07 ais523: is he just going to release it with all your modifications and claim it as his own? 14:21:12 although I'd appreciate more testing on platforms I don't have access too, the fact that it's running fine in DOSBox indicates that it is at least not completely nonportable 14:21:14 elliott: I doubt it 14:21:23 ais523: maybe he'll ignore all your modifications instead 14:21:28 elliott: of course 14:21:28 I doubt that too 14:21:31 ais523, what happened to my patches? 14:21:34 he's likely to just call me comaintainer 14:21:37 an RFC like that can only be written on one day 14:21:41 Vorpal: the Mac Classic patches, I still haven't really looked at 14:21:45 the other patches are in there, though 14:21:54 ais523, so the generic ones got integrated? Nice. 14:22:11 Vorpal: I mean, the ones in the Mac Classic bundle, even the generic ones, I haven't really looked at 14:22:22 the patches you submitted via unrelated means are in there, and have been for a while 14:22:54 ais523: are you going to stop releasing your own versions? 14:22:56 please don't 14:23:02 i couldn't stomach downloading esr software 14:23:03 :) 14:23:07 fizzie, a /32? ... okay that is just silly. 14:23:10 ais523: call it the -mm tree 14:23:26 elliott, why -mm, his name is nothing like that 14:23:26 elliott: I don't really get the concept of revulsion at software based on who technically owns it 14:23:47 Vorpal: http://en.wikipedia.org/wiki/Mm_tree 14:23:51 ais523, technically a lot of people own different parts of it 14:23:53 elliott: The idea of IPv6 is to have *so incredibly many more* addresses that they can do wasteful things with the allocations just to make routing easier. 14:23:53 ais523: it was more a joke 14:24:03 ais523: but, your releases are more likely to be frequent and debugged than esr's, I suspect 14:24:03 Vorpal: Well, you know, it's an ISP. They're quite "high up" when it comes to routing. 14:24:06 elliott, I know what the -mm tree is... 14:24:06 even if it's mostly your changes 14:24:09 in the real one 14:24:13 Vorpal: he isn't called Mandrew Morton 14:24:17 "Historically, the -mm tree focused on new developments for the memory management part of the kernel (mm)." 14:24:26 -- but it extended to general-everything 14:24:29 elliott, well, -mm still makes no sense for ais523 here 14:24:32 and so it ceased being an acronym, really 14:24:37 Vorpal: "experimental development tree" 14:24:49 elliott: actually, ESR's big new feature for the upcoming release is a testsuite 14:25:18 although he just used existing programs, and had to ask me what many of them did 14:26:01 then I wrote a fuzztest for the optimiser 14:26:04 which actually caught quite a few bugs 14:27:14 ais523: he can't actually code intercal, can he? 14:27:33 so, will you still release your EXPERIMENTAL AIS523 BRANCH or will it all be through esr now? 14:27:52 ais523, where is the darcs repo for it? 14:28:15 (or other repo for that matter) 14:28:39 Vorpal: see, esr actually follows the cathedral model (afaik) 14:28:44 OK, webplat now has proper fallthrough and POWAHJUMP 14:28:49 elliott, that would be extremely ironic 14:28:50 Gregor: You're... making it? 14:28:55 Vorpal: indeed 14:28:56 elliott: I'M AWESOME 14:28:59 http://codu.org/webplat/ 14:29:11 IANA has seemingly allocated 5.116 /12s of IPv6 unicast address space. There are 512 /12s total. So IANA IPv6 pool is at 99%. 14:29:12 It... I see nothing. 14:29:18 Oh isee 14:29:20 it's a bookmark 14:29:21 *i see 14:29:55 Vorpal: http://gitorious.org/intercal/intercal 14:30:02 Gregor: I can't fall off the page, dude. Force html to full height/width and migrate margins to body or whatever 14:30:03 = git://gitorious.org/intercal/intercal.git 14:30:05 So I can fall properly :P 14:30:10 And, uh, body full height urgh i don't know 14:30:17 elliott: You CAN fall off the page. 14:30:24 Gregor: Not off /webplat/ you can't. 14:30:25 You can drop through elements. 14:30:29 Yes, off /webplat/ you can. 14:30:33 Because you can drop through elements. 14:30:34 Gregor: I triwed. 14:30:35 *tried. 14:30:41 I stop a bit below the last paragraph. 14:30:49 Ohohoh 14:30:54 Because of heights. 14:30:55 Frowny face means you fell off :P 14:31:08 Gregor: You... suck 14:31:28 HEY MAN LOOK AT THIS THING I MADE IN ABOUT TWO HOURS THAT ISN'T ENTIRELY PERFECT 14:31:42 Gregor, 2 hours? that long 14:31:44 huh 14:31:57 ... 14:32:02 Vorpal has never coded anything. 14:32:05 Gregor, I would expect full featured 3D by then! 14:32:11 Gregor, and custom music and so on 14:32:23 elliott, you fail at humour 14:32:25 Gregor: Plz add a textbox with the link for those of us who don't have bookmark bars :P 14:32:36 Vorpal: 14:32:38 Gregor, 2 hours? that long 14:32:39 huh 14:32:40 Not funny, just stupid. 14:32:43 Big difference. 14:32:47 -!- oerjan has quit (Quit: Grmbl). 14:32:49 elliott, you didn't include the next two lines 14:32:53 elliott: Right-click, copy link or whatever :P 14:32:57 Vorpal: 14:32:58 Gregor, 2 hours? that long 14:32:58 huh 14:32:58 ... 14:32:58 Vorpal has never coded anything. 14:33:02 Gregor: For some reason I can't copy that particular link. 14:33:02 elliott, not here 14:33:07 Vorpal: Here. 14:33:10 Gregor, 2 hours? that long 14:33:11 huh 14:33:11 Gregor, I would expect full featured 3D by then! 14:33:14 ... 14:33:14 Vorpal has never coded anything. 14:33:16 elliott, you fail at humour 14:33:19 OMG NETWORK LAG 14:33:20 HOW AMAZING 14:33:27 elliott, indeed 14:33:35 Gregor: Invisible elements should so fall-through. 14:33:41 (Like the ones on google.co.uk) 14:33:49 That's the idea. 14:33:56 However, it's not always easy to tell whether they're visible or not. 14:34:31 Gregor: Are you meant to be able to double-jump? 14:34:49 Yes. 14:34:58 Double-jump is also power-jump, it lets you jump up through certain elements. 14:35:01 is this turning web pages into platform games? 14:35:10 that's kind-of nifty 14:35:11 ais523: Eventually :P 14:35:18 Gregor: I... it freezes reddit. 14:35:26 Holy shit horizontal scrollbar what 14:35:32 elliott, ? 14:35:41 See above 14:35:45 God damn it elliott, stop blaming me for the fact that this shit is super-complicated to do :P 14:35:52 I'm not :P 14:35:56 But you totally need to be perfect. 14:36:05 Vorpal: what do you think of the repo? 14:36:05 Gregor, why does it show "h1" in the upper corner for

