00:06:10 -!- variable has joined. 00:08:37 -!- zzo38 has quit (Remote host closed the connection). 00:18:06 -!- MDude has joined. 00:24:07 -!- edwardk has quit (Ping timeout: 240 seconds). 00:24:42 -!- edwardk has joined. 00:38:02 -!- derdon has quit (Remote host closed the connection). 00:40:41 -!- nortti_ has quit (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )). 00:45:32 olsner: oh right, because you can always turn any object into an array of char, and Good Things are guaranteed to happen 00:47:51 -!- Slereah_ has quit (Read error: Connection reset by peer). 00:48:07 -!- Slereah has joined. 01:00:50 -!- copumpkin has quit (Ping timeout: 248 seconds). 01:13:03 -!- pikhq_ has joined. 01:13:06 -!- pikhq has quit (Ping timeout: 255 seconds). 01:14:09 pikhq_, if I'm holding data in a variable, and I want to persist it to a file, is it a terrible idea to just put a variable trace on that variable and have it write the variable's value out to a file? 01:18:41 -!- edwardk has quit (Quit: Computer has gone to sleep.). 01:24:54 -!- edwardk has joined. 01:31:26 -!- Jafet has quit (Quit: Leaving.). 01:43:32 -!- zzo38 has joined. 01:49:19 -!- copumpkin has joined. 01:51:05 -!- edwardk has quit (Quit: Computer has gone to sleep.). 01:52:37 -!- edwardk has joined. 01:54:08 -!- edwardk has quit (Client Quit). 02:03:42 -!- Tod-Autojoined has joined. 02:03:54 -!- Slereah_ has joined. 02:06:56 -!- itidus20 has joined. 02:06:56 -!- soundnfury has quit (Remote host closed the connection). 02:06:57 -!- azaq23 has quit (Max SendQ exceeded). 02:07:01 -!- Gregor has quit (Excess Flood). 02:07:01 -!- copumpkin has quit (Ping timeout: 264 seconds). 02:07:01 -!- pikhq_ has quit (Ping timeout: 264 seconds). 02:07:01 -!- Slereah has quit (Ping timeout: 264 seconds). 02:07:02 -!- oklopol has quit (Read error: Connection reset by peer). 02:07:13 -!- oklopol has joined. 02:07:18 -!- MDude has quit (Ping timeout: 264 seconds). 02:07:18 -!- kallisti_ has quit (Ping timeout: 264 seconds). 02:07:19 -!- TodPunk has quit (Ping timeout: 264 seconds). 02:07:34 Eek, I think Tcl is making me think in terms of global variables more :/ 02:07:37 -!- MDude has joined. 02:07:54 -!- variable has quit (Ping timeout: 264 seconds). 02:07:59 -!- pikhq has joined. 02:08:00 -!- kallisti has joined. 02:08:01 -!- kallisti has quit (Changing host). 02:08:01 -!- kallisti has joined. 02:08:07 -!- Slereah_ has quit (Ping timeout: 240 seconds). 02:08:16 -!- copumpkin has joined. 02:08:21 -!- variable has joined. 02:08:21 -!- variable has quit (Changing host). 02:08:21 -!- variable has joined. 02:08:21 -!- itidus21 has quit (Ping timeout: 245 seconds). 02:08:52 -!- azaq23 has joined. 02:15:56 Oh wow, I never realise YouTube free energy videos would be so hilarious. 02:17:58 -!- Jafet has joined. 02:27:42 -!- MDude has quit (Ping timeout: 264 seconds). 02:27:57 -!- MDude has joined. 02:37:19 -!- Gregor has joined. 02:38:39 -!- david_werecat has quit (Ping timeout: 260 seconds). 02:46:49 -!- augur has quit (Remote host closed the connection). 03:04:21 -!- itidus20 has changed nick to itidus21. 03:08:25 -!- oerjan has joined. 03:13:27 Sgeo_: *wince* 03:13:48 pikhq, at my variable tracing thoughts? 03:13:54 Yeah. 03:14:05 * variable traces Sgeo_ thoughts 03:15:44 Blah 03:21:59 -!- Phantom_Hoover has quit (Read error: Connection reset by peer). 03:22:06 > let e('(':'\\':v:'.':l)=let(x,')':t)=e$d l in("(\\"++v:'.':x++")",t);e('(':s)=let(x,t)=e s;(y,')':u)=e$d t in(a x y,u);e s=splitAt 1$d s;d=dropWhile(==' ');a('(':'\\':v:'.':l)s=let f x|x==v=s|1>0=[x]in init l>>=f;a f x="("++f++" "++x++")" in fst$e "((\\x. (x x)) (\\x. (x x)))" 03:22:08 "((\\x.(x x)) (\\x.(x x)))" 03:23:42 -!- augur has joined. 03:24:11 > let e('(':'\\':v:'.':l)=let(x,')':t)=e$d l in("(\\"++v:'.':x++")",t);e('(':s)=let(x,t)=e s;(y,')':u)=e$d t in(a x y,u);e s=splitAt 1$d s;d=dropWhile(==' ');a('(':'\\':v:'.':l)s=let f x|x==v=s|1>0=[x]in init l>>=f;a f x="("++f++" "++x++")" in fst$e ""((\\x. (\\y. x)) (\\z. z)" 03:24:12 : 03:24:12 lexical error in string/character literal at end o... 03:24:16 oops 03:24:30 > let e('(':'\\':v:'.':l)=let(x,')':t)=e$d l in("(\\"++v:'.':x++")",t);e('(':s)=let(x,t)=e s;(y,')':u)=e$d t in(a x y,u);e s=splitAt 1$d s;d=dropWhile(==' ');a('(':'\\':v:'.':l)s=let f x|x==v=s|1>0=[x]in init l>>=f;a f x="("++f++" "++x++")" in fst$e "((\\x. (\\y. x)) (\\z. z))" 03:24:33 "(\\y.(\\z.z))" 03:24:53 fancy 03:25:11 Well, it's only short because substitution isn't hygienic 03:25:17 ah. 03:25:53 pikhq, what are good uses of variable tracing? 03:26:01 And is my idea really that terrible :/ 03:26:20 Sgeo_: Debugging, inspiring drinking. 03:26:42 haskell's tracing function uses unsafePerformIO, just a hint 03:26:53 Different thing 03:27:11 Tcl's tracing lets you run code when a specified variable is read from or written to or unset 03:27:18 oerjan: Tcl variable tracing is where you set up a callback that is triggered every time a variable is mutated. 03:27:58 ok 03:29:10 Is there a command like this in some standard library somewhere? 03:29:17 Or, well, some thingy like tcllib 03:30:09 proc debugline args { set result [uplevel 1 $args]; puts "$args -> $result"; return $result } 03:30:15 Wait, that doesn't quite work, does it 03:33:16 Meh, just a command to stick in front of a line to print the line and its result 03:33:28 I could probably write it myself, just wondering if it pre-exists 03:43:56 -!- coppro has quit (Ping timeout: 252 seconds). 03:44:35 -!- coppro has joined. 03:46:55 -!- david_werecat has joined. 04:12:58 pikhq, what sort of headaches does variable tracing cause? 04:21:16 -!- copumpkin has quit (Ping timeout: 255 seconds). 04:21:48 -!- copumpkin has joined. 04:30:37 Sgeo_: all sorts 04:31:29 -!- VorpalPhone has joined. 04:35:06 bassett's headache liquorice 04:50:18 -!- azaq23 has quit (Quit: Leaving.). 05:06:06 -!- VorpalPhone has quit (Read error: Connection reset by peer). 05:08:46 -!- Jafet has quit (Ping timeout: 245 seconds). 05:09:29 -!- Jafet has joined. 05:35:18 so 05:35:28 salt is essential for the survival of many animals. 05:35:57 but it doesn't seem that abundant in nature? Is it just consumed from saltwater and saltwater fish and then propoagates through the food chain of carnivores? 05:36:45 s/salt/sodium/ 05:39:30 spirity: I guess? Salt is essential for the survival of animals, but not in large quantities, so... 05:45:38 seems to be the case. 05:45:50 apparently herbivores get along fine without too much salt. 05:46:33 very few plants contain sodium 05:47:27 Though, herbivores in particular will often explicitly seek out salt licks. 05:57:47 -!- pikhq_ has joined. 05:57:59 -!- pikhq has quit (Ping timeout: 248 seconds). 06:01:41 Peppercorns (dried black pepper) are, by monetary value, the most widely traded spice in the world, accounting for 20 percent of all spice imports in 2002. The price of pepper can be volatile, and this figure fluctuates a great deal year to year; for example, pepper made up 39 percent of all spice imports in 1998. 06:01:46 that's really weird to me. 06:01:46 * pikhq_ cannot explain his enjoyment of this series 06:01:51 black pepper isn't even that great of a spice. 06:01:55 pikhq_: which 06:01:58 MLP 06:02:33 And, yeah. It's not that pepper is a very *expensive* spice, it's just used a lot. 06:03:05 I bet by mass it's on the order of 50% of the spice trade. 06:03:24 I typically use red pepper in place of black pepper, for things where it makes sense. 06:03:33 like soups and such. 06:04:00 * spirity enjoys spicey foods. 06:05:37 * pikhq_ concurs 06:08:02 -!- Vorpal has joined. 06:12:37 so I just noticed 06:13:00 that most western european cuisines aren't really known for spicy foods. 06:13:35 compared to say, korean, thai, south america, indian. 06:14:41 Or even North American. At least, we've got a nice tendency to adopt spicy foods from other cultures... 06:15:03 is authentic Italian spicy? 06:15:21 Well. I guess Tex-Mex is purely American, and reasonably spicy... 06:15:40 America has all the cuisines. 06:15:44 True. 06:15:58 That's kinda the only distinctive property it has. 06:16:06 (cuisine-wise) 06:17:51 yep 06:18:05 pizza? sure. but it could use some pineapples you know. 06:18:09 * spirity american logic. 06:19:13 Quite right. 06:19:18 And extra cheese. 06:19:55 Epic Meal Time is actually the essence of American cuisine in its purest form. 06:20:00 pikhq_, what's the worst that will happen if I use variable traces the way I said? 06:20:08 Though it's Canadian, I agree. 06:20:15 Sgeo_: You will confuse yourself greatly. 06:20:16 -!- david_werecat has quit (Read error: No route to host). 06:20:34 pikhq_: I thought they were from Vermont or something. 06:20:38 -!- david_werecat has joined. 06:21:07 -!- oerjan has quit (Quit: leaving). 06:21:37 spirity: Montreal. 06:21:52 Vermont's *like* Canada. 06:22:08 oh. okay. 06:22:16 that makes perfect sense actually 06:22:21 americans would never do that to food. 06:22:28 Nah, we totally would. 06:22:42 well, maybe. 06:22:45 'Cept with more cheese. 06:22:48 but we wouldn't make an internet show about it 06:22:55 we'd just eat it and move on with our lives 06:23:10 Yeah, we'd make a restaurant instead. 06:23:47 (see: Texas) 06:28:15 -!- Nisstyre has quit (Read error: Connection reset by peer). 06:28:34 -!- yours_truly has joined. 06:29:34 -!- yours_truly has quit (Read error: Connection reset by peer). 06:30:02 -!- pikhq has joined. 06:33:07 -!- pikhq_ has quit (Ping timeout: 240 seconds). 06:38:07 pikhq, not totally sure why I'd be confused, especially in the script in question is a simple one. 06:39:07 Sgeo_: Basically, variable tracing in Tcl is on par with ptrace for Linux binaries. It'll work, but it's Nasty™. 06:39:34 I don't know anything about prtrace 06:49:17 -!- asiekierka has joined. 06:49:56 -!- lifthrasiir has quit (Ping timeout: 248 seconds). 06:57:34 -!- lifthrasiir has joined. 06:59:34 -!- zzo38 has quit (Remote host closed the connection). 07:15:41 -!- Nisstyre has joined. 07:19:33 -!- epicmonkey has joined. 07:20:03 -!- Taneb has joined. 07:32:30 Hello 07:33:08 hi 07:35:07 -!- epicmonkey has quit (Ping timeout: 240 seconds). 08:31:09 hi 08:39:55 -!- epicmonkey has joined. 08:43:32 -!- david_werecat has quit (Ping timeout: 248 seconds). 08:45:06 -!- AnotherTest has joined. 08:45:46 mroman: thanks for the reference table :D 08:56:01 fuck ant 08:59:37 yes 09:00:45 I have no option though, existing project for which I'm writing a small patch. Also Android SDK... 09:01:19 AnotherTest: No problem ;) 09:01:33 (but you need to pull from the repos. Since I added some functions) 09:01:45 also... you can no go into shell mode by calling the binary with --shell 09:01:49 *now 09:04:54 I think I made something that calculates the Fibonacci sequence 09:05:02 but I'll test it first 09:11:37 it doesn't 09:12:04 -!- derdon has joined. 09:15:11 mroman, what is it you are working on? 09:17:49 Vorpal: http://esolangs.org/wiki/Burlesque 09:17:59 or... if you prefer a satiric website about it 09:18:10 http://mroman.ch/burlesque/index.html 09:20:39 mroman: I made something which, for a number n, n * 2^n 09:20:50 Not exactly the intention :p 09:21:39 ri^^{\/^^.+\/1.-}{}w!vv 09:23:30 But it does n * 2^n :) 09:23:58 Indeed 09:26:55 I added it to the examples 09:27:49 I'll do a decent Fibonacci later :p 09:27:58 I need a more light weight browser for this computer 09:30:43 Although producing n * 2 ^ n will get easier once I added all builtins from Stlang 09:31:18 "obfuscated n*n^2" then 09:31:31 But good example for using while, thx. 09:32:04 10^^2\/**.* 09:34:36 I guess the biggest issue for fibonacci is, that Burlesque has no explicit recursion :) 09:48:16 Vorpal: Graphical or Text? 09:49:32 Vorpal: http://www.netsurf-browser.org/ 09:49:43 Small as a mouse, fast as a cheetah! 09:49:59 mroman, graphical 09:50:01 And note that cheetah sounds like cheater in some dialects ;) 09:50:10 mroman, similar capabilities to firefox, html5 and so on 09:50:17 It doesn't fully support everything from html. 09:50:18 Vorpal: if you want javascript you can use hv3. itis pretty buggy though 09:50:23 Vorpal: Ok 09:50:32 So you wan't a lightweight browser that is not lightweight? 09:50:37 it is not a super old computer. but firefox + android emulator = painful 09:50:47 Vorpal: also what os are yo using? 09:51:00 nortti, 64-bit Ubuntu on this laptop, it has 2 GB RAM 09:51:16 hmm.. 09:51:21 10.04 LTS 09:51:34 how is that computer slow/old? 09:51:35 http://twotoasts.de/index.php/midori/ then? 09:51:42 It uses WebKit so... 09:51:49 WebKit should support all the regular stuff. 09:52:00 nortti, not by itself. But when I run other stuff apart from firefox I can get it to swap trash every now and then 09:52:02 which is not fun 09:52:17 and I don't want to have like only 2 tabs open 09:52:30 I generally have around 30-40 tabs open 09:52:43 Vorpal: kazehakase? 09:52:47 nortti, what? 09:52:57 Vorpal: kazehakase 09:53:00 mroman, will check that out 09:53:03 nortti, is it a web browser? 09:53:11 Vorpal: yes 09:53:57 Vorpal: it is directly available from ubuntu repo 09:54:00 hm okay 09:54:07 (or was last time I checked) 09:54:10 -!- nooga has joined. 09:54:16 also surf from suckless 09:54:24 nortti, not in the version of ubuntu I run 09:54:25 or xxxterm :P 09:54:28 which is 10.04 LTS 09:54:31 hmm 09:54:37 yes I need to update to the new ubuntu release soon 09:54:44 but eh, I'm going xubuntu then I think 09:54:56 why not lubuntu? 09:55:02 LXDE? 09:55:07 maybe I'll give that a try 09:55:07 yes 09:55:12 never used LXDE 09:55:23 but I'm quite fond of Gnome 2, and xfce is almost as good 09:55:30 http://kazehakasa.sourceforge.jp 09:55:35 right 09:56:59 oh. Line Mode Browser is still developed 09:57:06 hm? 09:57:09 what is that 09:58:00 kinda like old lynx fused with ed 09:58:11 second web browser ever developed 09:58:51 hm 10:00:15 also Amaya seems to be pretty light (w3c testbed browser) 10:01:39 hm 10:01:47 hmm. Dillo is still developed? 