00:00:12 oerjan: Fjordgata ? 00:00:35 could be 00:00:51 fizzie, "according to EXIF data this image is rotated\nWould you like gimp to rotate it into the standard orientation? [Rotate] [Keep orientation]" 00:00:53 which do I want 00:01:04 TIAS 00:01:12 ok. Not rotate 00:01:13 Just try whichever; but I think you want to keep it. 00:01:15 anyway, somewhere near, because i've been only in that area ;d 00:01:27 huh 00:01:35 sec 00:01:52 "www.connectiontonature.com" what is "tonature"? 00:01:59 aspell doesn't know it 00:02:00 connection-to-nature. 00:02:03 oh 00:02:05 duh 00:02:23 It's the Gallery2-scaled version, and it has already been rotated; I didn't know that it embeds the original photo's EXIF data in the resized copies too. 00:02:32 fizzie, I read it as "connect-ion-tonature" 00:02:35 for some reason 00:02:38 * oerjan googles the company name visible 00:03:16 http://img24.imageshack.us/img24/6056/img0439anj.jpg 00:03:34 seems to be Pirsenteret 00:03:52 that's a bit further out towards the see 00:04:54 ah 00:05:04 yeah this looks like the same from the other side: http://images.google.no/imgres?imgurl=http://www.gregus.no/Inngangsparti.jpg&imgrefurl=http://www.gregus.no/Hovedside.htm&usg=__ACpOnKhddK_AY54nKR-Xw8NGIV4=&h=300&w=400&sz=30&hl=no&start=10&um=1&tbnid=GgQJp-VBb52YfM:&tbnh=93&tbnw=124&prev=/images%3Fq%3Dpirsenteret%26hl%3Dno%26sa%3DN%26um%3D1 00:05:22 er, http://www.gregus.no/Hovedside.htm 00:05:41 fallen bicycle 00:05:42 :D 00:06:12 http://img22.imageshack.us/img22/4783/img0442d.jpg which one is me? ;p 00:07:37 The guy with the briefcase-like thing and a brown jacket? 00:07:37 the ugly one 00:07:44 oh burn 00:07:53 fizzie: :D 00:08:15 no ;p 00:08:39 ehird, you may be surprised, but I really like dvtm 00:08:49 just needs to be a bit more flexible 00:08:49 AnMaster: shut down x and use it then 00:08:54 my bet is on the guy with the camera, since that's obviously a tourist 00:09:06 ehird, I use it inside konsole with great success! 00:09:07 ;P 00:09:10 teh red haired one 00:09:14 flexible how 00:09:25 ehird, no-recompile 00:09:35 scriptable 00:09:45 like I said above: tmonad 00:10:06 AnMaster: alias dvtm-config="cd /usr/src/dvtm && $EDITOR config.h && make install" 00:10:06 then 00:10:11 $ sudo dvtm-config 00:10:15 easy 00:10:17 ehird, haha 00:10:25 and there's not exactly much to script :P 00:10:26 aliases are evil 00:10:32 use functions 00:10:39 I like aliases. 00:10:44 well ok, not evil 00:10:47 that last picture just barely fails to show my favorite restaurant 00:10:49 but they do have some issues 00:10:51 `alias x=y` is shorthand for `x () { y $* }` imo 00:10:55 if it's not it should be :P 00:10:58 ehird, no it isn't 00:11:04 well it should be 00:11:07 I'd really have liked something like dvtm back in the fb-console age; but it seems to be a recent innovation. 00:11:12 ehird, no it shouldn't 00:11:17 why not 00:11:18 that function can't handle spaces 00:11:22 x () { y "$@" } 00:11:26 ... 00:11:30 that doesn't handle multiple arguments 00:11:35 ehird, yes it does 00:11:37 also, $* gets it quoted, I'm pretty sure 00:11:39 It does; that's the magic of $@. 00:11:39 also, wtf? how? 00:11:42 ehird, no it doesn't 00:11:43 that's gross 00:11:48 That's what $@ does. 00:11:56 "$@" expands into: "arg 1" "arg 2" ... quoted separately. 00:12:03 thus reinforcing my need for the orthogonal shell! 00:12:18 While "$*" would be all-in-one-quotes. 00:12:27 $ find . -name \*.jpg -exec {x| echo $x; rm $x} \; 00:12:28 comes to 00:12:40 I just need to bother making Tclsh into a usable shell. 00:12:41 oerjan: next time, dinner is my treat 00:12:51 ehird, say arguments 1 is : "a b" argument 2 is: "c". Then "$*" -> "a b c", "$@" -> "a b" "c". without quotes both expand to "a" "b" "c" 00:12:53 $ find . -name \*.jpg -exec /tmp/osh/closures/1d874r \; 00:13:03 or somethin 00:13:04 g 00:13:05 yes, closure 00:13:08 {...} is a closure 00:13:10 ehird, did that explain it? 00:13:10 functions: 00:13:18 $ fn hello {x y|echo "hello!"} 00:13:21 $ fn a b 00:13:24 err 00:13:25 $ hello a b 00:13:36 fn defining a closure as a command obviously 00:13:38 ehird, what 00:13:39 Oh, wait. It almost is... 00:13:43 nooga: :D 00:13:53 AnMaster: Design for an orthogonal shell with closures. 00:13:54 For instance, 00:13:56 {} in find is NOT the same as function 00:13:57 oh 00:14:02 And, {} is shell syntax 00:14:04 ehird, new shell right 00:14:04 in this case 00:14:08 $ each * {x|echo $x} 00:14:10 comes down to 00:14:12 Just need to do a bit of fiddling with tclreadline. 00:14:25 $ each file1 file2 /tmp/osh.3478234/closure.349723 00:14:33 ehird, list comprehension? 00:14:34 each would be a builtin shell command, but needn't be 00:14:36 and it'd come down to: 00:14:49 foreach (args[0...-2]) { args[-1](it) } 00:14:52 so you could also do 00:14:55 $ each * rm 00:14:59 to be a pointlessly verbose form of `rm *` 00:15:09 and, since the closure files are executable, you can pass them to find, etc 00:15:13 to do complex operations with find 00:15:15 foreach...? 00:15:22 AnMaster: it's pseudoC 00:15:25 ehird, foreach is what a for does in bash 00:15:26 I was demonstrating how each would be implemented 00:15:29 by the shel 00:15:29 l 00:15:34 nuff 00:15:39 for file in *; do echo "$file"; done 00:15:41 AnMaster: IT'S NOT SHELL SYNTAX 00:15:42 is a foreach loop 00:15:43 ehird, true 00:15:44 goddamn 00:15:44 gnight 00:15:46 ehird, but 00:15:53 bash as C style for too 00:15:53 I was telling you how "each" was implemented in the shell implementation language 00:16:01 and I know bash kthx 00:16:11 -!- nooga has quit ("Lost terminal"). 00:16:13 for ((i=0;i<10;i++)); do echo "$i"; done 00:16:22 :P 00:16:31 I don't personally like the $*/$@ either; it feels a bit ugly that they do such strange things inside quotes; but I don't think my dislike is acute enough to switch shells. 00:17:04 In My Shell(TM), it'd just be {*a| rm $~a} 00:17:16 $~foo being "foo expanded out (i.e. not quoted)" 00:17:18 because quoting is default 00:17:25 and *a meaning "rest of arguments" 00:17:26 so 00:17:33 Incidentally, speaking of bash... how do you return a string from a bash shell-function? 00:17:41 $ fn justLikeRm {a*| rm $~a} 00:17:44 $ justLikeRm a b c 00:17:48 fizzie: echo 00:19:02 So do I need to call it inside $() or backticks if I want to stuff the "return value" in a variable? Guess so. 00:19:11 Yep. 00:19:14 foo=$(bashSucks) 00:19:28 Oh wait. 00:19:33 fizzie: You can do foo=(bashSucks). 00:19:43 Or not. 00:19:48 Silly me. 00:19:50 That's the array syntax. 00:19:55 fizzie, I do like the $*/$@ thing 00:19:56 Why you using Bash anyway 00:20:09 it allows you do do crazy things 00:20:13 How do I return a value from a function that prints out diagnostics to the user to stdout? (I guess by using a different pipe.) 00:20:23 fizzie: You don't. 00:20:25 Oh, I certainly like that it exists, but I don't like the idea of it. 00:20:29 Incidentally, speaking of bash... how do you return a string from a bash shell-function? 00:20:32 there are two ways 00:20:34 echo and $() 00:20:39 is the more common, and slow way 00:20:43 and it creates a subshell 00:20:50 so setting global variables won't work 00:20:52 HOWEVER 00:20:55 And the amazing, envbot approved way... 00:20:57 there are some other ways 00:21:02 Is $("@$plaineval func) 00:21:09 Which invokes an internal bash shell exploit. 00:21:13 no 00:21:17 WHAT! 00:21:21 I will come to the envbot way soon 00:21:29 "Factor VM ported to C++" WHY GOD WHY‽‽‽‽‽ 00:21:52 anyway one method is eval, like eval "\$$varname=\$varwithvalue" 00:21:56 but that is bugprone 00:22:01 so I recommend the envbot way: 00:22:14 printf -v "$varname" "%s" "$value" 00:22:35 And you return via global variables? 00:22:40 Neat. Your function calls must be ugly. 00:22:42 ehird, that is yet another way 00:22:45 but no I don't 00:22:45 That is fugly. 00:22:50 How do you return then? 00:22:51 ehird, bash has dynamic scoping 00:22:53 Vomiting over a baby? 00:22:57 so I return to a local in the caller 00:22:59 :D 00:23:07 Oh. *Oh*. 00:23:10 * pikhq vomits 00:23:11 AnMaster: How? 00:23:39 -!- puzzlet has quit (Read error: 60 (Operation timed out)). 00:23:39 That's some of the most horrid dynamic scoping I've seen. 00:24:02 So do you take the return-variable name as an argument or something? Or is it just mandated that all callers of that function must have a fixed-name parameter? 00:24:03 ehird, caller() { local myvar; callee "myvar" "other arguments"; } callee() { printf -v "$1" "%s" "$2"; } 00:24:04 is that clear 00:24:11 Ah, an answer. 00:24:11 And I regularly program in a language that lets functions do things in the caller's scope. 00:24:17 AnMaster: You're an awful person. 00:24:24 An awful, awful person. 00:24:27 ehird, intercal is worse 00:24:50 this is an esolang channel 00:24:55 what do you expect 00:25:03 caller() { callee "x="; echo "$x" }; callee() { eval "$1 hello" } 00:25:07 Flexible/ 00:25:10 s/\/$/./ 00:25:14 um 00:25:26 If it doesn't work that's just because it's flexible 00:25:33 Even shorter: 00:25:38 caller() { callee echo; } 00:25:39 ehird, eval is bug prone though. Not good in an IRC bot 00:25:45 Truly a beautiful example of the continuation-vomiting style. 00:25:49 for obvious reasons 00:25:59 AnMaster: your OS doesn't have cheap jails? 00:26:05 -!- puzzlet has joined. 00:26:07 ehird, it does. But it shouldn't be needed 00:26:08 proc caller {} {callee myvar;puts $myvar};proc callee {var} {uplevel "set $var flimble"} 00:26:19 actually, I should write some commands for extremely-cheap chroot jail creation and usage 00:26:22 That's what you're doing, in Tcl-ese... 00:26:25 And *eeew*. 00:26:27 pikhq, is it? 00:26:30 they'd be useful 00:26:32 I don't know TCL 00:26:39 like 00:26:51 ehird, that thing TravelGreaseGod use? 00:26:54 in EgoBot 00:26:56 Sticking it in a language that has dynamic scoping as a normal thing to do doesn't make it better. 00:27:01 $ cjcreate ~/hello 00:27:01 AnMaster: Definitely. 00:27:09 $ cjadd myamazlinglibrary ~/hello 00:27:15 hmm 00:27:16 make that 00:27:19 $ cjadd ~/hello myamazlinglibrary 00:27:23 uplevel executes its argument in the caller's scope. 00:27:26 $ cjexec ls ~/hello 00:27:31 the last one being equivalent to: 00:27:38 pikhq, hey it is a working and fast solution 00:27:40 $ chroot ~/hello ls 00:27:43 don't complain 00:27:46 anyway 00:27:50 (that's how the control structures are implemented) 00:27:52 it doesn't work for arrays 00:27:53 except setuid root, possibly 00:27:53 hmm 00:27:58 why does chroot() require roo? 00:27:58 t 00:28:00 I always forget 00:28:02 so you need eval to do that 00:28:04 pikhq, ^ 00:28:05 because you can escape a chroot jail with it? 00:28:17 ehird, that could be one reason 00:28:17 AnMaster: That's ugly. 00:28:24 pikhq, and bug prone 00:28:32 I'll make the chroot()-using tools setuid root, but then disallow chrooting below / 00:28:33 Yuh. 00:28:37 Problem solved 00:28:44 Cheap, easy chroot jails in a few commands. 00:28:48 You have no excuse not to use them ;-) 00:29:02 What if my OS supports jail()? 00:29:03 ehird, I do. Because I need to interact with existing files 00:29:05 for some stuff 00:29:15 like package database search for a gentoo related channel 00:29:25 AnMaster: It'd have tools to give pathways into the parent, probably. 00:29:34 pikhq, have you seen bashdoc btw? 00:29:34 "Make hardlink in jail", for instance. 00:29:35 jail() is fun 00:29:36 pikhq: The BSDs? 00:29:46 (mine doesn't) 00:29:48 Hmm, this OS X box has no jail() 00:30:02 AnMaster: No, I've not touched bash since I learned of zsh. 00:30:02 What does jail() give over chroot 00:30:12 jail is actually secure 00:30:14 Shame it keeps the bad of bash along with the good, though. 00:30:21 pikhq, http://envbot.kuonet.org/trac/browser/anmaster-trunk/tools/bashdoc (note that I didn't write it originally, it is a modified version of the bashdoc program from sourcemage) 00:30:34 a jail is almost a vm 00:30:47 pikhq, you will love this: http://envbot.kuonet.org/trac/browser/anmaster-trunk/lib/hash.sh 00:30:59 bsmntbombdood: chroot can be perfectly secure 00:31:03 ehird: no, it can't 00:31:06 you just don't let the jailed commands run as root 00:31:23 ehird, a jail is secure as root. 00:31:30 so? 00:31:38 why can't a chroot be secure 00:31:39 A jail also can have a different IP address. 00:31:40 a jail has different processes, different superuser, different filesystem 00:31:45 different network interface 00:31:53 err 00:31:56 not exactly 00:32:05 Well? 00:32:17 a jail does limit network, and superuser, and processes though 00:32:20 unlike a chroot 00:32:33 http://www.freebsd.org/cgi/man.cgi?query=jail&apropos=0&sektion=0&manpath=FreeBSD+7.2-RELEASE&format=html 00:32:33 file system could be just a subdir 00:32:38 Network can be handled in other ways, don't give the jail superuser, and who cares about processes 00:32:39 bsmntbombdood, indeed 00:32:46 chroot handles like 90% of cases :P 00:32:53 ehird, you are soo wrong 00:33:00 Argument by assertion. 00:33:02 a jail as it's own /sbin/init 00:33:10 ehird, read the link bsmntbombdood provided 00:33:10 so? 00:33:13 I did. 00:33:19 I'm just saying that a chroot() is usually fine 00:33:32 ehird, in less than 30 seconds? 00:33:33 read jail(2) also 00:33:35 I don't believe that 00:33:39 If you're paranoid about the chroot security. 00:33:43 AnMaster: what? 00:33:57 pikhq: that's why I'm gonna make a toolchain to automate it :) 00:34:16 what are you doing? 00:34:24 ehird, how well does your befunge interpreter work atm? 00:34:24 what? 00:34:40 AnMaster: can you stop asking me that? it's an ongoing project that i haven't worked on for a few days 00:34:46 ok 00:34:46 are you trying to prove something? 00:34:49 Also, root escalation in a chroot is dangerous. root excalation in a jail is useless. 00:34:57 it seems like you always bring it up when I diasgree with you 00:35:02 *disagree 00:35:05 ehird, no... what would I try to prove? 00:35:10 pikhq: Root escalation is fucked up anyway 00:35:11 and rare 00:35:27 Jails can't create device nodes, can't mount or umount filesystems, can't modify their network configuration... 00:35:39 ...execute instructions... 00:35:42 can't see processes outside the jail 00:35:53 you could do virtual servers with jail right? 00:35:55 I'm really failing to see actual issues here 00:35:59 bsmntbombdood: yep 00:36:00 that's one use 00:36:03 they'd be super fas 00:36:03 t 00:36:06 bsmntbombdood: That's a main use case. 00:36:15 bsmntbombdood: you'd need a script that renice's them a lot though 00:36:19 *renices 00:36:22 to stop hoggers 00:36:28 how's the ulimits work? 00:36:41 you want renice for adaptive speeds 00:36:47 so that people paying more get more of the idle time etc 00:37:01 if I had a dedi and was virtual servering it out I'd use jail(), probably 00:37:04 unless I wanted a linux 00:38:05 what about ram? 00:38:22 bsmntbombdood: adjust ulimits 00:38:34 obviously 00:38:36 it'd be an interesting project ... controlling a bunch of jail()s 00:38:46 anyone want to donate a dedi for it? :-) 00:39:59 damn, now that idea's perlocating in my head 00:40:09 Percolating. 00:40:16 no... 00:40:18 pikhq: maybe it's locating perl 00:40:19 perlocating 00:40:22 Perlocating. 00:40:39 Perl ocation. 00:40:44 compare: pythonocating 00:40:46 How does one ocate? 00:41:06 pikhq, using an ocater on an ocatee. duh 00:41:36 * pikhq needs to invent a new concept in CS just to call it "ocation". 00:42:12 anyway important hint for the future: Should you ever need to write a Swedish name, song title or other word: Don't drop any dots. In the best case it will look rather silly. In the worst case it will either be incomprehensible or mean something totally different. <<< Ja val, forstatt 00:42:26 Ocation: adjusting limits for various non-virtualized process jails automatically. 00:42:59 oerjan, "yes whale, "? 00:43:24 oerjan, quite a good example indeed 00:43:48 might also mean "yes election, " 00:43:55 or 00:44:03 "yes choice, " 00:44:18 i don't see anything in the man page about ulimits 00:44:27 ehird: I approve. 00:44:30 bsmntbombdood: it just jails a process 00:44:32 bsmntbombdood: so ulimit that process 00:45:28 pikhq: It leads to the catchy name for the "v"ps host based on it: Ocator. :-P 00:46:51 i dunno how ulimits work with children 00:47:07 "No Person except a natural born Citizen, or a Citizen of the United States, at the time of the Adoption of this Constitution" 00:47:08 Hmm... 00:47:13 Those commas... 00:47:18 "No Person except a natural born Citizen at the time of the Adoption of this Constitution" 00:47:38 Holy shit all the presidents born after the ratification of the constitution have been invalid 00:47:47 :D 00:48:05 pikhq, did you check http://envbot.kuonet.org/trac/browser/anmaster-trunk/lib/hash.sh 00:48:18 it is truly horrible 00:48:23 it seems children just inherit the same ulimits 00:48:33 Yes. 00:48:36 bsmntbombdood: they're not shared? 00:48:37 which means you can only set the ram limit per process, and not per jail 00:48:39 I hate you. 00:48:41 pikhq, so you like it then? :) 00:48:42 then ulimits are useless 00:48:43 logic terrorists: endangering america by tying the constitution into knots 00:48:45 just fork and use some more 00:48:48 pikhq, intercal is worse 00:48:50 remember that 00:49:01 Yes, but Intercal is meant to be bad. 00:49:10 pikhq, so is this program 00:49:13 bash is not meant. 00:49:14 bsmntbombdood: just watch the process for children and adjust the ulimits based on their usage 00:49:20 bsmntbombdood: you'll be dynamically adjusting them anyway 00:49:30 i don't know if that's practical 00:49:34 pikhq, yeah I rather suspected that bash wasn't meant, rather it was an accident. 00:49:47 bsmntbombdood: well, is here an alternative to ulimits? 00:50:04 Yeah, that's bash. 00:50:07 ehird: it seems like jails ought to support soemthing like that 00:50:08 And, really, all shells. 00:50:10 pikhq, anyway you hate IOCCC then? 00:50:15 (except for csh. That is evil) 00:50:19 bsmntbombdood: mm 00:50:19 C wasn't meant for that 00:50:32 AnMaster: No, I merely joke about hating you. 00:50:33 ;) 00:50:35 bsmntbombdood: One thing's for sure is that a jail() based "V"PS would be way faster than Xen and UML and all of that crap 00:50:40 pikhq, so you are a fan then? 00:50:41 :D 00:50:47 * pikhq nods 00:50:47 ehird: probably 00:50:59 -s securelevel 00:51:00 Sets the kern.securelevel sysctl variable to the specified 00:51:01 value inside the newly created jail. 00:51:03 Oy, what's that do then 00:51:08 pikhq, do you dare look at module loading/unloading? 00:51:24 * pikhq shudders 00:51:24 http://envbot.kuonet.org/trac/browser/anmaster-trunk/lib/modules.sh 00:51:45 securelevel doesn't have anything to do with ulimits 00:51:51 ehird: Xen is very freaking fast. 00:52:05 pikhq: As fast as jail()? 00:52:09 I differ to beg. 00:52:17 with hardware support it should be close 00:52:17 The difference is negligible, I'm sure. 00:52:24 pikhq, it is noticable 00:52:26 Neglegible? 00:52:27 Absolutely not 00:52:30 I used both freebsd jails and xen 00:52:32 I have a Xen server 00:52:35 It's not a fast as the host 00:52:36 By far 00:52:40 It's really fucking fast for a VM 00:52:41 Hmm. 00:52:43 But it's obviously a VM 00:52:46 jails are as fast as the host 00:52:46 whereas a jail() has 0 overhead 00:52:52 Probably the emulated devices. 00:52:57 the only limits being what the ocator sets 00:53:20 ehird, quotas are a PITA in jails though 00:53:21 In fact, I'm rather certain of that; good God, the context switches that it goes through for disk access. 00:53:29 unless you have one partition / jail 00:53:42 AnMaster: Slicehost do one partition 00:53:48 It's not a big deal for most VPSes 00:53:53 Ring 3 to ring 1 to ring 0 to ring 1 to ring 3 to ring 1 to ring 0 to ring 1 to ring 3. 