but not, say, "p" or such for such blocks 14:36:08 Gregor: Okay on reddit.com you definitely start out dead. 14:36:25 Vorpal: It shows that for everything you touch. 14:36:28 Vorpal: what do you think of the repo? ;; subtle, subtle. 14:36:32 Vorpal: Including "P" and whatnot. If you don't see it, your browser asplode? 14:36:33 ais523, well, it is git://, would prefer something I could look at in the browser 14:36:49 Vorpal: as in http://gitorious.org/intercal/intercal/trees/master 14:36:50 Gregor, hm 14:36:53 ? 14:36:55 I'm not sure what you mean 14:37:01 ais523, ah yeah that is what i meant 14:37:03 (? on a separate line to stop it ending up as part of the URL) 14:37:09 (really, I should have used angle brackets) 14:37:16 Gregor: Slashdot keyboard shortcuts clash :( 14:37:23 :P 14:37:35 ais523, well what I think of the repo: "it's a repo, yeah" 14:37:46 that's... quite a thought 14:38:10 Gregor: Somehow loading it twice seems to help. 14:38:13 Sometimes. 14:38:22 ais523, well, what am I supposed to think of a repo. It is just a repo, with a .git in it and so on! 14:38:23 elliott: Help ... what? 14:38:28 Gregor: Help it ... work. 14:38:34 Gregor: Sweet, it turns "free encyclopedia" into "freeencyclopedia" on Wikipedia's main page. 14:38:36 ais523, would have preferred darcs of course 14:38:42 Vorpal: normally, people would have some opinions on the contents 14:38:49 I thought that the Wikipedians had resolved the long-term dispute about that line with a neologism. 14:38:49 ais523, I haven't had time to look at it 14:38:53 Which is very them. 14:38:53 elliott: It actually works pretty well on Wikipedia entries too. 14:39:00 I'd have preferred darcs too really, but I don't have the experience in repo reconstruction that esr does 14:39:06 apparently, he's had to do it quite a lot before 14:39:15 ais523, I expect it contains c-intercal. 14:39:27 looks somewhat cleaned up though 14:39:31 and there's nothing wrong with git; it just makes things a bit more complex and frustrating than darcs does 14:39:33 ais523, did you drop the prebuilt thingies? 14:39:41 ais523, if so, you broke mac 14:39:43 it needs them 14:39:49 Vorpal: not from the distribution tarball 14:39:51 but they aren't in the repo 14:39:53 ais523, ah 14:40:21 ais523, I presume you tested with cfunge? I don't have time to right atm 14:40:36 might do it in the weekend 14:40:43 Vorpal: I did 14:41:41 http://gitorious.org/intercal/intercal/blobs/master/src/perpet.c <-- hm, "Copyright (C) 1996 Eric S. Raymond", shouldn't it include me and you as well? I'm no legal expert however 14:42:20 Vorpal: copyright notices in C-INTERCAL have always just stated the original author of the file 14:42:25 ah okay 14:42:27 I'm not entirely sure if that's technically correct or not, but it's what happened 14:42:31 (also, the original date too) 14:42:43 unravel.c, for instance, just has an Alex Smith copyright 14:42:49 arguably because nobody else understands it enough to patch it 14:45:12 ais523, what is that file for now again? 14:45:23 runtime support for threading 14:47:28 ah 14:48:15 ais523, which file had the path code? 14:48:24 "path code"? 14:48:27 -!- BeholdMyGlory has joined. 14:48:46 ais523, that one of my patches fixed, it was generating // and such in paths 14:48:57 oh, probably either perpet.c or uncommon.c 14:49:14 not perpet, and I can't find it in uncommon 14:49:26 maybe it was cesspool? 14:49:40 that seems a little implausible, cesspool mostly doesn't care about files 14:49:59 hm 14:50:08 elliott: as for authorship of the next release, the NEWS file is split up by who actually made the change 14:50:20 esr and Vorpal have about the same number of changes; Joris is just behind, I have considerably more 14:50:24 ais523: what i mean is -- will you still release tarballs via your own channels? 14:50:30 or will you only be a contributor to esr's releases? 14:50:33 I don't have any channels 14:50:38 ais523, I only had like a hanful of changes iirc? 14:50:39 ais523: Well, sure you do. 14:50:42 handful* 14:50:48 ais523: You release C-INTERCAL. 14:50:52 mostly, I just announced the new release on Usenet, emailed it to anyone who asked, then linked to it once someone put it online 14:51:03 it's an unusual way to do software releases 14:51:11 ais523: esr will obviously go by his own schedule and want to put his own changes in before release. 14:51:17 his own changes are in already 14:51:19 So will you keep releasing your "development" releases however? 14:51:25 ais523: How do you know he'll never make more? 14:51:29 elliott: I never did release development releases 14:51:36 ais523: I know that. 14:51:40 apart from that beta a while back, because I wanted something in time for april fool's 14:51:48 ais523: I meant that to keep the peace with esr they'd be called development releases. It was a joke. 14:51:49 Then there's APNIC policy that anyone that has IPv4 allocation from APNIC but no IPv6 addresses can get a /32 without any justification and For IPv4 allocation but no IPv6 addresses can get a /48 without any justification. 14:51:50 with the repo online, it seems implausible that an actual release is needed before it's finished 14:51:55 ais523, hm did you integrate my patch from the mac stuff to fix path code? 14:52:02 after all, it was generic 14:52:09 Vorpal: as I said, I haven't really looked at those patches 14:52:22 ais523: So the choice is between waiting for esr to decide this is a great time to release and fishing through the repo for a semi-stable version? 14:52:23 Woooooo. 14:52:24 ais523, a lot of them were generic and at least one or two fixed serious bugs 14:52:26 -!- catseye has joined. 14:52:37 elliott: well, the current repo version is pretty stable 14:52:50 The first is 79 228 162 514 264 337 593 543 950 336 addresses... 14:52:52 the only thing I really want to do before release is check the documentation for inaccuracies and typos 14:52:58 ais523: Now, sure... Wait until esr decides it needs a fun fun rewrite of several components. 14:53:04 he wouldn't dare 14:53:07 as in, code-wise 14:53:10 I doubt most people would 14:53:20 even I don't dare to try to rewrite anything in C-INTERCAL nowadays 14:53:23 ais523: Dude, esr is stupid enough to do anything. 14:53:26 He wrote *fetchmail*. 14:53:40 you seem to have a pretty low opinion... 14:53:57 I mean, you can be annoyed at someone, or not like what they're doing 14:53:58 Well, yes, I do :p 14:53:59 ais523, and include the generic fixes, after all, without one of them, the generated C code is technically not valid. 