10:02:27 * Sgeo_ wonders in what circumstances command prefixes are not suitable alternatives to having closures 10:03:02 PDAs can't have random memory access? 10:03:35 Would that imply that it is impossible to do an arbitary effect at an arbitary point? 10:04:35 Although if you have enough operations you can pop, put the element in a list, pop, add the element to the list 10:04:44 and go back to any location of the stack, change it 10:04:50 and deconstruct the list 10:04:55 to reconstruct the stack 10:05:36 That is, if you can have arbitary data on the stack, right? 10:08:00 -!- pikhq_ has joined. 10:08:20 -!- pikhq has quit (Ping timeout: 248 seconds). 10:09:33 hmm. Dillo is still developed? <-- is it? nice 10:11:04 still netsurf seems to be the best lightweight browser unless you are really into surfing the net with your 486 in which case I can recommend links2 from experience 10:11:20 heh 10:11:49 feature-wise it seems pretty limited though 10:11:55 yeah 10:12:16 but links2 is good enough for me as my main browser 10:12:18 nortti, you know what I need? A desktop version of the mobile chrome browser 10:12:25 why? 10:12:26 that is pretty memory efficient iirc 10:12:34 uses like 50 MB on phone 10:12:43 while desktop chrome uses way more 10:12:44 wow. so much? 10:12:55 nortti, with a few tabs open yes 10:13:01 wow 10:13:12 few tabs = 10 or so 10:13:13 browsers sure are bloated nowadays 10:13:28 nortti, I do want HTML5 support and so on though 10:13:35 because a lot of websites need it to be useful 10:14:12 nortti, with 1 tab open (front page of xda-developers.com) it uses 31.52 MB 10:14:18 http://kmandla.wordpress.com/2010/05/24/the-1-2kb-python-browser-script/ 10:14:33 there. it supports html5 10:15:12 "But there are no tricks or gimmicks at work here, except to say that the browser is all of about 1.5Kb on disk, written in python and has webkit as its core." <-- except that I'm pretty sure webkit is larger than 1.5Kb 10:15:42 yeah. only the size of browser script is counted 10:15:59 not really fair 10:16:04 chrome embeds webkit after all 10:16:18 btw how can I measure memory use of program under linux 10:17:10 Vorpal: or xxxterm 10:17:16 what is that 10:17:29 btw how can I measure memory use of program under linux <-- I would use htop? 10:17:32 or ps 10:17:36 or top 10:17:40 xxterm is webbrowser 10:17:42 ah 10:17:49 *xxxterm 10:18:14 Vorpal: I mean in megabytes. htop only show percentages 10:18:30 nortti, no? 10:18:36 RES, VIRT and SHR 10:18:55 what do they mean? 10:19:03 well SHR is shared memory iirc 10:19:29 22460 10680 2212 S 0.0 17.9 0:01.81 links -g xda-developers.com 10:19:31 RES is resident, VIRT is address space allocated (not same as memory allocated obviously on linux) 10:19:45 nortti, bytes or kbytes I guess 10:19:49 don't know the unit there 10:19:56 kbytes 10:19:58 and I can't see the column headers either 10:20:37 VIRT RES SHR S CPU% MEM% TIME+ Command 10:20:51 anyway, memory usage is complicated given that stuff like shared libraries (libc.so or whatever) will exist in one copy in memory, shared between all processes on the system using it 10:21:01 well, except for .data and so on 10:21:09 .text and .rodata segments will be shared 10:21:16 Vorpal: I think that my links is staticaly linked 10:22:05 nortti, VIRT is not really interesting. It tells you how much it has allocated (mmap or sbrk. NOT malloc. Your libc provides malloc which uses something like mmap or sbrk to allocate memory from the kernel) 10:22:26 but linux won't actually allocate any physical memory to the program until the program tries to access said memory 10:22:41 before that it only exists potentially 10:22:48 so is SHR the right reading? 10:22:49 nortti, thus VIRT is utterly useless for most purposes 10:23:08 nortti, SHR is shared from what I remember. So that is not all that interesting 10:23:11 also it was dynamicaly linked 10:23:20 * Madoka-Kaname hugs nortti 10:23:31 Madoka-Kaname: why? 10:23:41 anyway it is memory that is shared by more than one process, I'm not sure if that includes ro shared or just rw shared 10:23:43 Because I like hugs. 10:23:47 probably both 10:23:49 Vorpal: what does RES mean? 10:23:52 nortti, resident 10:24:02 nortti, I'm trying to remember how to interpret that 10:24:26 nortti, I don't remember if it includes mmaped files or not 10:24:30 which makes a huge difference 10:24:50 ok. also links is running on 1280x1024x32 framebuffer 10:25:06 does that have any effect in RAM use? 10:25:08 nortti, it might be that RES doesn't include swapped out stuff? 10:25:28 nortti, the ps man page contains the following note: 10:25:30 The SIZE and RSS fields don't count some parts of a process including the page tables, kernel stack, struct thread_info, and struct task_struct. 10:25:30 This is usually at least 20 KiB of memory that is always resident. SIZE is the virtual size of the process (code+data+stack). 10:25:43 not sure about htop 10:26:11 but yeah, the virtual size will include pages you allocated but never touched 10:26:14 which is common 10:26:26 your libc is going to do that for pooling purposes 10:26:43 the top man page: 10:26:45 q: RES -- Resident size (kb) 10:26:45 The non-swapped physical memory a task has used. 10:27:00 also hm: 10:27:01 o: VIRT -- Virtual Image (kb) 10:27:01 The total amount of virtual memory used by the task. It includes all code, data and shared libraries plus pages that have been swapped out. 10:27:01 VIRT = SWAP + RES. 10:27:01 p: SWAP -- Swapped size (kb) 10:27:02 The swapped out portion of a task's total virtual memory image. 10:27:17 HOWEVER, I know for a fact that VIRT includes the untouched memory 10:27:24 hmm.. 10:27:42 does that really go into RES? 10:27:53 pretty sure it doesn't 10:28:05 yeah it doesn't in htop at least 10:28:10 20555 arvid 20 0 858M 173M 39056 S 0.0 9.4 0:31.42 /usr/lib/firefox/firefox http://esolangs.org/wiki/Burlesque 10:28:21 see, 858M VIRT, 173M RES 10:28:59 well I can at least say that running 4 instances of links2 with Xvesa on Pentium 100MHz with 40MB of RAM works fine 10:29:16 20555 9.4 858m 685m 173m 68 364m 38m 326 0 S 20 0 2 firefox 10:29:27 err 10:29:29 PID %MEM VIRT SWAP RES CODE DATA SHR nFLT nDRT S PR NI %CPU COMMAND 10:29:29 20555 9.4 858m 685m 173m 68 364m 38m 326 0 S 20 0 1 firefox 10:29:30 there 10:29:34 from top 10:29:36 wow 10:29:51 however, that SWAP includes stuff not actually swapped out 10:29:58 what? 10:30:02 because my total swap usage is 214 MB 10:30:12 according to free -m 10:30:15 oh come the fucking on 10:30:28 Right, I now have a computer 10:30:33 Now to find some OS's 10:30:39 nortti, I think SWAP there includes swapped out stuff and also the allocated but not actually ever used stuff 10:30:54 Taneb: haiku, syllable, netbsd 10:30:55 Taneb: does your computer have an adder circuit? 10:31:14 nortti, yes memory usage under Linux is /complex/ 10:31:26 now you know that 10:31:31 ok 10:32:05 nortti, for a single process just reading /proc//maps and adding up the relevant segments might be saner :P 10:32:33 $ cat /proc/20555/maps | wc -l 10:32:33 752 10:32:35 that is firefox 10:32:36 lol 10:32:44 a /LOT/ of mmaped stuff 10:33:14 lets filter out ro mappings 10:34:04 hrrm 10:34:28 itidus21, quite possibly! 10:34:47 juhani@T20-slitaz$ cat /proc/5337/maps | wc -l 10:34:47 42 10:34:49 it shares most libraries with other apps, but some of them only firefox uses 10:34:54 hkallasj@valdieu:~ cat /proc/737/maps | wc -l 10:34:55 that is links 10:34:55 869 10:35:02 That's also a Firefox. 10:35:11 nortti, that doesn't tell you the amount of memory, just number of mapped regions 10:35:29 7fc4fcc87000-7fc4fce86000 ---p 00018000 fc:00 22921 /lib/libpthread-2.11.1.so 10:35:34 what is the point of THAT mapping? 10:35:36 juhani@T20-slitaz$ cat /proc/5451/maps | wc -l 10:35:36 126 10:35:41 that is hv3 10:35:42 a private mapping with no permissions 10:35:42 why 10:35:57 nortti, the interesting bit is really the size of the allocations 10:36:14 Sometimes there are ---p guard pages, I suppose. 10:36:34 7fc4e3c70000-7fc4e3c80000 rwxp 00000000 00:00 0 <-- oooh I think firefox is JITing something eh? 10:36:40 It's not like the mapping costs much, and it catches writes before they go mungle something else. 10:37:00 fizzie, right 10:37:09 Could be something completely different too. 10:37:26 $ cat /proc/20555/maps | grep -Ev '[-r]-[-x][sp]' | wc -l 10:37:26 238 10:37:26 That one for example is a bit big. 10:37:43 (For a guard page, I mean.) 10:37:52 what size is it 10:37:55 I saw a few single-page ---p nameless things. 10:37:56 too lazy to calculate that 10:38:05 The 00018000 is the size, isn't it? 10:38:11 oh right 10:38:28 with my os I'm going to have this syscall: therealgoddamnedmemoryuse(PID) 10:38:33 Or maybe not. 10:38:39 It just seemd to be in the right ballpark. 10:38:45 fizzie, so 24 pages? 10:38:48 hm 10:39:09 Oh, it's the offset field. 10:39:10 fizzie, anyway it could be that the kernel has that mmaped to the process but the process hasn't yet used it 10:39:13 For files. 10:39:16 oh okay 10:39:22 so not the size then 10:39:25 Right. 10:39:40 ~ cat /proc/737/maps | grep -c rwx 10:39:40 76 10:39:45 Quite a lot of rwx in there too. 10:39:53 fizzie, if you put it as no access you are going to get a page fault on access, thus you can map it lazily 10:39:54 Software these days, so complicated. 10:40:04 I guess that could be what is going on 10:40:06 Yes, it could be that. 10:40:23 with links: 10:40:24 juhani@T20-slitaz$ cat /proc/5337/maps | grep -c rwx 10:40:24 1 10:40:37 nortti, the conclusion is that calculating memory usage on Linux (or any modern OS) is utterly hard 10:41:50 what... firefox has no [heap 10:41:53 [heap] 10:41:55 how does that work 10:42:06 what the crap 10:42:22 does firefox not use sbrk? And bypass glibc malloc? 10:42:40 I would assume glibc performs some internal allocations using sbrk at least 10:42:42 how weird 10:42:50 fizzie, go figure that out ^ 10:43:39 hm all the low address mappings are weird for firefox compared to other processes 10:43:41 I'm guessing something related to their portability layers, but I'm no FF developer. 10:43:54 $ cat /proc/$$/maps 10:43:54 00400000-004db000 r-xp 00000000 fc:00 3255 /bin/bash 10:43:54 [...] 10:44:04 firefox: 10:44:06 7fc4c2a00000-7fc4c2c00000 rw-p 00000000 00:00 0 10:44:10 is the lowest one 10:44:32 I've got a couple usual ones in my Firefox process. 10:44:33 00400000-00410000 r-xp 00000000 fd:00 1194314 /usr/lib64/firefox/firefox 10:44:36 in fact firefox itself: 10:44:37 00610000-00611000 rw-p 00010000 fd:00 1194314 /usr/lib64/firefox/firefox 10:44:38 7fc4fd0ae000-7fc4fd0bf000 r-xp 00000000 fc:00 430071 /usr/lib/firefox/firefox 10:44:50 fizzie, which version of firefox is that? 10:45:00 14.0.1 here 10:45:27 The about window says "Firefox ESR 10.0.6", I don't have a clue what that means. 10:45:31 Older, anyway. 10:45:51 00400000-00417000 r-xp 00000000 08:03 2914597 /usr/lib/firefox-beta/firefox 10:45:51 You can get a fancy memory allocation tree thing on Firefox with about:memory. 10:45:52 00616000-00617000 rw-p 00016000 08:03 2914597 /usr/lib/firefox-beta/firefox 10:46:25 144 MB 10:46:30 81 MB for js 10:46:46 14 MB for sqlite... 10:46:50 about:memory?verbose for more detail 10:47:34 pressing the "minimize memory usage" button at the bottom made it 133 MB 10:48:00 and loading that verbose version made it 146 MB 10:48:13 and I only have 8 tabs open 10:48:20 not very heavy pages either 10:48:31 Vorpal: The "heap-unclassified" memory is apparently called 'dark matter' by Firefox people. 10:48:32 hm can I see if any addons uses a lot of memory? 10:48:40 fizzie, dark matter eh? 10:48:41 what is it then 10:48:52 I don't know, and maybe they don't either. :p 10:48:58 and that is 32 MB approx 10:49:26 Things that don't have proper about:memory "reporters", apparently. 10:49:46 77.40 MB (28.44%) -- heap-unclassified 10:49:58 compartment([System Principal], 0x7fc4e9ff1000) uses the majority 10:50:06 this is under js 10:50:09 46 MB 10:50:12 single largest post 10:50:33 hm a post under that called gc-heap is 32 MB 10:51:08 31 MB when I closed all the other tabs 10:51:42 I would like to know what "System Principal" is 10:52:50 libxul (whatever that is) has a massive .text segment it seems 10:52:57 18 MB 10:53:07 Is the a Haskell Platform for Haiku? 10:53:11 XUL is the user interface stuff Firefox uses. 10:53:37 ah 10:53:47 XML User Interface Language. 10:53:48 Taneb, s/the/there/? 10:53:51 ouch 10:53:55 why not XUIL 10:53:58 Vorpal, yes 10:54:01 Maybe it was taken. 10:54:04 It rings a bell. 10:54:08 Maybe the U stands for UI 10:54:31 imho XUL suck. I am still bit angry at mozilla discotinuing gecko embedding 10:55:33 There are all kinds of "there is only XUL" jokes. 10:55:37 -!- copumpkin has quit (Ping timeout: 255 seconds). 10:55:43 like? 10:55:52 XUL's XML namespace URI is http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul for example. 10:56:01 :P 10:56:07 hm it did list one of the addons in that memory list (noscript) but that was just 6 kb 10:56:09 -!- copumpkin has joined. 10:56:21 > keymaster/gatekeeper 10:56:22 Not in scope: `keymaster'Not in scope: `gatekeeper' 10:56:38 oh wait, there are two entries for noscript, so ~300 KB 10:57:08 yeah the total of the addon related stuff listed there is less than 5 MB 10:57:09 It says "There is no data. There is only XUL!" if you actually HTTP-fetch that. 10:57:49 fizzie, ouch 10:58:55 XUL (pronounced "zool") 10:58:58 ....wot 10:59:05 since when do x's make a z sound? 10:59:07 :P 10:59:28 The pronunciation is also part of the joke. 10:59:46 I don't get it. 10:59:59 spirity: never watched ghostbusters? 11:00:02 It's a reference to the "There is no Dana, only Zuul" line. 11:00:04 In that. 11:00:13 btw, mobile firefox is a joke on a phone with 1 GB RAM. It works. Barely 11:00:21 oh. okay. 11:00:34 but you end up with so much other stuff getting killed by the system so everything gets slow when switching apps 11:00:42 Vorpal: wow. microb works on n810 11:00:47 nortti, microb? 11:00:51 spirity, never seen a xylophone? 11:01:12 fizzie, what was "zuul"? 