00:00:09 <Sgeo> > ['a'..'z'] \\ (nub . sort . map toLower) "The quick, yet lazy brown fox jumpwedover my bix, doglike sphynx of qwertz."
00:00:53 <alise> oerjan: that was me testing my new keyboard :D
00:01:09 <oerjan> Sgeo: "yet" is redundant
00:01:10 <alise> a blend of jackdaws love... and the quick brown... obvs
00:01:37 <Sgeo> oerjan, letterly, sure, but not semantically
00:06:04 <alise> http://esolangs.org/wiki/Image:Sprites.GIF Sherby Tepu
00:06:18 <oerjan> <alise> Things that annoy me: People who get annoyed by too many things.
00:06:30 <oerjan> reminds me of a calvin & hobbes cartoon
00:07:24 <oerjan> (calvin makes a list of things that annoy him, and hobbes suggests "negative people". or something like that.)
00:09:05 <oerjan> alas i read it in norwegian so even if it were on the internet it'd probably be hard to find)
00:09:45 <pikhq> oerjan: I recall that.
00:10:07 <pikhq> That's about the right phrasing.
00:11:18 <oerjan> ah "negative people" was the right phrase
00:14:15 <oerjan> curse bill watterson and his excessive copyright protectivism
00:15:47 <pikhq> Better than being a corporate whore, though. I can at least *respect* a guy who refuses any merchandising on principle.
00:19:34 <oerjan> <Phantom_Hoover> Things that annoy me: People who feel the need to get a sense of self-superiority by not complaining about things.
00:19:43 <oerjan> we prefer to kill you in your sleep instead.
00:26:23 <alise> i propose we create war
00:29:59 <alise> oerjan: NO AWESOME
00:30:34 <oerjan> rubbish, you should be shot for making such an idiotic suggestion
00:31:05 <alise> hey oerjan, debracketise this for me, i'm too tired:
00:31:28 <alise> (((A and B) or C) and D)
00:38:34 <alise> Vagrant report: 63 lines, AI is still broken because cheater is moody
00:39:14 <quintopia> holy sit i've been gone two hours and my last message is still on the screen
00:40:53 <quintopia> man, the kind of AI you're doing is easy. why not just do it yourself?
00:40:58 <oerjan> yeah you just brag about your big screen
00:41:00 -!- augur has joined.
00:41:22 <quintopia> oerjan: that was more a jab at everyone's inactivity...
00:41:55 <alise> quintopia: i'm lazy and obfuscated python is a bitch
00:42:08 <alise> i've forgotten what my variables are for
00:42:33 <alise> i swear this is reducible
00:43:06 <oerjan> quintopia: you clearly cannot have been here during the times when it's _really_ silent and all you see is quit/join messages for hours...
00:43:08 <quintopia> alise: it is indeed, reducible, using cheater's o function you refuse to borrow
00:43:28 <quintopia> oerjan: those are off-peak hours. the two hours i was gone were near-peak
00:44:04 <alise> quintopia: what are you, cheater's lawyer? i didn't see that copysign could help
00:44:07 <alise> how can it help here?
00:44:52 <quintopia> i dunno python exactly, so this shall be pseudocode
00:46:33 <quintopia> if abs(X-x)>17 X-=o(1,X-x) should do the work of both the first two lines.
00:46:43 <quintopia> there may be a sign error in there, but i think that's it
00:46:54 <alise> err, wait, what does copysign do again?
00:47:15 <alise> quintopia: then that's the same as cmp(X,x) I think
00:48:09 <quintopia> rename it to a letter to save space :D
00:49:15 <alise> quintopia: it works, thanks!
00:49:17 <quintopia> at the cost of the define operation
00:49:25 <alise> if abs(X-x)>17:X+=cmp(x,X)
00:49:26 <alise> if abs(Y-y)>5:X+=cmp(y,Y)
00:49:44 <alise> quintopia: well i use cmp elsewhere, albeit in broken ai code
00:49:56 <alise> but 1, I have very few letters left, and 2, i use range more often (four times)
00:49:58 <alise> and that has a longer name
00:50:00 <alise> so it'd take priority
00:50:21 <oerjan> alise: 17< and 5< to save a space, surely?