14:54:06 but you seem to have the sort of hatred of ESR that most people reserve for Microsoft 14:54:42 hah 14:55:24 ais523: His sociopolitical opinions are beyond idiotic and childish. He took the Jargon File, made it stupider, and then had it eclipse the original. He appointed himself leader of the Open Source Movement by parroting things that are either obvious or just wrong, and got immense fame and respect for it. 14:55:26 And he wrote fetchmail. 14:55:31 Oh, and he's a bigot. 14:55:53 So... idiotic, intolerant, self-promoting and fetchmail-writing. I mean, it does add up... 14:56:19 http://geekz.co.uk/lovesraymond/archive/show-them-the-code 14:56:23 (relevant comic) 14:56:30 from what I've learnt from working with him, his main issue is doing without things without thinking of the implications 14:56:39 like the botched attempt to simplify the build system 14:56:45 it did motivate me to simplify it properly, though 14:56:49 ais523: oh yeah, he also wrote that article about how you can channel gods to be great at sex. 14:56:52 although i can't find it now 14:57:05 that was... possibly the worst sequence of bytes i have ever experienced 14:57:11 Oh, and there is also AS number exhaustion... There is BGP extension for 32-bit AS numbers (standard BGP ASes are 16 bits)... 14:58:57 "doing without things without thinking of the implications"... isn't that "generalized fail"? 14:59:07 ais523: oh, he also flamed the Fedora dudes because he did something stupid and it broke his system, then pledged his allegiance to Linspire 14:59:09 then Ubuntu 14:59:23 people pledge allegiance to distros? 14:59:24 because it can't possibly be *my* fault, I'm the leader of the Open Source Movement! 14:59:26 does that even make sense? 14:59:27 his allegience counts for so much 14:59:27 ais523: esr does. 15:03:34 elliott: opinions: deal between Facebook and Bing. Good idea? 15:03:39 ais523: WHAT 15:03:43 this is news to me 15:03:45 really? 15:03:51 yes, which surprises me too 15:04:18 Google news gives loads of results 15:04:46 presumably, some more trustworthy than others 15:05:03 ais523: oh dear, bing results based on facebook connections? 15:05:10 good thing bing is *already* useless 15:05:22 elliott: I think the idea is that it's linked to the facebook "like" buttons 15:05:27 yeah nothx 15:05:42 so that if some of your friends "like" something, it's more likely to show up in your search results 15:05:49 this does seem a little ridiculous 15:05:56 it may simply be another search mode, dunno 15:05:58 article wasn't claer 15:05:59 *clear 15:06:33 (incidentally, what bing's advertised as doing is exactly what I don't want a search engine to do; the reason I block cookies from Google, and nowhere else, is that it tries to customize search results based on searches and I prefer search engines to act literalistically) 15:07:50 Let's invade the Linux Mark Institute! 15:09:21 ais523: the internet -- protecting you from the internet since about 1999 15:10:26 the internet doesn't really make much sense, but then neither does anything else in life 15:10:49 it's probably best to just assume it's a force of nature, and try to figure out how it behaves scientifically 15:12:49 the internet shouldn't work really, the fact that it does is impressive 15:13:07 and it's a wonderful autonomous mind, really, we just need to cut off those little bits of centralisation 15:13:11 (e.g. root servers) 15:13:41 hmm, is there anything that works both in practice and in theory? 15:13:46 most things seem to work only in one world or the other 15:13:58 ais523: logic? of course applying logic in practice is a bit difficult 15:14:08 ais523: anyway, if there's a disparity all it means is that your theory isn't good enough 15:14:14 if it can't explain why it works, it's not the right theory 15:14:17 if it can't explain why it doesn't work, it's not the right theory 15:14:24 meh, the real world doesn't really follow logical rules, as seen by my conversation with comex in ##nomic yesterday 15:14:24 ais523: logic? of course applying logic in practice is a bit difficult 15:14:26 well, in some situations 15:14:30 ais523: ? 15:14:44 I posted it to a-b so people not in the channel at the time coudl read it 15:14:46 *could read it 15:15:01 which thread? 15:15:07 ais523: i ...can't think of any, surprisingly 15:15:18 elliott: the one about the judgement to Bucky's CFJ 15:15:36 we were talking about logical paradoxes involving belief 15:15:38 ais523: have you seen esr's response to microsoft sending him a job offer, btw? 15:15:42 no, I haven't 15:15:47 also, Microsoft sent him a job offer? 15:15:50 ais523: imagine the most egotistical three-year-old you can 15:15:54 well, some microsoft recruiter 15:16:02 ais523: I can link it if you want 15:16:09 meh, imagining's probably more fun 15:16:20 ais523: 15:16:21 [[If you had bothered to do five seconds of background checking, you 15:16:21 might have discovered that I am the guy who responded to Craig 15:16:21 Mundie's "Who are you?" with "I'm your worst nightmare", and that I've 15:16:21 in fact been something pretty close to your company's worst nightmare 15:16:21 since about 1997.]] 15:16:21 and it's likely to be a famous enough incident that I could find it by myself 15:16:26 are you sure you can imagine things as crazy as that? 15:16:28 (it gets worse, btw) 15:16:31 oh, right 15:16:44 ah, you've seen it? 15:16:55 no, I was just replying to your "you can't possibly imagine it" comment 15:16:57 -!- tombom has joined. 15:17:03 ah 15:17:08 I've seen the Halloween documents, though 15:17:23 so Microsoft were at least aware of him, unless he edited them to put mentions of himself in 15:17:34 (now I've suggested that, it suddenly becomes vaguely plausible...!) 15:17:37 which are those again? 15:17:53 leaked Microsoft internal memos discussing open source and Linux, from ages ago 15:17:57 they were leaked to ESR and he published them 15:18:11 i think I may have seen them 15:18:20 with a bunch of comments added for legal reasons, in order to make it technically speaking journalism 15:18:22 ais523: boolean logic and logic gates come out *pretty* close, but only because you have things like coding theory to keep them away from the persistent misbehaving of the universe 15:18:26 ais523: heh 15:19:36 catseye: and I have enough training to know the differences 15:19:47 it's actually quite easy to obtain boolean values between true and false in practice 15:19:56 ais523: how do they behave? 