11:01:20 Vorpal: or was it micro-b. the meamo browser based on gecko and hildon 11:01:25 nortti, ah 11:01:30 Taneb: what are those? 11:01:33 nortti: It's MicroB. 11:01:36 nortti, what is n810? I thought it was n900? 11:01:40 Musical instrument 11:01:46 Percussion-y keyboard-y 11:01:47 Vorpal: Those are the predecessor non-phone tablets. 11:01:51 ah 11:01:51 You hit bits of metal with sticks 11:02:12 Vorpal: nokia produced four maemo devices. 770, n800, n810 and n900 11:02:12 And Firefox itself seemed to work passably on the N900 (which has 256M of RAM) last I tried. 11:02:17 nortti, ah 11:02:33 fizzie, well, firefox used to work better on this laptop too 11:02:45 I think modern firefox is more bloated than a few versions ago 11:03:01 true 11:04:01 http://www.areweslimyet.com/ 11:04:29 it especialy shows when you do same kind of switch like I did. (Bon Echo+patches -> TenFourFox 4) My computer was so much faster but browser was a bit _slower_ 11:04:31 Deewiant: I read that at least thrice as "are we slime yet". 11:04:45 nortti, hm? 11:05:07 Bon Echo=Firefox 2 11:05:23 TenFourFox 4=Firefox 4 11:05:39 oh. I did that switch in 2011 11:05:46 summer 11:06:50 ah 11:07:34 well actually the Bon Echo I used was based on Iceweasel 2 but the difference isn't that big 11:07:34 -!- MDude has quit (Read error: Connection reset by peer). 11:07:44 -!- MDude has joined. 11:08:00 Deewiant, what is that TP5? 11:08:08 Vorpal: It does have a FAQ button. 11:08:13 ah right 11:08:18 (It's 100 popular pages selected from Alexa top 500.) 11:08:44 (Scrubbed to be self-contained and so on.) 11:10:34 "But if Firefox has freed all the memory it allocated during the test, why is it using more memory after the test? Our data shows that most of the difference is due to heap fragmentation. Before the test, all the objects on our heap are tightly packed. But after the test, our heap uses twice as much space for the same amount of storage, because the objects on the heap now have gaps between them." <-- so 11:10:34 use a compacting GC... 11:11:47 What, with C++? 11:13:47 "This is all well and good, but my Firefox still leaks like a sieve" is the best FAQ. 11:14:03 :P 11:14:07 It's also probably the most FAQ. 11:15:06 Silverfish are a thing that actually exists? 11:16:26 firefox doesn't exactly leak I think... It just uses a lot of memory. The amount doesn't increase that much over time, only a bit 11:16:39 yes, but you can't get your omega 3 from them 11:16:57 Sgeo_, yes but they are not fish as such 11:17:04 some sort of bugs 11:17:10 check wp 11:17:27 oh insects 11:17:53 i like my reference 11:18:12 to the way fish is sold as a provider of omega 3 11:18:40 took a look on what chrome on my desktop uses. 1.6 GB "private memory" (I'm booted to windows atm, I have no clue how the memory works under that) 11:18:57 wait, misread, 165 MB 11:18:58 XD 11:19:06 or wait a second 11:19:09 :P 11:19:27 well 175 MB is what windows task manager claims in the private column 11:19:46 oh no, it's both 165mb and 1.6gb in a quantum superposition 11:19:59 1,633,060k is what chrome claims under chrome://memory-redirect (redirect? what?) 11:20:09 ... 11:20:52 I have about 30 tabs or so open. Some quite heavy pages 11:21:06 my advice is get comfortable with browser-refreshing, and deprioritizing open tabs. 11:21:16 itidus21, what? 11:21:26 deprioritizing tabs? 11:21:30 itidus21, anyway on my desktop even 1.6 GB would be a non-issue 11:21:36 "chrome://memory 11:21:38 This will redirect to “chrome://memory-redirect/”." 11:21:40 by browser refreshing i mean closing a browser, and opening it again 11:21:40 That's kinda weird. 11:21:48 fizzie, so does about:memory in chrome 11:21:49 hmm 11:21:52 the issue is firefox on my laptop 11:21:57 with 2 GB RAM 11:22:14 my desktop with 16 GB RAM... I just don't care if my browser is using a GB RAM or not 11:22:15 and by deprioritizing open tabs i mean, occasionally go through your tabs and close the unnecessary ones 11:22:57 Vorpal: Yeah, well, http://sprunge.us/hcYF 11:23:16 ouch 11:23:21 $ free -m 11:23:22 total used free shared buffers cached 11:23:22 Mem: 1838 1044 793 0 47 381 11:23:22 -/+ buffers/cache: 615 1223 11:23:22 Swap: 2045 214 1830 11:23:23 on my laptop 11:23:24 I once made a mistake of accidentally starting up MATLAB locally on this box. I barely lived to tell the tale. 11:23:30 heh 11:23:45 Maybe I should just run the browser on some other computer too, really. 11:23:53 Vorpal: well try these: Kazehakase, Amaya, xxxterm and http://kmandla.wordpress.com/2010/05/24/the-1-2kb-python-browser-script/ 11:23:58 wait, why do I have 2045 MB swap? 11:24:17 The box I run MATLAB at the moment has 48G of memory, I'm sure nobody would notice even a Firefox. 11:24:45 /dev/sda2 193 205 98280 83 Linux 11:24:45 Partition 2 does not end on cylinder boundary. 11:24:46 Admittedly it has 46G of those in use. 11:24:47 interesting 11:24:53 it doesn't? 11:24:58 hmm 11:25:03 and why is that important... 11:25:10 fizzie, hah 11:25:30 3 1680MB 3824MB 2144MB primary linux-swap(v1) 11:25:31 okay? 11:25:38 that last MB number is the size 11:25:40 from parted 11:25:57 is this a 1000 vs. 1024 thingy? 11:26:04 seems my firefox mem use is constantly creeping up nonstop 11:26:21 itidus21, slowly or quickly? 11:26:34 slowly... 4kb at a time 11:26:36 itidus21, use about:memory and the buttons at the bottom 11:26:51 and that is one page at a time 11:27:22 hm how do you setup a compiler toolchain to compile native android programs. Like command line tools... 11:27:46 Vorpal: https://developer.android.com/tools/sdk/ndk/index.html 11:28:02 Deewiant, I thought that could only do *.so for JNI stuff? 11:29:14 I do not know 11:31:04 what was the name of that software that allowed you to connect to existing X11 sessions 11:31:06 hm 11:31:12 I'm under the impression that http://source.android.com/ have a prebuilt toolchain, and in any case the source distribution should have everything necessary. 11:31:55 It may be somewhat more involved than just doing regular Android development or using the NDK. 11:31:56 fizzie, I recently checked that out to find a specific stock icon... Wow did it take ages... 11:32:10 16 GB or something like that IIRC 11:32:18 removed the source tree after I was done 11:33:02 it claims it is 6 GB, I guess git is to blame for the extra bulk 11:33:30 "For Gingerbread (2.3.x) and newer versions, including the master branch, a 64-bit environment is required. Older versions can be compiled on 32-bit systems." heh, why 11:33:47 and you need the sun jdk specifically? 11:33:49 wow 11:34:13 Vorpal: look at $ndk_root/build/tools/make-standalone-toolchain.sh 11:34:20 ah thanks 11:35:34 Okay, I accidentally installed a copy of Rome: Total War on my Haiku memory stick 11:36:07 :P 11:36:16 Vorpal: Can you even run plain old regular native apps on those things? 11:36:35 fizzie, from a shell? sure 11:36:52 there are terminal emulators for it, and you can use adb shell from a computer to connect to it 11:37:29 fizzie, it is by default a somewhat crippled environment. I installed busybox by doing cat /mnt/sdcard/busybox > /system/xbin/busybox because there was no cp 11:37:38 Taneb, how? 11:37:39 but if you are coding native apps for android remember that bionic libc is broken 11:38:01 nortti, well I just wanted to compile a few useful tools 11:38:11 Vorpal: like? 11:38:12 such as an scp, maybe a dropbox one or something 11:38:15 err 11:38:17 dropbear* 11:38:24 Busybear. 11:38:31 Vorpal: you can try statical linking 11:38:37 sure 11:38:39 Pedobear. 11:38:47 there is the dropbear ssh server 11:38:53 don't know if it provides a client 11:39:04 Dropbear has a client too IIRC. 11:39:08 I suspect openssh's scp would be painful to get to work 11:39:09 I think it does 11:39:32 I had ssh on this computer when I only had dropbear installed 11:39:37 Vorpal, the thingy I was writing the image doesn't like writing images, and prefers to write copies of CD's? 11:39:42 -!- copumpkin has quit (Ping timeout: 264 seconds). 11:39:50 Taneb, ? 11:40:07 The image writer copied the CD I had in onto the memory stick instead of the ISO 11:40:13 -!- copumpkin has joined. 11:40:32 I see 11:41:15 I should probably install 'less' on the phone one day, it's so often I try to |less accidentally. 11:41:32 doesn't android have more? 11:41:44 I don't know what Android has. 11:41:54 I have a 'more', but I often try 'less' instead. 11:41:59 oh. you have maemo 11:42:05 Yes, it's a N900. 11:42:05 nortti, nope! 11:42:18 at least meamo has busybox 11:42:18 I symlinked less to my copy of busybox of course 11:42:32 which is busybox with mostly everything in it 11:42:47 Possibly my aversion to 'more' is leftover paranoia from a DOS box where MORE.COM rebooted the computer. 11:42:56 fizzie, it did? 11:43:01 It was somehow broken. 11:43:05 Vorpal: does installing busybox require rooting the phone 11:43:06 anyway more is inferior to less 11:43:06 It doesn't normally. 11:43:43 nortti, no, but you need to put the executable somewhere that has proper permissions (so not fat32 sdcard...) 11:43:54 How do you get the CLI on an android? 11:44:05 Taneb, adb shell on the computer? adb is a program from the SDK 11:44:16 Okay 11:44:20 or you use an on-phone terminal emulator, there are a bunch of open source one on the market 11:44:37 nortti, non-root adb shell does however not have write access to any such location! 11:44:43 You know, if your right hand is too far to the right when you type "Okay", you get "Play" 11:44:46 so you use a terminal emulator on the phone 11:44:56 and put busybox in the private app dir of that app 11:45:02 and then make it world readable 11:45:07 and possibly writable 11:45:13 then you can use it from adb shell 11:45:16 ok 11:45:49 There's a busybox in my VDSL2 modem (a GPL violation, in fact, I believe) that's really crappy. It's almost like they haven't designed it for shell use. (They haven't.) 11:45:53 then I rooted my phone and just put it in the path instead 11:46:02 much more convenient 11:46:34 Taneb, wait, which hand do you use for y? 11:46:42 Left 11:46:45 Vorpal: what terminal emulator should I use? where is the private app dir? 11:46:49 Taneb, I use my right hand for that 11:46:57 since it is on the right half of the keyboard 11:47:03 I never really learnt to type properly? 11:47:06 ah 11:47:29 nortti, well, I used the one called "Terminal Emulator" XD 11:47:54 :P 11:48:07 Vorpal: where is the app dir located? 11:48:10 sec 11:48:22 nortti, oh it is "Android Terminal Emulator" on google play 11:48:26 ok 11:48:27 blue icon 11:48:32 kay 11:48:51 nortti, and /data/data/jackpal.androidterm/ 11:48:59 for that one 11:49:06 nortti, for a different one the last bit will vary of course 11:49:13 ok 11:49:17 since that is the package name of the app 11:49:49 I use sl4a or connectbot as my terminal emulator 11:49:54 nortti, there is a dedicated "install busybox" app on the market too, haven't used it myself 11:50:00 ok 11:50:04 nortti, you could do the same for connectbot (which I also use, for ssh) 11:50:09 just a different package name? 11:50:15 ok 11:50:26 let me check what it is, I have the code open in eclipse (wrote a patch for connectbot today) 11:50:51 org.connectbot apparently 11:51:26 nortti, does connectbot have sdcard access permissions though? 11:51:33 no idea 11:51:36 oh yes it does 11:51:39 well my copy does anyway 11:51:44 but that is git head 11:52:01 -!- Jafet has quit (Quit: Leaving.). 11:52:25 nortti, I got my busybox copy from the S3 root toolkit files. Seemed easiest at the time 11:53:01 most pointless HTML file ever: /opt/android-ndk-r8b/docs/OVERVIEW.html 11:53:10 it is just a text file wrapped in
11:53:10  XD
11:53:29  :P
11:53:56  there is no non-html version though
11:57:00  Everything has to be html, don't you know that ;)
11:57:32  OSses of the future are just a kernel and a browser .
11:57:48  do not want
11:58:02  Future does not care what you want ;)
11:58:14  NDK_DIR=`dirname $0`
11:58:14  NDK_DIR=`dirname $NDK_DIR`
11:58:14  NDK_DIR=`dirname $NDK_DIR`
11:58:15  ouch
11:58:44  mroman: I'll build my OWN computer with µcontrollers! 
11:58:45  this won't work when invoked with a relative path I think
11:58:47  ouch
11:59:14  nortti, wirebonding and TTL logic?
11:59:29  yes
11:59:43  err not wirebonding
11:59:46  that is the wrong word
11:59:48  (01:13:04 PM) fizzie: What, with C++? <- C++11 allows GC
11:59:53  wirewrapping?
12:00:02  nortti, yes that is it
12:00:14  AnotherTest, compacting GC?
12:00:16  AnotherTest, HOW?
12:01:01  anyway I think you could include a GC in C/C++... The compiler has to embed type information though so you can know what is a pointer and what isn't
12:01:21  and it would still be easy to break it by doing stuff many programs do but which isn't strictly conforming
12:01:41  People actually use that Boehm GC sometimes, I believe.
12:01:52  It's one of the conservative guesswork mostly-works ones.
12:02:38  "However C++11 adds a few restrictions to implementations so that some behavior that would prevent garbage collection to work is now disallowed. This includes in particular common ways to "hide" pointers from a possible garbage collector, like applying xor to it."
12:02:44  http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2011/n3242.pdf
12:02:44  Yeah, I recall they did something like that.
12:02:46  fizzie, boehm gc is not compacting
12:02:52  Vorpal: I didn't say it was.
12:03:06  You just said "include a GC".
12:03:14  right
12:03:19  fizzie, I meant a compacting one
12:03:23  that is actually not slow
12:03:36  I used boehm, it is slow
12:03:50  old versions of cfunge supported it as an option
12:04:17 -!- monqy has joined.
12:04:27  A 1334-page PDF is the best response to a question like that.
12:05:02  It's a working paper for the standard
12:05:22  I'm however 99% sure that cfunge has no actual memory leaks. There are some stuff that will never shrink due to poor fingerprint design or the stuff being kept around in a pool for the future, but no actual leaks.