00:53:53 they're too small to warrant partitioning 00:54:00 pikhq: jails or xen 00:54:03 ehird, then quotas doesn't work. And I need per-user quotas on my servers 00:54:05 Xen. 00:54:09 AnMaster: sure quotas work... 00:54:13 why wouldn't they 00:54:26 pikhq, ring 1? 00:54:29 What sort of jail goes through that many context switches for disk access? 00:54:30 if you use 1 partition per user you can't overcommit disk space 00:54:41 bsmntbombdood: this is arguably a good thing. 00:54:45 AnMaster: Xen runs the OS in ring 1 instead of the usual ring 0. 00:54:51 if you tell someone they can have N gigabytes, and they can't, you suck dicks. 00:55:00 pikhq, so it can't use SYSENTER/SYSEXIT stuff? 00:55:05 slower 00:55:12 That's the syscall stuff. 00:55:16 besides, disk is cheap 00:55:16 ehird: you can still overcommit 00:55:24 err SYSCALL/SYSENTER, same thing really iirc 00:55:26 a regular 10GB VM ... you can host 1,000 users with just 10TB of storage 00:55:35 one is intel, one is amd 00:55:44 10TB is el cheapo 00:55:45 And actually, I think that SYSENTER and SYSEXIT in Xen go to the hypervisor which sends it to the kernel. 00:55:57 i mean 00:56:08 10gb is tiny 00:56:14 pikhq, ouch 00:56:23 BADLY ouch 00:56:25 bsmntbombdood: let's say 24gb 00:56:29 That's for the paravirtualisation, rather. 00:56:39 pikhq, what about with hardware support then 00:56:40 For the hardware-assisted stuff, um. 00:56:42 2tb WD drive on newegg: $279.99 00:56:51 still a lot of emulation overhead 00:57:01 Actually, I think that just lets the hypervisor trap it instead of the kernel needing to be patched to handle it. 00:57:26 okay, so 00:57:34 The solution on *Linux* is to make the entry gate not do sysenter. 00:57:37 1000 users with 24gib storage each 00:57:40 = 24000gb 00:57:59 which is 12 x 2TB drives 00:58:07 So $3359 00:58:14 That's dirt cheap for a host 00:58:30 (Linux, in addition to supporting int 8h and sysenter, has an "entry gate", which is just a function that does whatever's fastest) 00:58:31 $3k to host a thousand users is basically fine, assuming you don't intend to grow to megahuge size 00:58:42 bsmntbombdood: conclusion - overcommitting is mostly unneccessary 00:59:25 and also, 24GiB is basically more than most VPS users need 00:59:50 you can get by with 8 drives to serve 1,000 users on 16gib 00:59:58 = $2,239 01:00:16 and really, I don't think anyone could argue that that's prohibitive for a VPS host 01:00:22 compared to the price of the CPUs 01:00:24 ehird: If you do grow to megahuge size, you're getting paid well enough to expand. ;) 01:00:30 yeah, exactly 01:00:38 ehird, you won't have 1000 users per server, more like 100 per server at most 01:00:38 let's say $25/mo for a 24gib vps 01:00:43 that 01:00:48 's $25k a month 01:00:57 which simply dwarfs the $3k cost for storage 01:00:58 AnMaster: of course 01:01:01 AnMaster: Yeah; that's because of CPU usage, not HD usage. 01:01:02 I was just being hypothetical 01:01:05 Which is his point. 01:01:17 even 100 sounds a bit much 01:01:29 and not just cpu, ram too 01:01:30 If it's light usage, 100 is feasible. 01:01:33 yeah, that's some pretty shitty vps 01:01:42 bsmntbombdood: what? 01:01:47 (light, like "Geocities". ;)) 01:01:48 $25/mo for 24gib is really cheap 01:01:58 more usual is $25 for 10gb 01:02:02 gib? 01:02:07 ehird: 100 users per server. 01:02:12 AnMaster: Harddrive maker bytes. 01:02:19 that is 24 gibbering bytes! 01:02:21 :D 01:02:25 ehird: That would be a GB. 01:02:38 pikhq, GiBi? 01:02:47 pikhq: nobody says gibibyte 01:02:49 GiB is the unambiguous way of saying that you're using the binary prefixes. 01:02:58 hmm 01:03:01 alright then :P 01:03:03 ehird, KDE's file info dialog 01:03:07 binary prefixes are the only sensible ones for storage anyway 01:03:12 A gibibyte is 1024 mebibytes, which is 1024 kibibytes, which is 1024 bytes. 01:03:13 ;) 01:03:18 bsmntbombdood: It's not what harddrive makers use 01:03:27 Anyway, $25 for 24gb storage is cheap & reasonable. 01:03:32 on a vps 01:04:23 Hmm... 01:04:32 like you said, disk is cheap 01:04:34 It occurs to me that being a VPS hoster is terribly profitable. 01:04:36 err 01:04:40 you can't compare vps's based on disk space 01:04:45 both 10-based and binary are WAY off 01:05:03 If you have 100 users paying $25/mo, you're earning $2,500 a month. 01:05:06 ternary-based is the only sane way 01:05:20 For a few hundred in dedi maintenance costs. 01:05:31 And 100 users would be a pretty small VPS host... 01:05:32 ehird: half of that will go to colocation and bandwidth, a quarter to hardware 01:05:43 hmm 01:05:49 I don't seem to remember colo being awfully expensive 01:06:15 anyway, $625 in profit is still quite nice for such a small userbase 01:07:29 so start your business 01:07:31 night (not IRCing from bed tonight 01:07:32 ) 01:07:40 bsmntbombdood: I'm considering it :-) 01:08:00 can a 13 year old even legally start a business 01:08:05 i sort of doubt it 01:13:09 Hey, I thought of another advantage of jails. 01:13:25 Since it doesn't have to do virtualization, the tax on the server is less = cheaper to run 01:13:53 -!- Gracenotes has joined. 01:14:18 AnMaster: balanced ternary, of course, otherwise you'd get an imbalance of positive charge 01:14:43 oh he left 01:15:00 oerjan: think I should start a VPS hosting business? 01:15:50 "The game cannot continue: Please change your display settings to be 'High Color (16 bit)'" 01:17:40 erm 01:18:39 i don't have the foggiest clue 01:19:04 well apart from the possible 13 year old problems 01:25:49 ehird: The trick is getting a few hundred users in the first place. ;) 01:26:07 Once you hit the break-even point, you're good. 01:26:20 Mm. 01:26:54 http://static.cbslocal.com/station/wcbs/img/flyover.jpg looks very CGIish 01:26:59 pikhq: My first plan would be getting people I know on and having them do a bit of evangelizing ;-) 01:27:58 -!- puzzlet has quit (Remote closed the connection). 01:28:09 -!- puzzlet has joined. 01:30:01 pikhq: Hmm... I'm trying to decide between Quad-Core 3.2GHz Intel Nehalems, Quad-Core 2.7GHz AMD Opterons and 6-Core 2.4GHz Dunningtons for the hypothetical server... 01:30:26 The 6-core not being the most computationally powerful processor (it's last-generation, I beleive) and also being the most expensive, but I get 12 cores total then. 01:30:38 The Nehalem being the best computational-wise, but I could only have two of them due to Intel being silly. 01:30:58 !slashes /a\\a/a\\aa\\a/a\a 01:30:58 The Opteron not being the best computational-wise too, but it's cheap and I can have four of them. 01:31:06 !help 01:31:06 Supported commands: addinterp bf_txtgen daemons delinterp fyb help info kill userinterps 1l 2l adjust asm axo bch befunge befunge98 bf bf16 bf32 bf8 bfbignum boolfuck c chiqrsx9p cintercal clcintercal cxx dimensifuck echo forth glass glypho hello kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor rot13 sadol sceql sh slashes test trigger udage01 underload unlambda whirl yodawg 01:31:24 You want cheap. 01:31:30 The cores: Nehalem - 8 (16 threads), Opteron - 16, Dunnington - 16 01:31:33 pikhq: True enough. 01:31:35 !slashes a\a 01:31:35 aa 01:31:41 You want to be able to crap out a hundred of these without too much effort if you *really* need to. ;) 01:31:59 !slashes /a\\a/a\\aa\\a/a\\a 01:31:59 aa 01:32:01 pikhq: OTOH, the range is $1,699 (nehalem) to $1,825 (opteron) to $2,399 (dunnington) 01:32:10 pikhq: The dunnington is probably not the best option. 01:32:26 Opteron or Nehalem. 01:32:26 !slashes /a\\a/b/a\a 01:32:26 b 01:32:32 Nehalem would give better performance and is cheaper; but the Opteron lets me have more cores, which is always a good thing when you have a lot of users. 01:32:54 Actually, in this case Opteron would be better performance. 01:32:57 And also, saying "With the latest Intel Nehalem(TM) technology" is more catchy than "Using a regular AMD server CPU" 01:33:03 pikhq: Hmm... 01:33:05 The Opteron does NUMA. 01:33:06 Better parallelism? 01:33:07 Ah. 01:33:15 Opteron it is. 01:33:31 Now how many would I want in one server... 01:33:31 hmm 01:33:40 !slashes /a\\a/xx//x/a\\a/a\a 01:33:40 aaaa 01:33:44 With a 4-processor mobo, I wonder if there'd be any issues using 3? 01:33:55 AMD's not the best when it comes to consumer gear. Their architecture scales like crazy, though. 01:34:03 I honestly don't know. 01:34:14 That would seem to be the right balance between parallelism and "I Can Shit These Out Of My Butt If I Need To" 01:34:18 hmm 01:34:38 !slashes /*/\/a\\\\a\/xx\/\/x\/a\\\\a\//**a\a 01:34:38 aaaa 01:35:06 ¡This cheese is burning me! 01:35:14 !slashes /*/\/a\\\\a\/xx\/\/x\/a\\\\a\//***a\a 01:35:14 aaaa 01:35:24 Oh, Nehalem supports NUMA now. 01:35:28 -!- coppro has joined. 01:35:33 pikhq: Nehalem can give me 16 cores for $3398 whereas Opteron can give me 12 cores for $5475 01:35:34 !slashes /*/ffff/***a\a 01:35:34 ffffffffffffaa 01:35:37 Leaning to Nehalem now 01:35:39 It's the first Intel x86 chip that does. 01:35:46 err 01:35:49 Nehalem can give me 16 THREADS 01:35:49 rather 01:36:00 which is probably more important than cores for VPSes; they can't max them out 01:36:04 Hyperthreading is a hack to get around poor pipelining. 01:36:14 That's true, but it works, doesn't it. 01:36:30 Somewhat. 01:37:05 pikhq: Also, I'm not sure what the effects of having three Opterons in a 4-CPU mobo would be. 01:37:46 I think I'll go Nehalem. It's cheaper, after all. 01:37:58 !slashes /a\\a/xx//x/a\\a//a\\a/xx//x/a\\a//a\\a/xx//x/a\\a/a\a 01:37:59 aaaa 01:39:28 Oh, *damn*... Apparently, hyperthreading tends to increase cache thrashing by about 50%... 01:39:40 Holy shit. 01:39:44 Won't be doing that then. 01:39:50 Esp. since the Nehalem's L2 is puny. 01:40:03 !slashes /a\\a/xx//x/a\\a//\//|//a\\a/xx//x/a\\a//a\\a/xx//x/a\\a/a\a 01:40:04 |aa|aaaa||aa|aa||aa|aaaa||aa|aa|aaaa 01:40:06 And it tends to screw with scheduling. 01:40:17 pikhq: Still, $7300 for four Opterons is pretty costly. 01:40:32 You don't need to have four of the damned things. 01:40:36 True. 01:40:46 But I'll probably want more than 8 cores. 01:41:01 pikhq: remember, the jails are running instructions direct on the processor 01:41:02 *facepalm* 01:41:18 The solution is to wait for Intel to start making the 8-core chips. 01:41:33 possibly. 01:41:47 pikhq: ya think using 3 Opterons in a 4-slot mobo would work? 01:42:01 It sounds like the kind of thing that should work but will fry your computer ;-) 01:42:15 * pikhq checketh 01:42:20 * pikhq also checketh down 01:42:27 !slashes /*/\/a\\\\a\/x\\x\/\/x\\x\/a\\\\aa\\\\a\//***a\a 01:42:27 aaaaaaaaaaaaaaaa 01:42:36 pikhq: Checketh down? 01:42:46 !slashes /*/\/a\\\\a\/x\\x\/\/x\\x\/a\\\\aa\\\\a\//**********a\a 01:42:46 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 01:42:47 3 opterons would give me 12 cores, which should be very much adequate 01:43:29 what are you speccing now? 01:43:53 bsmntbombdood: hypothetical server for VPSery. 01:44:14 In other words, something practical. 01:44:20 Yes. 01:44:33 Practical, but performing, with an emphasis on parallelism over raw throughput. 01:44:49 Talk to IBM about a mainframe. :p 01:44:59 pikhq: Hm, er... do the Opterons support DDR3? 01:45:03 I'm not getting that vibe 01:45:55 "Next-gen Opteron/DDR3 in 2009 - News and Siteseeing" 01:45:58 Answer: no. 01:46:05 pikhq: Back to Nehalem, I suppose. 01:47:20 * pikhq finds it increasingly tempting to upgrade his personal system... 01:47:44 1 core -> 4 cores. 1 GB -> 4 GB or more. Um. Yeah... 01:47:59 how about 12 gb? 01:48:08 bsmntbombdood: Student budget. 01:48:09 -!- Judofyr has quit (Remote closed the connection). 01:48:18 We're talking about $200. 01:48:44 And I need a new motherboard in there. 01:48:54 (just *had* to skimp on my last one, didn't I?) 01:49:13 a new proc, motherboard and ram for $200? 01:49:27 Yeah. 01:49:39 Low-end, obviously. 01:51:46 Keep in mind that I've been dealing with an old single-core processor... This will still seem like a major upgrade to me. 01:54:28 -!- puzzlet_ has joined. 01:55:20 -!- puzzlet has quit (Remote closed the connection). 01:57:43 http://james-iry.blogspot.com/2009/05/brief-incomplete-and-mostly-wrong.html :D 01:58:28 I... *Just* found that link from somewhere else. 01:58:38 probably 01:58:38 I was literally about to paste it. 01:58:50 BWAHAHAHA 01:58:51 DAMN YOU OERJAN!!! :p 01:59:38 (this link pasted for the 5% of #esoteric members not following reddit) 02:00:01 or something 02:00:56 Not from reddit. 02:01:00 Different blog. 02:01:02 ah 02:01:03 * pikhq should follow reddit, though. 02:01:41 * pikhq loves the description of Haskell 02:02:01 * oerjan hasn't got to that yet 02:02:25 i stopped at the prolog description to paste 02:05:21 also, perl 02:08:03 That part is not a joke. 02:08:07 It is truth. 02:08:15 (true but false) anyone? 02:08:44 "Later still, in an effort to cash in on the popularity of skin diseases the language is renamed ECMAScript. 02:09:11 what skin disease is that, anyhow? 02:09:58 It sounds vaguely like eczema. 02:10:38 And sounds a lot like some sort of skin disease. 02:10:43 (unspecified) 02:10:56 gar newegg hasn't given me a tracking number yet 02:12:59 they are far too easy on java 02:14:30 Nah. They are far, far too easy on C#. 02:27:05 http://www.reddit.com/r/programming/comments/8itq7/a_brief_incomplete_and_mostly_wrong_history_of/c09f5ur 02:27:35 http://www.huffingtonpost.com/2009/05/08/obama-mustard-attack-beco_n_199953.html 02:27:51 Republicans accuse Obama of elitism because he orders spicy mustard on his burger. 02:27:55 I freaking hate humanity. 02:34:35 i heard that 02:43:46 heh. I'm watching the Daily Show episode on that at the moment... 02:45:37 for my once-weekly-entertainment-spree >_> 02:55:00 Eh, you're forgiven. The Daily Show is worth watching. 03:07:13 -!- psygnisfive has joined. 03:16:10 Well, I think I've outlived the humor of this /nick. 03:16:13 -!- TravelGreaseGod has changed nick to GregorR. 03:16:48 * pikhq bows before the former diety of traveled grease 03:18:40 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 03:19:24 -!- bsmntbombdood has joined. 03:20:32 Hwaet! 03:20:50 Whoot, I'm on OOPSLA. 03:21:06 oopsla? 03:21:12 how can you be /on/ oopsla? 03:21:26 Sneaky 'o' key, pushing its way in on the 'i' key's territory. 03:22:06 what I don't like about haskell is exceptions :< 03:22:18 though I love the type 03:22:23 throw :: (Exception e) => e -> a 03:22:33 What I don't like about haskell is that I don't know enough math to understand monads. :p 03:23:25 What I don't like about C is that I don't know enough math to understand pointer arithmetic...... oh wait... 03:23:33 Hahaha 03:23:40 Gracenotes: You may want to revisit the third grade :P 03:24:19 heh. for pointers, it's mainly the syntax that gets you at first 03:25:07 easy peasy once you're comfortable. same with monads, and uncountable other language features 03:25:20 ... or should I say countably infinite? :o 03:25:31 Uncountably finite? 03:25:32 i don't understand not understanding p ointers 03:25:44 I don't understand not understanding pointers 03:25:47 I do understand not understanding monads 03:25:49 the syntax mainly 03:25:52 I don't understand not understanding not understanding pointers. 03:25:55 i don't understand not understanding monads 03:25:56 this is because I understand the former, and not the latter 03:25:59 lvalues vs. rvalues 03:26:09 other things that are not intuitive whatsoever at first 03:26:12 BWAHAHAHA *CACKLE* *COUGH* 03:26:17 bsmntbombdood: It's one of those things that you wrap your head around and then it makes perfect, perfect sense. 03:26:35 oerjan: You've got an Erdos number smaller than mine; shaddup. 03:26:41 now monads, like typeclasses in general, are abstract. They're like uber design patterns 03:26:48 gregorr: how can you be IN oopsla 03:26:55 well, that's not true. I understand monads, in the sense of how to use the list and maybe monads, declare a new monad (+google), whatever 03:26:58 just use specific cases of them. you'' be fine 03:27:02 but I don't have an "intuitive" grasp of monads yet 03:27:08 psygnisfive: "published" in 03:27:16 list and maybe are a good start. Try State now :) 03:27:21 oh i see. i didnt realize oopsla was a publication 03:27:33 State is a pretty big hurdle but getting past it is significant 03:27:41 in... understanding. 03:27:42 stuff. 03:28:01 and then I have to figure out monoids 03:28:02 :< 03:28:10 GregorR: Oh, you have an Erdos number now? :p 03:28:13 and why do I want to learn a state monad in Haskell 03:28:20 pikhq: Probably a very high one :P 03:29:35 Oh, you're at a conference. 03:29:49 OOPSLA is a conference, not a journal. XD 03:29:59 GregorR: Well, then, your Erdos number is 7. 03:30:00 it is essentially a monad wrapping around an action, not a data structure. It shows even more so how they chain actions. and you basically get understanding Reader and Writer for free 03:30:10 Making its way through two other members of this chat room. 03:30:12 :p 03:30:24 ? 03:30:30 How does that establish it as 7? 03:30:45 Assuming you count working on a program together, sorry. 03:31:21 I have an Erdos number of .. undefined 03:31:22 Oerjan has an Erdos number of 5, he and I collaborated for PEBBLE, giving me an Erdos of 6. You and I collaborated for PEBBLE, giving you an Erdos number of 7. :p 03:31:34 no papers whatsoever. undergrad! *cringes* 03:31:59 I submit homework for you TAs to grade! *mwahahaha* 03:32:09 huh? 03:32:13 Gracenotes: I'm an undergrad myself. 03:32:16 pikhq: 4, thank you very much 03:32:24 why would working on a program together count? 03:32:26 oerjan: My apologies. 03:32:28 oh, nice :D when did you publish your first paper? 03:32:37 * oerjan had to check again 03:32:52 Have yet to. I said "assuming you count working on a program together". ;) 03:33:00 4 is not bad. 03:33:23 bsmntbombdood: "Published work". 03:33:37 itym "paper" 03:33:47 SWEET 03:33:48 So far, Free Realms is leaving a bad taste in my mouth 03:33:49 I'm 4 8-D 03:33:56 Some work published work of research. 03:34:03 GregorR: Oh? 03:34:22 bsmntbombdood: The question is, does PEBBLE count as a research project? :p 03:34:32 Me -> Jan Vitek -> Nir Shavit -> Michael E. Saks -> Erdos 03:34:49 Huh. 03:35:21 NOOOOOOOOOOOOO 03:35:24 Well, then. If PEBBLE doesn't count as a research project, does Plof? 03:35:41 pikhq: Only whenif I publish anything about it :P 03:35:45 I mean, it's a tiny bit unique as far as programming languages go. 03:35:56 * oerjan feels his #esoteric 1337-ness slipping 03:36:17 GregorR: Plof itself might count as a published work. 03:37:09 pikhq: ... sorry, when did you publish a paper? O-o 03:37:19 Gracenotes: that's what i said 03:37:29 Gracenotes: I have yet to, as I said previously. 03:37:37 It really only counts if it's a research paper :P 03:37:45 GregorR: Yeah, shaddup. 03:37:56 pikhq: ah, okay. Doing any research, by chance? 03:38:06 * Gracenotes hopes to get involved with that >_> 03:38:08 Not yet. 03:38:09 He's an undergrad :P 03:38:13 Inevitably will be. 03:38:19 * pikhq plans to do grad school 03:38:33 Go Purdue whooooooooo 03:38:36 okay. just wondrin. People going to research universities can get started somewhat early-ish, I've heard 03:38:50 Though my university does allow for undergrad research. 