15:19:59 sometimes it seems like half of electronic engineering is trying to stop that happening 15:20:01 (ok, stupidly general question, I know) 15:20:06 elliott: they generally drive your power consumption through the roof 15:20:20 and act like 0 or 1 depending on minor physical details, apart from that 15:20:30 sometimes they screw up timing, too 15:21:03 ais523: is bucky uorygl? 15:21:10 no 15:21:14 ok 15:21:15 uorygl = Warrigal, or Tanner L. Swett 15:21:20 yes, i know 15:21:24 he just changes his nick a lot 15:21:30 Bucky = Bucky, and uses the pseudonym John Smith on the lists 15:21:43 I loved it when everyone tried to avoid saying my name directly because of the profanity 15:24:15 Woo I broke the bookmarklet. 15:24:36 ais523: dear god @ that conversation 15:24:57 the book I referenced is almost entirely about that sort of paradox 15:25:11 it ends up proving Gödel's incompleteness theorem twice 15:25:22 a much simpler way than the original proof 15:25:39 ais523: was G.'s judgement correct? 15:25:43 about your/my win 15:25:48 well, your 15:25:50 oh, let me check 15:26:24 I think it falls into a gray area in the rules 15:26:47 he 217ed an entirely new theory of logic 15:27:19 ais523: yikes 15:27:22 ais523: appeal? 15:27:37 the annoying thing is, I can't actually see a flaw in the reasoning 15:27:51 ais523: just appeal it for being judge activism, like they do in America 15:41:00 -!- FireFly has joined. 15:52:16 catseye: hi 15:53:08 elliott: misping? 15:53:15 or, hmm, that doesn't seem to make sense in any context 15:53:19 nope 15:53:32 why would you say hi to someone who's been in the channel for a while, and not said anything for a while, when you haven't said anything for a while? 15:53:48 ais523: to ask him something that i don't want to bother typing unless he's there :P 15:54:24 I only do that sort of thing in /msg, in order to not send people demands out of context where I'm not there when they reply to me 15:54:56 ais523: wait, how does the former aid the latter? 15:55:12 because if they reply and I'm online at the time, I can instantly reply back with my actual question 15:55:19 and then for their rereply, I generally am online 15:55:24 well, that's what i'm going to do, no? 15:55:28 i don't see what /msg has to do with it 15:55:34 it doesn't, really 15:55:46 well, apart from loggedness of the channel I suppose 16:03:52 oerjan: for logreading reference, the company wine/debian/ubuntu were referring to is Blizzard 16:04:36 17:06:46 nothing beats "cat - > myprogram.c" 16:04:39 useless use of - award 16:04:56 18:33:02 enum 4 0 0 doif cage lt 4 setv va00 4 subv va00 cage ages va00 endi next 16:04:57 wat. 16:05:24 incidentally, "useless use of cat" is not useless when using sh interactively 16:05:31 it lets you write the line from left to right as you think of it 16:05:34 which saves cursor motion 16:05:36 18:46:10 Wait what? 16:05:37 18:46:17 catseye is now a Creatures person? 16:05:37 18:46:54 Or, at least, is mocking the "Age all creatures to adult" code 16:05:37 18:47:06 But... catseye provided absolutely no context. 16:05:37 lawl 16:05:38 ais523: indeed 16:05:44 ais523: it can also improve readability 16:05:53 I like your "useless use of - award", though 16:05:57 ais523: like writing haskell code with points, sometime :) 16:06:10 *sometimes 16:06:47 "Rememeber, nearly all cases where you have: 16:06:47 cat file | some_command and its args ... 16:06:47 you can rewrite it as: 16:06:47 ais523: ^ wow. 16:06:57 elliott@dinky:~$ elliott@dinky:~$ 16:06:58 holy carp. 16:07:00 elliott: that's a bashism 16:07:05 yeah, i figured 16:07:11 ais523: is this a bashism?: 16:07:12 $ >foo 16:07:15 in place of "touch foo" 16:07:21 I think so, less sure on that one 16:07:48 hmm, would >foo blank or touch an existing file? 16:07:51 the first seems more logical 16:07:57 ais523: blank 16:07:59 after all, :> foo would blank the file 16:08:02 ais523: i mean in place of using touch to create a file 16:08:20 people use touch with the explicit intention to /create/ a file? 16:08:24 ais523: err, yes 16:08:26 ais523: I do, for instance 16:08:28 I thought they used it either to update the timestamp, or to ensure that it exists 16:08:37 ais523: how would you create a blank file? 16:08:39 ensuring that a file exists isn't quite the same as creating it 16:08:47 elliott: generally speaking, I wouldn't 16:08:53 ais523: and if you would? 16:09:12 the only times I've done that, it was via a GUI, so right-click | New... 16:09:23 ais523: ok, ">foo" works in both dash and pdksh 16:09:27 portable enough for me! 16:09:36 I would use touch to create an empty file via the commandline, but I don't think that's ever come up 16:09:39 ais523: and so does " hmm, interesting 16:10:04 ais523: and any sh-alike that can do less than dash is just pitiful :) 16:10:18 * elliott installs ash, just to check 16:10:31 ais523: works in ash, too 16:10:32 both of them 16:10:40 busybox sh? 16:10:44 you can't get more bareboens than that 16:10:49 hmm, I'll test it, I have it installed 16:10:51 ORIGINAL BOURNE SHELL 16:10:57 ais523: wow 16:11:00 ubuntu ships with busybox 16:11:12 ais523: both work 16:11:12 of course it does, how else do you recover from a broken sh? 16:11:18 heh, true 16:11:22 but then why "I have it installed"? 16:11:24 elliott: ais523: and? 16:11:33 elliott: I mean, I instaleld it explicitly 16:11:34 that's to be expected 16:11:38 why? 16:11:40 ais523: you're feeding /etc/passwd to nothing 16:11:43 oh 16:11:51 ais523: " *work 16:11:56 the idea is to replace 16:11:58 you're right, cat x | f 16:12:03 with 16:12:05 with f hey, that means sh uses a special indicator to denote whether to be RPN or PN 16:12:36 omg 16:12:37 it can even do infix 16:12:40 ais523: y 16:12:52 for some definition of ifnix 16:12:53 *infix 16:12:57 well 16:12:59 ais523: more like 16:13:01 f it's only convention that commands tend to interpret their first arg as a filename to use instead of stdin 16:13:23 which, alas, doesn't seem to work :) 16:13:29 ick < hello.i or whatever will fail, for instance 16:13:30 even if you say cat - - 16:13:35 ais523: indeed 16:14:03 -!- MigoMipo has joined. 16:14:38 HEEEEEY 16:14:38 elliott: cat - /dev/fd/4 < /etc/passwd 4< /usr/share/common-licenses/GPL 16:14:48 webplat kinda works on Facebook now :) 16:14:50 ais523: i approve 16:14:59 come to think of it, 3 would work too 16:15:02 (0, 1, 2 are taken) 16:15:19 ais523: having to list your operands is a bit silly, though :) 16:15:20 /dev/fd is a brilliant invention 16:15:39 (and can incidentally be used to make suid shellscripts secure; some BSDs do that, although IIRC Linux doesn't) 16:15:48 how? 16:16:12 you open the file first, then check the fd to see if it was suid when it was opened 16:16:21 if it was, you pass the file descriptor to sh or whatever as an argument 16:16:27 no level of symlink hackery can get around that 16:16:28 heh 16:16:36 ais523: you still have to trust sh, though 16:16:40 and the script itself 16:16:42 especially if the program takes input 16:16:43 i wouldn't 16:16:50 oh, indeed 16:17:04 but the point is that it isn't inherently insecure, like it used to be 16:17:12 ("if you don't trust it don't suid it!" making proper shell scripts is almost impossible!) 