12:07:06  http://www2.research.att.com/~bs/C++0xFAQ.html#gc-abi
12:07:17  is probably a better answer
12:08:45  " There are legitimate reasons to disguise pointers (e.g. the xor trick in exceptionally memory-constrained applications), but not as many as some programmers think. "
12:08:50  what xor trick?
12:09:02  and why would it help with memory-constrained applications
12:09:07  Vorpal: the xor trick
12:09:30  olsner, yes
12:09:32  what is that
12:09:49  Vorpal: You can do a doubly-linked list with just one pointer.
12:10:01  fizzie, oh, nice
12:10:06  right
12:10:11  Vorpal: Assuming you always know one of the neighbours, the one you're coming from.
12:10:34  well obviously
12:10:59 -!- MoALTz has joined.
12:11:05  It also means if you give someone a plain "Node*" they can't iterate in either direction, of course.
12:11:10  But them's the breaks.
12:11:26  indeed
12:11:36  fizzie, kind of useful in some embedded systems I guess
12:12:01  I haven't seen it live, but it's quite often mentioned as an example in these discussions.
12:12:05 -!- ogrom has joined.
12:13:17  nortti, you were putting busybox on your phone?
12:13:21  nortti, how did that work out?
12:13:23  You can remove the pointer overhead by as much just by unrolling two (or more by unrolling a few more) list elements in a single node, anyway.
12:13:35  s/remove/lower/ or whatever.
12:13:37 -!- Jafet has joined.
12:13:40  fizzie, true
12:13:44  Reduce is the word I was thinking of, probably.
12:13:51 -!- ogrom has quit (Client Quit).
12:15:12  huh, now MTP started working under linux
12:15:16  when I started rythmbox
12:15:18  wow
12:15:43  not that it started working /IN/ rythmbox, no it opened nautilus with the phone
12:15:50  err the phone with nautilus
12:16:07  and it only shows the content after I close rythmbox
12:21:14  AnotherTest: You can generate fibonacci sequence by chaining ^^++[+[-
12:21:34  blsq ) {2 3}{^^++[+[-}e!
12:21:35  {3 5}
12:22:00  blsq ) {2 3}{^^++[+[-}2.*\[e!
12:22:01  {5 8}
12:22:29  :)
12:22:57  `run tclsh
12:23:07  bash: tclsh: command not found
12:23:19  blsq ) {0 1}{^^++[+[-}9.*\[e!-]
12:23:20  34
12:23:26  ^- 9th fibonacci number
12:23:43  blsq ) {0 1}{^^++[+[-}10.*\[e!-]
12:23:44  55
12:23:50  ^- 10th fibonacci number and so on.
12:24:55  It calculates the sum of the list, appends to sum to the list, and removes the first element of the list.
12:25:06  *the sum to
12:25:26  how would you calculate a square root?
12:26:51  blsq ) 1{1.+}{^^.*81==1.-}w!
12:26:51  9
12:27:18  increment it, duplicate, multiply, compare with the number
12:27:25  We could use that to generate the nth fibonacci number
12:27:43  for(i=1;i*i != 81;i++) 
12:27:56  Hey, no wonder this box is slow; a gigabyte of memory, and someone else has a full gnome desktop also running in a Xvnc on it.
12:28:06  1.- is the not, here :)
12:28:11  although it's not really a not of course.
12:29:35  What kind of a language is that?
12:30:01   increment it, duplicate, multiply, compare with the number <-- that sounds like a slow square root implementation
12:30:16  Vorpal: Indeed.
12:30:24  mroman, besides what will it do when the square root is not an integer?
12:30:38  fizzie, talk to them about it?
12:31:02  Stlang has a sqrt builtin, so Burlesque get's one too
12:31:22  It's just not available *yet*.
12:31:25  I see
12:31:33  mroman: if you can round, fibonacci would get really easy
12:31:38  anyway how is sqrt normally done on real hardware?
12:31:43  floating point sqrt that is
12:31:46  You could just use the formula
12:31:54  Vorpal:
12:32:01  My calculator uses newton's method
12:32:05  hm okay
12:32:08  I think
12:32:16  what does glibc do
12:32:22  The newton tangens thingy?
12:32:31  I'm not familiar with that one
12:32:34  Vorpal: I don't know who it is.
12:32:53  AnotherTest: round floats?
12:33:06  I mean, I know the user name of course, but I hardly know anyone here.
12:33:07  mroman: yes, for Fibonacci
12:33:30  The tails/replicate version works, but will get slower of course :)
12:33:49  I'd need to add Floats first :)
12:33:56  ah okay
12:34:01  But
12:34:08  Gimme 10 minutes and it's done :D
12:34:24  also, could you make integers that can be larger
12:34:30  like haskell's Integer
12:34:36  I could switch to Integer
12:35:05  because ri^^{\/^^.+\/1.-}{}w!vv with 100000 gives me 0
12:35:45  and 31 gives me
12:35:51  -2147483648
12:36:03  which both look like an overflow :(
12:36:10  Yes.
12:36:12  Int is bounded.
12:37:29  although certain haskell functions onyl take Int
12:37:42  so If I use Integer I have to "downsize" it to Int for some functions
12:38:17  mroman, or avoid those functions
12:38:30  implementing replacements for them
12:38:47  or that, yes.
12:39:39  mroman: I'm writing HELP (Handy esoteric language preprocessed) for blsq :D
12:39:58  *preprocessor
12:42:26  Doesn't Data.List have generic functions
12:42:30  which work on Integral stuff.
12:43:05  ah
12:43:08  genericTake
12:47:26  https://github.com/FMNSSun/Burlesque/commit/0fe0340d664617e65fc3f92a2555485c24ed7a48 <- Integer
12:49:08 -!- Jafet has quit (Quit: Leaving.).
12:52:50 -!- asiekierka has quit (Remote host closed the connection).
12:58:48  hm NEON registers are mapped into the same memory as the FPU registers on ARM
12:58:48  ouch
12:58:51  that is like MMX
12:58:53  terrible
12:59:19 -!- kallisti has quit (Ping timeout: 248 seconds).
12:59:41  blsq ) 81.0 0.5 **
12:59:42  9.0
13:06:45 -!- kallisti has joined.
13:06:45 -!- kallisti has quit (Changing host).
13:06:45 -!- kallisti has joined.
13:11:41 -!- nooga has quit (Ping timeout: 245 seconds).
13:18:55  AnotherTest: blsq ) 3.1459 3 r_
13:18:55  3.146
13:19:18  Does HELP support #define,#include stuff?
13:19:58  Vorpal: I haven't tried it yet
13:21:59  mroman: so far it's a simple macro preprocessor, I plan to add more features such as #include
13:23:03 -!- Tod-Autojoined has changed nick to TodPunk.
13:24:27  mroman: eventually it should have much more than cpp
13:24:47  using the closed fibonacci formula is not really practical on any physical machine :)
13:24:56  it diverges pretty soon.
13:25:05  somewhere around 40
13:25:21  :(
13:25:33  What did you expect?
13:25:36  nortti, ah
13:25:43  cpus have limited precision.
13:26:21  nortti, I'm going to work on getting various useful tools compiled for my phone. If you want copies once I get them working I will happily share if your ABI is compatible
13:26:33  I'm going for ARMv7 with NEON here
13:26:34  If you stored it in ExtraLongFloat?
13:26:44  well that'd be BigDouble
13:26:54  We have arbitrary precision floating point code.
13:27:07  Would only push the problem a little bit further into the future ;)
13:27:12  Vorpal: what processor you have them compiled to?
13:27:18  Madoka-Kaname: Like gmp?
13:27:42  Something like that
13:27:45  gmp also only has a fixed precision afaik.
13:27:45  nortti, Well I read the NDK docs now and it seems I want ARMv7 with NEON
13:27:54  nortti, since my phone supports that
13:27:56  you can initialize a float with 100000000000000 bit precision
13:27:58  but it's still fixed.
13:28:02  I don't think you can even target more specifically than that
13:28:06  Vorpal: well my phone is ARMv6 ...
13:28:23  nortti, ouch. You have to do your own compiling then (or find suitable versions)
13:28:33  mroman, You could do something like
13:28:44  log2(n)*32 bits percision
13:28:55  nortti, how old is it?
13:29:10  I bought it new in spring 2011
13:29:14  hm
13:29:18  nortti, low end I presume?
13:29:23  yeah
13:29:27  HTC Wildfire
13:29:37  pretty sure that even the N900 has an ARMv7
13:29:40  fizzie, is that right?
13:31:22  I no longer use it as my main phone because after I accidentaly dropped it in sink full of water, disassembled it and let it dry it no longer recognises my SIM card
13:32:13  nortti, oh
13:32:18  nortti, so what is your main phone?
13:32:28 -!- edwardk has joined.
13:32:31  Vorpal: Yes.
13:32:39  fizzie, and how old is that model?
13:32:45  oh and does it have NEON?
13:32:57  Vorpal: some old Nokia phone I picked up at second hand shop for 8 euros
13:33:15  Vorpal: It does have NEON, and it's from about 2010 I suppose.
13:33:36  fizzie, ah
13:33:44  nortti, yeah your phone was definitely low end :P
13:33:51  yeah
13:34:09  nortti, no hardware fpu
13:34:24  It's the same OMAP3 platform model that's in e.g. Motorola Droid and Palm Pre, if you know those things.
13:34:25  (or even if it has one, android ndk won't use it)
13:34:38  fizzie, not really no
13:35:17  note to self: mash the volume down button on the computer that the headphones are actually connected to next time
13:36:48  okay, now to compile file(1)
13:37:07  first step: finding the source
13:37:23  Vorpal: try stealing the sash version. it is self contained
13:37:38  my busybox lacks file
13:37:39  hm
13:37:46  nortti, any features lacking?
13:38:14  Vorpal: well it has builtin database but nothing that I can tell
13:38:52  also building sash can be good thing to do regardless. it is bit nicer than android toolbox
13:39:10  nortti, I do have busybox on it, and I very much like busybox
13:39:19  it is just that my busybox lacks file
13:39:50  hm apt-get source file might work
13:40:40  Vorpal: oh. I thought that you were installing busybox. sash has for example builtin gzip/gunzip, chmod, cp and tar
13:40:55  nortti, no I installed busybox earlier with a prebuilt version of it
13:41:02  ok
13:41:03  like two weeks ago
13:41:24  but file is such a useful command when poking around :P
13:41:38  true
13:41:58  yay, massive ubuntu patch from apt-get source
13:41:58  actually my busybox on  my main computer also lacks file
13:42:15  74 patch files
13:42:23  file is just a shell script that invokes sash
13:42:24  I'm not sure busybox even has file
13:42:51  I want my GNU file (or whatever, is the usual file GNU?)
13:48:11 -!- ogrom has joined.
13:49:30  So
13:49:43  Now I'm gonna write me a syntax highlighter html-format for Burlesque
13:49:49  so I can write a cookbook for it :D
13:49:53  why?
13:50:05 -!- MoALTz has quit (Read error: Connection reset by peer).
13:50:41 -!- MoALTz has joined.
13:50:53  dammit, what is the --build and --host for android
13:51:43  are you trying to cross-compile some autotools thing for Android?
13:51:46  olsner, yes
13:51:50  olsner, file
13:52:07  and the --host from the cross compiler's -v output was unrecognized
13:52:19  write a small shell script to run adb pull and run your host's file command on it
13:52:30  that is arm-linux-androideabi
13:52:43  olsner, I want it from the terminal emulator on the device :P
13:53:53  fwiw, this problem sounds reasonably googlable :)
13:54:01  hm
13:54:03  true
13:57:00  Vorpal is actually incapable of googling
13:57:17  if you look through the logs you'll see many instances of this unfortunate syndrome.
13:57:30  `? Vorpal
13:57:33  Vorpal is really boring. Seriously, you have no idea.
13:58:09  `? olsner
13:58:12  olsner? ¯\(°_o)/¯
13:58:31  `ls wisdom
13:58:34  ​? \ ais523 \ augur \ banach-tarski \ c \ cakeprophet \ category \ coffee \ comonad \ coppro \ egobot \ elliott \ endofunctor \ esoteric \ europe \ everyone \ finland \ finns \ fizzie \ flower \ friendship \ functor \ fungot \ glogbot \ gregor \ hackego \ haskell \ hexham \ ievan \ intercal \ internationale \ itidus20 \ itidus21 \ kallisti \ lens \ lifthrasiir \ mad \ misspellings of croissant \ monad \ monads \ monoid
13:58:34  aha, there is a newer config.sub
13:58:40  `? ?
13:58:44  ​? is wisdom
13:58:54  `? misspellings of croissant
13:58:57  misspellings of crosant? ¯\(°_o)/¯
13:59:01  `ls wisdom/lifthrasiir
13:59:04  wisdom/lifthrasiir
13:59:11  `cat wisdom/lifthrasiir
13:59:14  lifthrasiir is shunned by the rest of his country for being no good at League of Legends.
13:59:17 -!- MoALTz has quit (Ping timeout: 255 seconds).
13:59:19  huh.
13:59:26  `cat bin/?
13:59:30  ​#!/bin/sh \ topic=$(echo "$1" | tr A-Z a-z) \ [ -e "wisdom/$topic" ] || { echo "$1? ¯\(°_o)/¯"; exit 1; } \ cat "wisdom/$topic" \
13:59:40 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
13:59:53  `? banach-tarski
13:59:56  ​"Banach-Tarski" is an anagram of "Banach-Tarski Banach-Tarski".
14:01:16  `? coffee
14:01:18  hmm, isn't banach-tarski effectively saying the same thing as 2*Inf = Inf?
14:01:19  Coffee is a strange hot brown liquid, often consumed, sometimes with milk and sugar. It contains chemicals considered stimulants.
14:01:33  `? finns
14:01:36  Finns are helpful, albeit grossly overpopulated (cf. 'Finland').
14:01:39  olsner: but there's more balls involved.
14:02:28  dammit, I selected --enable-static and got a dynamically linked binary
14:02:29  why
14:02:52  --enable-static is short for --enable-static=no
14:02:57  try --static
14:03:03  olsner, really?
14:03:16  Vorpal: yes, I googled it for you
14:03:21  how do you read the list of dynamic libraries without using ldd (which obviously won't work cross-platform)
14:03:27  olsner, I don't believe it
14:03:32  Vorpal: then don't!
14:03:45  oh it is only the library
14:03:47  hm
14:04:22  Good Guy Greg should be a religion.
14:04:31  nortti, unknown option. I guess I need to mess with CFLAGS and/or LDFLAGS
14:04:35  for atheists
14:04:56  Vorpal: are those arguments for configure?
14:04:56  hmm, a way to list of dynamic libraries without ldd sounds useful... obviously the information is in there, in some section related to dynamic linking
14:05:01  nortti, yes
14:05:14  Vorpal: CFLAGS=--static
14:05:19  nortti, and it is -static for gcc I'm pretty sure (as in, just one -)=
14:05:23  s/=/?/
14:05:28  olsner: you'll need to find it for Vorpal
14:05:32  otherwise there's no hope for him
14:05:33  yeah it is
14:05:38  Vorpal: well gcc seems to accept both
14:05:43 -!- edwardk has quit (Quit: Computer has gone to sleep.).
14:06:12  spirity: ahh modern religion might prove be kick ass, hence they won't actually call it religion
14:06:50  well Good Guy Greg inspires what is essentially humanism.
14:07:24 * spirity should switch majors to English and then right a thesis on Good Guy Greg.
14:07:45  Well Good Guy is becoming a group now
14:08:15 -!- azaq23 has joined.
14:08:26 -!- azaq23 has quit (Max SendQ exceeded).
14:08:34  “Being a Humanist means trying to behave decently without expectation of rewards or punishment after you are dead.”