03:44:04 execState (tick >>= (\x -> tick >>= (\y -> return y))) 5 03:44:11 h m m 03:46:51 execState (tick >>= (\x -> return 42)) 8 03:46:54 returns 9 still 03:47:00 wat 03:47:29 what's the definition of tick? 03:47:35 tick = tock 03:47:35 oh right 03:47:46 from the example 03:47:47 comex: try runState 03:47:47 tick :: State Int Int 03:47:47 tick = do n <- get 03:47:47 put (n+1) 03:47:49 return n 03:47:59 oerjan: what's the difference 03:48:04 execState only returns the final state 03:48:11 evalState only the final result 03:48:17 runState returns both in a tuple 03:48:21 iirc 03:48:41 quite 03:49:16 and the way this works with lazy state is pretty interesting too :) 03:52:49 *Main> execState (tick >>= (\x -> tock >>= (\y -> return y))) 5 03:52:58 where tock is multiply by two 03:53:05 performs tick FIRST 03:53:08 wtf 03:53:19 of course 03:53:40 that's as designed 03:53:46 well, I guess it makes more sense in do notation that way... 03:54:22 i think i actually saw someone make a ReversedState monad 03:54:31 which does the opposite 03:54:55 it probably needs one to be careful with deadlocks, though 03:55:20 since you would then have information passing both ways 03:55:42 which both ways? 03:55:59 the \x and \y are always passed to the right 03:56:12 as you can tell just by scope 03:56:31 but that monad somehow managed to make state change backwards 03:57:44 I still don't see why not run the inner function first 03:57:49 but I don't get states either so :p 03:58:16 monads differ in whether they mostly run the left or right part of >>= first 03:58:27 as in, I don't see where you would use a state instead of a function (a, b) -> (a, b) 04:00:04 comex: here's something that might help show you why it must be like that 04:00:05 mainly because every example I've seen is of the form 04:00:12 do foo <- get 04:00:13 put bar 04:00:15 return qux 04:00:36 define tack n = do m <- get; put (m+n); return m 04:00:55 but why not replace it with f foo = (bar, qux) 04:00:57 (just to keep that tradition ;) ) 04:01:14 oh i'm still on a previous question 04:01:33 well, yeah 04:01:38 with do notation it looks all nice and forwards 04:01:52 comex: indeed f foo = (bar, qux) is essentially what it is inside 04:02:52 now try execState (tick >>= (\x -> tack x >>= (\y -> return y))) 5 04:03:02 well, I have yet to see an example that doesn't have exactly those two lines 04:03:03 :< 04:03:25 you then see that tack x cannot possibly run before you know what x is 04:03:44 (well, ignoring laziness, which might allow it anyhow) 04:03:54 which means it must run after the tick 04:05:11 comex: as for why not replace it with f foo = (bar, qux) 04:05:14 ah, that makes sense 04:05:19 you're not getting a meaningful return value 04:05:22 but you are getting a meaningful parameter 04:05:46 well if you do that you could not use all the nice monad combinators that work on _all_ monads for it 04:06:31 for example, with import Control.Monad you can do: 04:07:10 runState (replicateM 5 tick) 10 04:07:47 http://haskell.org/ghc/docs/latest/html/libraries/base/src/Control-Monad.html#sequence 04:07:51 :< 04:08:10 -!- bsmntbombdood has quit (Success). 04:08:18 yeah replicateM is just ordinary replicate + sequence 04:08:50 -!- bsmntbombdood has joined. 04:09:11 try runState (sequence [tick, tock, tick, tock]) 5 04:10:15 actually you can even try: 04:10:48 evalState (sequence (cycle [tick, tock])) 5 04:10:51 (i think) 04:11:20 sucks! 04:11:29 hm? 04:11:51 er maybe i should have warned you about infinite output :D 04:11:59 bah 04:12:05 all the cool kids are learning haskell 04:12:32 * GregorR wishes there was more Haskell at Purdue. 04:12:39 I didn't actually run it :p 04:12:51 ok, now I did 04:12:59 take 20 $ evalState (sequence (cycle [tick, tock])) 5 04:13:07 if you want to abbreviate a bit 04:13:09 ...Free Realms sent me 11 emails saying that my name has been approved 04:13:26 Sgeo: Well, they reaaaaaaaaaaaaally approve of it. 04:13:47 Make that 12 04:13:50 I want a picture :( 04:13:51 comex: ok if you hadn't run it then what was it that sucked? :D 04:13:55 but I'm going to bed, maybe I'll run it tomorrow 04:14:01 oerjan: well, I knew it was going to do something infinite! 04:14:10 They're sending an email every 1-2 minutes 04:14:16 s/run/understand 04:14:25 maybe I'll understand it tomorrow :) 04:14:29 Wait, no, there's a 13 minute gap 04:14:40 comex: well this is a gradual process 04:15:01 comex, of the emails? 04:16:15 Sgeo: did you ask to be approved? 04:16:50 oerjan, since I used a custom name, it had to be approved before it could be used 04:29:45 So, I'm wondering if the American stereotype of a Swedish accent has anything to do with reality. 04:30:01 As far as I can tell, there's no similarity whatsoever. 04:30:42 It's not even just taken to the extreme or something, it just has nothing to do with anything. 04:32:50 The American stereotype of a Swedish accent has more to do with the Muppets than anything else. 04:33:00 :) 04:34:32 börk börk börk 04:34:41 i'd hit bjork 04:37:41 oerjan: Ah; wasn't aware that anyone overseas would've seen that. 04:37:43 :) 04:38:28 the muppet show was shown in norway already when we had a single state channel 04:38:52 which means that pretty much _everyone_ of that age has seen it 04:39:21 Ah. 04:39:29 mind you that was a while ago and i may not remember that much of it 04:39:43 So, that's a piece of Americana that's about as well-known in Norway. Got it. 04:41:58 So what's the reverse? 04:42:07 What piece of Norweigicana is known in the US? 04:42:16 ... 04:42:17 Hmm. 04:42:31 the nobel price and, well, that's about it 04:42:42 Beowulf. 04:42:54 beowulf is not particularly norwegian 04:43:02 (not necessarily Norwegian. Close?) 04:43:10 but then, i don't know beowulf 04:43:24 * pikhq wikipedes 04:43:32 ok i vaguely recall it's mostly set in denmark and sweden or something 04:43:41 and that it's in old english 04:44:01 Scandinavian, at least. 04:44:17 henrik ibsen, i hope 04:44:48 might be too class-dependent... 04:45:12 "The Scream" 04:45:34 i'm sure there must be _something_ from after the 19th century 04:45:48 Hmm. 04:45:55 Start naming things. 04:45:55 :p 04:46:12 the A-HA band 04:46:26 No. 04:46:41 this may be complicated by the fact that for ten years or so, _i_ haven't got much of a clue what's hot in norway :D 04:46:55 (maybe make that 15) 04:47:01 :D 04:47:23 * GregorR 's Americanness seeps into Europe 04:47:28 Age of Conan MMORPG 04:47:47 Henrik Ibsen isn't well-known amongst the general populace, but decently known amongst literature buffs. 04:47:51 (i don't play MMORPGs, but i know that's made in norway) 04:48:06 Age of Conan MMORPG is known *of*... 04:48:17 Not the most popular MMO, but it has a niche. 04:48:43 "The Scream" is something people don't realise they know. :p 04:49:38 Jarlsberg cheese :D 04:49:49 it's our main diary export i think 04:49:51 Didn't know that was a kind of cheese. 04:49:55 *dairy 04:50:09 We're Americans - we don't know things from other continents. :p 04:50:11 well i saw it when i was in the us 04:50:17 iirc 04:50:35 Doesn't mean I know what it is. XD 04:50:48 lutefisk 04:51:30 (if it's known, then it's probably as something to scare children with :D) 04:51:38 (mind you i love it) 04:51:39 Never had it, which is kinda surprising considering the amount of family I have in Minnesota. 04:52:01 (Minnesota was settled by Norwegian immigrants) 04:52:12 i know, vaguely 04:52:29 And it's something to scare children with, definitely. 04:52:36 (good God, it's *caustic*!) 04:52:51 possibly whale meat, although we try not to remind you of it :D 04:53:05 That'd be illegal. 04:53:31 Except for those common Norwegian-Amerindians :P 04:53:44 I think Norwegian cuisine is the most likely to be known in the US... 04:53:57 Except for the few fans of Norwegian metal, I guess? :p 04:54:08 brown cheese just to complete the stereotypically bad norwegian cuisine 04:54:11 * GregorR wonders what Norwegian cuisine is :P 04:54:26 GregorR: Visit Minnesota. You'll know. 04:54:31 (strangely enough norway has had several internationally award-winning chefs) 04:54:32 I don't like cheese :P 04:54:45 Or fish, just to cover both of those bases. 04:54:50 but i doubt they're cuisine is very traditional 04:54:53 *their 04:55:00 That covers a good chunk of Norwegian cuisine. 04:55:05 indeed 04:55:10 * pikhq shoves sushi down GregorR's throat 04:55:21 it doesn't cover fårikål though 04:55:32 which was once voted the national dish 04:55:38 (sheep in cabbage) 04:55:52 (with whole peppers) 04:56:23 Actually sounds rather tasty, if a bit simple. 04:56:34 (that would be the relatively small black kind of pepper) 04:56:48 Ah, yeah... I was about to say. 04:56:55 pikhq: I like those sushi things that don't contain fish. 04:57:04 Kinda surprising to see capsicum in Norway. 04:57:15 GregorR: Mmm... Fair enough. 04:57:20 um is that what it is? 04:57:39 Capsicum is the genus of chili peppers. 04:57:42 Fårikål is a traditional Norwegian dish, consisting of pieces of mutton with bone, cabbage, whole black pepper and a little wheat flour, cooked for several hours in a casserole, traditionally served with potatoes boiled in their jackets. 04:57:44 This sounds good. 04:57:51 well that is not what you use in fårikål 04:57:57 I know. 04:58:06 it's a very seasonal dish, btw 04:58:09 Now to find one of those popular on-every-corner Norwegian restaurants. 04:58:12 autumn 04:58:22 Americans have no conception of seasonality. 04:58:26 :P 04:58:43 how unseasonable 04:59:12 Mother Nature is America's bitch. :p 04:59:19 Pretty much. 04:59:44 I would make a joke about eating an extremely-out-of-season fruit, but I literally have no conception of what fruits are in season at what time. 05:00:05 A hint: they're out of season in the winter. 05:00:18 ;) 05:00:25 There are fruits that are in season in the winter! Just not many. 05:00:27 GregorR, same her 05:00:29 here 05:00:34 it depends what you mean 05:00:49 for example few norwegians eat figs and dates other than christmas 05:01:20 of course those are generaly dried 05:01:20 Hmm. Apparently pears and tangerines are in season during the winter. 05:01:27 *lly 05:01:38 (Wikipedia FTW) 05:01:48 Free Realms is still spamming me 05:02:01 Sgeo: Hahahahaha 05:02:11 maybe google to see if it's a known problem? 05:02:12 Although there was a 45 minute gap between the latest junk 05:03:05 Sgeo: Years from now, you'll be receiving it still, but so rarely that you always forget about it until you get another one. 05:03:58 If it sends an email every time a separate mod approves it, then it might pick up tomorrow, or keep going for weeks, until it runs dry 05:08:44 * pikhq observes that American cuisine is very, very freaking odd... 05:09:09 pikhq: American cuisine: Take something simple, put cheese on it. 05:09:27 GregorR: That covers some things. 05:09:35 *cough* Cheeseburgers. 05:10:01 GregorR, I'm eating Pasta and cheese tonight... just like most nights 05:10:32 Doesn't cover how somehow Italian, Mexican, and Chinese foods have somehow morphed into being a major part of our national cuisine... 05:10:44 pikhq: 'struth. 05:11:11 Or, for that matter, British or Native American foods. 05:11:22 (or freaking adoration of corn. :p) 05:11:27 s/or/our/ 05:11:51 British "cuisine" is not dissimilar to American "cuisine" 05:12:06 Less focus on cheese, more focus on flavorless and fattening. 05:12:29 I thought that American cuisine had cornered the market on fattening. 05:12:52 The Brits may have such things as frying bread in bacon grease, but we, *we* have fast-food. 05:13:23 One day in an English pub I had eggs not-actually-over-hard-but-we'll-forgive-them-they're-British, baked beans, the liver of a whiskey-fed-chicken, mushrooms and fried tomato. 05:14:55 Also, not all British cuisine these days is flavorless; remember, curry has become a national dish for them. ;) 05:15:22 The McDonalds' there have curry for fries (chips) 05:17:42 -!- puzzlet_ has quit (Remote closed the connection). 05:17:48 -!- puzzlet has joined. 05:18:44 Also, they have tea. Any other culinary sins they make are forgiven by that alone. 05:19:03 * pikhq is fond of tea, in case you couldn't tell 05:19:16 Bleh 05:19:55 Do I sense someone who has been mistreated by American tea 'sensibilities'? 05:20:31 I don't like tea. 05:20:42 I'm hyposmic, so it tastes like bitter water to me. 05:20:50 (Along with coffee, beer) 05:21:50 * pikhq notes that bad tea, coffee, and beer *is* bitter water. 05:22:17 My insane tea friends took me about seven or eight times to this deluxe absurdly-expensive tea place. 05:22:19 But with hyposmia, good tea, coffee, and beer might be a bit lost on you. :/ 05:22:26 hyposmic? 05:22:38 Reduced sense of smell. 05:22:57 GregorR: So, nothing with subtle flavors for you. 05:23:04 Yuh 05:23:22 Or at least, no subtle flavors if there's also a significant bitter flavor to overwhelm :P 05:23:23 I hope you're at least fond of spicy foods. :p 05:23:28 Is it possible to have hyposmia and not know it? 05:23:30 I wurve spicy foods. 05:23:43 Good man. 05:23:46 Sgeo: Hyposmia is a symptom, not a disease. 05:23:59 Sgeo: Which is to say, if you think you're hyposmic, you're hyposmic :P 05:24:16 Sgeo: But if you've never cooked a rotten chicken only for your roommate to come home and go "holy shit WTF is that smell", you're probably not. 05:24:18 Or maybe I'm just bad with tastes. I notice other smells just fine 05:24:28 * pikhq likes his food with either subtle flavors or incredibly strong flavors 05:24:40 Mmm, salt and vinegar chips... 05:24:44 (or at least, what I think is just fine) 05:24:45 GregorR: Damn. 05:24:48 Just damn. 05:26:10 pikhq: I don't like inviting friends over, because the experience often exposes odors I was not aware of ... 05:26:45 Meanwhile I, if anything, exhibit hyperosmia. 05:26:46 -!- puzzlet has quit (Remote closed the connection). 05:26:54 -!- puzzlet has joined. 05:27:21 ("fun" when combined with autism) 05:32:09 Or maybe I'm just bad with tastes. I notice other smells just fine 05:32:13 Also, I'm not bad with tastes :P 05:32:30 The subset of food which really /requires/ smell (a subset I am uniquely qualified to identify) is quite small. 05:32:37 (Not true for drinks) 05:33:06 I'd imagine soda would be just fine. 05:33:08 Beyond that, well... 05:33:22 But how would you distinguish between different sodas? 05:33:22 Yeah, I drink soda. I wurve the bubblitude :P 05:33:28 Sgeo: .......... srsly? 05:33:46 Sgeo: I can tell the difference between all sodas, and have a distinct (and distinctive!) favorite. 05:34:04 Sgeo: Soda flavor is anything but subtle. 05:34:07 I thought taste only existed for very general things ("sweet" etc) 05:34:24 Scent gives extra detail on flavors. 05:34:24 And beyond that is smell, or maybe I'm just very confused and tired right now 05:34:37 Try tasting stuff with your nose plugged. 05:35:19 GregorR: Your favorite being? 05:35:23 Based on descriptions I've heard from people who've tried that, I think I've got a one-sense-weak-makes-the-other-stronger thing going on :P 05:35:25 Moxie! 05:35:31 It's distinctively different! (And better) 05:35:33 ... Moxie? 05:35:39 Never had it. 05:35:44 Of course not, you're not a Mainer :P 05:35:48 Or heard of it. 05:35:51 Of course not, you're not a Mainer :P 05:36:01 It's the oldest continually produced soda in the US. 05:36:30 Huh. 05:37:17 * pikhq is rather fond of Dr. Pepper & Mountain Dew... 05:37:35 Surely you've heard the English term "moxie", though? 05:37:40 And also a *good* ginger ale. (most of it isn't) 05:37:44 Yeah. 05:37:52 That term comes from the soda, not vice-versa. 05:37:57 Wow. 05:42:07 http://en.wikipedia.org/wiki/Moxie , btw 05:42:47 I looked it up. 05:42:54 And kinda want to try it. 05:43:11 If for no other reason then that a soda I haven't tried is just wrong. :p 05:43:13 Unless you know somebody who imports it locally (e.g. me :P ), it's too expensive to try. 05:43:33 Next time I'm in New England. 05:43:39 (with any luck, next summer) 05:43:43 It's difficult to find outside of Maine proper. 05:44:16 There's a lot of stuff that's hard to find in Boston. :p 05:45:18 Hey, sure enough. The Star Market in Cambridge has it. 05:45:26 Easy enoguh. 05:46:00 As does the one in Newton, and that odd grocery store in Somerville... 05:46:09 Well, then. 05:46:17 There are a few places here and there you can find it. That's noreast enough that I'm not sure whether you'd be getting authentic Moxie (look for orange) or the just-as-good-if-not-better Real Soda brand Moxie. 05:46:47 Doubt I'll find it here in Missouri, though. 05:46:59 Prooooooooobably not :P 05:47:07 Actually easier to find in Oregon I'd bet :P 05:47:12 ... But there's a shop in Montana that sells it. WTF? 05:47:31 ... Ah. Missoula. The only town in that hellhole worth living in. 05:47:32 It probably all comes down to "somebody from Maine moved there and opened a store" ;) 05:47:38 Odd coincidence, that. 05:47:46 Missoula is a college town. 05:47:54 So, probably. 05:50:09 Jeeze; $24 for 24 cans. 05:50:19 Yeah, just a little pricy. 05:50:25 Eh, you can get a better deal than that (before shipping) 05:50:37 Still. 05:51:00 I think it's a bit cheaper to be fond of Dr. Pepper without HFCS. 05:51:16 Yeah, my Moxie lust is expensive :P 05:51:19 (at least that I could reasonably drive down and load my car up with) 05:51:43 Dr. Pepper without corn syrup is a bit difficult to find... 05:52:08 There's one distributor that makes it. 05:52:20 I can find it easily, but only one bottle at a time and overpriced :P 05:52:53 * pikhq got addicted for that short time it was available nation-wide... 05:52:59 * Sgeo joins Reddit 05:53:31 Now, the only way to get it sanely is to go down to Texas and clean out a store. :p 05:57:05 * GregorR wonders how "go down to Texas" and "sanely" can be in the same sentence. 05:59:19 It's not far into Texas, and it's in the name of one of the few American products that doesn't use corn syrup. 05:59:30 Making it worth it. 05:59:31 :p 05:59:40 And I use that smily a lot. 05:59:53 Hey now, there are lots of American products that don't use corn syrup. 05:59:59 Few of them are beverages, but still. 06:00:34 And few of them are food products for that matter. 06:00:59 i think some day OOPSLA should be held in Upsala 06:01:24 I think that we should cut back on the sugar tarriff and the corn subsidies. 06:01:31 For example, a 2009 Ford F-150 contains no corn syrup, as that would be too efficient of an alternative fuel. 06:01:35 And we wurve oil. 06:01:41 :) 06:01:58 i agree, pikhq. 06:02:37 but thats partially because corn takes more energy to turn into fuel than it produces 06:03:18 As a side benefit: corn syrup tastes awful. 06:03:22 it does 06:03:26 pure sugar is better 06:04:03 Of course, this is America; screw taste, the almighty dollar is in charge. 06:04:33 Oh, and also subversion of the free market. 06:04:40 (as we are wont to do here) 06:05:23 afk all 06:05:24 is the turtlre? 06:05:45 NO - THE ACCUMULATOR 06:08:05 Hmm, that's interesting... In mice, HFCS has been shown to increase the absorption of fat. 06:08:20 hm. 06:08:25 So, we're manipulating the free market to make Americans fat. *Great*. 06:08:45 Oh, it also seems to supress mice's tendency to stop themselves from overeating. 06:09:10 yes 06:09:11 this is true 06:09:24 as it does not release the hormones that make you feel full 06:09:37 Explains a lot. 06:09:48 it also doesnt cause the release of insulin, which leads to elevated blood sugar levels, and a higher risk of diabetes 06:10:00 Also explains a lot. 06:10:09 yep 06:10:13 We're really fucking ourselves over here. 06:10:20 yep! 06:10:36 And we're not even getting better-tasting food out of it! 06:11:02 yeah 06:11:09 if were gonna kill ourselves, we shold atleast enjoy it 06:11:18 Seriously. 06:18:51 Really annoying when you consider that we've got the ability to grow a *lot* of sugar beet... 06:19:04 And a decent bit of sugar cane as well. 06:20:43 Both of which would make for a really good ethanol crop if corn weren't so damned subsidised. 06:39:27 -!- ais523 has joined. 06:43:59 -!- ais523 has quit (Read error: 104 (Connection reset by peer)). 06:44:37 -!- ais523 has joined. 