16:17:37 apparently it used to be common in university computer labs to gain root privileges via a race condition exploit on "eject", which was a suid shell script that arbitrary people could run 16:28:46 a binary now, it seems 16:41:40 elliott: What's one of the pages you had that had weird behavior with webplat? 16:55:58 -!- atrapado has quit (Quit: Abandonando). 16:56:42 -!- BeholdMyGlory has quit (Remote host closed the connection). 16:57:56 -!- FireFly has quit (Quit: swatted to death). 17:01:33 Gregor: reddit 17:01:42 20:26:33 --- join: awilcox (AWilcox@awos/maintainer/awilcox) joined #esoteric 17:01:42 20:26:41 elliot, python is awesome 17:01:42 20:26:48 --- part: awilcox left #esoteric 17:01:52 one, it's blatantly obvious greasemonkey told you to come in here or something 17:02:01 elliott: Wowsa. 17:02:04 two, you can't even spell my name, why should i listen to your language opniions? 17:02:06 elliott: That is superbroken. 17:02:07 two, that's not how you spell elliott? 17:02:17 three, you just came in there to say that, you're amazingly idiotic if you think that'll change my mind 17:02:24 four, you're therefore just saying it for the hell of it 17:02:29 five, do you seriously have nothing better to do with your time? 17:02:35 eliotte: How do you spell elliott? 17:02:41 elliott: it took all of 15 seconds 17:02:51 five isn't really a valid point, some of the others are though 17:02:53 six, "AWOS is my attempt at creating an Operating System". clearly not 17:03:07 seven, python sucks. 17:03:09 the end! 17:03:15 ais523: and the OS one is hypocritical :P 17:03:49 actually it is 17:03:50 "* A universal operating system that can run DOS, Windows (3.1/9x/NT), Linux and BSD applications." 17:03:57 the mark of an idiot OS non-developer (more or less) 17:04:06 they *all* want compatibility with everything 17:04:46 six point one, "kernel.kdevelop" dear god you have no taste in both languages and IDEs 17:05:09 elliott: you dislike kdevelop? 17:05:16 ais523: well, :) 17:05:20 I've never used it, but you're the first person I've heard say anything bad about it 17:05:29 ais523: it's not bad it's just... 17:05:30 which is nice, because now I get to hear both sides of the argument in an attempt to form an opinion 17:05:32 an IDE-style IDE 17:05:35 for KDE 17:05:45 so it's, like, the epitome of boringness 17:05:53 hmm 17:05:57 elliott: /me can't even begin to guess why reddit is so hideously broken right now X-D 17:06:03 really, for big enterprisey projects you need an IDE 17:06:13 ais523: enterprisey is an epithet. 17:06:14 the major purpose of an IDE is helping people get their head around enterprisiness 17:06:24 elliott, hm I think I identified the main issue I have when coding in haskell. 17:06:41 Vorpal: yourself? 17:07:31 elliott, harhar. The issue is that I don't really have any sort of "feeling" for when lazyness matters. I keep thinking along strict evaluation lines a lot. 17:07:41 Vorpal: don't think along evaluation lines at all 17:07:49 oops, bbl missed what the time was 17:07:53 if you're trying to get eval order in your head when writing Haskell, you're doing it wrong 17:08:00 (I read most langs in eval order; not so with Haskell) 17:09:36 yeah, that 17:09:55 "I decided earlier today that my OS will be open source, under an Apache 1-like license. This was a big step for me; I wasn't sure if I wanted to be closed or open." 17:10:15 your amateur OS is totally important enough that people will pay attention to it even if they can't study the code. 17:10:43 wait, Apache 1? 17:10:48 Apache abandoned that license because it was insane 17:11:22 ais523: but that does not stop insane internet people. 17:11:33 I suppose so 17:11:41 who come into #esoteric presumably at the behest of GreaseMonkey (he's hosted on a "wilcoxtech" server) to tell me how awesome Python is because I dissed it 17:11:52 without even offering any sort of argument and not even insulting my mother 17:12:02 at least someone cares about my opinions enough :) 17:12:04 (I've had to reimplement libraries before because the existing libraries were Apache 1-licensed) 17:13:24 20:55:17 Sgeo: In English, "or" = xor *only* if talking to someone who is actually familiar with xor. :P 17:13:26 well, no 17:13:30 it's xor even then, they just don't know it 17:13:59 I'm familiar with xor, but normally interpret "or" as inclusive 17:14:03 although it depends on context 17:14:29 (even more, I interpret it as alternative, i.e. "do you want A or B" doesn't normally cause the answer "yes") 17:16:41 ais523: "did you do it, or did he?" 17:16:48 hmm 17:16:54 ais523: "did you do fly, or did he fly?" 17:16:56 if both of you flew, uh 17:16:57 anyway 17:17:01 ok so it makes little sense 17:17:11 elliott: "both" 17:17:24 ais523: clearly we must introduce Both Logic 17:17:32 elliott: it already exists 17:17:42 half the people in this department, if you ask them about truth values 17:17:52 ais523: that's unknown logic, probably 17:17:54 not BOTH LOGIC 17:17:56 will draw a little diamond, with true on the left, false on the right, neither on the bottom, both on the top 17:19:35 bottom is, incidentally, the same bottom that's the return value of an infinite loop 17:21:08 -!- MigoMipo has quit (Read error: Connection reset by peer). 17:21:23 ais523: and both is the type with one element 17:21:29 (confusingly, not the type with all elements) 17:21:35 sort of 17:21:37 well 17:21:38 uh 17:21:41 it depends :D 17:21:53 ais523: anyway, actual both logic is quite an interesting idea imo 17:21:55 the issue is 17:21:56 F + F = F 17:21:56 F + T = F 17:21:56 T + T = B 17:22:02 what's T+B? what's F+B? 17:22:09 where + is like souped-up and here 17:22:12 elliott: these are the sort of people who quantify over systems of logic 17:22:21 well 17:22:23 T+B = B, I think 17:22:27 F+B is probably F 17:22:30 but it might be T 17:22:34 ("both is just THAT TRUE!") 17:22:44 -!- cpressey has joined. 