14:08:37  ― Kurt Vonnegut 
14:08:58 -!- azaq23 has joined.
14:09:46  yay, my file works
14:10:10  yay
14:10:18  Vorpal: your file is just, like, data man.
14:10:34  spirity: command named file
14:10:43  spirity: on the surface it's easy
14:11:17  nortti, it expects to find the magic data under a specific place though, since that is how normal file works. I decided to go for putting all my custom stuff in /data/arvid, since I found it unlikely to collide with anything else
14:11:22  unlike your mom, who is easy at all depths.
14:11:28  so it is useless to you
14:11:47  (for a start, different ABI, also your phone is not rooted, so you can't use that path)
14:11:57  ok
14:11:58  there are a lot of appeals to nature, to genetics, to other things, to explain human desires
14:12:04  or is it rooted
14:12:08  I thought you said it wasn't
14:12:11  not sure any more
14:12:12  I'll just compile my own cross compiler
14:12:20  nortti, eh, android ndk includes it
14:12:25  you just have to set it up
14:12:31  and, society becomes reified
14:12:34  Vorpal: how much disk space it requires?
14:12:36  like run a script, export a few env variables
14:12:37  sec
14:14:00  nortti, 561 MB for the NDK itself (when uncompressed). The script then generates the standalone toolchain directory tree (98 MB), after that you can throw away the rest of the NDK. You also need a few basic bits of the SDK to push your files onto the device probably
14:14:11  ok
14:14:18  not sure how large those bits of the SDK are
14:14:25  since I have a lot of the SDK installed
14:14:26  what bits?
14:14:32  for multiple ABI versions
14:14:32  sec
14:14:40 -!- Phantom_Hoover has joined.
14:15:01  nortti, the platform tools and sdk tools
14:15:08  why?
14:15:14  nortti, to get adb for adb shell?
14:15:24  and iirc part of the NDK docs said it was needed
14:15:41  nortti, right you might want to save the docs directory from the NDK as well :P
14:15:54  Vorpal: does it need usb connection to phone?
14:16:10  nortti, well how else would you push files easily?
14:16:20  I guess you could send them over some other way
14:16:22  I don't know
14:16:40  nortti, but no the NDK doesn't need it when compiling
14:16:45  then I have a problem. I don't reliable usb port
14:16:52  nortti, huh?
14:16:54  +have
14:17:01  well, send it over however you want then
14:17:14  and use whatever you use on the device for the shell :P
14:17:15  only usb port on this machine sometimes refuses to work
14:17:21  ok
14:17:43  nortti, just be aware of that you need numeric modes for the standard chmod on android :P
14:17:49  yes it is pretty terrible
14:18:24  hm /sbin/ueventd is a symlink to /init
14:18:27  interesting
14:18:57  hm what is ueventd, thought it was udev, but doesn't appear so
14:19:09  spirity: i suppose my confusion is to how an individual gains more benefits by reproducing and by providing riches and power to his/her offspring than he/she benefits by living in harmony
14:19:21  Vorpal: what other modes there are other than numeric?
14:19:25 -!- MoALTz has joined.
14:19:36  in other words, in order to have our individuality are we secretly in debt to our collective?
14:19:42  nortti, symbolic, like u+x
14:19:46  oh
14:19:46  they don't work
14:19:50 -!- copumpkin has joined.
14:19:53  you need to do 775 or whatever
14:20:06  I didn't know that kind of modes existed
14:20:12  what, really?
14:20:17  I always use numeric mode
14:20:19  yes
14:20:19  are we secretly required to help our own species compete and prosper against other species?
14:20:22  nortti, that way you can do difference to existing
14:20:28  can't with numeric
14:20:35  oh
14:20:52  nortti, like chmod -R g+r . to recursively add group read to a directory hierarchy
14:20:53  is our means of doing this to blow up earth? :D
14:20:54  read the man page :P
14:21:09  ok. I'll just install man
14:21:14  nortti, "lol"
14:21:17  and man-pages
14:21:26  for the actual data files
14:21:31 -!- monqy has left.
14:21:32  ok
14:21:37  or whatever your system calls that package
14:21:38  done
14:21:47  now you have manuals
14:21:54  hooray
14:22:39  well I already had them but I had to install retawq to use man
14:23:01  wtf is retawq
14:23:15  nortti, wait are you still on netbsd?
14:23:22  not sure the ndk will work on that
14:23:26  Vorpal: no. on slitaz
14:23:27  <-- not so good guy
14:23:34  what is slitaz?
14:23:36  is it linux?
14:23:40  Vorpal: yes
14:24:03  Vorpal: netbsd was in my virtual machine
14:24:28  okay so anyone knows how to do what ldd does without using ldd? Since ldd is not cross-platform
14:24:32  I guess readelf can do it
14:24:38  but I don't know which bit
14:24:44  hm
14:25:53  basically what i mean is that it seems to me that what is a crime among individuals is often not against the goals of a species at large
14:26:20  like?
14:26:47  that, stealing is just about being competitive with resources
14:26:58  ah readelf -d
14:27:08  "NEEDED" lines
14:27:45  itidus21, google altruism
14:29:20  Vorpal: I'm not sure if that was intended as a counterargument
14:29:24  but it's not really one.
14:30:05  it kind of is?
14:30:25  how to do what ldd does without using ldd? set the magic environment variable and run the program, then it will print the ldd output and exit instead of starting up
14:31:07  spirity, rather than trying to outsmart the other individuals to get the most food or whatever for yourself, a /truly/ altruistic approach would be to share so that everyone gets enough to survive at least
14:31:48  I think...
14:31:54  I think I've broken my memory stick
14:31:58  Vorpal: do I need to specify what android version to use?
14:32:09  true altruism might involve selecting the people that are useful for everyone and feeding them first
14:32:21 -!- MoALTz has quit (Quit: Leaving).
14:32:23  olsner, no it won't, because dynamic linker is stupid on target platform :P
14:32:40  i think where humanity is heading is to focus not on killing the excess, but having less babies, based on a change of priority to benefit those living humans rather than increasing their number
14:32:48  olsner, android doesn't use ld.so, it uses a custom inferior one (/system/bin/linker)
14:32:57  nortti, err the API version yes
14:33:10  nortti, what does about device say for android version?
14:33:18  I can convert it to the API version for you
14:33:23  Any ideas what http://imgur.com/M1RjT means?
14:33:26  Vorpal: 2.2.1
14:33:52  sec
14:34:21  Taneb: that's what mount gives you when something is wrong, could mean anything
14:34:33  nortti, $NDK/build/tools/make-standalone-toolchain.sh --platform=android-8 --install-dir=/tmp/my-android-toolchain
14:34:52  nortti, that should be the right one
14:35:00  we're at a point as a species that, in an emergency we could probably reach 50 billion humans in 10 years with appropriate use of laws and medical technology
14:35:16  we no longer need our obsession with sex
14:35:26  nortti, and then move /tmp/my-android-toolchain to wheverever you want
14:35:28  olsner, can just format it?
14:35:30  and read the rest of the document
14:35:31  ok
14:35:53  (or specify your final target location for it right away)
14:36:09  arvid@dragon /opt/android-ndk-r8b $ ls platforms/
14:36:09  android-14  android-3  android-4  android-5  android-8  android-9
14:36:17  not all API versions have new NDK versions
14:36:30  ok
14:36:45  Taneb: fwiw, it's usually *not* a broken file system that's the cause
14:36:55  olsner, I wanted to format it anyway
14:37:06 -!- MoALTz has joined.
14:37:13  ah, then formatting before trying to mount it is a good idea
14:37:25  nortti, anyway, you might want to save a copy of docs/STANDALONE-TOOLCHAIN.html at least even if you delete the rest of the NDK
14:37:41  ok
14:37:42  maybe the entire docs/
14:37:45  it isn't that large
14:38:12  i hope i didn't make monqy leave, i have /leave paranoia when someone leaves after i have started 
14:38:27  but i don't think he is so fickle as that
14:38:30 -!- MDude has quit (Ping timeout: 264 seconds).
14:38:44  Taneb, sudo file -sL /dev/sdb1 
14:38:47  what does it say
14:39:08  "data"
14:39:11  wow
14:39:17  Taneb, sudo file -sL /dev/sdb* ? 
14:39:32  /dev/sdb1: data
14:39:37  Wait
14:39:38  Taneb, and for /dev/sdb itself?
14:39:41  I forgot to take off the 1
14:39:51  # ISO 9660 CD-ROM filesystem data 'bootimg                         ' (bootable)
14:40:03  eh, how does that even have a sdb1 then
14:40:11  since there is obviously no partition table
14:40:14  HOW DOES THAT WORK?
14:40:23  I DON'T KNOW
14:40:52  Taneb, did you not unplug and replug it after you messed up the partition table?
14:40:55  that could explain it
14:41:08  I've repeatedly unplugged and replugged it?
14:41:12  huh
14:41:12  okay
14:41:15  then I have no idea
14:41:38  Taneb, anyway I don't think bootable USB sticks work like that
14:41:44  if that is what it is supposed to be
14:44:24  not sure though
14:45:18  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: can't open name: no such file
14:45:20  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: OPTIONS_text_--arch=: not found
14:45:24  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: OPTIONS_abstract_Specify: not found
14:45:25  what
14:45:26  eh
14:45:36  what did you specify?
14:45:40  what is your command line
14:45:54  I don't think you specify --arch to it? Since it defaults to arm
14:46:02  SYSROOT=/home/juhani/src/android-ndk-r8b/platforms/android-8/arch-arm/ $NDK/build/tools/make-standalone-toolchain.sh --platform=android-8 --install-dir=/tmp/my-android-toolchain
14:46:22  err that is not right
14:46:24  isn't that the script to *compile* a toolchain from scratch? the ndk already has a built toolchain you can use
14:46:39  nortti, drop the sysroot stuff, that is for the alternative approach
14:47:14  ok
14:47:37  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: can't open name: no such file
14:47:40  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: OPTIONS_text_--toolchain=: not found
14:47:43  /home/juhani/src/android-ndk-r8b/build/tools/make-standalone-toolchain.sh: eval: line 1: OPTIONS_abstract_Specify: not found
14:47:45  nortti, okay, maybe you need bash for this?
14:47:46  expr: syntax error
14:47:47  I don't know
14:47:52  or at least dash
14:48:04  same thing with bash
14:48:09  what then I'm clueless
14:48:17 -!- edwardk has joined.
14:48:17  trying busybox instead of toybox
14:48:23  nortti, that might help yes
14:49:05  nortti, personally I used the full gnu thingy, I guess I could export one and upload it to dropbox for you if that doesn't work. Going to take a while though, my upload is rather slow
14:49:11  same thing
14:49:43  Vorpal: are your binaries PIII compatible?
14:49:57  nortti, well, the script forces 32-bit binaries
14:50:03  ok
14:50:06  as long as the NDK works on PIII then sure
14:50:43  How do I get on this memory stick
14:51:21  25 MB file to upload, *tries to recompress as xz*
14:52:01  xz -9 running...
14:52:37  Using plain xz would probably end up being faster :-P
14:52:48  hm true
14:52:53  and still smaller than bzip2
14:53:08 -!- pikhq_ has quit (Ping timeout: 248 seconds).
14:53:10  -9 only differs from -8 for files bigger than 32 megs anyway
14:53:11 -!- pikhq has joined.
14:53:38  Deewiant, well the file is 96 MB :P
14:53:47  25 MB from bzip2
14:53:53  Ah
14:54:03  ah 16 MB from plain xz
14:54:07  good enough
14:54:30 * Vorpal moves it to the dropbox directory
14:55:49  Deewiant, anyway the default is -6 not -8
14:56:18  but yeah xz is slow
14:56:35  Yes I know, I was just pointing out that -9 is uselessly much anyway
14:56:42  Although of course it's easier to just use -9 always
14:56:45  (Or -9e :-P)
14:57:05  Deewiant, -e increases the work on the uncompressing side, no?
14:57:14  oh I guess not
14:57:18  No
14:57:29  Unless it compresses worse, of course
14:57:39  does higher number mean more memory when uncompressing then?
14:57:44  Yes
14:57:45  right
14:57:51  guess -6 is good for nortti then :P
14:57:59  what with his low spec machine
14:58:06  yeah
14:58:24  Vorpal: your phone is probably more powerful
14:58:35  nortti, http://dl.dropbox.com/u/87474461/arm-linux-androideabi-4.6.tar.xz
14:58:48  nortti, yes, quad core 1.4 GHz ARMv7 with NEON :P
14:58:53  :D
14:58:55  it has more cores than my laptop
14:59:06  (Core 2 Duo @ 2.26 GHz)
14:59:20  yeah. my computer is single core 700 MHz Pentium III
14:59:39  nortti, still my desktop beats both: Quad core Core i7 (Sandy Bridge) at 3.4 GHz
14:59:53  wow
14:59:55  that has hyperthreading too
15:00:09  so 8 logical cores, 4 real ones
15:00:32  ok. well I'm moving to 500MHz g3 with 640MB of RAM in few weeks
15:00:34  nortti, anyway tell me when you finished downloading, I want to delete that file when I'm done
15:00:39  g3?
15:00:40  wow
15:00:44  poor you
15:00:47  I'm done
15:00:52  nortti, why would you use a G3?
15:01:01  because it was cheap
15:01:15  nortti, not going to drop your current machine I hope
15:01:23  why?
15:01:39  my current machine has 64MB of RAM
15:01:50  no hope of android programming on PPC I'm afraid, without making your own toolchain from scratch
15:02:04  yeah?
15:02:09  don't think the SDK for java supports that either
15:02:23  so. I hate java
15:02:26  oh well
15:02:34  nortti, why did you ever get an android phone then? ;P
15:02:48  because other choice was nokia
15:02:55  symbian? right
15:02:59  with symbian s60
15:04:18  the symbian that the last nokia symbian phones used was large improvement over symbian s60
15:04:47  nortti, you might need to download new config.sub or config.guess for autotools projects, if they complain about androideabi being an unknown system
15:04:58  ok
15:05:06  I used --host=arm-linux-androideabi to ./configure
15:06:00  also set --prefix to wherever you intend to put the stuff on the phone, then use DESTDIR=some/path/here make install
15:06:10  so you don't try to install to that absolute path
15:08:03  where configure is located?
15:08:27  oh. you're talking about busybox
15:08:57  nortti, no about ./configure style projects
15:09:02  ok
15:09:04  nortti, like GNU file
15:09:08  busybox uses a different system
15:09:15  ok
15:09:15  basically like the kernel config
15:09:19  I have no idea how you set that up
15:09:27  hmm. I'll try
15:09:35  I'll google :PO
15:09:37  :P*
15:09:47 -!- nooga has joined.
15:09:47  like first hit is their FAQ
15:10:15 -!- zzo38 has joined.
15:11:21 -!- Sgeo__ has joined.
15:14:22 -!- Sgeo_ has quit (Ping timeout: 252 seconds).
15:16:41  mroman: HELP will also allow comments I think :)
15:17:20  (that is, comments that do not take time to process)
15:17:50  AnotherTest, comments that do not take time to process?
15:17:51  what
15:18:07  of course it will take time to parse them
15:18:33  AnotherTest, you can't write a lexer which magically skips comments!
15:18:51  include/libbb.h:256:7: error: size of array 'BUG_off_t_size_is_misdetected' is negative
15:18:52  Vorpal: they're preprocessed out long before execution
15:18:56  uhm. what?
15:19:02  AnotherTest, right, you didn't specify at which stage
15:19:04  nortti, what
15:19:10  nortti, what are you trying to compile?