06:45:22 Maybe I should start eating HFCS foods 06:45:31 Well, the diabetes bit doesn't sound good 06:49:42 * oerjan figures his infinite loop in slashes should work but it may be easier to use haskell to generate it 06:50:22 oerjan: you're trying to write a nontrivial /// infinite loop 06:50:26 yep 06:50:45 I spent hours working on that problem once, my conclusions were that it was obviously possible, but would require someone who made fewer silly mistakes than me 06:50:54 ah :) 06:51:05 /// ? 06:51:30 http://esoteric.voxelperfect.net/wiki/Slashes 07:14:21 okay, i got dead code elimination pass correct... now esotope-bfc prints only PUTS("Hello world!") for hello.b. 07:15:58 in allcaps? 07:16:04 is that a wrapper macro? 07:16:14 ais523: defined as #define PUTS(s) fwrite(s, 1, sizeof(s)-1, stdout) 07:16:31 also, does it optimise 99bob into printing a constant string? 07:16:52 partially. 07:17:42 propagation pass is (as i mentioned days ago) ad-hoc, so it misses some cases 07:19:13 for example, it doesn't propagate Output[] node yet so there are lots of code looks like p[3] += 72; PUTC(p[3]); 07:20:45 -!- oerjan has quit ("Good night"). 07:34:21 7 more spam messages from guess who 07:57:08 wow, someone was reminicing about old-fashioned copy protection schemes on Slashdot, and apparently one of them involved rot13ing the binary 07:57:15 amongst other things 07:57:20 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 07:57:24 I have to admit, I never even thought of rot13ing binary data... 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:03:32 memfrob 08:08:59 memfrob would make more sense 08:09:05 its purpose is to hide strings in executables, right? 08:11:18 probably more for the lulz 08:13:14 -!- coppro has quit (Read error: 110 (Connection timed out)). 08:13:29 There was a demo of a game they told me to poke at, to see how it was made; the data files of that were memfrob()'d (except not with the constant 42) .zip files. 08:14:06 heh, XOR encryption 08:14:28 I remember recommending that to a friend about 6 or 7 years ago to use on his website 08:14:40 because it was stronger than the nonencryption he was using, but I would still be able to crack i 08:14:42 *it 08:14:54 Not quite sure what purpose it served; I guess it was there to discourage any "non-approved" modificating, or the extraction of graphical resources, but those both seem to be rather lamey reasons. 08:18:17 And there was that actual "binary rot-0x1b modulo 256 applied to .tar archives" ehird's router-box-or-whatever had for configuration dumps. 08:19:01 the impression that I got from the Slashdot comment was that the binary had actually changed all the opcodes and data that happened to correspond to letters of the alphabet round 13 alphabetical places 08:19:14 and left everything else untouched 08:19:16 Yes, that's pretty senseless. 08:19:22 quite a few common x86 instructions correspond to capital letters, though 08:19:44 in realmode, at least, the lowercase letters all seem to be invalid opcodes 08:23:29 -!- bsmntbombdood has quit (Remote closed the connection). 08:25:41 -!- daffa has joined. 08:25:59 What's up, all? 08:26:11 I'm having a completely crazy sleep pattern atm 08:26:22 so I took advantage of the 24-hour opening of the University atm to come in at 6am 08:26:46 -!- daffa has changed nick to evincar. 08:27:38 Sounds like fun. 08:28:16 well, it's a case of "would I rather be lying in bed doing nothing wondering if 6am is too late to try to go to sleep, or come online and do nothing?" 08:29:06 Fair enough. 08:29:36 It's only 3:30 where I am. Been programming, and debating going for a run. 08:29:38 esoteric-wise, the most interesting things I've done recently are write a BF Joust interp, and a fast Underload interp in a sane language 08:29:55 Ah, nice. 08:30:05 BF just never goes away. 08:30:12 because it's ridiculous that nearly all the Underload interps were in esolangs 08:30:14 although rather fun 08:30:25 Ha, true. 08:31:00 I dunno, I've been working on a legit language lately, so I figured it was time to take a break and see what the other half is doing. 08:31:50 I had a great idea for a rather esoteric language, but unfortunately it seems too useful to count as an esolang in the truest sense of the term. 08:31:56 More like domain-specific. 08:32:08 I don't mind esolangs being useful 08:32:19 in fact, I've been aiming towards that sort of thing quite a lot recently 08:32:43 Underlambda is (will be) designed to be useful, Thutu turned out to be useful by mistake, and I've been progressively making INTERCAL more useful and hoping nobody notices 08:32:54 Heh. 08:33:01 Well, I guess it amounts to preference. 08:34:02 My first esolang was rather useful, come to think of it, for batch scripting and lightweight graphics applications... 08:34:13 ...made in good ol' QuickBASIC 4.5, of course. 08:34:17 Such a dinosaur. 08:35:49 I was stuck with QBasic ages ago 08:36:44 It was a certain flavour of esoteria in its own right. 08:37:37 Consider the things that the QB community did with that language in the early 00's that should never have been done with it. 08:37:51 you can say that about any language, more or less 08:38:00 I know when I was younger I implemented pointers in QBasic 08:38:05 actual memory pointers 08:38:08 Oh, I loved doing that. 08:38:09 that's esoteric in its own way 08:38:17 Yeah. 08:38:31 It feels so good to get a linked list working in QB. 08:38:41 And at the same time vile and disgusting. 08:39:01 I couldn't figure out how to extract the value of a number, so in the end I blitted into a temporary variable and read the value from there 08:39:12 I was so programming-naive then I didn't even know about floating point 08:39:19 so no wonder I couldn't figure out the byte order 08:39:25 Haha... 08:39:40 ...but programming knowledge builds on itself. 08:39:59 I've come a long way in the...almost ten years it's been. 08:40:11 Since I started getting interested, that is. 08:40:13 a bit more than that for me 08:40:18 I started programming when I was 6, I'm 22 now 08:40:20 I figured. 08:40:24 Ah. 08:40:30 Pascal and BBC Basic were my first languages 08:40:40 and, as you can probably guess, 6502 asm was my first asm 08:40:46 Yep yep. 08:41:18 We recently HTMLized the QB manual. 08:41:24 Or QBasic manual, anyway. 08:41:26 we as in #esoteric? 08:41:29 I was a QB/Pascal child, myself, but I got into C rather early on, and then C++ has pretty much stuck with me ever since, despite Perl, Python, and what have you. 08:41:35 I know there was a discussion about it here a while ago 08:41:38 The dos .HLP decompiler didn't work on the QuickBasic one. 08:41:42 I even managed to get what I think is a legal copy 08:41:47 at least, it came from the Microsoft website 08:41:50 and it runs in DOSBox 08:41:59 (also, I have a spare Win98 license around, so I'll count it against that) 08:42:06 http://zem.fi/~fis/qbc.html 08:42:25 That's the one with CSS rules to make the colors correct. 08:42:34 Neat. 08:42:38 This is the 1.1 manual? 08:42:57 Doesn't look like 4.5. 08:43:25 It's the QBasic one; like I said, real qb45 manual didn't get through the "helphelp" tool. 08:43:40 Ah, gotcha. 08:46:50 It did get a couple of pages correct, but the rest of the output was just "^B^D^B^E^B^F^B^K^B^L^B^T^B^T\p^B\uP,P,P,P,iallimmedia\p" and so on. Must be an incompatible version. 08:48:08 * ais523 laments the death of winhelp 08:48:22 * evincar dons a black arm band. 08:48:23 even winhlp32 is pretty much dead nowadays 08:48:35 and you can't get hold of compilers for winhelp nowadays, I don't think 08:48:51 that's actually a good point, I'll have to see if I have one lying around on a hard disk somewhere 08:49:00 -!- puzzlet has quit (Read error: 60 (Operation timed out)). 08:49:01 the source code was a really weird format 08:49:03 it was RTF-based 08:49:04 -!- puzzlet has joined. 08:49:14 and you put metadata in footnotes, or as hidden text 08:49:27 Strange. 08:49:35 a link, for instance, was double-underlined-text, followed by the link target as hidden text 08:49:48 I think it was double-underlined, anyway, I'm not sure, it was so long ago! 08:51:44 * ais523 opens up an old winhelp source file in OpenOffice to look 08:52:11 yep, double-underline 08:53:07 oh, and it had a scripting language too 08:53:25 which had a slightly weird syntax, it was mostly lots of nested camelcase functions 08:53:42 it looks functional looking at the source, but I suspect it may be imperative written with a functional synax 08:53:51 and you put the commands in footnotes 08:54:09 I don't think it's TC, but only because you can't have a loop not containing user input, and there's no infinite storage 08:55:34 There's no way to hack around the former? 08:55:44 Guess it's a moot point. 08:55:46 no, it basically has no control structures but if 08:55:53 Ah. 08:56:48 Hey, you mind if I rattle off my recent idea and get some feedback? 08:56:58 I don't mind at all 08:57:07 All right. 08:57:49 IfThen(Not(IsMark("mpk2pa")), "JI(`m2loadgm.hlp', `lerror')") 08:58:10 Ew. 08:58:18 * ais523 boggles at programming languages that use ` for open quote and ' for close quote 08:58:22 The "whc" program reads WinHelp project files (*.HPJ) and Rich-Text Format 08:58:22 file (*.RTF) and converts them into a WinHelp file with a .HLP extension. 08:58:30 That's included still in the OpenWatcom package. 08:58:35 ah, that sounds about right 08:58:41 I still have a few HPJs lying around somewhere 08:59:05 -- 08:59:07 they basically correspond to Makefiles 08:59:07 Who to Yell at if "whc" Explodes in your Face 08:59:07 --------------------------------------------- 08:59:07 That would be me: 08:59:07 Peter Milley 08:59:10 If I'm not at Watcom, try: 08:59:10 email: pmilley@undergrad.math.uwaterloo.ca 08:59:12 -- 08:59:17 I'm not quite sure he's still interested in whc bugs. 08:59:24 can it use memory outside the bottom 640K? 08:59:27 if so, it's better than the one I have 08:59:33 it keeps crashing on large images 08:59:50 It should build on Linux, so I would hope so. 08:59:53 ooh, wait, I think you can do an infinite loop 08:59:54 -!- M0ny has joined. 09:00:18 I've just reseen the source of one of my most complex attempts to write a WinHelp program 09:00:26 Heh. 09:00:27 it has a lot of JI commands forming a chain 09:00:36 There's no unreseeing it. 09:00:40 in order to do more complicated coding than would fit into the footnotes of one page 09:00:49 I expect you could fit the JIs into a load 09:00:50 *loop 09:01:00 which would be much the same as an HTML page refreshing itself in the middle of loading to loop 09:01:13 Frightening. 09:01:18 which is stupid, but this is #esoteric we're talking about and it's a lot saner than some control structures I've seen 09:01:24 True. 09:01:39 Reminds me of the abuses of the C preprocessor that you see in the IOCCC. 09:01:43 yes 09:01:57 Anyway, my idea. 09:01:57 even more amusingly, JI requires you to specify the literal filename of the target of the jump 09:02:02 D: 09:02:06 That Watcom's WHC is a bit limited, though: supports only .bmp and .shg images, not .mrb/.dib/.wmf; no redefining of window styles, whatever that means; no support of the built-in table commands, and "no disk caching, and as a result is a RAM hog, especially with the 2MB help files people compile here at Watcom!" 09:02:16 I guess that last point isn't such a great thing any longer. 09:02:20 don't worry, I'm perfectly capable of having two conversations at once, with the same people on occasion 09:02:35 Heh...all right. 09:02:48 err, .dib = .bmp with a different extension, so they could support that one pretty easily 09:03:02 and how nostalgic to see .shg there 09:03:18 "Support for Windows bitmaps and Windows SHG files is available, but other image formats supported by Microsoft's compiler (MRB, DIB, and WMF) are not supported." 09:03:29 .shg is basically a .bmp, but with WinHelp commands embedded on rectangular and elliptical areas 09:03:32 I did think .bmp files were device-independent-bitmaps at heart. 09:03:55 So my original concept was an "emotional" programming language. Roughing that out, I decided to design an arbitrary system for creating "emotional automata". Basically I mashed together the idea of the cellular automaton with that of a neural network, added some nondeterminism, and behold. 09:04:12 well, this is sounding pretty esoteric 09:04:29 although "temperamental" might be a better description, by the sound of it 09:04:40 Well, I wanted something for simulation purposes. 09:05:05 "Given these behaviours and arbitrary simulated inputs, how does this agent operate?" 09:05:31 have you seen evfunge? 09:05:35 Nope. 09:05:42 unfortunately, neither have I 09:05:53 it's been discussed here, but all we know of is a webpage describing it 09:05:58 A program consists of a "space", which can be Cartesian, graph-based, or fully-connected; "agents", which are effectively just neurons in the network; and "traits", which provide the capabilities of the agent. 09:06:21 A funge for genetic programming, I'd assume? 09:06:25 yes 09:06:33 Hmm... 09:06:35 apparently it worked rather well, but I haven't seen the source 09:06:45 either of evfunge itself, or of the programs it generated 09:07:01 Hmm... 09:07:26 aargh, I just found wininiss 09:07:36 possibly one of my most ridiculous utility functions ever 09:07:56 Oh? 09:08:00 it was a rather silly FFI, that worked from programs I didn't write to programs I did 09:08:18 it was an executable that took data from its command-line arguments and recorded them in the main Windows configuration file, in a custom setting 09:08:23 from where the program I wrote could read them 09:08:34 at the time, I was rather worried it might wear out the hard disk 09:08:43 which was probably not what I should have been worried about 09:08:49 Heh. 09:09:38 I should look through my "programs written in things not meant to be programmed" directory more often 09:09:50 it's a rather related field to esoprogramming, although not technically the same 09:09:56 oandx.bmp is a classic 09:10:16 Ah, right...so..."traits" express "capabilities", which are responses to traits of other agents; "responses", which are responses to sensory input (i.e., events); and "impulses", which are nondeterministic. So you've got a full range. 09:10:32 Being? 09:10:40 a bitmap that plays noughts and crosses 09:10:54 Should've guessed. 09:10:58 How'd you pull that one off? 09:11:11 I also have the same program as .bat, .dot, .hlp, and .ppt 09:11:19 and by relying on the way floodfill works 09:11:22 and a lot of visual trickery 09:11:34 it's mostly about looking cleverer than it is 09:12:13 Heh. I think I can guess. 09:14:07 Every agent runs in its own thread, and is active and listening to all of its inputs unless otherwise stated. When no agent is active and it can be determined that no agent will become active again (e.g., via an impulse), the program ends. 09:15:39 Writing a program consists of defining a space, declaring agents, and linking up their inputs and outputs into the initial state, if they're in a non-Cartesian space. 09:17:01 I'm feeling a syntax like LISP or Unlambda. 09:18:28 Um...thoughts? Questions? I've rambled an outline, at least. 09:18:33 Unlambda's LISP with half the brackets 09:18:46 Hence the mention. 09:18:48 and it seems a relatively standard general idea 09:19:54 I think targeting simulation applications is going to make it worth it. 09:20:48 evincar 09:20:50 "There. I've made my robot. Now give it some Mars-like terrain and weather, and keep blowing up the little guy until he learns from his mistakes. Then give me an analysis." 09:20:58 Yeah? 09:20:59 it sounds like youve just invented normal agents. 09:22:04 *shrug* 09:23:48 So you're saying it's not necessary? Been done this way before? 09:24:00 it just sounds like youve reinvented agent based modelling. 09:25:29 Well...yes. That's exactly what I've done. I'm just wondering if it's a good idea to make a programming language out of the method. 09:26:56 there already is one. its called agent based modelling. :P 09:27:07 or, after a fashion, object oriented programming. 09:30:00 I was thinking esoteric language, but whatever. And over OO it's more of an architecture or design pattern than straight up part of the paradigm. 09:30:23 please rephrase that last part 09:30:27 im not sure i know what you mean 09:31:24 I can write a program in an object-oriented language that uses the concept of agent-based modelling, but there's nothing about object-oriented programming that is specifically geared toward that technique. 09:31:46 after a fashion there really is. 09:32:25 its just that when we code things in the OO paradigm, we're doing it not for modelling so much as to achieve a particular goal 09:32:43 and so the emergent behavior of modelling doesnt really arise. 09:32:49 OO as in true Smalltalk OO, or OO as in what people normally mean by OO? 09:33:04 true OO smalltalk, lets say. 09:33:12 I may be biased by C++. 09:33:25 C++ is rarely programmed in an OO fashion. 09:33:32 -!- dbc has quit (Client Quit). 09:33:33 I know. 09:33:46 and C++'s OO is like machine code 09:33:56 It tends to be more like a...hierarchical C. 09:34:02 it's so ridiculously do-everything-by-hand that you can hardly get a proper OO system running 09:34:04 Bad term for it. 09:34:07 Yeah. 09:34:45 Well, anyway. 09:37:42 I would say that OO is "more generic", that is, it applies to a wider problem domain than, and is a superset of, agent-based modelling, so I would tend to view it as a conceptual or architectural decision rather than a programming paradigm outright. 09:38:09 ::shrug:: 09:38:13 i dont see any clear distinction. 09:38:41 I mean, I can write a genetic algorithm or agent simulation in C, or assembly, that takes advantage of far different techniques than an object-oriented language while accomplishing the same goal. 09:38:54 an agent-based model seems like merely a particular kind of OO program geared towards discovering emergent phenomena 09:39:15 sure, but you can write ANY program in ANY TC language 09:39:32 Yes, but it's not necessarily OO. 09:39:34 Aagh. 09:39:39 We're chasing ourselves around in circles. 09:39:40 you can write time-ordered code in haskell 09:40:23 i just dont see how what youve suggested is novel. :P 09:40:24 Yes...and this is in favour of what I've been saying. 09:40:34 I don't mean the concept. 09:40:51 I mean the funky esoteric programming language geared toward that concept. 09:41:03 That's where I could be innovating. 09:41:10 ok 09:41:23 Whew. 09:41:25 well what does that language look like and why is it novel in looking like that 09:41:37 That's what I'm working on. 09:41:41 ^_^` 09:41:47 lol 09:41:52 Ha...wow. 09:42:08 We seriously just had a million-line discussion for that. 09:42:49 * evincar gets a glass of milk. 09:42:54 milk?! 09:43:09 I don't have lemonade. 09:43:11 http://blog.milkboys.org/topics/photo/milk-monday/ 09:43:23 Um. 09:43:28 milk! 09:43:33 ... 09:43:35 Convenient? 09:43:40 sexy. 09:44:07 The guy without the shirt isn't bad-looking. 09:44:29 The girl would be better-looking without the...eh...well...milk. 09:44:33 :P 09:44:35 thats no girl 09:44:42 Ah. 09:44:48 thats tokio hotel 09:44:51 A suitably effeminate boi, then. 09:45:08 http://www.deedeecus.com/wp-content/uploads/2008/05/tokio-hotel-mtv.jpg 09:45:12 there are no females in that picture 09:45:31 http://www.cartoondollemporium.com/forum/pics/cdeblog/tokio%20hotel.jpg 09:45:36 I am both frightened and intrigued. 09:45:44 he's gay. 09:45:45 * evincar whistles. 09:45:52 No shit. 09:45:52 im pretty sure he's not actually 09:46:00 6_o 09:46:26 hes just bill kaulitx. 09:46:28 You're telling me a boy who takes care of himself like that is not fond of a more familiar species? 09:46:29 kaulitz* 09:46:42 well given that males and females are of the same species! 09:46:54 *facepalm* 09:46:58 ;P 09:47:05 My subtlety is lost. 09:47:17 Like my marbles, at this hour. 09:47:23 i deny your implicatures and substitute my own! 09:47:45 He likes the male variety! 