00:50:34 <alise> oerjan: how does that save a space, exactly?
00:50:44 <quintopia> alise: can you use extended ascii in var names?
00:51:20 <quintopia> alise: what do those lines do that they shouldn't?
00:51:40 <alise> X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y)
00:51:58 <alise> this works, unbelievably
00:52:03 <alise> take that, four lines
00:52:18 <alise> quintopia: yeah i abuse integers-as-booleans and vice versa so much in this program :P
00:52:25 <alise> q=w[v];s.addch(B-Y,A-X,choice(T)if U and r(0,2)==0 and q-32 else q)
00:52:29 <alise> q-32 for q!=32? why not
00:52:35 <alise> def Q(x,p=0):global V;v=V;V=lambda:(v()and C(x)or p)and s.getkey()
00:52:39 <alise> because fuck you, using def is too long
00:53:06 <quintopia> then i'm not seeing how that works
00:53:11 <oerjan> doesn't python have && ?
00:53:21 <alise> negatives are true too
00:53:39 <alise> it *does* have bitwise and
00:53:50 <alise> which is shockingly defined on booleans
00:54:10 <quintopia> alise: i don't see how 17<abs(X-x)and cmp(x,X) ever returns -1?
00:54:18 <quintopia> explain to this python non-knowing person
00:54:25 <alise> quintopia: cmp is -1 if less
00:54:41 <alise> True and x = x, even if x is a horse, not a boolean
00:55:01 <alise> TypeError: unsupported operand type(s) for &: 'int' and 'NoneType'
00:55:13 <quintopia> but never seen and return non-bools
00:56:55 <alise> i wonder if there's a better way to write
00:57:04 <alise> i.e. inc S only if S>=0
00:58:35 -!- cpressey has joined.
00:59:06 <alise> quintopia: you mean -=
00:59:27 <quintopia> well, your code said minus, but your text afterward said increment
00:59:49 <alise> also er >0 obviously
01:00:19 <alise> assignments aren't expressions
01:00:39 <alise> i am pretty proud with how tiny this code is
01:00:47 <alise> i scroll through it all and think "that's it??"
01:00:56 <quintopia> i've never done python, but it already seems really strange
01:00:59 <alise> quintopia: yes, but you can put anything that doesn't have a ; in it on the right of =
01:01:12 <alise> no, python's basically just the most boring language ever, on purpose
01:01:22 <alise> you have to be cunning, like me, to do perverse things
01:01:28 <quintopia> duck typing isn't all that boring actually
01:01:59 <oerjan> that's because you're a quack
01:02:02 <pikhq> Yeah, Python focuses on not being surprising. And it succeeds at that.
01:02:20 <quintopia> alise: can i see a current snapshot?
01:02:32 <pikhq> Problem being that interesting things are often surprising. :)
01:03:02 <quintopia> pikhq: you can be interesting and surprising even given the most mundane material to work with
01:03:03 <alise> no python violates POLS a lot
01:03:09 <alise> Ruby doesn't so much
01:03:26 <alise> python is very surprising, but if you do boring stuff it's ... boring like you expect
01:03:26 <quintopia> ruby i have played with. i like it a lot. it's neat.
01:03:31 <alise> quintopia: http://pastie.org/1204286.txt?key=iqxs2n6fgidpua9mnwpbrw
01:03:35 <alise> quintopia: complete with broken ai
01:03:38 <alise> (from q=w.copy() to w=q)
01:03:52 <alise> quintopia: debug.py:
01:04:00 <alise> def hook(t,v,b):endwin();sys.__excepthook__(t,v,b)
01:04:00 <alise> sys.excepthook=hook
01:04:00 <alise> execfile('vagrant.py')
01:04:09 <alise> if you modify the code at all, this is vital; it lets you see exceptions
01:04:20 <alise> the normal script calls endwin() and they get sucked up before they're displayed
01:04:51 <pikhq> alise: But it's quite surprising for a modern language to actually not violate POLS. :P
01:05:17 <alise> quintopia: endwin() resets the terminal and all the curses stuff goes
01:05:26 <alise> but it's in an exit handler
01:05:32 <alise> which is called *after* the exception is displayed
01:05:34 <alise> meaning it gets sucked up
01:06:12 <quintopia> well, i wasn't planning on running it just yet...just look
01:06:30 <alise> quintopia: it only wipes your hard drive if you want
01:06:33 <alise> if you do, controls are trivial
01:06:37 <alise> vikeys to move including diagonals
01:06:48 <alise> everything else pauses
01:07:06 <alise> ! is potion, # wall (cube of impenetrable yet transparent glass -- you can see beyond it!)