15:19:15  busybox
15:19:15  Vorpal: currently you would have to pop a value from the stack
15:19:27  nortti, oh I'm configuring it atm. Spending ages in menuconfig for it XD
15:19:31  Vorpal: but you're right, I should have mentioned that
15:19:34  (yeah I'm that sort of person)
15:20:02  Vorpal: make defconfig
15:20:31  nortti, default config? Too boring
15:20:40  besides some of these options look bad on android
15:20:44  since there is no /var
15:20:53  you obviously can't do stuff that requires that
15:21:50  ok
15:24:21  should I use devpts?
15:25:16  nortti, well on my phone the answer is yes
15:25:26  check if you have /dev/pts (and something in that) on your own phone
15:25:51  nortti, also avoid the suid stuff, especially if you aren't rooted, you are not going to have /etc/busybox.conf in any case
15:26:18  mroman: 1) would you be able to make an online shell (I could maybe do it if you don't want to) 2) haskeline is not part of haskell-platform
15:26:28  also Build options, set the sysroot and so on (/opt/android-ndk-standalone/arm-linux-androideabi-4.6/sysroot for me)
15:26:34  Path to BusyBox executable (BUSYBOX_EXEC_PATH) [/proc/self/exe] ?
15:26:43  nortti, sure, that is standard linux behaviour
15:28:23  intallation prefix?
15:28:43  mroman: never mind 2, I installed libghc6-haskeline-dev
15:28:50  nortti, hm? where is that?
15:29:11  anyway, I like how the config tool warns you that enabling a specific feature will use ~60 bytes more
15:29:12  XD
15:29:45  BusyBox installation prefix (PREFIX) [./_install]
15:30:20  hm no idea
15:30:29  anyway I can't find the applet selection list
15:30:30  where is it
15:31:22  oh right
15:31:26  that wasn't the top menu XD
15:31:40  Use termios to manipulate the screen (FEATURE_USE_TERMIOS) [Y/n/?]
15:32:08  no idea, I'm jumping around in menu config, not using plain config
15:32:24  probably should be enabled
15:32:36  at least after reading the help for it
15:34:44  why is patch under editors?
15:34:55  I haven't gotten that far yet
15:38:20  nortti, bbiab food
15:44:59  Vorpal: still getting include/libbb.h:256:7: error: size of array 'BUG_off_t_size_is_misdetected' is negative error
15:52:58  hm
15:53:10  nortti, I will try to compile it myself soon
15:59:19  Non-empty list comonad can be coapplicative, isn't it?
15:59:34  ./lib/portability.h:84:1: error: unknown type name 'ssize_t'
15:59:34  ./lib/portability.h:84:34: error: unknown type name 'size_t'
15:59:34  ./lib/portability.h:84:56: error: unknown type name 'FILE'
15:59:37 -!- edwardk_ has joined.
15:59:40  nortti, huh
15:59:43  that is screwed up
15:59:52  trying to compile toybox
16:00:04  hm
16:00:12  edwardk_: Did you have _ on your name before, or just now?
16:00:13  well I run into the libbb thing too
16:00:43  the client puts it on if it can't get my nick, due to a stuck connection, etc.
16:00:51 -!- edwardk has quit (Ping timeout: 245 seconds).
16:00:52 -!- edwardk_ has changed nick to edwardk.
16:01:05  nortti, oh googling that error suggests some small patches may be needed
16:01:15  ok
16:01:45  nortti, or hm configs/android_ndk_defconfig?
16:02:08  what?
16:02:15  nortti, in the busybox sources
16:02:17  that file existsw
16:02:20  exists*
16:02:20  ok
16:02:24  edwardk: Do you know coapplicative?
16:02:29  nortti, I do want a few extra applets though
16:02:37  anyway google the libbbb error
16:02:42  i've seen various attempts by people to specify one
16:02:47  first hit was on github, very useful
16:03:11  but as you don't have comonoids in hask its less relevant ;)
16:03:20  and we also don't have coexponentials
16:03:47 -!- edwardk has left.
16:11:21  nortti, yeah https://github.com/tias/android-busybox-ndk works for me so far (still compiling) Need to adjust the API version from 9 to 8 for your phone though
16:13:18 -!- epicmonkey has quit (Ping timeout: 264 seconds).
16:13:28 -!- asiekierka has joined.
16:14:20  yay this busybox has lsof
16:14:25  the one I was using did not
16:14:31 -!- asiekierka has quit (Client Quit).
16:17:57 -!- ogrom has quit (Quit: Left).
16:25:23 -!- ais523 has joined.
16:28:44 -!- Madoka-Kaname has quit (Ping timeout: 260 seconds).
16:30:54  Vorpal: should I be able to install busybox if I download the file to sdcard using ftp with es explorer and the install it using cat to /data/org.connectbot ?
16:31:30 -!- oerjan has joined.
16:32:17 -!- Madoka-Kaname has joined.
16:33:21  nortti, sure, and then chmod as required
16:33:45  nortti, oh and use busybox ln rather than just ln if you want to link the applets
16:33:52  I forgot why
16:33:59  but android ln was slightly broken
16:34:05  don't remember the exact details
16:38:36  nortti, btw why haven't you rooted the phone?
16:44:57 -!- Taneb has quit (Ping timeout: 255 seconds).
16:45:13  nortti, did it work?
16:45:18  AnotherTest: I think I could hack something together using Ajax and CGI, yes.
16:46:41  or probably without Ajax too :)
16:51:46  But I can't host it, no.
16:52:05  Not without lots of efforts.
16:52:38  I would need to make sure that the process terminates after a time, terminates if it consumes to much time etc.
16:52:55  I'm not sure apaches cgi module has enough support for that.
16:53:23  mroman, you could write a cgi wrapper for that?
16:53:48  I could.
16:54:04  If I knew the right unix tools to launch processes with restrictions :)
16:54:40  oh
16:54:44  apache has RLimitMEM
16:54:45  cool.
16:55:23  Ok. Then I can host it.
16:57:49 -!- Taneb has joined.
16:59:37  Well, it is NOW TIME TO INSTALL HAIKU
16:59:39  brb
17:00:09 -!- MoALTz_ has joined.
17:00:11 -!- nooga_ has joined.
17:00:28 -!- nooga has quit (*.net *.split).
17:00:28 -!- MoALTz has quit (*.net *.split).
17:02:16 -!- quintopia has quit (Read error: Operation timed out).
17:02:35 -!- nooga has joined.
17:02:35 -!- MoALTz has joined.
17:02:57 -!- nooga has quit (Ping timeout: 247 seconds).
17:02:57 -!- MoALTz has quit (Ping timeout: 247 seconds).
17:03:31  Taneb, why haiku?
17:03:34  isn't that beos?
17:06:44  Yeah
17:07:03  I want all my OS's to be from different families
17:07:16  Vorpal: compile failed
17:07:19  I've got a friend with a copy of Windows 7 Ultimate
17:07:26  And I'm used to Ubuntu
17:07:29  And I craved a third
17:07:40  Taneb: so windows, linux, haiku, syllable, aros?
17:08:23  Vorpal: networking/telnet.c:39:25: fatal error: arpa/telnet.h: No such file or directory
17:08:32  Syllable is unix-like
17:08:38  is it?
17:08:46  Wikipedia thinks it is
17:08:50  oh
17:09:03  is SkyOS still available?
17:11:23  I... don't think so
17:11:29  Taneb: Visopsys?
17:12:12  Yes
17:12:25  It looks very 90's
17:12:27  okk. now istall it and aros
17:12:34  *ok
17:13:46 -!- nooga_ has quit (Ping timeout: 245 seconds).
17:16:25   Vorpal: networking/telnet.c:39:25: fatal error: arpa/telnet.h: No such file or directory <-- did you use the config provided by that github project and the patches from it?
17:16:28  that worked for me
17:20:05  At least I *could* host it if RLimitMEM would do what I expect it to do :(
17:27:14  mroman: "Berlesque programmers are very rare. " typo :D?
17:28:14  oh
17:28:15  yeah :(
17:28:44  HELP can only do something like # test := ri{^^\/1.-}{5.-}w!1.- at the moment
17:29:19  I'm going to add \ to allow it spanning over multiple lines
17:29:46  Or I could use a delimiter character
17:30:05  Not sure what would be best
17:30:50  AnotherTest: http://eso.mroman.ch/cgi/burlesque.cgi
17:31:57  Nice
17:32:03  very nice
17:32:27  Also, should I use ';' as a delimiter ?
17:32:29 -!- MDude has joined.
17:32:45  Delimiter for what?
17:32:48  Or should I use '\' to use multiple lines
17:32:51  For a defintion
17:33:02  # macro := something here
17:33:07  # foo := bar ; # foo2 = fooooo
17:33:11  ^- like that?
17:33:20  no, like:
17:33:35  # foo := bar
17:33:35  and bar bar
17:34:01  (a multi-line macro or whatever it should be called)
17:34:26  I'd use indentation for that
17:34:27  but the example you gave would then be valid too
17:34:32  # foo := bar
17:34:35   still foo
17:35:02  what if someone had burleque that was shaped?
17:35:19  maybe they would need a space on the start of the line...
17:35:22  Then you have to put a space in front of each line
17:35:35  If your code line starts with a space, then you just have two spaces
17:35:46  # foo := python code
17:35:50   if bar:
17:35:54    foobar
17:35:59  ^- like that.
17:36:11  ah okay
17:36:42  Vorpal: YES
17:36:46  *yes
17:41:02  nortti, ?
17:41:05  mroman: is a single space fine, or should it be a larger amount of spaces?
17:41:06  yes to what
17:41:33  20:16 < Vorpal>  Vorpal: networking/telnet.c:39:25: fatal error: arpa/telnet.h: No such file or directory <-- did you use the config provided by that github project and the patches from it?
17:41:39  nortti, I assume you adjusted the sysroot path in make menuconfig for it?
17:41:44  yes
17:41:59  well actually with vi .config
17:43:01  nortti, there were some other differences due to newer version
17:43:09  like?
17:43:15  arvid@dragon /opt/android-ndk-standalone/arm-linux-androideabi-4.6/sysroot/usr/include $ find . -iname 'telnet.h'
17:43:16  ./arpa/telnet.h
17:43:31  nortti, LSOF had to be enabled, since it is new in last busybox version
17:43:37  ok
17:46:58  Well, it refuses to get past the third symbol on the boot-up screen
17:47:14  Taneb, qemu?
17:47:39  I... don't understand what you mean by that
17:48:58 -!- elliott_ has joined.
17:49:00  /usr/bin/ld: note: 'SDL_GetKeyState' is defined in DSO /lib64/libSDL-1.2.so.0 so try adding it to the linker command line
17:49:01  Whoa, fancy.
17:49:01  oerjan, I'm up to 2008
17:49:40  youtube is breaking for me today. Randomly jumping back to the start of the video or sometimes just stopping playing
17:50:01  elliott_, which linker is that? gnu ld or gold?
17:50:07  Taneb: darn you're fast
17:50:10  and did it scan all your *.so=
17:50:12  s/=/?/
17:50:15  or what
17:50:17  gold is a GNU ld, too.
17:50:20  oerjan, I have practisew
17:50:24  But it's not gold.
17:50:24  *-w
17:50:26  ah
17:50:31  elliott_, fancy still
17:50:39  they should call them gnuld and gnold
17:50:47  elliott_, what version and/or how?
17:50:53  olsner: new ld, gn old
17:51:06  Vorpal: 2.22.0.20120323
17:58:53 -!- oonbotti has quit (Quit: oonbotti).
18:00:59  AnotherTest: A single spaces is fine I guess.
18:01:22  Link-time typechecking is what would be fancy. Does that happen yet? 
18:02:18  fizzie: As a replacement for, or in addition to, the normal kind?
18:03:48  In addition to. The sort of thing that would catch if you have "int f(void); f();" in one translation unit (a term so old-fashioned-sounding) and "int f(int x) { return x; }" in another.
18:04:41  Sounds like something at least LLVM-style "compile to LLVM intermediate bytecode and then do whatevers at link-time" would catch.
18:04:54  elliott_, hm
18:05:41  fizzie: Right. That essentially amounts to a header file checker in practice, surely? And you usually include them in such a way that it'd get caught anyway.
18:06:08  elliott_, what if you forgot to put static in front of some local functions?
18:06:15  then that could happen
18:07:01  Header files, schmeader files. I just heard about some dude who just declares external functions only inside functions using them, because that's "data hiding".
18:07:15  ouch
18:07:16  (Including printf.)
18:07:47  fizzie: Heh. Where is this?
18:08:05  Possibly it was a comp.lang.c thread, I don't think I could re-find it.
18:08:20  It wasn't "just" just.
18:09:03  as long as everything uses the wrong data, the real data is indeed hidden
18:13:25  does C let you declare external functions inside functions? istr that's a c++ feature
18:13:51  I'm pretty sure you can in C too.
18:15:53  "The declaration of an identifier for a function that has block scope shall have no explicit storage-class specifier other than extern.
18:16:15 -!- MoALTz_ has quit (Quit: Leaving).
18:16:50  hi ais523
18:17:00  Yeah, that's legal C.
18:17:49  stupid idea anyway, local declarations of external functions just invites getting it wrong
18:17:52  @ping
18:17:52  pong
18:17:59 -!- Taneb has quit (Quit: Leaving).
18:18:16 -!- Taneb has joined.
18:18:21  bye ais523
18:18:46  wat
18:19:12 * oerjan scans for invisible ais523 talk
18:19:14 -!- oonbotti has joined.
18:19:22  olsner: Data hiding! 
18:19:30  oerjan: there is none
18:19:31  Practically any monad or comonad can be made using Const as a base
18:19:41  Proxy/Empty = Const ()
18:19:48  Identity = Cofree Proxy
18:19:56  Maybe = Free Proxy
18:20:24  NonEmpty = Cofree Maybe
18:20:24  fizzie: true, that's a REALLY GOOD use for that feature
18:20:32  [] = Compose Maybe NonEmpty
18:20:52 -!- ogrom has joined.
18:20:57  Actually, this isn't true at all
18:21:04  A lot of the abstract ones can
18:21:23  You can't make anything with an (->) in it.
18:21:27  Unless you allow (->).
18:21:30  Obviously.
18:21:32  Yeah
18:21:48  That was a lot less awesome a revelation then I first entailed
18:22:17  Either is the tagged union of Const and Identity?
18:42:05  `words --help
18:42:10  Usage: words [-dhNo] [DATASETS...] [NUMBER_OF_WORDS] \  \ options: \   -l, --list             list valid datasets \   -d, --debug            debugging output \   -N, --dont-normalize   don't normalize frequencies when combining \                          multiple Markov models; this has the effect \                          of making larger datasets more influential \   -o, --target-offset    change the target length
18:42:14  `words -l
18:42:17  valid datasets: --eng-1M --eng-all --eng-fiction --eng-gb --eng-us --french --german --hebrew --russian --spanish --irish --german-medical --bulgarian --catalan --swedish --brazilian --canadian-english-insane --manx --italian --ogerman --portuguese --polish --gaelic --finnish --norwegian --esolangs \ default: --eng-1M
18:42:21  `words --eng-1M 50
18:42:26  toox sanami tible austol dun unier ellit shutzotheich herian eling impumo bungazy veneffer fida eccart buonapolyzan baadenking pell corcen cord resposed oratine kunk more chaggr
18:43:27  sanami
18:43:31  that seems more japanese
18:43:43  chaggr, check + swagger
18:43:45  “shutzotheich” doesn't strike me as very English either.
18:43:59  short for "check your swagger", said to someone when they are super uncool but acting like they are not
18:44:17  Austol is clearly some medication
18:44:28  a brand name for buonapolyzan
18:44:39  “Yo dawg, chaggr!” “Me chaggr? You chaggr, you just said 'Yo dawg'!”