09:48:04 -!- tombom has joined. 09:48:17 Heh...well, to be perfectly honest, I think it's rather silly to be guessing at someone's sexual preferences based on appearances alone. 09:48:19 he's said himself that hes not gay 09:48:56 given that hes german, and a popstar, and german, looking gay is expected 09:49:06 Hm... 09:49:13 Heh...well, to be perfectly honest, I think it's rather silly to be guessing at someone's sexual preferences. 09:49:17 :p 09:49:26 ...European power pop does that to a person. 09:49:50 Look at ex-Ozone Dan Balan, RadU, and Arsenium. 09:49:58 Less the latter. 09:50:29 And RadU, the young-looking, effeminate, "cute one" is the oldest, married, and maybe with a kid, I forget. 09:50:51 ... 09:50:56 This is a hell of a tangent. 09:52:40 the thing about bill kaulitz is like 09:52:45 hes REALLY femmy, even without makeup 09:53:49 also, he could just be a lesbian in a boys body. 09:53:52 Crap. I just realised that I'm built just like him. 09:53:55 :/ 09:54:39 Blarg. 09:55:22 So, how about that...anything else? 09:55:27 The cake is a lie. 09:55:42 his cake is, anyway 09:55:45 Is there a Portal-based esolang yet? 09:55:55 mm 09:55:58 you mean like with teleports? 09:56:03 I have a feeling something awesome could be done with that. 09:56:06 or with program-created teleports? 09:56:16 Um. Why not both? 09:56:27 ehhh see the problem with that is that if you have teleports, you get no interesting constraints 09:56:32 it just becomes a linear language 09:56:41 in the simplest case 09:56:47 And of course the nondeterminism that arises from the question of whether the cake is or is not a lie. 09:57:00 Hmm...explain? 09:57:13 Oh, I see what you mean. 09:57:15 well, it depends on how you use teleports, i guess 09:57:15 INTERCAL control flow feels a bit portally 09:57:20 although it isn't exactly the same 09:57:24 True. 09:57:28 if teleports are like 09:57:40 site-to-site beaming 09:57:41 And program flow is only one object that can go through a portal, remember. 09:57:43 on certain conditions 09:57:49 its just a fancy version of befunge 09:57:55 Oh, I wasn't thinking that. 09:57:58 that bypasses the constraints of befunge 09:58:06 in that befunge forces you to work in 2 dimensions 09:58:10 Right. 09:58:20 which means paths will cross 09:58:26 so you have to code around that 09:58:33 once you go into three dimensions, theres no such restriction 09:58:44 and teleporting essentially does the same thing 09:58:49 I don't know that a Portal language could be a fungeoid, but it seems like a logical conclusion. 09:59:07 well, portals only matter when you have something that moves about a space 09:59:13 True. 09:59:28 Hey >:| 09:59:31 hey 09:59:32 Portal language was my idea! 09:59:40 YOU KLEPTO THIEF 09:59:42 Haha. 09:59:54 Hey, I haven't been here in *ages*. 10:00:02 Independent innovation. 10:00:11 slereah, what do you do at slereah.place_of_employment? 10:00:18 For the record, I also had the idea of a Lemmings language, so HANDS OFF OF THAT 10:00:30 psygnisfive : I program stuff and shit 10:00:36 Do equations and such 10:00:43 ah 10:01:02 Science you know 10:01:07 But of course. 10:01:13 The Science. 10:01:17 Right know I'm trying to model a parton jet 10:01:23 a parton jet? 10:02:39 http://en.wikipedia.org/wiki/Jet_(particle_physics) 10:03:01 parton? 10:03:12 gluon + quark 10:03:22 a free quark??? 10:03:40 Or gluon 10:03:52 Once free, it decays rapidly 10:03:58 Into a shitload of particles 10:04:05 All going in the same general direction 10:04:08 That's the jet 10:04:09 oh i see 10:04:57 Worst part about work is that they're rationning Mathematica 10:05:04 You only get some time per day 10:05:19 sounds like you need a bootleg copy 10:05:59 I have one, but at home 10:06:05 ok 10:13:23 Well, I'm no longer productive. 10:13:30 Time to turn in for a few hours. 10:13:56 It's been interesting. 10:15:30 -!- evincar has quit ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]"). 10:46:13 ehird (when you're here): more Perl6 stealing ideas from INTERCAL: it uses just-in-case compilation for procedures that aren't declared yet 10:46:33 as in, compiles as if they're declared later, then errors out if it doesn't find the declaration 10:57:59 -!- MigoMipo has joined. 11:21:24 -!- WangZeDong has joined. 11:23:22 ok, Perl6 has reverse-subtract as an operator 11:23:26 along with reverse everything else 11:24:18 > say(2 R- 5); 11:24:20 3 11:24:39 > say(1 R+ 2); 11:24:40 3 11:24:46 and there you have the ever-useful reverse-addition operator 11:25:47 Substraction? 11:28:25 its sort of subtraction 11:28:39 its subtraction with the arguments reversed 11:28:45 ais523: why not a meta-operator for exchanging operands? 11:28:54 that's exactly how it's implemetned 11:29:07 and why things like reverse-addition exist 11:29:14 i see. 11:29:21 whats the POINT of Rop? 11:30:59 > say [<](); 11:31:00 1 11:31:05 there we have it, a less-than with no arguments 11:31:18 to be precise, a reducing less-than 11:31:24 a what 11:31:28 > say [>](); 11:31:31 aww 11:31:35 you tricked me! 11:31:41 youre not running a bot on your computer :( 11:31:51 ais, simulate a bot! 11:32:01 1 11:32:04 GASP 11:32:08 apparently, both less than and greater than are true 11:32:09 ok im off to bed 11:32:13 if applied to nothing at all 11:32:23 just like [+]() is 0 and [*]() is 1 11:32:28 -!- psygnisfive has changed nick to augur[sleep]. 11:32:47 hm 11:32:48 thats odd 11:33:13 because 0 and 1 are the identity values for those operators 11:33:18 yes 11:33:24 while > and < have no identity operator 11:33:24 -!- tombom has quit (Read error: 54 (Connection reset by peer)). 11:33:26 that's something like foldl or foldr in many languages, but took a form of n-ary operator 11:33:28 it's meant to be the identity value you get if you apply them to no operands 11:33:38 and I'd say "true" is the identity value of less-than 11:33:43 because 4 < 5 < 6 < 7 is true 11:33:43 -!- tombom has joined. 11:33:51 so [<]() is true 11:33:55 -!- Slereah has quit (Read error: 110 (Connection timed out)). 11:34:02 lifthrasiir: yep, it's a folded-operator syntax 11:34:15 but the chained < is not an accurate representation of < mathematically 11:34:25 as in, chaining a less-than is an implicit "and" in pretty much all usage 11:34:30 and the identity for and is truth 11:34:54 true 11:35:05 but < cannot by its very nature have an identity 11:35:10 not in the traditional sense of an identity 11:35:36 since its not an operation from the domain back to the domain 11:37:31 yes 11:37:35 the identity's on the return value, there 11:37:50 as in, a < b < c is parsed by Perl6 as (a < b) && (b < c) 11:37:56 so it uses the identity for && 11:37:56 true 11:38:13 but i still cant reeeaaally comprehend what an identity for < would be 11:38:29 because identities are not conceptualized as relating to chaining at all. 11:39:18 ais523, did you highlight me? 11:39:29 I don't think so 11:39:30 * AnMaster notes someone highlighted him but it is out of scrollback 11:39:36 not today, anyway 11:39:44 well if it was anything important I'm sure that person will re-highlight 11:39:45 I highlighted you the day before yesterday, IIRC 11:39:49 but I think you were online then 11:39:50 well,* 11:39:58 and I was talking about you, not to you, then 11:39:59 ais523, yes, this happened since yesterday evening 11:40:09 in the context of people who ehird tended to disagree with 11:40:16 ok but really, night 11:40:25 night 11:40:30 augur[sleep], cya 11:41:10 odd... 11:41:32 > say((1,3,5)X+(2,4,6)); 11:41:34 3575797911 11:42:11 strangely, that took it a while to calculate 11:42:12 when X runs for a long time (several weeks) it's memory usage sometimes start to rapidly grow. /proc/pid/maps indicate it is the heap that grows. However this doesn't always happen. 11:42:19 I wonder how much is coing on behind the scenes there 11:42:24 This time it happened after just 5 days 11:42:25 AnMaster: memory leak only in certain circumstances, I wonder? 11:42:36 ais523, I don't know. 11:42:37 it could be that they're very rare circumstances 11:42:42 but once they happen, they continue happening 11:42:50 that would give apparently random lengths of time before it happened 11:42:52 ais523, it happens about 2/3 of the times X runs for a few weeks without restart. 11:43:00 also, I tend to restart X every few hours 11:43:05 usually it doesn't happen after just a few days 11:43:11 due to not staying on one internet connection for all thatl ong 11:44:02 ais523, why would you need to restart X when changing internet connection?? 11:44:12 because I turn the computer off 11:50:25 -!- BeholdMyGlory has joined. 11:58:17 Perl6 has one of the silliest lambda syntaxes ever: 11:58:28 -> $var { #(code here...) } 11:58:37 the #() is an inline comment, btw 12:04:31 ais523: so does #«foo»... 12:04:38 yes, I know 12:04:46 in fact, you can use any of the bracket chars in Unicode 12:05:01 and if i remembered correctly, there are anonymous arguments $^a, $^b etc. 12:05:26 actually, $^anything works to implicitly give arguments to the closure you're in 12:05:38 and all blocks are treated as 0-arg lambdas 12:05:44 unless they have more than 0 arguments 12:06:41 -!- MigoMipo_ has joined. 12:16:28 -!- FireFly has joined. 12:16:33 the #() is an inline comment, btw 12:16:34 o 12:16:35 m 12:16:36 g 12:16:47 why are inline comments in Perl so surprising? 12:16:49 -!- tombom has quit ("Peace and Protection 4.22.2"). 12:16:52 ais523, the syntax for them 12:16:59 what if you write at the start of a line: 12:17:02 #( blah blah 12:17:05 syntax error 12:17:08 yes, seriously 12:17:14 surprising 12:17:23 # ( blah blah would be a one-line comment 12:17:31 and space #( blah blah 12:17:33 ais523, only if it ends with a matching space right? 12:17:35 ;P 12:17:35 would be the start of an inline comment 12:17:40 which can last multiple lines 12:17:46 and space isn't a sort of paren 12:17:49 true 12:18:12 but it would match sblahblah 12:18:13 :P 12:18:36 ais523, see what I mean? 12:19:01 err, but s doesn't allow any char nowadays 12:19:05 ais523, = 12:19:05 in fact, it never used to 12:19:06 ? 12:19:10 it didn't allow letters, for instance 12:19:26 in perl6 it doesn't allow : or # either 12:19:26 hm ok 12:19:31 pretty sure sed does however 12:19:38 because it's legal to mix quote operators 12:19:50 ah hm no 12:19:54 any non-letter 12:20:08 wait no 12:20:09 it does 12:20:19 $ sed 'sxfooxbarx' 12:20:19 foo 12:20:19 bar 12:20:19 ^C 12:20:32 :) 12:21:01 this may be clearer: 12:21:02 $ sed 'sxfooxbarx' <<< foo 12:21:03 bar 12:22:27 what about NUL? 12:22:44 $ sed $'s\0foo\0bar\0' <<< foo 12:22:45 sed: -e expression #1, char 1: unterminated `s' command 12:22:45 hah 12:22:49 hm 12:23:13 you can use $'' to interpolate backslash-0? 12:23:29 actually I'm not sure bash will handle it correctly 12:23:30 * AnMaster checks 12:24:07 $ sed -f /tmp/blah <<< foo 12:24:07 sed: file /tmp/blah line 1: unterminated `s' command 12:24:20 -!- MigoMipo has quit (Read error: 110 (Connection timed out)). 12:24:21 and there I'm sure the null bytes are there 12:24:35 ais523, and yes $'' interprets like echo -e 12:25:00 and, obviously, I seriously doubt newline works as a separator 12:25:03 I'm less sure about tab 12:25:09 obviously 12:25:23 $ sed $'s\tfoo\tbar\t' <<< foo 12:25:23 bar 12:25:25 tab works 12:25:33 so does space 12:28:01 I wonder how backslash reacts to escape codes in the s\\\ itself 12:28:18 try it yourself, I'm a bit preocupied 12:32:16 ok 12:32:21 I might try late 12:32:22 *later 12:44:41 ais523, seems \ doesn't work then 12:44:42 just checked 12:46:27 ah, ok 12:55:24 -!- asiekierka has joined. 12:55:27 Hi 12:55:33 I implemented Deadfish in C64 basic O_O 12:55:40 and I think I didn't see this channel more crowded 12:56:12 oh, 1/10th of you all is bots 12:56:20 egobot, fungot, clog, thutubot 12:56:20 asiekierka: which was the first thing that popped into my brain 12:56:32 fungot: uh... clog? 12:56:33 asiekierka: what will scheme42 be exactly? some kind of priorities between your rules. ha ha ha you changed some words. 12:56:42 ...:( 12:57:09 hi asiekierka 12:57:20 also, it's often more crowded than this, you must be in a weird timezone 12:57:41 GMT+1 is not weird 12:57:53 but i'm not in the US, which has GMT-6 or whatever :P 12:58:14 ais523: Did you check out DEADFISH 64? 12:58:18 I saw it 12:58:23 it looks like portable BASIC to me 12:58:28 C64 Basic 12:58:33 presumably it would work on other basic interps too, though 12:58:46 probably 12:58:56 that's like saying that a particular standard strict-C89 program is x86 C 12:59:10 Well, I wrote it on a C64 12:59:15 and only tested on a C64 12:59:19 And it has a subset of Deadfish~, too 12:59:23 ah, ok 12:59:59 if you click the external link to the webpage of Deadfish 64 (on the bottom of the Deadfish page) you can see what features does it have 13:00:07 The one in the implementations only implements basic deadfish 13:00:08 :D 13:00:16 and h 13:04:40 what does h do? 13:05:07 * ais523 thinks it's an interesting comment on programmer psychology that deadfish has such appeal 13:08:15 Halt 13:08:20 see deadfish~ 13:08:20 :P 13:08:24 and the dead64 webpage 13:08:29 -!- asiekierka has quit. 13:18:36 -!- MigoMipo_ has quit (Read error: 110 (Connection timed out)). 13:35:07 -!- ais523_ has joined. 13:53:56 -!- puzzlet has quit (Remote closed the connection). 13:54:12 -!- puzzlet has joined. 13:54:30 -!- ais523 has quit (Read error: 110 (Connection timed out)). 14:08:09 -!- oklopol has joined. 14:08:33 so i saw evilsort and decided to have some fun http://www.vjn.fi/pb/p114666125.txt 14:09:01 (not evilsort, my own verysort thingie, in case you don't read python or don't, for some reason, know evilsort) 14:09:57 it's competitive sorting, i just started writing code and that happened 14:10:08 seems to be pretty fast 14:10:27 i mean for a pessimal algo, which of course is a bad thing. 14:10:44 i mean it can easily sort lists of seven elements 14:10:48 -!- Gracenotes has quit (Remote closed the connection). 14:11:28 OH MY GOD I'M A GENIUS, wait a mo i know just what to do to fix that... 14:17:57 -!- okloduk has joined. 14:23:55 what is evilsort? 14:24:27 When you sort a list by asking Satan 14:25:22 wow, that's quite a sorting algorithm 14:25:40 is it O(n!), or worse? 14:25:55 It is O(0), but it will cost you YOUR SOUL 14:26:01 Or the program's soul. 14:26:03 I'm not sure 14:26:04 I was refering to verysort 14:26:06 -!- WangZeDong has changed nick to Slereah. 14:29:18 -!- oklofok has joined. 14:30:35 so umm 14:30:40 i fixed it 14:30:51 but i'm not sure it works, because i can only test up to lists of length 3. 14:31:21 http://www.vjn.fi/pb/p214351221.txt <<< sorry, even more okloisms here 14:31:58 but basically what it does is pick a list at random 14:32:03 and order the rest randomly 14:32:13 and then take an element from each list 14:32:37 and if the chosen list has that element, and the one to the right of it in the correct order, and all the others don't, then it gets a point. 14:33:01 best has the list with the best score, it's returned if it's sorted 14:33:22 what i wanted to add to evilsort was nondeterminism and not using polynomial space 14:34:42 i guess i also added much more pessimality, but i guess this one looks somewhat pessimal already, that is, kinda hard to justify making sure it beats everyone else before giving it a point, which is a bad thing. 14:35:18 but i guess you could call it a tournament to the death, but with the dead guys getting another chance as opponents. 14:37:55 btw it's actually somewhat nontrivial that there even always is a terminating shuffling, if the elements aren't all equal 14:37:57 oklofok, sorting by genetic algorithm? 14:38:01 or what do you mean 14:38:05 AnMaster: you could call it that 14:38:31 -!- oklopol has quit (Read error: 110 (Connection timed out)). 14:38:39 hm 14:38:51 but genetic algos don't usually have explicit "fights" 14:38:55 true 14:39:01 -!- okloduk has quit (Read error: 110 (Connection timed out)). 14:39:20 well, incrediblysort has more like a deathmatch than a fight, as i guess i explained already 14:39:30 bbl 14:39:36 byes 14:40:35 btw not only doesn't a list of four elements get sorted in a reasonable time, it's not very likely that anyone will even get any points in a quadrillion years :D 14:41:00 the probability of getting a point is 1/2^(n!) 14:41:50 assuming elements not equal, if they can be equal, it's probably much easier 14:42:48 yep, it runs in an instant :<<<<<< 14:43:11 i should add like an epsilon, "to make it stable" :P 14:43:11 well, you probably have no idea what i'm talking about 14:43:24 pizza and family guy, as halfway mentioned -> 14:44:12 that is, epsilon is a bad term, more like.......... umm..... random skew 14:44:44 except not random or it wouldn't be stable 14:44:44 it's just i can't find another justification for it 14:57:50 -!- ais523_ has quit (Read error: 104 (Connection reset by peer)). 15:08:06 -!- nooga has joined. 15:08:20 -!- puzzlet has quit (Remote closed the connection). 15:08:24 -!- puzzlet has joined. 15:08:43 anyone wrote something using GTK2? 15:13:44 -!- oklodok has joined. 15:29:39 -!- oklodok has quit (Read error: 60 (Operation timed out)). 15:30:23 ... 15:36:23 -!- oklofok has quit (Read error: 110 (Connection timed out)). 15:47:47 back 16:07:35 nooga: In the distant past. 16:12:08 10:31 ais523: there we have it, a less-than with no arguments 16:12:08 10:31 ais523: to be precise, a reducing less-than 16:12:24 #;1> (<) 16:12:25 #t 16:12:26 #;2> (>) 16:12:28 #t 16:12:30 r5rs 16:12:32 05:04 pikhq: Oh, and also subversion of the free market. 16:12:34 05:04 pikhq: (as we are wont to do here) 16:12:36 the free market sucks 16:12:48 And now, back to assembling this hypothetical server for the VPSness. 16:13:59 (in case anyone wasn't here yesterday: I've semi-come up with a semi-novel way to do VPSes fast and easily and now I'm toying with the idea of getting a dedi to sell them.) 16:15:09 GregorR: what does moxie taste of 16:15:31 ehird: It's not comparable to any other soda, really. 16:15:42 GregorR: compare it to something that isn't a soda then 16:16:01 "For those without access to Moxie, the flavor can be approximated (and adjusted to taste) by adding Angostura bitters to root beer, or by mixing Campari with Coca-Cola, or by mixing a shot of Jägermeister in a glass of Coca-cola (6-8 oz.)." 16:16:06 Uh. But of course! 16:16:28 Yeah, Idonno, can't help ya :P 16:16:40 GregorR: Just give me some :-P 16:17:24 ehird: If you were in W. Lafayette, and I had some right now, I might :P 16:20:47 GregorR: This online store is offering me six cans for $7.99 plus a probably inordinately huge amount of shipping to the UK, and I'm trying to decide whether my curiosity extends that far. 16:21:16 That would probably be some ultra-expensive shipping. 16:21:32 It's really good, I'd recommend that everyone try it, but it's not necessarily worth the price. 16:21:35 "To obtain an accurate shipping amount, enter shipping postal code (US only) " 16:21:37 GregorR: $infinity 16:21:46 lol 16:21:52 Unless that means they just can't estimate it for non-US 16:24:29 -!- ais523 has joined. 16:25:04 ais523: Whar's the bf joust bundle for EgoBot? :P 16:25:05 05:04 pikhq: Oh, and also subversion of the free market. <-- is the repo public? 16:25:22 That was so unfunny it took me seconds to get it. 16:25:40 svn co https://svn.free-market.co.su/trunk 16:25:51 GregorR, .su? Sudan? 16:25:55 AnMaster: soviet union 16:25:55 GregorR: well, put it this way 16:25:57 aha 16:26:00 But they don't exactly have a free market! 