17:22:50 hi cpressey 17:22:55 hi elliott 17:23:03 cpressey: any netbsd luck? 17:23:33 elliott: no, mostly backing things up, now that i have something that can burn dvds. 17:23:40 tomorrow perhaps 17:23:55 wait no 17:23:57 T+B = T 17:24:07 F + F = F 17:24:07 F + T = F 17:24:07 F + B = F 17:24:07 T + B = T 17:24:07 T + T = B 17:24:08 i think 17:24:20 i am actually planning on installing 10.10 here at work so i don't have to install it at home :D 17:24:36 22:21:44 It does not work nicely in Chrome /anywhere/ 17:24:37 it does 17:24:42 at least it did when i tried 17:24:47 cpressey: how does that make sense xD 17:25:35 if i have access to a 10.10 install *somewhere*, i won't be as tempted to install it on my home machine where I COULD be installing netbsd instead! 17:25:57 also, getting sick of virtualbox 17:26:13 things will go smoother here if windows is not involved 17:26:39 (except for scheduling meetings; outlook actually kind of works for that, while nothing else seems to) 17:27:11 what about talking to people in person? that's how I schedule meetings 17:27:46 omg civilization would fall apart 17:27:47 23:28:24 Right, if a node has non-whitespace text children, OR it has no non-inline children at all, then it's a platform. 17:27:52 Gregor: bet wrapper divs fuck it up 17:28:07 ais523: it's more a matter of it being retained on a calendar as a reminder 17:28:23 well add it to your calendar then? 17:28:26 cpressey works for the MAN 17:28:29 Evolution does that just fine for me 17:28:31 he sold out 17:28:43 even directly from emails sent from someone else's Outlook or whatever the Mac equivalent is 17:29:04 ais523: maybe all the non-outlook-using people here are making excuses, then :) 17:29:30 it's possibly because everyone but me seems to hate Evolution for some reason 17:29:33 00:08:21 Damn it jQuery, stop solving all my problems! 17:29:43 [John Resig's guiding spirit disappears from Gregor's soul] 17:29:47 but I find it a pretty nice Outlook substitute, and get irritated at Thunderbird 17:29:58 i hate both! but you could have guessed that 17:30:36 ais523: wow, best imitation of zzo38's style of speech yet found: 17:30:46 "I agree. jQuery is really the bset, it solves all kinds of browser problems and is good, as well" 17:30:48 *best, 17:30:50 elliott: do you hate all email programs? 17:30:58 "and X, as well" at the end of a long sentence seems to help a lot 17:31:03 ais523: Well ... um ... maybe ... yes 17:31:10 elliott: I saw the comment before the introductory sentence, and thought "hey that sounds just like zzo38" 17:31:36 it's a really charming form of grammar 17:31:47 if unparsable 17:32:04 (sometimes) 17:32:43 01:40:20 gregor: http://filebin.ca/qxuvnd 17:32:45 what file format is this? 17:32:50 the lack of a filename doesn't help... 17:33:02 (why can't filebin specify the filename in the relevant header? sheesh) 17:34:19 filebin specifies it in the URL, optionally 17:34:26 does file(1) help identify the file? 17:34:49 ais523: yes, but... it should really use the correct http header 17:34:51 Content-Disposition or whatever 17:34:55 it already does it to force a download 17:34:59 it just needs ; filename=foo 17:35:22 Friend: so have you decided what you are going to do about the websites? 17:35:22 Mark Zuckerberg: yea i’m going to fuck them 17:35:22 Mark Zuckerberg: probably in the year 17:35:22 Mark Zuckerberg: *ear 17:35:24 [on ConnectU] 17:35:33 "On September 20, 2010, Facebook confirmed the authenticity of these leaked instant messages in a New Yorker article." 17:35:38 ais523: it's probably a gif 17:35:50 since that fits what quintopia said about it afterwards 17:35:51 and the context 17:35:52 "probably"? 17:36:07 GIFs have a reliable magic number, don't they? 17:36:14 or are you trying to determine what the file is without downloading it? 17:36:28 ais523: i already dismissed the download window and ~/Downloads is huge :P 17:36:42 you mean you don't move files to more sensible locations immediately after downloading them? 17:36:59 what location could possibly be more sensible than ~/Downloads? 17:37:25 ais523: i totally would but i'm lazy as hell. also -- why not just pop up a download box if you do that? 17:37:30 for "save as" 17:37:42 cpressey: the fact that a file originally came from the Internet is pretty irrelevant to it, generally 17:37:59 elliott: oh, I do normally, but sometimes the file downloads in the wrong place for no apparent reason 17:38:31 ais523: I used to have ~/Saved/YYYY-MM; anything I wanted to keep would go there. 17:38:39 For instance, I'd put installation ISOs there and the like. 17:38:43 And just about anything I didn't want to delete. 17:38:52 It saved me having to categorise what was essentially a heap of random stuff. 17:38:54 hmm, if I don't want to keep something, I generally just delete it 17:38:57 ais523: i usually extract the contents to somewhere sensible, and leave the archive in the downloads directory, pretending it's a cache. of course, if it's not an archive, that's a different matter 17:39:01 ais523: ditto 17:39:10 "somewhere sensible", though, is usually ~/build 17:39:10 but if I did want to keep it, it went into ~/Saved/YYYY-MM 17:39:19 which barely counts 17:39:27 although, ~/Saved/YYYYQN/ would probably be more reasonable 17:39:29 e.g. 2010Q3 17:39:35 for third quarter of 2010 17:40:34 fun thing to do: compare infoboxes on Wikipedia for two identical twins 17:42:01 I figured new potential problem with IPv4 depletion: After the first RIR (APNIC) has been depleted, "RIR shopping" likely commerces, especially for Asian users. This will fragment the routing tables even more (and IPv4 routing table fragmentation is already bad). 17:43:32 Ilari: how, exactly, does the internet stll manage to work? 17:43:45 *still 17:44:06 elliott: it's working a lot slower than in theory it should 17:44:18 ais523: yes, but it still works remarkably well 17:44:22 try tracerouting your connections some time, and geolocating each of the hops 17:44:34 the number of times the route actually starts off in entirely the wrong direction is quite worrying 17:45:03 ipv6 isn't going to happen, is it... 17:45:14 although i have to note that ipv4 depletion was predicted to happen very very soon years ago 17:45:15 Routing table fragmentation is router-internal problem. 17:46:08 elliott: it's probably going to happen eventually in 2012, and cause the end of the world 17:46:22 (Note: Mayan calendar actually rolls over to a new cycle in 2220, it was originally misinterpreted) 17:46:32 ais523: and cycles aren't the end of the world, etc. etc. 17:46:43 still, i totally have plans to make plans for December 21st, 2012. 