18:45:08  actually it's "chagger"
18:45:13  chaggr is a site for posting people who need to chagger
18:45:30  chagger.com is free; the opportunity is open
18:45:44  `words --en-gb
18:45:47  Unknown option: en-gb
18:45:48  lol
18:46:11  `words --en-gb 20
18:46:15  Unknown option: en-gb
18:46:19  um
18:46:23  Oh.
18:46:26  `words --eng-gb
18:46:31  coonting
18:46:38  Hot.
18:46:41  coonting
18:46:48   toox sanami tible austol dun unier ellit shutzotheich herian eling impumo bungazy veneffer fida eccart buonapolyzan baadenking pell corcen cord resposed oratine kunk more chaggr
18:46:57  so, kunk is clearly some sort of bodily fluid
18:46:58  `words --eng-gb 10
18:47:03  resposed is... almost a word
18:47:03  arry abylosson tear hamma sae haul vening riggle fam fren
18:47:15  it is rest + repose
18:47:24  nonono
18:47:28  it's like disposed
18:47:36  oratine is clearly oral sex involving ovaltine
18:47:44  no idea what "more" could be
18:48:14  o tempora o mores
18:48:21 -!- pikhq has quit (Ping timeout: 245 seconds).
18:48:25 -!- pikhq_ has joined.
18:48:27 -!- epicmonkey has joined.
18:48:28  tear is an actual word.
18:48:40  very funny
18:48:43  you're such an abylosson
18:48:44  whatever HackEgo's words is supposed to do o_O
18:48:54  `words --eng-gb 1
18:48:59  inde
18:49:01  `words --eng-gb 3
18:49:03  `words --finnish 50
18:49:06  fizzie: interpret
18:49:07  vassani nojalollakunne ostamistänsä kahlavisikilot maksitovaamupaljaa automaitsevimpana tutteluaakkoamaalta ahneellaistumies leveämpisi hautovilla haustannelastammenensä aforimittava muottaviini munkeroiksesi toilleen tautteellistan humiin koettamani alailevissan mutojärkemilta valleen noustasi rastuvinämme auttaan syventuvampuilta
18:49:07  osta ofu drough
18:49:17  `words --german 3
18:49:21  tierungsrationszys schafte ooooo
18:49:26  ooooo
18:49:30  ooooo!
18:49:34  `words --german 10
18:49:39  händehn besitz cadichenhügerennahmenti dischemigungsorganie dart goeterialennenbild romethornehmentaltenringe hedrige suizialerbonde humen
18:49:42  that is the funniest thing `words has ever produced
18:49:45  ooooo
18:49:59  besitz und dart are real words.
18:50:17  `words --german 10
18:50:19  `words --english 50
18:50:21  beform cblichsgewaffi haftend ingebirgenen kinerungso arrictheil phle osenentcrationenzwängi konzeugführ wirdinatur
18:50:22  Unknown option: english
18:50:22  besitz is obviously equivalent to the english 'besits'
18:50:22  that suizialerbonde is clearly on hard times
18:50:34  haftend is also a real word :)
18:50:37  `words --english-1M 50
18:50:39  Unknown option: english-1m
18:50:43  mroman: you're meant to interpret the other ones
18:50:44  What does the 1M mean?
18:50:47  (easier in an agglutinative language)
18:50:49  Phantom_Hoover: one million
18:50:49  oh I see.
18:50:55  `words 50
18:51:01  pollum akhald quotehgion cov sce virger umth zozing veracored fecreying fielle depper origg weymoy farge aflat hem olar wrhydroo couperup tak exta amen flementa mythm
18:51:12  pollum
18:51:13  mythm
18:51:20  depper sounds like destroying stuff.
18:51:25  schni schna cblichsgewaffi
18:51:32  `words --norwegian 50
18:51:36  frikkvene ene rehorka gring piadentning tekirkne betakettets gruppene aborgesengen hjemmige veser ulykkelbevis bolitidelret solvestudenesoppesighet kretøyelstyr kjøttemperar teatets blyvendedunde lørdende kontria omskegaens krisligstfrier påstethets rådetssitusganiske skattlede
18:51:52  oerjan: tolke
18:51:55  `words --french 10
18:52:00  comprem tolorré rétre plarimasthugl ocumulai mera sexuale rémono eds kils
18:52:02  Norwegian? Hmm, that sounds like a made up language to me.
18:52:09  French? Hmm, that sounds like a made up language to me.
18:52:10  sexuale!
18:52:20  elliott_, I'm not convinced by "plarimasthugl"
18:52:44  which one was that from
18:52:46   Norwegian? Hmm, that sounds like a made up language to me. <-- nah, looks fine to me
18:52:51  some silly bits though
18:52:56  `words --latin
18:52:56  elliott_, French
18:52:59  Unknown option: latin
18:53:02  :(
18:53:25  yeah "gruppene" is clearly nonsense
18:53:44  `words 50
18:53:48  oerjan, what about "rådetssitusganiske"?
18:53:49  subsershi vervan nide free begli cort esfilk pectonlyfr bon nontile ago obsecum fcendo achener reggia fosteclick off panum line johnsgrui couve abdue cocke refen patzeroidate
18:53:57 -!- Gregor has set topic: Vive le plarimasthugl! | http://codu.org/logs/_esoteric/.
18:54:19  Vorpal: perfectly normal bureaucratic term
18:54:23  Phantom_Hoover: what's subsershi
18:54:26  oerjan, meaning?
18:54:49  I'm gently idling in this channel, and suddenly the topic gets changed to "Vive le plarimasthugl!".
18:55:07  boily, yes and?
18:55:15  it's the special mushroom that town councils consume while in situ.
18:55:19  plarimasthugl?
18:55:21  oerjan, oh okay
18:55:30  nortti, read the 30 last lines or so
18:55:51  the plarimasthugl just crashed the Java mental stack I was juggling with.
18:55:55  ok
18:56:15  boily, how did you notice this if you were idling?
18:56:16  boily isn't very good with french, you see
18:56:51  elliott_, the small-scale components of sershi.
18:57:01  Vorpal: multiple channels open, and  against my better judgment I checked this one.
18:57:13  oerjan: :p
18:57:28  never use judgment in this channel, it can cause brain damage
18:57:32  ago obsecum fcendo (latin)
18:58:00  quidquid latine obsecum sit, altum fcenditur
18:58:37 -!- ogrom has quit (Quit: Left).
18:58:50  Note to self: refreshDungeonCell
18:59:04  Note to self: plotCharWithColor
18:59:32  yeah a dirty dungeon cell is just trouble
19:00:21  Also check whether full screen "ticks" are done anywhere; plotCharWithColor might be bad for animation. Look into what termcast and so on do.
19:00:52  plotCharcoalBlack
19:02:46  oerjan, could you add a lot of dye to the charcoal
19:02:50  so it wasn't black any more
19:03:58  yes but how would you ignite it when it's full of dye
19:04:17  oerjan, inflammable dye obviously
19:04:57  Must be some sort of a dynamic dungeon, if it needs refreshing.
19:10:47 -!- oerjan has quit (Quit: Good night).
19:17:43 -!- quintopia has joined.
19:18:48  note to self: displayLevel
19:19:04  note to self: displayLoops, waitForAcknowledgement (research)
19:19:53  also figure out wtf plotCharToBuffer is
19:19:56  Do you know if you can somehow get internet on Famicom?
19:20:22  also note: commitDraws(), plotChar()
19:20:24  It may be difficult.
19:21:54  further notes: figure out whether platform/platformdependent.c is per-platform or not, if not look into plotChar()
19:25:27  elliott_, why are you using irc as a source navigation history?
19:25:54  i don't want to remember another workspace number and nobody is saying anything interesting
19:45:49  zzo38: I think I heard something about a modem.
19:46:55  21:45 < monochrom> and yes I hate basic and c use "=" for assignment. x=x+1, therefore 0=1?
19:47:09  Somebody feel the need for creating a language where assignment is ==?
19:47:18  x==x+1
19:48:00  No, syntax is boring.
19:48:06  Hence x is infinity?
19:48:45  A language where assignment is done using the roots of equations?
19:51:09 -!- asiekierka has joined.
19:51:49  A language where every program IS an equation.
19:52:10  x+y==5 asigns 5-y to x, and 5-x to y
19:52:48  print x==5 prints iff if it prints 5
19:52:54  *-if
19:54:27   Hence x is infinity? <-- or -inf?
19:54:32  that should work too
19:54:37  It's infinite either way
19:54:42  true
19:55:22  anyway pascal uses := for assignment, I don't remember what it uses for equals
19:55:27 * mroman Enjoying a C,Haskell bashing in #haskell
19:55:28  Plain =, IIRC.
19:55:34  ah
19:56:31  Why is there no englisch word for unterjubeln o_O
19:56:41  What does it mean?
19:56:53  And Scheme = is of course equality too.
19:56:57  I would like to see a realistic post-nuclear war game. Not like fallout with it's mutants. But what it would actually be like
19:57:15  Vorpal, like that Raymond Briggs "comic"?
19:57:23  Taneb, I don't know, what is that?
19:57:29  When the wind blows
19:57:33  Taneb, what is that
19:57:42  It's about an elderly couple in a nuclear war
19:57:42  It means to slip somebody something
19:57:46  or something like that.
19:57:49  They die of radiation poisoning
19:58:19  Taneb, hm, well I was thinking set maybe 200 years after a nuclear war, people coming out of bunkers and so on.
19:58:24  If I didn't do something
19:58:32  but you convince other that I did
19:58:41  then you're unterjubeln me something
19:58:42  Vorpal, they do come out of bunkers, just after a weekend
19:58:50  Taneb, that sounds like too early
19:58:57  Yeah, hence why they die
19:59:00  to palm sth. off on sb. [coll.] [to dispose of sth.]    
19:59:00  to pin sth. on sb. [coll.] [to lay the blame]
19:59:00  to plant sth. on sb. [coll.] 
19:59:00  to slip sb. sth. [coll.] [secretively administer]
19:59:00  to plant sth. on sb.
19:59:04  If you ask one dictionary.
19:59:06  Raymond Briggs is not a happy man
19:59:06  Taneb, right, so not very interesting
19:59:16  Taneb, would plants have reclaimed the affected areas after a couple of hundred years? Or would it all be dead dead dead?
19:59:22  He also wrote the Snowman
19:59:23  Or if I go to the market and the clerk charges me a tomato which I didn't buy
19:59:33  Taneb, no clue what that work of fiction is either
19:59:38  leo.org didn't have a translation.
19:59:45  Ask Phantom_Hoover
19:59:53  I've never actually seen or read it
19:59:58  brb
20:00:35  I've never read the Snowman nor do I know anything of Briggs beyond the name.
20:00:43  okay
20:01:08  well the point remains, what would actually have happened in a situation like that of Fallout. You wouldn't have mutant two-headed cows for sure...
20:01:28  As far as realistic postnuclear stuff goes, there's always Threads if you're sure you have no suicidal tendencies.
20:02:02  Phantom_Hoover, well, I would like a video game made about a realistic postnuclear setting.
20:02:03  Also the unrealistic nature of Fallout comes straight from the retrofuturistic 50s style.
20:02:11  and yes I'm quite happy to be alive
20:02:16  Phantom_Hoover, indeed it does
20:02:42  Phantom_Hoover, is Threads a book or something found online or what?
20:02:42  Although FWIW most of the mutants you see are actually the result of genetic engineering.
20:02:51  British film from the 80s.
20:02:53  ah
20:02:57  It's on Google videos, I think.
20:03:11  google videos? Is that still around? You mean youtube right?
20:03:16  Back
20:03:17  or do you mean their video store?
20:03:24  ...no, I mean Google videos.
20:03:24  I don't think that is available in Sweden
20:03:33  It's a weird artefact, yes.
20:04:07  google videos seems to be a video search engine nowdays...?
20:05:22  ah it is in the process of being shutdown, and Threads seems to be on youtube
20:08:11  youtube, being a service which has a monopoly, capitalizes on it's monopoly by asking for mobile phone numbers during signups
20:08:17 -!- elliott_ has quit (Read error: Operation timed out).
20:08:17 -!- epicmonkey has quit (Read error: Operation timed out).
20:08:31 -!- epicmonkey has joined.
20:08:55  itidus21, it doesn't quite have a monopoly, other hosts are used at times.
20:08:56  my understanding is that you can sometimes sign up without phone
20:09:05  Phantom_Hoover, okay I read the plot summary on Wikipedia, ouch
20:09:15  yeah I see what you mean about suicidal tendencies
20:09:26  They're just not as often used, largely because YouTube has a better service.
20:09:28  Phantom_Hoover, anyway what I'm more interested in is the more long term effects
20:09:40  roger
20:09:41  there is vimeo iirc
20:09:44  and some other one
20:09:45  So... you mean Fallout, but realistic.
20:09:50  Phantom_Hoover, yes
20:09:51 -!- elliott_ has joined.
20:09:56  Phantom_Hoover, the Snowman is where "I'm walking through the air" comes from, iirc
20:10:13  Yeah Vimeo seems to attract artsy people a lot for some reason.
20:10:26  Taneb, ohhh, I thought you were talking about some post-apocalyptic thing.
20:11:00  Vorpal, again, nothing I know of for games; A Canticle for Leibowitz is a book that fits that description.
20:11:02  Phantom_Hoover, oh and blip.tv?
20:11:58  I don't really know what its niche is.
20:12:18  Viddler's used a lot for Let's Plays, possibly because it allows longer videos so you don't have to segment as aggressively.
20:12:31  Nah, that's When The Wind Blows
20:12:51  Phantom_Hoover, My impression is that people who do movie reviews on youtube but get screwed over by copyright infringement due to showing a 10 second clip move it to blip.tv
20:13:02  the sample size is rather small for that impression though
20:13:27  Phantom_Hoover, and youtube allows multi-hour videos since a year or two
20:14:10  But not to default accounts.
20:14:19  Phantom_Hoover, oh I thought it did that nowdays?
20:14:24  what sort of account then?
20:14:47  Well I dunno, but I've heard repeated references to people wanting to get some kind of upgrade to upload longer videos.
20:15:00  hm
20:15:27  http://support.google.com/youtube/bin/answer.py?hl=en&answer=71673
20:15:36  BtW which Fallout games have you played?
20:19:33  let's plays are pretty cool
20:20:08  much more entertaining than the 8bit game generation itself was
20:21:14  They're very, very subject to Sturgeon's Law though.
20:22:03  The good ones, where the game is accompanied by insightful commentaries are fun; the ones that consist of someone playing through and waffling are not.
20:22:17  itidus21: youtube is far from a monopoly
20:22:23  once again you have no idea what you're talking about
20:22:44  shocking
20:22:46  Well I can see why you might call it a monopoly.
20:23:00  !
20:23:05   BtW which Fallout games have you played? <-- 3 and new vegas, haven't finished either, NV was better than 3 by far. Should get around to play some of the original ones
20:23:08  it probably has a majority market share, but far from 100%
20:23:12  well YouTube is probably not a monopoly in the direct sense
20:23:13  thats my good deed for the day
20:23:22  kmc: Microsoft didn't have 100% either and they were/(are?) considered a monopoly
20:23:30  100% is basically never going to happen
20:23:33  consistancy in my faults
20:23:34  for anything, I suspect
20:23:43  Well no, it can in some situations.
20:23:53  The BBC had a monopoly on broadcasting in the UK, for instance.
20:24:22  And there are any number of other infrastructural monopolies where the cost of setting up make competition very thin on the ground.
20:24:30   The BBC had a monopoly on broadcasting in the UK, for instance.