16:26:02 I last woke up at 5pm yesterday 16:26:05 Er, didn't. 16:26:12 and went to University at 6am today because I couldn't sleep 16:26:21 ais523: try Uberman's. It can't be as fucked up as your current schedule 16:26:22 now I'm back closer to home, so I've been up for almost 24 hours 16:26:27 and didn't sleep properly before then 16:26:29 ( http://www.kuro5hin.org/story/2002/4/15/103358/720 ) 16:26:33 ais523: And you used all that extra time to write a bf joust interp for EgoBot, right? ;) 16:26:49 no, I used most of it wondering why time had apparently broken down into discrete steps 16:26:55 I suspect I'm so tired I'm hallucinating 16:27:01 uh uh 16:27:07 either that, or I'm just seeing the fundamental cellular automaton behind the universe 16:27:18 ais523: just make sure things don't turn into 2D 16:27:19 ais523, in what way did you notice it 16:27:23 or you might start moving diagonally 16:27:30 AnMaster: as in, it was like the universe's frame-rate was rather low for some reason 16:27:31 AnMaster: you can't describe that kind of stuff... 16:27:45 ais523, heh 16:27:47 you know, too low to give a decent illusion of reality 16:27:53 right 16:28:00 ais523, sure you weren't dreaming? 16:28:08 -!- oerjan has joined. 16:28:08 and no, I've spent most of the time reading the Perl6 docs, I think they make more sense when sleep-deprived 16:28:20 AnMaster: I was technically awake, but probably not awake enough to do anything in practice 16:28:21 ais523, aha, That is the cause clearly. 16:28:42 GregorR: I have a BF Joust interp that pits two programs against each other, and returns the result to stdout and to exit status 16:28:48 what else would it need? 16:29:00 oerjan, IWC was slightly amusing today. 16:29:04 I was rather hoping the existing FYB scripts would fit to BF Joust well 16:29:07 ais523: could you think inbetween the discrete steps? if not, that's some feat to notice it :P 16:29:12 ais523: Just an adaptation of FYB's report.c 16:29:17 ais523: And scmds/fyb 16:29:35 link to EgoBot source? 16:29:40 !info 16:29:41 EgoBot is a bot for running programs in esoteric programming languages. If you'd like to add support for your language to EgoBot, check out the source via mercurial at https://codu.org/projects/egobot/hg/ 16:29:43 there 16:30:15 * ais523 installs Mercurial 16:30:36 I may as well have all three of the leading DVCSes, plus the one I actually use 16:30:43 -!- coppro has joined. 16:31:05 (The scmds/fyb script is sort of stupid because it was written before fyb exited with a useful status code, so I had to parse the human-readable output :P ) 16:31:10 AnMaster: slight *groan* 16:31:23 oerjan, agreed 16:31:46 GregorR, fyb? 16:31:56 http://codu.org/eso/fyb/README 16:31:58 fuck yor brane. 16:32:01 aha that language 16:32:02 hm 16:32:02 right 16:32:03 fuk 16:32:53 argh, Ubuntu, stop telling me at is broken, it's been broken for two versions now and I've reported the bug and there are lots of helpful comments indicating what's causing it and it should be a one-line fix and you still haven't fixed it 16:33:01 oerjan, the last sqrt(-garfield) (from yesterday iirc?) was rather *groan* too IMO 16:33:43 well... 16:33:58 ais523: why are you using ubuntu again? :P 16:34:17 it's the OS that the computer came with 16:34:55 "This comic has been produced by applying the identity function simultaneously to every pixel of the comic of 2002-09-05. " 16:34:56 :-D 16:35:02 ais523: is installing an OS hard? 16:35:17 very meta 16:35:23 yes if you don't have enough backup media you trust to hand to back up all your files first 16:35:34 besides, in theory this thing might still be in warranty 16:35:44 ais523: oh, you don't have a /home partition? 16:35:49 ;-) 16:35:54 also, installing an OS breaks the warranty? 16:35:56 what kind of shitty warranty is that 16:36:14 it's standard excuse for more or less every major PC manufacturer ever 16:36:14 and mezzacotta today advocates killing children, i see 16:36:38 actually, I have no idea what would happen if you tried wiping Mac OS X on a Mac, installing Windows/Linux as the only OS, and then asking them to fix a hardware problem 16:36:39 ais523: so you buy a computer whose warranty involves not using it 16:36:42 it's just like DRM! 16:36:53 well, I don't plan to use the warranty 16:37:02 half the hardware's malfunctioning prettily as it is 16:37:04 also, I think as long as you didn't imply it was anything to do with windows they'd help 16:37:06 the touchpad hasn't worked for years 16:37:07 Boot Camp is supported 16:37:10 although I'm not sure removing OS X is 16:37:20 well 16:37:23 you can boot into an OS X install CD 16:37:27 and remove your OS X partition 16:37:33 so for a tenuous definition of supported it is 16:37:45 boot camp doesn't actually do anything but partition, heh 16:38:10 really, "boot camp" is just the EFI's bios emulation 16:40:42 ais523, ehird: They'd "fix" it, but when it came back it'd only have OS X installed :P 16:41:01 GregorR: I find that unlikely. 16:41:22 At least, if you had Windows AND OS X they'd stay since that's officially supported, and I don't recall seeing removing OS X be classed as unsupported 16:42:14 ehird, Is putting the computer in lava explicitly classed as unsupported? 16:42:30 that's totally irrelevant 16:42:34 ehird, just wondering 16:43:01 I imagine that destroying the computer is classed as unsupported, though. 16:43:12 right, that covers a lot of different cases. 16:43:38 AnMaster: Lawyers wrote it. They have thought of everything ;-) 16:43:47 yeah :/ 16:44:39 You know... 16:44:41 http://xkcd.com/373/ 16:44:57 kerlo: how's that relevant to anything 16:45:01 That comic is just as true if you replace "Claims of Supernatural Powers" with "Theories of Physics". 16:45:30 kerlo: it's called casual conversation, s/confirmed/seems to work in practice for blah blah balh/ 16:45:45 -!- Judofyr has joined. 16:47:25 * AnMaster wonders how hard it would be to make a web browser using XeTeX as rendering backend. 16:47:45 Hard. Also impractical. 16:47:56 Maybe WebKit or Gecko let you pug in a backend. That'd be hard though so I doubt it. 16:47:58 you mean: only static pages? 16:48:02 We're all about practicality here in #esoteric 16:48:04 .......... no? 16:48:14 It's just that TeX isn't anything like HTML+CSS. 16:48:21 +JS 16:48:31 +VBScript :P 16:48:36 pages using JS would be hard 16:48:36 Oh gawd. 16:48:44 AnMaster: you'd just re-render it every time 16:48:49 yeah. 16:48:51 that's basically what browsers do, 'cept incrementally. 16:48:59 what about forms? 16:49:01 GregorR: I wrote some VBScript-using pages when I was young and idiotic. 16:49:04 Despite not knowing VBScript or JS. 16:49:12 AnMaster: Tough shit. It's just not suited. 16:49:12 ehird: Be ashamed. 16:49:22 ehird, pdf forms duh! 16:49:23 I tried writing in VBScript, but then gave up and went back to batch files 16:49:30 AnMaster: That's not TeX. 16:49:48 the only VBScript program I ever got working could be translated into bash as echo $(($1)) 16:49:49 ehird, can't you do it using some package for LaTeX? 16:49:50 hm 16:49:55 actually, the bash script is probably more featureful 16:50:40 I wouldn't be surprised if something similar to hyperref (for pdf TOC and links in pdf) existed for pdf forms too 16:50:54 just use JI! 16:50:58 JI? 16:51:04 WinHelp command 16:51:13 ais523, what has that got to do with TeX? 16:51:16 it's useful by WinHelp standards, but not by regular programming standards 16:51:21 and 16:51:22 and because you can use it to implement links 16:51:25 what does it do in WinHelp 16:51:43 oh, links? Is that what it does. OK 16:52:18 it's basically a goto, but you need to specify the filename of the file to jump to 16:52:24 even if it happens to be the one you're in at the time 16:56:20 AnMaster: PDF is extremely powerful, actually. 16:56:29 It's basically a superset of Postscript... 16:56:35 Which is Turing-complete. 16:56:35 ;) 16:57:05 pikhq, iirc pdf is a subset, not a superset? 16:57:09 Superset. 16:57:15 mhm 16:57:17 Preview.app translates .ps files to PDF before rendering. 16:57:24 Although maybe PDF doesn't have the code bits. 16:57:27 And it just renders it. 16:57:28 Dunno. 16:57:49 PDF contains Postscript compiled to bytecode, some headers, and sometimes some other file sections. 16:58:02 (in essence) 16:58:05 Wait, isn't it proprietary? 16:58:09 Surely they couldn't use PS. 16:58:16 from what I remember: pdf was originally a subset, but later on they added more features to pdf, so they are actually two intersecting formats, each having a few bits the other doesn't. 16:58:19 No... 16:58:26 Postscript is a standard designed by Adobe. 16:58:30 but yes I might have been wrong 16:58:33 yes,* 16:58:34 PDF is also a standard designed by Adobe. 16:59:10 translating postscript to pdf is like running ick -F 16:59:15 ... Adobe made Postscript? 16:59:19 I actually didn't know. 16:59:19 Yes. 16:59:28 ehird, odd you didn't know 16:59:30 ... wait, yes I knew that. 16:59:38 But on the other hand I didn't know it as a sort of relation, just as a standalone fact. 16:59:44 Thus confusion. 16:59:46 Hey. 16:59:48 You're wrong 16:59:54 hm wait 16:59:58 nope 17:00:00 you're right :-P 17:00:03 AnMaster, you're right. PDF and PS have a common subset. 17:00:06 XD 17:00:16 pikhq, but both have features the other doesn't. Right 17:00:22 is pdf tc too? 17:00:23 Heh, an Apple printer was the first PS printer 17:00:28 AnMaster: I don't think so 17:00:32 aha 17:00:34 Still Turing-complete. 17:00:40 AnMaster: That's what a common subset means. 17:00:51 I thought PDF was rather light on control structures 17:00:57 The use of PostScript did not come cheap. At an introductory price of US$6,995, the LaserWriter was more expensive than PC laser printers of comparable print speed and quality. The LaserWriter's high cost was largely due to the extra processing power needed to run the PostScript interpreter. As it was a complete programming language, PostScript came saddled with the overhead of a complex software rasterizer program (running inside the printer). Powering 17:00:59 ais523, "me too" 17:00:59 the LaserWriter was a Motorola 68000 CPU running at 12 MHz, 512KB of workspace RAM, and a 1 MB framebuffer. At introduction, the LaserWriter had the most processing power in Apple's product line — more than an 8 MHz Macintosh. 17:01:04 Just use a LaserWriter to do your computing! 17:01:07 Erm, wait... 17:01:07 It's extra-powerful ;-) 17:01:14 the conditionals are removed from PDF. 17:01:21 *Why* are the conditionals removed from PDF?!? 17:01:21 Damn the old macs were slow. 17:02:20 ehird, that's still faster than the C64 afaik 17:02:24 Duh. 17:02:29 The C64 is ancient. 17:02:33 AnMaster: but the c64 didn't have a nice GUI. 17:02:34 true 17:02:40 The C64 wasn't used for desktop publishing. 17:02:41 just trying to add a perspective to this 17:02:55 8mhz is slow however you slice it :P 17:03:08 by the standards back then Macs were not that slow. 17:03:20 I know this. 17:03:29 AnMaster: also, the NTSC c64 was 1.02MHz 17:03:33 admittedly a different architechture 17:03:37 but that's not massively slower 17:03:43 ehird, and 8 mhz is faster than many embedded microcontrollers. 17:03:46 even today 17:03:51 Yes, but they're not desktop computers. 17:03:58 They're embedded microcontrollers. 17:03:59 that's true. 17:04:20 I couldn't write a non-trivial program that ran under the constraints of the original Macintosh, probably 17:04:34 ehird, in 20 years or so people will look at intel nehlam (spelling?) the same way. 17:04:40 *nehalem 17:04:43 And true enough. 17:04:47 I could. 17:04:48 AnMaster: closer than 20 years 17:04:50 moore's law 17:04:55 meh 17:05:29 * pikhq is freaking crazy, and knows of some scary, scary ways to do multitasking on systems with low memory. 17:05:46 I've done multitasking on a microcontroller without an OS before 17:05:57 ... Not as crazy as the guy who made a multitasking GUI on the C64, mind. 17:06:01 ais523, what microcontroller? 17:06:01 by hooking all the tasks but one into signal handlers 17:06:07 AnMaster: dsPIC 17:06:39 ... Or the UNIX for C64, for that matter. 17:07:20 ais523, what does the ds mean 17:07:32 pikhq, there is unix on c64!? 17:07:36 AnMaster: as in DSP 17:07:48 aha 17:07:59 ais523, that is a rather high end PIC right 17:08:01 also, someone's likely to have ported ucLinux or something to the C64 by now 17:08:05 probably just to claim they could 17:08:12 AnMaster: higher enough it's technically something different 17:08:16 it's 16-bit, for one thing 17:08:43 luxury! 17:08:51 http://hld.c64.org/poldi/lunix/lunix.html 17:08:52 Lunix! 17:08:52 AnMaster: Lunix. 17:08:54 For the c64. 17:08:56 pikhq: snap 17:08:58 Written in assembly. 17:09:02 ehird: :) 17:09:03 This just makes "Lunix" to mean "Linux" even funnier. 17:09:19 yes 17:09:49 Contiki, though, is just freaking crazy. 17:10:12 anyone got a link to Contiki. I remember reading about it some years ago 17:10:27 http://upload.wikimedia.org/wikipedia/en/5/5c/Contiki-C64.png 17:10:33 Wikipedia. 17:10:39 official website? 17:10:47 http://www.sics.se/contiki/ 17:10:57 .se! 17:11:46 why do people keep linking to the images themselves on Wikipedia, not the image description page/ 17:11:47 I assume it needs some form of changed hardware to support network? 17:11:55 ais523: because we don't like the UIcruft. 17:11:58 it's like people are going out of their way to violate the attribution requirements of the image 17:12:00 ais523, because that loads more slowly? 17:12:11 ais523, and often it is scaled down 17:12:14 almost all of the time the cruft around the image is worthless 17:12:17 and yeah, scaled down 17:12:18 so you HAVE to click the image to see what it means 17:12:27 and feel free to sue me for not attributing 17:12:28 and see the details 17:12:42 ehird: I can't, it has to be the image's copyright holder who does that 17:12:50 ais523: I'll wait here. 17:13:49 ais523, there should be a stripped down format then. Just a minimal html page with no css, first a link to the image description page, then the image below in full size 17:13:57 and everyone would be happy 17:14:14 I wouldn't be, it's still cruft :-P 17:14:21 For example, you can't do 17:14:23 ehird, much *less* cruft though 17:14:23 imageviewer url 17:14:29 When it's an image we're linking to 17:14:30 agreed 17:14:43 ah, it's no good, I'm going home 17:14:46 it should have same file name with .html instead of .png/.jpg/.svg/.whatever 17:14:49 I'm too tired to think straight at all 17:14:50 ais523, what is no good? 17:14:53 ais523, ok cya 17:14:56 -!- ais523 has quit (Remote closed the connection). 17:15:04 ehird, then you could just have http://upload.wikimedia.org/wikipedia/en/5/5c/Contiki-C64.html 17:15:15 and it would be easy to change 17:15:17 wait 17:15:19 AnMaster: yes, but think of the plan 9 users 17:15:21 they can't do 17:15:22 make that .png.html 17:15:26 ehird, why not 17:15:26 hget url | plumb 17:15:31 (I think that's the right invocation) 17:15:32 hm 17:15:44 AnMaster: more convincing: 17:15:51 they can't right/middle click the link in IRC (I forget whic) 17:15:52 h 17:15:52 ehird, surely there is some text replace to strip .html off http://upload.wikimedia.org/wikipedia/en/5/5c/Contiki-C64.png.html 17:15:55 and get their image viewer up 17:16:00 ah 17:16:04 fair enough 17:16:17 but it would open in their browser instead 17:16:22 would that be so bad 17:16:32 Surely they have sed. 17:16:32 AnMaster: if they have one. plan 9 browsers suck. And yes, one tool for one job. 17:16:40 pikhq: 17:15 ehird: they can't right/middle click the link in IRC (I forget whic) 17:16:44 pikhq: lern2read 17:16:53 They can type! 17:17:01 ehird, ok then, add a watermark to the picture. That is what is left then 17:17:05 and that is horrible 17:17:11 pikhq: so we waste people's time for no gain 17:17:12 awesome 17:17:15 AnMaster: or just continue doing what we do 17:17:22 ehird, you could script the client to automatically do that for you for wikipedia links 17:17:23 * pikhq vejnas 17:17:25 ehird, true we could 17:17:37 * ehird jkdsfha 17:19:24 s|http://upload\.wikimedia\.org/(.+)\.([^.]+)\.html|http://upload\.wikimedia\.org/\1.\2| 17:19:27 or something 17:21:13 hm contiki is ported to other platforms than C64 is it? 17:21:55 Who knows. 17:22:11 http://www.sics.se/contiki/ doesn't mention C64 anywhere I can see 17:22:11 It's a very technobabbly page. 17:22:33 The Contiki Operating System - Home 17:22:33 Highly portable, multitasking OS for low memory networked embedded systems; typical install is 2K RAM, 40K ROM; event-driven kernel, programs load and ... 17:22:35 from googl 17:22:37 e 17:22:47 AnMaster: http://www.sics.se/contiki/perspective/run-the-commodore-64-version-of-contiki-anno-2004-in-your-browser.html 17:22:49 Today Contiki is mostly known as an operating system for networked embedded systems. A few years ago, however, Contiki's primary claim to fame was its Commodore 64 port. With the help of JAC64, a Java-based C64 emulator developed by my colleague and fellow Contiki developer Joakim Eriksson, you can now experience the C64 port of Contiki 1.2-devel1 again, directly in your web browser! Click here to enjoy it - unfortunately without networking support at pr 17:22:50 ehird, I checked about, and so on and it only talks about embedded chipsets and such 17:22:52 esent. 17:23:00 well hidden 17:23:02 Contiki is on a *lot* of platforms. 17:24:06 ;] 17:46:09 a 17:56:31 -!- oerjan has quit ("leaving"). 17:58:05 -!- puzzlet_ has joined. 17:58:22 -!- nooga has quit (Read error: 104 (Connection reset by peer)). 17:58:51 -!- puzzlet has quit (Remote closed the connection). 17:59:24 AnMaster: Hey. AMD have their own version of ICC-like stuff 17:59:28 http://developer.amd.com/CPU/OPEN64/Pages/default.aspx 17:59:33 mhm 17:59:49 Try cfunge with it :P 18:00:59 I will look at the url later. Atm I wouldn't want to browse anything, system is heavily loaded, generating a graph of the space usage under /home 18:02:06 AnMaster: Er... how big is your /home? 18:02:14 ehird, over 30 GB, why? 18:02:32 So's mine but it takes less than one minute to list all the files and usages of them 18:02:44 ehird, loading firefox at the same time as the disk is working so heavily would be insane. 18:03:00 w3m :-) 18:03:21 ehird, your computer is newer to begin with. 18:03:42 AnMaster: it's a disk-bound operation 18:03:45 I guess your disk might be slow 18:03:54 ehird, it's SATA, not SATA 2 18:04:12 * ehird refers to floppies and HDs as disks but optical media as discs; I wonder why? 18:04:18 and I have mostly lots of small files. Lots of directories 18:04:22 It just feels right. "Disk is cheap", "Disc is cheap" 18:04:31 "I'm ripping this disk", "I'm ripping this disc" 18:04:39 "I got it off this floppy disk", "I got it off this floppy disc" 18:06:06 trivia: sv:diskmaskin means en:dishwasher 18:06:50 I got it off this floppy dish 18:08:27 The distinction I heard once and nowadays make myself: a disk is computer storage, a disc is a round thing. 18:08:44 kerlo: HDs use discs. 18:08:55 also, it's "Compact Disc" 18:08:57 A frisbee is a flying disc. A flash card is a disk. A CD or a hard drive platter is both a disc and a disk. 18:09:17 sv:disk means en:dish as in dish washing, but not in the other meanings of dish 18:09:19 I seem to use disk as a plural: 18:09:23 Disk is cheap 18:09:39 If you're using "is", it's not a plural. 18:10:32 en:dishware is sv:servis, but a dirty set of dishware, may very well be sv:disk 18:10:46 actually I can't think of any exact corresponding word in English 18:11:52 Wiktionary essentially says "floppy disk, hard disk, compact disc: optical media are discs, other media are disks." 18:12:12 That's what I use 18:12:20 18:09 kerlo: If you're using "is", it's not a plural. 18:12:21 did both disk and disc existed in English before computers? 18:12:24 but it's referring to multiple disks 18:12:27 the words I mean 18:12:33 AnMaster: Disc existed. Disk did not, 18:12:36 s/,$/./ 18:12:36 ah 18:12:48 so what is the origin of disk then 18:12:56 Neologism, probably. 