17:46:55 elliott: no, although Mayan mythology did have it that they caused mass extinctions 17:46:57 ais523: i wonder how many suicides it'll get 17:46:59 even Y2k got a few 17:47:05 ugh, a depressing thought 17:47:18 besides, why would you bother committing suicide if the world's about to end anyway? 17:47:29 ais523: because otherwise CTHULHU WILL CONSUME YOUR SOUL TEN TIMES MORE PAINFULLY 17:47:34 or something to that effect 17:47:53 ais523: at least solace can be taken in the fact that it's very unlikely someone smarter than a sack of bricks would do it 17:48:01 (even insane intelligent people tend to be a bit more creative than that...) 17:49:23 meh, believing I'm sane would require a huge number of assumptions about the world I'm not willing to make 17:49:23 Funny how many think that there's going to be some disaster because Mayans predicted there would be... Well, Mayans made no prediction of that day being special (apart of prediction of big party). 17:49:48 Ilari: they did, according to mythology there was a mass extinction at the end of every cycle so far 17:49:51 ais523: wait, what has your sanity got to do with it? :) 17:49:55 admittedly, they didn't predict that the pattern would necessarily continue 17:50:14 elliott: the topic came up 17:50:17 also, it was 2020 as you said 17:50:19 did they predict their own civilization's extinction? 17:50:21 elliott: 2220 17:50:25 ais523: oh 17:50:26 even better 17:50:34 cpressey: I'm not sure 17:50:38 ais523: although not really, as i don't get to see the *second* day of insanity 17:50:40 their calendar didn't, but I'm not sure it was meant to 17:50:54 I think they though the cycle wouldn't continue (because now (IIRC 5th time), the creation actually succeeded). 17:51:01 elliott: ohhh i'm sure there will be *some* insanity come 2020. 17:51:09 cpressey: there's insanity every year! 17:51:23 the mayans weren't that hot anyway 17:51:34 we're smarter than them :P 17:51:43 hm, there wasn't much panic in 1996, even though that year was explicitly mentioned in some prophecies... i guess they were too obscure 17:51:49 I seem to remember there was an end-of-the-world scare in 1996 17:51:54 at school, at least 17:52:02 ais523: hm, interesting 17:52:08 HEAVEN'S GATE 17:52:12 (note: unrelated to 1996) 17:52:13 ohhh right 17:52:13 cpressey: that was Nostradamus's prediction, IIRC, or at least one interpretation of what he wrote 17:52:17 i forgot about them! 17:52:18 that was 1997 17:52:19 not 1996 17:52:20 the sneaker peopl! 17:52:21 23:28:24 Right, if a node has non-whitespace text children, OR it has no non-inline children at all, then it's a platform. 17:52:21 Gregor: bet wrapper divs fuck it up 17:52:27 btw: 17:52:28 [[Exit Press Release: 17:52:29 "Away Team" Returns to Level Above Human]] 17:52:31 http://www.heavensgate.com/misc/pressrel.htm 17:52:32 Wrapper divs have non-inline children. 17:52:33 translated: 17:52:38 elliott: you wouldn't expect apocalypticists to get the year write, would you? 17:52:39 If they also have text, then yup, womped up. 17:52:40 "Here's a press release, we're going to go kill ourselves now." 17:52:43 And when it comes to disasters, there's considerable diference between disaster that takes down western civilization and an ELE. 17:52:54 ais523: I wouldn't expect them to get it write, no. What? 17:53:00 um, right 17:53:00 01:40:20 gregor: http://filebin.ca/qxuvnd <-- this is a .tar.gz. Learn to file(1) 17:53:13 Gregor: I don't see why I should go to the effort >:) 17:53:15 elliott: I type by thinking what I want to type auditorily, then my fingers translate it into letters 17:53:20 It opened with Archive Manager 17:53:27 but it gave up and cried after showing a file with the same name as the only entry. 17:53:34 ais523: that's... impressive... ly bad 17:53:42 it's not deliberate, it just sort of happened 17:53:47 ais523: can you not think non-audially? 17:53:49 then I have to go back and correct homophones that have crept in somehow 17:53:57 elliott: I can; but not words for some reason 17:53:59 (not "audibly", spellcheck! :P) 17:54:00 I normally just think in thoughts 17:54:13 ais523: so when you're thinking about whatever you... don't even have an internal monologue? 17:54:18 but they become sounds when I have to translate them to a language 17:54:23 you just float around pure concept-space? i find that unlikely 17:54:39 ais523: i had an experience the other day which made it clear to me that i think in both thoughts and words, like on two levels 17:54:40 elliott: I've started typing a sentence and then realised I didn't know what the next word in it was in English, or indeed any other language, before 17:54:43 although it didn't happen often 17:54:48 It only takes one very bad pandemic to take down civilization. That wouldn't be even near ELE. 17:54:54 that's proof that thoughts happen as thoughts, and only become words when you try to determine what they are 17:55:00 cpressey: what was yours? 17:55:12 cpressey died and experienced pure thought 17:55:16 then his boss woke him up 17:55:22 I N C E P T I O N 17:55:34 ais523: basically not being able to remember a word for something, in my internal monologue, but noticing that the thought behind it was still coherent 17:55:46 cpressey: same here 17:56:04 well, yes 17:56:05 obviously 17:56:07 sapir-whorf is false 17:56:08 but: 17:56:12 ais523: don't you hear an internal monologue? 17:56:14 in normal thoughts 17:56:15 I imagine that when people introspect, as in "what am I thinking", they translate their thoughts into English 17:56:16 but not otherwise 17:56:23 hmm 17:56:23 maybe 17:56:25 elliott: I can, but only when I choose to, or am thinking about it 17:56:31 it's hard to figure out what we do when not introspecting 17:56:33 for obvious reasons :) 17:56:39 I tend to use an internal monologue the same way other people use pencil and paper 17:56:43 to take down notes 17:56:51 ais523: i disapprove of your memory and wish i had it 17:56:58 (this is actually quite useful, because if I'm at a computer, I can actually write down the monologue as it happens) 17:57:38 Basically, what makes civilization vulernable is strong centralization and interdependence. 17:58:25 Ilari: at the same time, centralization and interdependence make civilization possible 17:58:34 stupid balancing acts 18:00:14 Actually, no... Its just the tendency that civilization goes towards stronger and stronger centralization and interdependence. Some centralization and interdependence is inherit, but very strong one isn't. 18:00:40 I didn't say "strong". 