20:24:38  Well OK but government-sanctioned monopolies don't really count :P
20:24:39  Systembolaget has a monopoly for selling alcoholic drinks in Sweden (for drinks above a certain strength that is)
20:24:44  Infrastructure is a fair point, though.
20:27:52  What is having a monopoly? BBC/something else?
20:27:55  100% market share just depends on how you define the market!
20:28:00  Like surjective functions.
20:28:03  OK
20:28:16  http://hackage.haskell.org/trac/ghc/ticket/368
20:28:17  uaaah
20:28:19  I hope not.
20:29:23  And what's kvm?
20:29:57  mroman, the thing that allows qemu to use hardware virtualisation is called kvm, might be a different kvm though
20:30:00  Nothing wrong with a Java backend.
20:30:08  my point of course was that for the tags  x  that was the only comment i had
20:30:15  There's a patch to very-old-GHC that adds one.
20:31:10  there should be a haskell-98 backend, so you could compile your code using lots of GHC extensions into pure haskell-98 code
20:31:30  Vorpal: That would require major whole-program transformations and so on.
20:31:51  elliott_, indeed
20:31:58  elliott_, or stupid output
20:32:00  that would work too
20:32:10  What?
20:32:31  elliott_, like compiling it into a state machine describing the program
20:32:41 -!- asiekierka has quit (Remote host closed the connection).
20:32:42  i. e. nothing even similar to the original program
20:32:55  That's what I'm saying it would require.
20:32:59  oh okay
20:33:00  right
20:33:16  night →
20:37:31 -!- Vorpal has quit (Ping timeout: 245 seconds).
20:39:01  ais523: hi
20:42:36  Wow, the next Muse album sounds really OTT
20:43:10  Even more than "Come on and swear it's something biblical" or "It's time the fat cats had a heart attack"
20:43:38  OTT?
20:43:44  Over the top
20:43:54  :)
20:44:05  Have you heard Survival?
20:44:11 -!- elliott_ has left ("Leaving").
20:44:45  http://www.youtube.com/watch?v=ZYMMnHW85mc
20:56:48 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
20:59:42 -!- Phantom_Hoover has joined.
21:03:01 -!- nortti_ has joined.
21:05:01  Well, goodnight
21:05:08 -!- oonbotti has quit (Remote host closed the connection).
21:06:17 -!- AnotherTest has quit (Quit: Leaving.).
21:08:06  Can they make a patch to compile Haskell codes to: MMIX, Z-machine, Glulx, Java, Z80, Famicom, brainfuck, CLC-INTERCAL, Csound, and hardware.
21:14:55  hardware?
21:15:15  I mean to build an electronic circuit
21:15:23  they can
21:15:41  but they have very peculiar priorities
21:17:52  What priority?
21:18:04  like collecting cars built by bugatti(spelling probly wrong), ferrari, and mclaren, and buying art at auction
21:18:21 -!- pikhq has joined.
21:18:31 -!- pikhq_ has quit (Ping timeout: 248 seconds).
21:18:31  so they could make a patch but they don't
21:19:33 -!- augur_ has joined.
21:20:00  What do you think is best way in Haskell for representing compiling programs which are not Haskell programs but which can be constructed and manipulated using Haskell codes?
21:20:17  its like, "what? put funds into the very same industry that made me rich? absurd! it defies the very rules of business"
21:21:53 -!- elliott__ has joined.
21:21:55  "You have an extremely simplistic view of hacking and viruses. A clever programmer can write a virus that will hop to just about any device you plug into an infected computer. So what happens is something connected to the internet gets infected. A physical device is then connected to the infected computer, becomes infected, is later removed and interfaced with a non-infected computer that isn't connected to the internet. Flash drive, hard drive, ipod
21:21:55  , whatever."
21:21:59  "Mac OSX and Linux don't automatically run scripts when you plug in a thumb drive. Only Windows does that."
21:22:03  "That functionality is not even remotely required for a virus to work as I described."
21:22:05  "How is a virus executed from a thumbdrive without use of something like autorun? If functionality such as autorun is not required how is this done?"
21:22:09  "Because all the files on the drive are already infected by the computer they came from. If you transfer any to the computer then the virus is on the computer."
21:22:17  this moron is highly upvoted btw
21:22:18 -!- augur has quit (Ping timeout: 264 seconds).
21:22:26  which moron
21:22:30  I see two
21:22:38  or are they all the same person
21:22:41  coppro: it alternates
21:22:50  the first one is the moron; the other ones are actually separate people each time but that's irrelevant
21:23:10  second one(s) is (are) not dumb enough to call a moron
21:23:23  or rather not sure enough of themselves
21:25:00 -!- augur_ has quit (Read error: Connection reset by peer).
21:26:11  elliott__: I suppose hypothetically a sufficiently clever hacker could know of a buffer overflow in the kernel's filesystem code. ... But this moron didn't say anything of the sort.
21:28:24  The virus could also be made from the hardware instead of software, if a USB memory is designed to corrupt files automatically or other things can also be done like that
21:31:25  Hang on
21:31:29  I said goodnight ages ago
21:31:30 -!- Taneb has quit (Quit: Leaving).
21:32:44  They could even send keyboard/mouse commands
21:34:08 -!- augur has joined.
21:34:18  yay, my os can be booted as a multiboot image using qemu -kernel now
21:39:31  Or even do this: They make a USB computer keyboard that records everything typed in its internal memory. It has a three year warranty. However, it is designed to break in two years. When the customer returns it under warranty, they can spy on everything the user typed.
21:39:53  olsner: isvyour os available anywhere?
21:40:57  nortti_: probably, but it's hardly useful for anything :)
21:41:05 -!- nooga has joined.
21:41:06  zzo38: it would help more if there was a law that they had to upgrade their old keyboards
21:41:52  I do know countermeasures against many of these things. To prevent a USB memory from sending keyboard/mouse command without your permission, you can fix the kernel to not accept a USB keyboard/mouse if there already is one connected.
21:42:06  (And that includes PS/2 keyboard/mouse too)
21:43:04  at some level your data isn't the most private thing, but rather, the identity of the user of the data
21:43:09 -!- Phantom_Hoover has quit (Remote host closed the connection).
21:43:09  For the keyboard break under warranty, well, if you suspect that then open it to fix it by yourself, or break it into too many pieces, or something
21:44:48 -!- Phantom_Hoover has joined.
21:44:58 -!- augur has quit (Remote host closed the connection).
21:45:55  i suppose i must assume that my biometric data gets collected whenever i sign a form using a stylus
21:46:19  If there is such a law you have to upgrade, then destroy your old one and build a new one yourself
21:46:39  itidus21: Well, it is possible, but it may or may not be true. Or maybe true in some cases only.
21:47:17  it could make signatures more reliable anyway
21:49:09  in any case its more and more common to sign via stylus
21:49:34  Yes it could make it more reliable.
21:49:53 -!- epicmonkey has quit (Ping timeout: 252 seconds).
21:50:49  But I still sign by my pens and use extra dots and shapes and so on so that I can claim I was threatened to sign it and put those dots there to warn you, or whatever
21:51:45  my signature tends to be just a scribble.. lacking real consistency..
21:51:53  zzo38: that argument seems somehow unconvincing
21:52:01  so i depend on the biometrics for it to have any meaning at all
21:52:53  OK
21:53:00  zzo38: the only way that would remotely have a chance of working is if it differs significantly from your normal signature.
21:53:20 -!- pikhq_ has joined.
21:53:28  however, if you always do that, it means it will look the same as all the other signatures
21:53:29  this always happens when zzo38 and itidus21 chat :P
21:53:31  I sign things in Japanese usually
21:54:03 -!- pikhq has quit (Ping timeout: 276 seconds).
21:56:21  Both English and Japanese often
21:56:32  yesod is very overbearing.
21:56:43  When I am under threat I will omit the Japanese
21:56:46  want to do something differently from the default? GOOD LUCK.
21:59:51  Who says you didn't just omit the Japanese to sneak your way out?
22:00:44 -!- boily has quit (Quit: WeeChat 0.3.8).
22:00:52  The key is to have a specific signature for every contract.
22:01:04  That also prevents somebody from faking your signature
22:01:13  because he doesn't know what your next signature looks like
22:01:27  Yes that is what I put the dots I explained for
22:01:36  `addquote  But I still sign by my pens and use extra dots and shapes and so on so that I can claim I was threatened to sign it and put those dots there to warn you, or whatever
22:01:39  853)  But I still sign by my pens and use extra dots and shapes and so on so that I can claim I was threatened to sign it and put those dots there to warn you, or whatever
22:01:44  The problem is, that probably doesn't hold in court :D
22:02:26  Yes, I know, but it is worth a try if there is some kind of threat
22:02:56  zzo38: how often are you threatened to sign things
22:02:59  and the best part is
22:03:03  Rarely
22:03:06  You don't have to fullfil the contract.
22:03:13 -!- kallisti has quit (Quit: leaving).
22:03:19  They can't prove that it is your signature if they have nothing to compare it.
22:03:57  zzo38: has it ever happened :P
22:04:06  Just once
22:04:25  that's more than I have
22:04:34  hmm, did google groups remove the old interface
22:04:36  that's annoying
22:05:28 -!- spirity has quit (Quit: leaving).
22:05:35 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
22:05:43  yeah. the new one doesn't work with links
22:05:44 -!- kallisti has joined.
22:05:45 -!- kallisti has quit (Changing host).
22:05:45 -!- kallisti has joined.
22:05:51 -!- kallisti has quit (Client Quit).
22:06:49 -!- kallisti has joined.
22:06:57 -!- kallisti has quit (Client Quit).
22:07:28 -!- kallisti has joined.
22:07:38 -!- kallisti has quit (Client Quit).
22:08:00  @tell kallisti i like how your hostname gets worse when you identify
22:08:00  Consider it noted.
22:09:53 -!- kallisti has joined.
22:09:54 -!- kallisti has quit (Changing host).
22:09:54 -!- kallisti has joined.
22:09:55 -!- zzo38 has quit (Remote host closed the connection).
22:10:17  I like how his connection is oscillating.
22:10:58  hmm, so spirity is kallisti?
22:11:17  @tell AnotherTest Do you have the source of HELP online available somewhere?
22:11:17  Consider it noted.
22:11:28  olsner: yes
22:11:29  kallisti: You have 1 new message. '/msg lambdabot @messages' to read it.
22:12:07  spirity is a better nick than kallisti btw
22:12:15  I agre
22:13:04  then why are you using kallisti
22:13:12  because
22:13:18  backwards compatability
22:13:23  with existing people
22:13:30  but I'll probably change it soon
22:14:23  kallisti: btw is there anything new at spirity.org?
22:14:28  hmm, plarimasthugl
22:14:57  nortti_: I have a completely empty gitweb page for privately hosted git repos.
22:15:06  ok
22:15:39  gitweb sucks btw
22:15:43  eventually there will be a  tabletop game frontend for IRC-based tabletop gaming.
22:15:47  elliott__: it's... functional
22:16:18 -!- MDude has quit (Ping timeout: 264 seconds).
22:17:23  i'd probably just pay for a private github account if i wanted private repos
22:18:12 -!- pikhq has joined.
22:18:18  elliott__: that's what I was doing.
22:18:31 -!- pikhq_ has quit (Ping timeout: 255 seconds).
22:18:33  but a dedicated server allows me to do other things. might as well save so money by hosting my own.
22:18:53  if you take that perspective: have fun configuring qmail/postfix/...
22:19:12  I've got postfix set up
22:19:27  it.... forwards specific addresses to gmail.
22:19:28  :>
22:19:53  but yes postfix is weird to configure.
22:19:56  why not just use google apps
22:20:13  doesn't that also cost money?
22:20:31  no
22:20:32  http://www.google.com/enterprise/apps/business/pricing.html
22:20:58  there are some annoying issues with google apps accounts sometimes getting older version of google products or not being able to use google+ and stuff i think though, so you'll want to keep a gmail account lying around
22:21:11  maybe that stuff got fixed though
22:21:11  or 2
22:21:25 * kallisti currently has something like 6 email addresses, for absolutely no good reason.
22:21:32  You can use G+ nowadays, but i don’t think you can use Youtube yet.
22:21:46  ideally I wouldn't use gmail at all
22:22:05  as part of my "prepare for incoming google takeover of the world" plan.
22:22:27  umh. what?
22:22:29  hmm, I wonder if free google apps gives you less space than gmail
22:22:34  since it is marketed as 10 gigs
22:22:43  but gmail is 10271.676317 gigs-and-counting-at-a-stupidly-slow-rate-by-now
22:23:03  nortti_: I don't like the stance Google has taken since their privacy policy change.
22:23:35  10 gigs? Wasn't it 5 gigs just recently
22:23:43  which is partially why I stopped using chrome in favor of firefox.
22:23:59  but I still have a google account fo gmail and youtube..
22:24:06  kallisti: what other email addresses you have?
22:24:37  kallisti: it is a bit pointless to take a stand on chrome since all its options to talk to google are disableable
22:24:52  and it is also open source (well, chromium, but same thing)
22:24:53  especially since firefox is awful
22:25:02  elliott__: that's probably true, but then I started liking firefox so it worked out regardless.
22:25:16  elliott__, does that include disabling autoupdates?
22:25:32  elliott__: how is firefox awful? well XUL but what else?
22:25:48  Sgeo__: chromium doesn't autoupdate
22:25:53  nortti_: everything
22:25:54  I won't attempt to argue that firefox is /better/ at /everything/, but it has some things going for it. chrome has some things going for it as well (particularly it's a bit faster with V8 and all)
22:27:01  I stay at my far away corner with links2, netsurf and hv3
22:27:24  for example, firefox has privacy extensions that aren't offered on google yet.
22:27:55 -!- copumpkin has joined.
22:27:57  also plugin achitecture was better on ff last tine I checked
22:28:09 * kallisti has 3 different extensions for blocking advertising agency trackers.
22:28:44  but also firefox was in version 11 last time I checked
22:28:52  it's fun to see how many marketing agencies are attempting to monitor your internet usage.
22:29:02  what version is it on now by the way?
22:29:15  latest stable release is 14
22:29:38  the version number is completely arbitrary. as of late they've started increment the major version almost every month.
22:29:41  *incrementing
22:29:53  It's once every six weeks.
22:29:56 -!- david_werecat has joined.
22:29:57  ah okay.
22:29:59  They have a policy about it.
22:30:51  Also there's some kind of a rolling scheme thing with the n+1, n+2 and maybe even n+3 (where n is current stable) having different terms for them, though I've forgotten about that. So they probably have a Firefox 17 build already existing.
22:30:56  oh. so ESR is 10, right
22:34:37  https://wiki.mozilla.org/RapidRelease/Calendar has a calendar, and it indeed was so that Firefox 17 is "central" now (while 16 is "aurora" and 15 is "beta"). They'll hit stable 18 (and "central" 21) at around the year's end.
22:35:30  ok
22:36:11  (And three digits in less than a decade.)
22:36:30 -!- augur has joined.
22:36:53  I'll be getting back to using tenfourfox in few weeks. their stable is firefox ESR and their unstable is firefox stable
22:38:35  Every seventh Firefox (so 17, 24, ...) is an ESR release, apparently.
22:40:05  3.6->10->17->24->31
22:47:51 -!- edwardk has joined.
22:52:01 -!- soundnfury has joined.
23:03:57 -!- elliott__ has changed nick to elliott.
23:04:03 -!- elliott has quit (Changing host).
23:04:03 -!- elliott has joined.
23:18:43 -!- david_werecat has quit (Ping timeout: 246 seconds).
23:21:50 -!- augur_ has joined.
23:23:57 -!- augur has quit (Read error: Connection reset by peer).
23:28:07 -!- nortti_ has quit (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )).
23:28:57 -!- augur_ has quit (Remote host closed the connection).