18:12:59 Or, to use an actual quote: "If the medium is optical, the variant disc is usually preferred . . . if referring to a physical drive or older media (3" or 5.25" diskettes) the k is used, but c is used for newer (optical based) media." 18:13:00 Technically, it's a floppy diskette, BTW. :p 18:13:12 why not discette 18:13:22 "Disk" is from Greek, "disc" is from Latin. 18:13:27 AnMaster: I don't think "cette" ever works. 18:13:30 -!- Sgeo has joined. 18:13:31 I think that forces it to be a k. 18:13:32 "discette" looks slightly French... 18:13:34 "Discette" would be pronounced wrong. 18:13:35 So probably disk came from diskette. 18:13:41 Where diskette comes from disc-ette. 18:13:42 Like "dissette". 18:13:50 heh 18:14:14 In Spanish, "disc" + "ette" would be "disquette". But "ette" isn't a Spanish suffix. 18:14:41 "disc" + "ito" would probably be "disquito", though, and it would mean "little disc". 18:15:36 " "Discette" would be pronounced wrong." <-- why then are there quite a few words in English that aren't pronounced as they are spelled. Since those exists, why would another such word hurt. 18:15:42 If the Spanish word for "disc" were "disc". It's actually "disco". The diminutive would still be "disquito", or possibly "disquecito" or possibly-possibly "discocito". 18:16:16 AnMaster: it's quite rare in English for a "c" to be pronounced the wrong way. 18:16:39 that is true, but weird spellings aren't unusual 18:16:45 "guitar" comes to mind for example. 18:17:08 But that's not a "c" that's being pronounced the wrong way. I think it's a copying of Spanish. 18:17:08 the "gui" in guitar is distinctly spanish feeling 18:17:20 ehird, yes it is probably an imported word 18:17:31 The Spanish pronunciation of "guitar" would be essentially the same. 18:17:51 ehird, when said in English it sounds more like "gitar" iirc? 18:18:01 AnMaster: Sort of. 18:18:12 it is "gitarr" in Swedish btw. 18:19:40 "gitar", stressed on the second syllable, with a hard g and a short i. 18:19:49 "Disc" comes from the Latin "discus". Disk also. 18:19:50 ;) 18:20:35 According to Wiktionary, "disc" comes from "disque" which comes from "discus" which comes from "diskos", and "disk" comes directly from "diskos". 18:21:00 Hmm. 18:21:03 * pikhq shakes a fist at English 18:21:50 Instead of speaking English, we should just speak an anglicised amalgamation of Old French, Latin, and Ancient Greek. 18:22:25 So, you mean English. 18:23:06 English contains things that are not from any of those. 18:23:11 mi.e ,Eli'at.xrd. 18:23:22 I only remember that Lojban now. 18:23:42 I'm pretty sure "wife" is quite Germanic, for one. 18:23:48 (though the French influences in English are from Anglo-Norman, not Old French) 18:24:00 Wiktionary says that everything is from Old French. 18:24:02 I wish Lojban had some sounds more suited to my name. 18:24:08 Eli'at.xrd is not very Elliott Hird at all. 18:24:18 It's more a-y than o-y, and my first name does not start with Bach's ch. 18:24:28 er 18:24:29 last name 18:24:50 Anglo-Norman, of course, being itself an amalgamation of Old French and some Germanic language... 18:25:13 (granted, the main Germanic influences on Norman are in its grammer) 18:25:55 grammer 18:26:13 XD 18:26:21 ehird: you're spelling it kind of weirdly. 18:26:32 I manage to misspell that word, and only that word, often. 18:26:33 mi'e .eli,at.xrd. 18:26:34 kerlo: It's what #lojban agreed on. 18:26:40 kerlo: Oh. Right. 18:26:42 Sorry. 18:26:46 That's just a typographical oddity. 18:26:55 I've forgotten Lojban, you see. 18:27:14 It seems "Elliott" would be .eli,yt., though. 18:27:15 I used "mi'e .Eli,at. xrd.", says google. 18:27:39 Unless you actually pronounce it with a short o. 18:27:50 So that it kind of rhymes with "polyglot". 18:28:36 It's time for a recording. 18:28:45 Cool. 18:28:48 * pikhq wants a Lojban programming language 18:29:19 pikhq: Been there, tried that 18:29:26 , failed. 18:29:31 It's still natural language. 18:29:33 You can write Unlambda in Lojban. 18:29:38 It just has an unambiguous grammar. 18:29:41 Seems like the grammar would be easy... 18:29:49 Getting sane semantics might be tricky. 18:29:56 ehird, it's a constructed language. 18:30:03 Yes. 18:30:03 I know. 18:30:09 Not a natural language. ;) 18:30:12 But it's constructed by humans, and to be a language for humans. 18:30:18 Right. 18:30:18 It's not a programming language or a formal language. 18:30:20 It's a natural language. 18:30:27 kerlo: http://filebin.ca/oceeca/eliatxrd.ogg 18:30:33 Released under the Share and Enjoy! license. 18:30:35 A natural language is the opposite of a constructed language. 18:30:39 The Share and Enjoy! License 18:30:41 1. Share and Enjoy! 18:31:01 ehird, are you saying that I should stick my head in a pig? 18:31:02 ``sk`sk = lo me sy be lo my ky bei lo me sy be lo ky 18:31:20 I think. 18:31:23 kerlo: that's using lojban's predicate stuff, right? 18:31:36 so if you say ```sii``sii, you'll crash the brains of Lojban speakers? 18:31:46 pikhq: I missed the joke. 18:31:58 It's impossible to say anything of worth in just about any language without using predicate stuff. 18:32:04 oh, wait 18:32:05 ehird: No, I'm doing a different one. 18:32:08 a related line from h2g2? 18:32:13 Yes. 18:32:29 Specific quotes from that book have a tendency to roll off me shortly after reading the whole thing. 18:32:32 * pikhq has that song stuck in his head now, dammit. 18:32:34 kerlo: So, what did I say in that ogg? 18:32:35 English has the same predicate stuff. That Lojban is pretty much precisely the same as the English "the S of the K of the S of the K", except with little words specifying associativity. 18:32:44 ehird: I haven't listened to ityet. 18:32:47 ...it yet. 18:32:49 ah 18:33:10 This is where you scold me for using Windows. 18:33:39 kerlo, stop using a shitty operating system. 18:34:44 Would you like me to attempt to justify my use of Windows to you? 18:34:55 kerlo: Yes, so I may laugh at you. 18:35:00 Okay. 18:35:27 As far as I remember, this is a complete list of things I've been unable to do under Windows: 1. Run Freenet reliably. 2. Play an OGG file. 18:36:35 kerlo: That's not a good argument. You need to argue why you won't use Linux, a technically superior operating system, not why you keep using Windows. 18:37:00 Or, y'know, OS X. 18:37:03 * pikhq notes that you also probably can't reliably play FLAC files 18:37:06 Or any other operating system on the planet apart from DOS. 18:37:06 ...well, things I've been unable to do under Windows that I would expect to be able to do under Linux. 18:37:37 Sound doesn't work well under Linux. YouTube videos don't play acceptably. 18:37:42 ehird: I dunno, DOS seems at least better-designed. 18:37:46 kerlo: Which Linux are we talking about here? 18:37:50 Ubuntu. 18:37:54 Here's how you get YouTube videos working on Ubuntu: 18:38:01 Accessories → Install/Remove. 18:38:02 kerlo: Sound works quite well under Linux, and I find YouTube works much nicer in Linux. 18:38:03 Search "Flash". 18:38:04 Tick Adobe flash. 18:38:05 youtube-dl and all that. 18:38:06 Click apply. 18:38:09 Refresh page. 18:38:13 That Was Hard(TM). 18:38:37 * pikhq doesn't use the Flash player for YouTube, but that's just because mplayer is much better at playing videos 18:39:01 YouTube videos actually play better without Flash? 18:39:28 kerlo: anyway, so, there's a refutation of the YouTube argument. If you're using a recent Linux (say with PulseAudio), there should be no sound problems either. 18:39:29 kerlo, what he suggests there is to *install* Flash. 18:39:31 Any other arguments? 18:39:44 Ubuntu by default ships with a crappy partial implementation of Flash. 18:39:51 No it doesn't, pikhq. 18:39:55 It doesn't do any such thing. 18:40:01 It ships without any flash plugin whatsoever. 18:40:05 Hmm. 18:40:10 I'm not used to operating systems under which "Add/Remove Programs" is capable of adding programs. 18:40:10 Must be thinking of Mandriva, then. 18:40:25 kerlo: OK, so your new argument is "It's easier, I don't like that." :P 18:40:32 That's the normal way of installing stuff on Linux. 18:40:33 ;) 18:40:42 That wasn't a new argument, that was jsut a comment. 18:40:48 I know. 18:40:51 I was Joe, King. 18:40:53 See you in a moment, when I'm running Linux again. 18:41:06 kerlo: that's one quick installer 18:41:31 It's not like I installed Linux in the past couple minutes. 18:41:44 kerlo: what version of Ubuntu is it? 18:41:55 If it's pre-pulseaudio, that may explain screwy sound. 18:42:06 If you've installed a non-Adobe Flash, that would explain YouTube. 18:42:11 I don't recall. Probably the newest. 18:42:19 9.04 18:42:20 ? 18:42:23 I have seen the PulseAudio thing. 18:42:25 You installed it after April 2x? 18:45:59 So here we are now, in containers. And by "in containers", I mean "running Linux". 18:46:22 The version turns out to be 8.10. 18:47:01 * Sgeo became a reddit person 18:47:05 The sound is set to "Autodetect", not "PulseAudio Sound Server". 18:47:37 kerlo: Don't change it. 18:48:12 When you say Accessories → Install/Remove, do you mean Administration → Synaptic Package Manager, by any chance? 18:48:23 No. 18:48:27 I mean Programs → Install/Remove. 18:48:56 And if it's not there, you've messed it up somehow and should just reinstall to exorcise whatever demons are in that installation. 18:49:15 * kerlo tries Applications → Add/Remove 18:49:23 Or that. 18:49:28 I haven't used Ubuntu in a day or two. 18:49:39 What does an asterisk next to e.g. "2 days ago" mean? 18:49:55 Sgeo: comment edited. 18:50:03 ah 18:50:06 ty 18:50:42 Will "Macromedia Flash plugin" do? 18:50:49 kerlo: Yes. 18:50:53 Sgeo: Tips for keeping your sanity on reddit: unsubscribe to politics, news, worldnews, worldpolitics, atheism. Unless you like hearing about how awesome Obama is, how stupid Christians are, and how awesome Obama is (what do you mean "world"). 18:52:07 Or just read /r/{programming,science,technology} which is a good chunk of what's left :-P 18:53:22 Anyway, the "o" of "Elliott" there actually sounds a lot like a short "i". 18:53:35 Which means it's a schwa, which is y, not a short "o", which is a. 18:53:42 kerlo: I suppose os. 18:53:43 *so 18:53:59 so .eli,yt.xrd. 18:54:01 kerlo: what about xrd? 18:54:05 That x is just so fake. 18:54:41 I don't think there's much you can do about that. I mean, you could make it .y'yd. or something. 18:54:54 That's not very Hird 18:54:55 Well, .y'YD. 18:55:11 kerlo: maybe .xird. is more accurate? 18:55:43 Probably not. 18:56:01 That's pretty much a long "e", after all. 18:56:08 * Sgeo wonders if ehird's comment will bring anyone into Agora 18:56:23 Sgeo: probably not. stop stalking me :-P 18:56:40 Anyway, I have to go eat lunch. The sound problem still exists; it's not exclusive to YouTube, apparently. See you when I get back. 18:58:40 I was mistaken. See you when I get there. 18:58:51 wat 18:59:22 And I guess I'll also see you before I leave. 18:59:33 It seems like buffer underruns, being slow and choppy. 18:59:47 That seems very unlikely 18:59:51 My suggestion: reinstall 18:59:55 With 9.04 18:59:59 Mmkay. 19:01:04 I don't suppose there's a convenient "uninstall Ubuntu" thing I can use that will keep my home folder and Firefox's configuration. 19:01:28 kerlo: Don't you have a flashdrive? 19:01:35 cp -R ~ /media/flashdrive/home 19:01:39 cp -R .firefox /media/flashdrive/ff 19:01:40 or sth 19:01:58 kerlo: I think ubuntu has an import thing but it's likely some of your settings are contributing to this 19:02:01 Well, it's only minimal stuff that I would have to do. 19:02:14 ? 19:02:35 All my data resides elsewhere; I'd just have to make new symbolic links to it. 19:02:49 Then go for that 19:02:53 * kerlo nods 19:14:06 bbl, possibly night. (not feeling well) 19:16:25 I could never sleep that early. 19:30:23 state is fun 19:31:04 how do I make this less ugly 19:31:07 http://pastie.org/private/er1czt3oce9nf7tuqcfpg 19:31:56 in particular it's way too polite 19:32:02 by not using state unneccessarily 19:32:16 and not using that fucked up "Wrap" type 19:32:25 does that thing even fulfil the monad laws 19:32:25 though that's kinda similar to how the HTTP module works 19:32:57 comex: isn't that the ((->) t) monad 19:33:03 maybe 19:33:08 ehird: never heard of it 19:34:20 * comex looks at Gtk2Hs documentation 19:34:23 I want to see how real modules do it 19:34:41 gtk2hs does it by being evil, comex. 19:34:49 whatever you're trying to do is probably wrong 19:35:00 I want to make a GUI toolkit :p 19:35:25 comex: you don't even know haskell properly, don't be stupid 19:36:58 :p 19:37:54 * Sgeo is devising ways to torture norns 19:38:52 * Sgeo pokes ehird 19:38:59 I am not a norn 19:40:35 I just figured you'd have interest in this 19:40:46 I do. 19:41:23 what about just declaring UIs functionally 19:41:36 My plan is constantly injecting full Disappointment and Punishment into all the norns in the world 19:41:42 been done, comex. 19:41:43 where the user defines a function blablastate -> UI 19:41:44 And sustaining their lives 19:41:45 I just suggest you learn hs first :P 19:41:47 also 19:41:48 that's not functional 19:41:53 that's just a manual monad 19:41:56 more functional than gtk2hs 19:42:02 no its not 19:42:07 yes, because 19:42:20 They wouldn't actually be in pain, but it would probably cause permanent brain damage after a while 19:42:20 you would declare the ui as a big list or something 19:42:22 no, it's just a manual monad 19:42:29 rather than by specifying operations to create it 19:42:57 As the bits that tell it "Ok, when I'm hungry, I should eat" eventually die from basically being told "WRONG!" 19:43:09 and just about everything rldr 19:43:11 else 19:43:12 [Button, Button, Button] whatever 19:43:13 rather than 19:43:17 button < - buttonNew olol 19:43:27 -!- olsner has joined. 19:44:02 Sgeo: Just give up the habit before Creatures 734723846823: Now They're Sentient 19:44:17 in other news 19:44:37 when I type :w on this remote file, the color goes away 19:44:58 I have to manually :syntax on 19:48:29 Hm 19:48:46 Maybe I should make a potion that, when eaten, causes the norn to have very violent dreams 19:49:10 A song. 19:49:10 "Mormons, mormons, mormons mormons mormons 19:49:12 Mormons, mormons, morrrr- 19:49:13 -mons, mormons, mormons mormons mormons 19:49:16 Mormons, 19:49:18 Mormons!" 19:49:19 When the norn wakes up, no matter what (e.g. hunger, bored, horny), its dream will have taught the norn to hit norns 19:49:56 Sgeo: Just make them reproduce all the time and die of overpopulation (does inbreeding have effects in Creatures?) 19:50:26 I don't think overpopulation is a serious issue, except slowing down the computer 19:50:37 As far as I'm aware, inbreeding has no effects 19:50:48 Sgeo: What, creatures take up no space? 19:51:43 Well, they'd feel crowded eventually 19:51:58 Sgeo: So you can have infinite creatures in infinite space 19:51:59 awesome 19:52:02 Yes, they take up space, but except for the crowdedness drive, they wouldn't notice 19:52:03 Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon MUSLIM MUSLIM! A Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon Mormon MUSLIM MUSLIM 19:52:36 GregorR: Obama! Non-natural-born citizeeeeeeeeeeeeeeeen 19:52:47 a mormon mormon mormon mormon mormon mormon mormon, mormon mormon mormon MUSLIN MUSLIN 19:52:54 ↑ Terrible 19:53:18 * GregorR lols at "muslin muslin" 20:03:01 ^choo mormon 20:03:01 mormon ormon rmon mon on n 20:03:32 What's "choo"? 20:03:52 There were a couple of variants on the "echo" commands on fungot. 20:03:52 fizzie: my hangman doesn't support unicode. 20:03:53 ^choo A magical butt. 20:03:54 A magical butt. magical butt. magical butt. agical butt. gical butt. ical butt. cal butt. al butt. l butt. butt. butt. utt. tt. t. . 20:03:58 20:03 fungot: fizzie: my hangman doesn't support unicode. 20:03:59 ehird: but we're rebuilding anyway, since it makes some sense. but basically, it's riastradh's sketchings i'm filling out nicely... 20:03:59 :DD 20:04:03 wow 20:04:06 it DOES make some sense 20:04:10 fizzie: what did you do to fungot?! 20:04:10 ehird: an alternative would be for. there was this http://www.sarg.ryerson.ca/dmason/ common/ fnord/ comments 20:04:17 fungot: What alternative? 20:04:18 ehird: applying ep1 to ep2 now. 20:04:26 fungot: That sounds like something liable to cause an explosion. 20:04:27 ehird: grrr... why doesn't ( cons ' a ' b 20:04:32 !addinterp c choo char buf[1024]; int i; fgets(buf, 1024, stdin); buf[strlen(buf)-1] = '\0'; for (i = 0; buf[i]; i++) printf("%s ", buf + i); 20:04:32 There is already an interpreter for c! 20:04:37 fungot: You're not cut out to be a Lisp programmer. 20:04:37 Err :P 20:04:41 ... 20:04:41 !addinterp choo c char buf[1024]; int i; fgets(buf, 1024, stdin); buf[strlen(buf)-1] = '\0'; for (i = 0; buf[i]; i++) printf("%s ", buf + i); 20:04:42 Interpreter choo installed. 20:04:45 !choo Hewwo 20:04:46 It hates me :-( 20:04:46 Hewwo ewwo wwo wo o 20:04:51 !choo IM OKA LOTJOIER 20:04:52 IM OKA LOTJOIER M OKA LOTJOIER OKA LOTJOIER OKA LOTJOIER KA LOTJOIER A LOTJOIER LOTJOIER LOTJOIER OTJOIER TJOIER JOIER OIER IER ER R 20:04:55 ^show choo 20:04:56 >,[>,]+32[<]>[[.>]<[<]>[-]>] 20:05:13 !delinterp choo 20:05:14 Interpreter choo deleted. 20:05:17 !addinterp choo bf >,[>,]+32[<]>[[.>]<[<]>[-]>] 20:05:17 Interpreter choo installed. 20:05:21 !choo Hewwo 20:05:21 Hewwo 20:05:25 afk 20:05:26 >_> 20:05:34 Heh, input has a newline. 20:05:36 !delinterp choo 20:05:36 Interpreter choo deleted. 20:05:40 !addinterp choo c char buf[1024]; int i; fgets(buf, 1024, stdin); buf[strlen(buf)-1] = '\0'; for (i = 0; buf[i]; i++) printf("%s ", buf + i); 20:05:40 Interpreter choo installed. 20:05:48 GregorR: '+32' 20:05:51 That means 32 +s. 20:06:00 Oh :P 20:06:04 !delinterp choo 20:06:04 Interpreter choo deleted. 20:06:14 !addinterp choo bf >,[>,]++++++++++++++++++++++++++++++++[<]>[[.>]<[<]>[-]>] 20:06:14 Interpreter choo installed. 20:06:16 !choo CHOO! 20:06:17 CHOO! 20:06:19 No, 'ts still no good because EgoBot sends a newline. 20:06:21 Ahem :P 20:06:29 20:06 EgoBot: HOO! 20:06:29 20:06 EgoBot: OO! 20:06:30 20:06 EgoBot: O! 20:06:58 Yeah, but that's still no good. 20:13:15 \Ö/ 20:13:46 !deilnterp choo 20:13:48 Err 20:13:51 !delinterp choo 20:13:51 Interpreter choo deleted. 20:14:26 !addinterp choo bf >,[>,]<++++++++++++++++++++++[<]>[[.>]<[<]>[-]>] 20:14:26 Interpreter choo installed. 20:14:30 !choo Hewwo 20:14:30 Hewwo ewwo wwo wo o 20:14:42 What a useless thing :P 20:14:43 !show choo 20:14:44 bf >,[>,]<++++++++++++++++++++++[<]>[[.>]<[<]>[-]>] 20:14:44 !choo Dij fog agju ufdhaiug njkknfjks klefjw sdiof 20:14:45 Dij fog agju ufdhaiug njkknfjks klefjw sdiof ij fog agju ufdhaiug njkknfjks klefjw sdiof j fog agju ufdhaiug njkknfjks klefjw sdiof fog agju ufdhaiug njkknfjks klefjw sdiof fog agju ufdhaiug njkknfjks klefjw sdiof og agju ufdhaiug njkknfjks klefjw sdiof g agju ufdhaiug njkknfjks klefjw sdiof agju ufdhaiug njkknfjks klefjw sdiof agju ufdhaiug njkknfjks klefjw sdiof gju ufdhaiug njkknfjks klefjw sdiof ju ufdhaiug njkknfjks klefjw sdiof u ufdhaiug njkk 20:17:10 !userinterps 20:17:10 Installed user interpreters: bfbignum chiqrsx9p choo echo hello rot13 slashes yodawg 20:17:18 !show slashes 20:17:19 perl #!/usr/bin/perl -w 20:17:31 Heh, I'm not going to get that output fully :P 20:18:07 Maybe I could make it ctcp send. 20:18:55 s/ctcp/dcc/ 20:19:47 GregorR: make it pastebin 20:20:14 Make it Codubin. 20:20:17 :p 20:20:34 Make it send it with smoke signals out of your chimney. 20:20:41 fizzie: PURRFECT 20:35:16 comex: it seems to me that the vast majority of your code there is unnecessary. 20:36:04 Do this instead: 20:36:24 data Sometype = SomeType Int Int Int deriving (Show, Eq) 20:36:46 j (SomeType _ _ c) = SomeType 5 9 c 20:36:54 k = j (SomeType 10 11 12) 20:36:58 (fin) 20:39:34 If you need a monad, it's data Wrap a = Wrap ((a, SomeType) -> SomeType), and already exists; it's called State SomeType. 