18:01:15 So, yes, some centralization and interdependence in inherent -- that was my point. 18:01:24 bbl (unless there's an ELE while I'm at lunch) 18:01:38 FOOM 18:02:29 Western civilization is much more than 100 years old. And 100 years ago there was centralization and interdependence, but MUCH less than today. 18:04:04 Good thing, because spanish flu really pushed the system. Over the breaking point in some places. And when the breaking point was exceeded, the results were VERY VERY ugly. 18:04:28 Ugly like ugliness. 18:04:47 Ilari: Modern Western civilisation isn't very old though (obviously). 18:04:51 I'd say less than 200 years. 18:05:03 Late 1800s onwards. 18:05:55 Like 90% of population died... That's actually worse than in cities that were written off during Black Plague (IIRC, about 75%). 18:06:32 elliott: Fixed Reddit. 18:06:48 Gregor fixed reddit singlehandedly. 18:06:51 Gregor: "reddit" btw. 18:06:57 ...link me again? >_> 18:07:04 http://codu.org/webplat/ 18:07:11 ReDdIt 18:08:17 Industrialization started in 18th century. 18:08:30 Ilari: Modern Western civilisation is a bit more than that. 18:08:45 Gregor: you frozen my reddit. 18:08:52 elliott: It's slow to load. 18:08:53 oh it fix. 18:08:55 elliott: But it works. 18:09:13 Gregor: Things that suck: Walls you can't see. Make them flash a border when you hit them and then fade that out or something? 18:09:26 Woot your thing fails on non-white background :P 18:09:36 Gregor: Okay, serious non-nitpicky suggestion: 18:09:47 Gregor: Have some sort of key that lets you jump into any element with child elements. 18:09:49 What ultimately kills civilizations: Running out of resources. 18:09:50 It doesn't "fail", it's explicitly set to black-on-white so it'll always be visible. I'll use the .pngs at some point. 18:09:53 Gregor: Text acts as floor and the like. 18:09:59 That way barriers wouldn't be an issue and it could be a game mechanic. 18:10:00 -!- sftp has joined. 18:10:23 * Gregor considers. 18:10:33 Usually they don't reach the amount of complexity such that one disaster could wipe it out in one shot. 18:11:12 Gregor: I mean, if you can jump into links, you can jump into elements :P 18:11:48 elliott: The reason why I didn't just say that any element with children is clear in the first place is that then whether text is a boundary or not is a mystery ... 18:12:06 elliott: Also, it does display a boundary when you jump into something that's a not-very-visible wall? 18:12:07 Gregor: ...text is water. 18:12:13 It's sort-of-solid-looking, sort of not. 18:12:15 So you can swim in it! 18:12:57 *brain axplote* 18:13:25 Gregor: You can't argue with the truth. 18:13:56 elliott: Also, it does display a boundary when you jump into something that's a not-very-visible wall? <-- Try it on reddit, then try and jump onto the "report" link. 18:14:08 You bash into the "submitted ..." line, which is full-width. 18:14:11 Despite not being visible there. 18:14:17 But no indication of this is given. 18:14:47 Gregor: Also, I'd just like to say that this game has hilarious physics: 18:14:51 *physics :P 18:15:00 * Gregor tries desperately to figure out where any of the stuff you're talking about on reddit is :P 18:15:43 Gregor: Oh, uh... you have to log in to see "report". 18:15:48 It's next to "share" basically. 18:16:09 Gregor: Aww, it resets Acid2. 18:16:16 So I can't see what the fuck it thinks of the crazy smiley. 18:16:19 I can't produce any weirdness around "share" 18:16:24 (Apart from "dear god what".) 18:16:27 Gregor: Just jump up, then. 18:16:30 You'll bash into whiteness. 18:16:44 From the very first position, move right until there's no links directly above you. 18:16:45 Jump. 18:16:48 Ohshitwhitewall 18:16:56 Gregor: Holy crap 18:16:59 http://www.webstandards.org/files/acid2/test.html 18:17:00 Try it here 18:17:05 And walk off the edge 18:17:47 javascript:(function(){var%20script=document.createElement('script');script.src='http://codu.org/webplat/jquery.js';document.getElementsByTagName('head')[0].appendChild(script);script=document.createElement('script');script.src='http://codu.org/webplat/webplat.js';document.getElementsByTagName('head')[0].appendChild(script);})() 18:17:49 (for my future usage) 18:20:42 elliott: That's ... wtf 18:20:50 Gregor: :D 18:21:03 WHO ARE YOU CRAZY ACID MAN??? 18:21:10 He's on Acid2. 18:21:13 *Acid, too. 18:21:18 (Good joke opportunity missed!) 18:21:21 *acid, too. 18:24:28 elliott: OHOHOHO, if you jump UP into something, not SIDEWAYS into something, it doesn't show the box? Is that what you mean? 18:24:41 Gregor: Yes. 18:24:45 Ahhh 18:24:48 Yeah, I'm aware of that problem. 18:24:48 Ohohohoho Santa 18:24:59 Gregor: Also it totally needs to be an MMORPG. 18:25:03 "omfg there are like 1,000 people on this page" 18:25:10 "FIRE ZE MISSILES" 18:25:38 Gregor: "5,000,000 free points and a never-seen-before power 10 weapon will be released by a GM on http://meatspin.com/ at 4pm UTC!" 18:25:48 lawl 18:25:55 6pm UTC: "Sorry, guys, there was a delay at the office. Unfortunately the servers have lost the code for now, so check back again another time!" 18:25:59 "In fact, every day!" 18:26:18 ... OMG OMG OMG ... you COULD have secret things to find ... they could even be handled in such a way that they couldn't be faked ... 18:26:36 Gregor: Y'know, it's time to ditch JS if you're getting that ambitious :P 18:26:48 NEVAR 18:26:48 WebSockets will save me! 18:27:05 Gregor: Have you considered how to handle following links yet? 18:27:07 JUST ASKIN' 18:27:18 Space will probably be generic "activate element" 18:27:21 WebSockets plus a port of WebKit to JavaScript? :P 18:27:24 Gregor: I mean technologically. 18:27:24 For activating input boxes, links, whatever. 18:27:32 Yes. 18:27:32 Proxy lawl 18:28:24 Basically, I may be able to get away with proxying the HTML page and adding an ever-unpopular tag. 18:28:40 Gregor: Oh great, so following a link forces a kludgy new page load and loss of seamlessnessness. 18:28:42 WOOO 18:28:51 Gregor: How about just using XMLHttpRequest and setting document.innerHTML or whatever? 18:28:54 That'd be less painful even for that... 18:29:02 I ... would that work ... 18:29:13 Gregor: Yes. Yes it would. 18:29:30 (That wouldn't work because you can't XHR past the SOP, but still) 18:29:36 Gregor: Proxy. Duh. 18:29:43 Fair enough. 18:29:43 That also inserts the