20:39:57 -!- sebbu has joined. 20:40:23 Bleh, /me can't figure out how DCC CHAT is supposed to work. 20:42:25 * kerlo curses his terminal 20:42:37 g - a cursed terminal 20:42:48 * kerlo downloads PuTTY 20:43:40 kerlo: I thought you were using linux 20:43:55 OK, so that's annoying. The IP needs to be in decimal form X_X 20:44:08 That is, one-decimal-number form. 20:44:13 The retarded form everybody forgets exists. 20:44:29 I am using Linux. 20:44:55 For sentimental reasons, I require an ssh client that does not send a username. 20:45:06 It's very nice to connect to an SSH server and be asked for a username. 20:45:20 -!- sebbu2 has quit (Read error: 60 (Operation timed out)). 20:46:23 Does EgoBot have a way to compose commands? 20:46:35 kerlo: why download putty 20:46:38 install it from the package manager 20:46:41 add/remove 20:46:45 You can do that? 20:46:51 Yes. 20:46:54 >_< 20:47:01 Hey, you can. 20:47:02 THAT'S LIKE HALF OF THE WHOLE POINT OF USING LINUX! 20:47:04 Just about everything is in the package manager. 20:47:10 YOU BARELY EVER INSTALL EVERYTHING MANUALLY! :P 20:47:17 Well, I knew that you could download just about everything using the package manager. 20:47:29 I just thought that PuTTY in particular was not available that way. 20:47:50 Right. PuTTY is made of magical unicorns, which are impossible to stick into a .deb 20:48:51 No, I seemed to remember reading something along the lines of "nope, we don't make PuTTY for *NIX; do it yourself". 20:49:56 That's why the distros make it themselves 20:50:35 That makes sense. 20:54:11 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/mandelbrot.b 20:54:13 AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDEGFFEEEEDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 20:54:46 Whoops :P 20:54:49 Forgot to update 20:55:07 "Santa Clara (CA) – Yesterday, we learned that Intel has begun phasing out the Core i7 940 processor, today we are told that the Extreme version 965 will be retired this year as well. " 20:55:17 WHAT THE FUCK ARE YOU DOING INTEL?! 20:55:36 STOP IT :| 20:56:04 Is there something wrong with phasing out first-gen processors? 20:56:17 Deewiant: They were only released a few months ago, and I want one :-P 20:56:26 I guess the 975 will be out sometime, though. 20:56:31 Most likely they'll release replacements. 20:56:53 It's not like they'll just say "nah, we're out of the high-end processor business, AMD can have that" 20:56:57 Lawl 20:57:16 Deewiant: Weeell, you have to go AMD if you want >8 cores. 20:57:39 "High-end desktop" 20:57:42 Sorry :-P 20:57:44 Wait, I think the last-gen Xeons had a single 6 core chip. 20:58:02 Deewiant: Yeah, I just picked up on it since I'm assembling such a system in my head :-P 20:58:09 ... or in an editor window as the case may be. 20:58:33 !befunge98 mycology/mycology.b98 20:58:38 Errr 20:58:50 !befunge98 http://rage.kuonet.org/~anmaster/mycology.b98 20:58:51 0 1 2 3 4 5 6 7 20:59:00 Whoo boy. 20:59:12 Uses DCC now :) 20:59:18 Works great. 20:59:27 !befunge98 http://rage.kuonet.org/~anmaster/mycology.b98 20:59:28 0 1 2 3 4 5 6 7 20:59:29 And no limits other than <4K total data. 20:59:29 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/mandelbrot.b 20:59:32 AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDEGFFEEEEDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 20:59:41 Still sends the first line to the channel though ;) 20:59:45 GregorR: my client appears not to do dcc chat. 20:59:55 ehird: Wow, you have a reaaaaaaaally lame client :P 21:00:01 Limechat. 21:00:05 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/mandelbrot.b 21:00:08 AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDEGFFEEEEDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 21:00:09 GregorR: Oh. 21:00:11 GregorR: Hm. 21:00:19 ? 21:00:23 GregorR: Does it wait until it's been accepted before it sends? 21:00:23 GregorR: No, this is odd, it hasn't said the first line of my mycology invocation yet. 21:00:26 I think I only got one line. 21:00:36 GregorR: 21:00 CTCP-query unknown(DCC CHAT) from EgoBot : chat 1077849409 10054 21:00:36 Since I had to look up the command to accept a DCC chat request :-P 21:00:49 Deewiant: There is no command to accept a DCC chat request. 21:01:06 ehird: Yeah, that's not gonna work :P 21:01:06 In my client there is. 21:01:22 Deewiant: Well that's laem. It should still work though, it only sends once you've connected. 21:01:34 http://en.wikipedia.org/wiki/Direct_Client-to-Client#DCC_CHAT 21:01:38 There's a handshake. 21:01:41 * Sgeo hates 16-bit color 21:01:42 GregorR: Just make it run a minimal HTTP server and proxy it from $SERVER, so if the program outputs multiple lines it goes: 21:01:44 first line 21:01:47 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/mandelbrot.b 21:01:50 AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDEGFFEEEEDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 21:01:50 http://codu.org/egobot/snippets/342348 21:01:52 Deewiant: The handshake is "I send you DCC CHAT, you connect" 21:01:57 GregorR: And then have people reload that to get more. 21:02:02 Now I got more. 21:02:05 Fun for everyone! 21:02:20 But then the session got lost or something, half way through 21:02:24 The 4K limit, mayhaps. 21:02:25 ehird: That would be so much more obnoxious. 21:02:36 GregorR: Yeah, exactly, I have to accept the connection, that's what I meant. 21:02:39 GregorR: How? It'd let other people view it, and not have their irc clients beep every 2 seconds. 21:02:52 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/mandelbrot.b 21:02:55 AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDEGFFEEEEDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 21:02:55 ehird: Every 2 seconds? It sends everything INSTANTLY 21:03:11 GregorR: My other reason, then. 21:03:24 I only get one line, AAAAAAAAAAAAAAABBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDDEEEFGIIGFFEEEDDDDDDDDCCCCCCCCCBBBBBBBBBBBBBBBBBBBBBBBBBB, if the chat wasn't already open 21:03:27 PMADDUBSWMultiply and Add Packed Signed and Unsigned BytesTake the bytes in registers A and B, multiply them together, add pairs, signed-saturate and store. IE [a0 a1 a2 ...] pmaddubsw [b0 b1 b2 ...] = [satsw(a0b0+a1b1) satsw(a2b2+a3b3) ...] 21:03:32 What an obvious operation. 21:03:55 Letting other people view it isn't compelling to me, that's really not the point, and I wanted to keep everything more-or-less in-IRC. 21:04:21 A combo would be nice, but a bit of a PITA. 21:04:21 GregorR: If someone runs a command in a public channel, I would assume they're displaying something to the world. They get the first line displayed at least. 21:04:27 If they don't want that why not just /msg the bot? 21:05:09 Well, right now because due to a bug that I haven't figured out, the bot doesn't accept commands over /msg :P 21:05:23 :P 21:05:55 I don't think people are always, or even often, showing something to the world, they're usually pokin' around. 21:06:28 Pokin' around's perfectly possible in /msg. Besides, if it's the fun sort of public pokin' around (innuendo not intended. Maybe.) then THE WORLD SHOULD SEE IT IN ALL IT'S GLOOOOOOOOORY 21:08:06 !bf http://esoteric.sange.fi/brainfuck/bf-source/prog/beer.b 21:08:07 99 Bottles of beer on the wall 21:08:23 Heh, it only got as far as 63 :P 21:09:33 And my "immortal" norn died 21:09:36 underwater 21:09:38 I wouldn't probably be in a very good speaking condition after `expr 99 - 63` beers either. 21:10:08 Sgeo: lol 21:10:15 It's not supposed to drown! 21:10:16 Sgeo: make an immortal norn suicidal 21:10:22 watch anguish 21:10:34 I don't think norns can be made suicidal. Violent, but not suicidal 21:10:43 OK, I have to know WTF norns are :P 21:11:30 http://en.wikipedia.org/wiki/Creatures_(artificial_life_program) 21:12:50 * Sgeo deactivates the Death by Low Energy gene 21:13:00 Huh 21:13:20 And death by low ATP, but even without that gene, 0 ATP kills organs, I think 21:13:51 * Sgeo adds full energy to his ATP creation gene 21:15:03 You sleep with the fishies now, tough guy 21:15:15 You do a good job of making it sound godawfully dull. 21:16:12 :P 21:16:21 GregorR: He's not playing the game regularly. 21:16:26 He's hax0ring the internal gene stuff. 21:16:32 I don't think that's supported 21:16:36 It is 21:16:43 But it's not normal gameplay 21:16:44 Well, it's not how you play the game regularly, at least. 21:16:47 Yeah. 21:17:05 Ahhhhhhhhhhh, OK 21:18:31 Who wants to write 8080 code? 21:18:38 WHO DOESN'T? 21:19:12 What about 4004 code? 21:19:19 740 kHz. 21:20:44 This analogy is like a really bad analogy: It's not clever and it makes no sense. 21:21:03 :D 21:21:04 http://www.e4004.szyc.org/ 21:21:42 SWEET! 21:21:52 Not as sweet as JSMIPS but still ;) 21:22:23 GregorR: @\0 21:22:25 An infinite loop. 21:22:29 (That's 40 00, hexwise.) 21:23:09 * 4-Bit Parallel CPU With 46 Instructions 21:23:14 Yep. 21:23:15 That's downright RISC ;) 21:24:10 GregorR: A thought for you while I brb: ARMjs. From the Simple Gargantuan Gadgets (SGG) chip company; an early competitor to Ligence. 21:25:03 Wow, that tells me so little :P 21:26:17 This might be considered torture: The norn will feel needs and make decisions like a regular norn, but will have a neurological issue that causes everything to come out as "hit norn" 21:26:31 lol 21:29:44 Ok, my "hyperviolent" norn, instead of doing what I want, is having sex with the victim 21:30:05 And things that the answer to all of life's problems is to go right 21:30:12 And she fell asleep 21:30:13 WTF 21:30:17 Oh I know 21:30:26 You die now 21:30:54 You are fucked up. 21:30:57 Hm 21:31:08 I was trying to clear all other attempted actions, instead, I set them all to active 21:31:16 pikhq, me? 21:31:53 Si. 21:32:36 lol at the violence machine 21:33:55 Sgeo: Any information on whether openc2e is goodish? 21:33:59 The violence machine is asleep 21:34:09 GregorR, it's not usable for norns yet, I think 21:34:13 It's been a while since I checked 21:34:32 So that makes it not usable with any of the games then, presumably :P 21:34:59 I think you can get DS running without the norns in them 21:35:02 But I'm not sure 21:35:04 Ask in Sine 21:38:26 -!- kar8nga has joined. 21:41:14 The norns walk around with frightened looks on their faces 21:41:17 Scared of Ms. Violence 21:42:46 "em like ms. violence" 21:42:47 wtf 21:43:17 "Ms. Violence" :P 21:43:39 I'm going to have to kill her 21:43:43 She keeps falling asleep 21:44:51 * GregorR adds that to the "list of comments to repeat endlessly out of context" 21:45:02 -!- KingOfKarlsruhe has joined. 21:45:33 She killed all the kids. The only other norn alive is near-immortal 21:46:00 Sounds like this was a really good decision on your part :P 21:46:16 What decision? 21:46:17 Which game in the series are you playing? I think I may actually own Creatures 1 :P 21:46:23 Sgeo: Making Ms. Violence 21:46:23 Docking Station 21:46:46 She can't eat. She doesn't need to eat, but she's hungry 21:46:50 Maybe I should fix that 21:47:00 Sgeo: I thought that was some kind of add-on (judging by Wikipedia), is that an actual (free??) game? 21:47:11 GregorR, yes, it's an actual free game 21:47:18 Oh! 21:47:19 * GregorR downloads. 21:47:36 It's also an add-on 21:47:40 But C3 is optional 21:48:38 21:24 ehird: GregorR: A thought for you while I brb: ARMjs. From the Simple Gargantuan Gadgets (SGG) chip company; an early competitor to Ligence. 21:48:39 21:25 GregorR: Wow, that tells me so little :P 21:48:45 Advanced→Simple 21:48:49 Micro→Gargantuan 21:48:54 Devices→Gadgets 21:48:56 Intel→Ligence 21:49:04 21:43 Sgeo: I'm going to have to kill her 21:49:04 21:43 Sgeo: She keeps falling asleep 21:49:06 lawl 21:49:13 The thing is, she doesn't WANT to hit norns. She is neurologically forced to, but she doens't want to 21:49:22 Sgeo: Hah 21:49:31 -!- jix has joined. 21:49:31 Sgeo: Do the same but with breeding. 21:49:38 I predict a fucked up Norn family. 21:50:23 I find Creatures boring because of the potential for the creatures to become immortal. 21:51:15 And then evolution just stops. 21:51:30 kerlo: Shouldn't you dislike the Singularity for the same reason? 21:52:37 I kind of do. 21:52:39 -!- tombom has joined. 21:57:30 kerlo: With immortality, you just switch to upgrading in-place. 21:57:47 -!- puzzlet_ has quit (Read error: 60 (Operation timed out)). 21:57:51 -!- puzzlet has joined. 21:58:11 That doesn't happen in Creatures. 21:58:39 -!- kar8nga has quit (Remote closed the connection). 21:58:51 ehird, making horny norns now 21:59:09 kerlo: Creatures isn't the singularity :P 22:02:00 They're going at it, and still no preg.. ok, there we go 22:04:21 Ms. Violence disapproves of this family's antics 22:04:33 haha, what's happening 22:06:28 I just fixed the violence genome, the new, untiring Ms. Violence is about to be born 22:06:56 Well, I said "kind of". 22:08:14 Ok, so adults aren't as fragile as children 22:08:31 Even stupid adults trying to do the murderer 22:08:35 $ /usr/local/bin/dockingstation 22:08:35 dirname: missing operand 22:08:36 Blargh 22:08:45 GregorR, there's some patch for that 22:09:16 try http://sgeo.diagonalfish.net/creatures/ds.diff (no guarantees) 22:09:45 ...and I somehow forgot that the superbreeders were immortals 22:09:47 -!- jix has quit ("PRYTA!"). 22:10:46 That's it, I had to kill them all 22:11:03 I'm going to send the violent norns to RANDOM PEOPLE! 22:11:14 " # Is the awk/ls magic portable? " 22:11:16 evidently not! 22:11:52 GregorR, is it working 22:12:28 Seems to be. 22:12:42 norns ? what does it mean ? 22:12:47 Gee, I can't log in :P 22:12:53 * GregorR goes to get food. 22:13:08 GregorR, make an account 22:15:50 ROFL at the fighting match 22:16:17 They like eachother 22:16:40 The girls are fighting the girls and the guys are fighting the guys 22:18:35 The astro girl's dead 22:19:23 This one is watching over the girl's corpse. Probably trying to kill it 22:20:23 Should I feel guilty about this? 22:21:02 -!- puzzlet has quit (Remote closed the connection). 22:21:06 -!- puzzlet has joined. 22:23:33 Sgeo: Are they alive? 22:23:57 The norns that I sent to other random people, who don't realize their nature? Yes, when I sent them 22:24:19 22:20 Sgeo: Should I feel guilty about this? 22:24:35 You shouldn't feel guilty about wrongdoing to things that aren't alive or valuable :-P 22:24:54 But it could hurt the norns that others care about 22:25:00 I don't care about them, but others might 22:25:13 Linux's market share isn't valuable! /me reboots into Windows. 22:25:14 :-P 22:28:57 kerlo: lawl 22:29:03 Sgeo: that's their problem. 22:29:05 it's a game 22:36:48 It's not exactly a competitive game. 22:37:57 -!- tombom has quit ("Peace and Protection 4.22.2"). 22:45:34 -!- BeholdMyGlory has quit (Remote closed the connection). 22:46:20 -!- iano has joined. 22:49:45 Sgeo: I already did make an account. 22:49:56 Um, hm 22:50:01 And you can't connect? 22:50:19 Try renaming server.cfg in the directory storing DS 22:50:25 Then restarting DS 22:51:26 -!- puzzlet has quit (Remote closed the connection). 22:51:31 -!- puzzlet has joined. 22:51:57 AFK 23:07:51 Sgeo: Thanks for the obscenely violent Norn :P 23:08:14 :-D 23:08:15 -!- Gracenotes has joined. 23:10:47 -!- coppro has quit ("The only thing I know is that I know nothing"). 23:10:59 GregorR, you received a norn from me? 23:12:03 Oh, I received a Norn from somebody, with you as the original breeder. 23:12:09 And it killed all my Norns. 23:12:12 It was pretty sweet. 23:12:29 "You should stop warping that norn around. It's sick, and it hits other norns compulsively. Also you have seriously screwed up its DNA. It's a ridiculously fast-ager." 23:13:14 -!- coppro has joined. 23:13:33 Sgeo: Do they not know you did it intentionally? :P 23:13:40 Sgeo: Spread around the uberbreeder. 23:14:04 I think the child of an uberbreeder and a normal norn might be deformed 23:14:13 Sgeo: Excellent 23:14:20 Well, not visibly deformed 23:14:32 But might, say, be completely unable to do anything 23:14:38 Do it 23:14:41 Or might be part immortal, and part suspectible 23:14:51 Also, it's not quite the best breedability yet 23:14:57 Oh just do it 23:14:58 I like havoc 23:15:22 I want to improve the fertility first 23:15:53 "Horny" is really the wrong word. They're not horny, they just compulsively do it 23:16:55 What game you guys playing? 23:17:03 pikhq: Sgeo's breeding mutant Norms in Creatures. 23:17:07 Well, engineering. 23:17:08 Hmm. 23:17:21 * pikhq has always wanted to try that game... 23:18:30 -!- FireFly has quit ("Later"). 23:19:35 * Sgeo does stuff that should increase fertility 23:20:07 * pikhq discovers that it appears to be gratis 23:20:10 Sgeo: How do you send a Norn through the portal? 23:20:16 Two ways 23:20:51 Put it in containment, config the Containment Chamber, click the face of the norn in the bar above the CC, and clikc the green checkmark 23:21:00 Or make a portal: There's a portal creator in that room 23:21:00 Ohhhhhhhhhh 23:21:01 -!- iano has quit. 23:21:20 I was using the portal creator, but I couldn't figure it out >_> 23:22:05 Ok, this should be hyper breeding 23:22:14 Actually, I should make the female's pregnancy go quicker 23:30:37 so far, it still isn't um.. what's the equivelentof birth, but for eggs? 23:31:00 -!- oerjan has joined. 23:31:10 Did she just give birth after I killed her? 23:31:32 ^^calculated to have maximum disturbing effect on oerjan 23:31:36 gestation? 23:31:59 Sgeo: it failed 23:32:41 although that may be because i had already changed to logreading, and so saw the preceding part immediately 23:33:23 afk 23:38:02 -!- M0ny has quit ("Read error: 182 (Connection reset by beer)"). 23:38:15 kisspop, dangit 23:40:40 Well, I'm bored with not effing up my Norns :P 23:40:57 GregorR: Learn CAOS!~ 23:41:01 s/~// 23:41:07 All commands are 4 alphanumerics long 23:41:12 So they fit into a 32-bit word 23:41:14 -!- jix has joined. 23:41:14 So they have a lookup table 23:41:22 And also they have a goto you can return from instead of functions. 23:41:34 ehird, this isn't CAOS that I'm doing 23:41:37 It is a torturous language to torture Norms. 23:41:41 This is the Genetics Kit 23:41:43 Sgeo: Yeah, but CAOS is easier :P 23:42:04 GregorR, Ctrl-Shift-C 23:42:16 In Spanish, "disc" + "ette" would be "disquette". But "ette" isn't a Spanish suffix. 23:42:29 i suspect it's from french, and does it the same way 23:42:32 Then: 23:42:50 enum 4 0 0 sway writ targ 0 1 0 2 0 3 0 4 next 23:42:58 Check the amount of pain they're in 23:45:31 Ultrabreeders, take 1 million 23:47:31 Ah, the sound of continuous sex 23:48:11 The lovers need to be separated in order for the egg to be laid :( 23:48:33 23:47 Sgeo: Ah, the sound of continuous sex 23:48:39 GregorR: there's an out of context thingy for you 23:52:32 My impression of bash.org: 23:52:35 on fire brb 23:52:45 :p 23:53:16 GregorR: 23:53:16 there's a small fire burning in my room 23:53:17 lemme guess im supposed to act suprised that you're telling us and not making any attempt to extinguish it, so i can submit it to bash where it will join the ranks of the other "SOMETHING CATOSTROPHIC HAPPENED SO I CAME TO TELL YOU GUYS ON IRC FIRST INSTEAD OF ATTEMPTING TO DEFUSE THE HOSTILE SITUATION" quotes that are grossly abundant, similar, and overrated. and despite a new one is submitted each week and only the location of the fire is alt 23:53:22 ered, loyal viewers firmly believe it is a unique and hilarious quotation, pledging support in the form of unneccesary votes 23:53:25 Bash quote 60469. 23:53:47 Note that there's at least another digit on quotes nowadays. 23:53:50 People just don't get the hint :P 23:57:25 I think something's broken 23:58:47 * Sgeo pokes GregorR