01:07:18 <alise> % is food (just walk into it; 1/15 chance of tripping balls)
01:07:32 <alise> Q is hopelessly confused, nonviolent monsters; if you want rid of one, walk into it
01:07:51 <alise> S is satiation, run out and you die quickly, eat to stop that
01:08:15 <alise> going into a ! will add to your potion-meter, displayed in parens next to HP; q to move 20 or whatever you have if it's less from potion meter to HP meter
01:08:27 <alise> most everything is random so i can't give specific values.
01:10:09 <alise> if S<1:L=max(L-25,0)
01:10:28 <alise> if S<1:L-=(L>0)*25
01:10:31 <alise> unbelievably, shorter
01:11:01 <cpressey> if you come to Python from a language with block scope you will be surprised, oh yes.
01:12:26 <alise> another simplification
01:12:34 <alise> quintopia: thanks for making my brain realise that works
01:12:41 <cpressey> seeing this makes me want to design a golfing language
01:12:43 <quintopia> alise: so there's no model behind the view? if you walk onto something, it ceases to exist?
01:13:01 <alise> quintopia: well there is one bit of model, your x and y
01:13:07 <alise> would be way too much work to find you every turn :P
01:13:17 <alise> quintopia: also you *could* store an inventory
01:13:21 <quintopia> cpressey: isn't that what perl is for?
01:13:23 <alise> but it'd be a bit of a bitch
01:13:39 <alise> if S<1:L-=(L>0)*25
01:13:39 <alise> if S<1 and L:L-=25
01:13:59 <alise> coolness or sanity... HMMMM
01:14:04 <cpressey> quintopia: it's what flogscript is for, literally, but i haven't checked it out yet.
01:15:03 <alise> cpressey: golfscript too
01:15:07 <alise> cpressey: no it does not
01:15:55 <alise> oh hm you forgot to include L in that conditional
01:16:03 <oerjan> alise: L-=25*(S<1 and L) perhaps?
01:16:08 <alise> L-=S<1 and(L>0)*25
01:16:57 <alise> after a ) or ' or " pretty much
01:16:59 <cpressey> oerjan's is shorter by one whole character!
01:17:11 <cpressey> and wtf does "1and" not work, guido?
01:17:34 <oerjan> alise: oh wait it nneds to be (L and S<1), not?
01:17:58 <alise> thank you good sir
01:18:08 <alise> ehird@dinky:~/Code/vagrant$ wc -c vagrant.py
01:18:15 <alise> should be around the same size with working AI, too
01:18:46 <quintopia> holy shit. you've come down a whole 3 characters on that one line in the last 15 minutes!\
01:19:12 <alise> how many pages of code is 58 lines? (one wraps)
01:19:32 <alise> whatever it is, it's remarkably playable. moreso when the monsters get a brain
01:19:38 <alise> quintopia: i dunno, people measure in pages of code all the time
01:20:00 <quintopia> usually, it's print out in 12pt monospace font
01:20:43 <oerjan> what's that o function again
01:21:01 <cpressey> istr a "page" being sixtysome, by some printer's reckoning
01:21:35 -!- Mathnerd314 has quit (Ping timeout: 252 seconds).
01:21:40 <alise> the math module did
01:21:47 <alise> oerjan: copysign(x,+n)=+x
01:22:01 <alise> copysign(x,n) = x sgn(n)
01:22:23 <cpressey> copysign is either a very lame function or a very cool function. i cannot decide.
01:23:14 <quintopia> since one can define it as x and x/abs(x)
01:23:52 <alise> i see quintopia has assimilated the crazy and notation :P
01:24:11 <alise> hallu in my game is so hardcore
01:24:13 <quintopia> which makes copysign x and x*x/abs(x)
01:24:17 <alise> now, this silliness shall continue, but tomorrow!
01:24:26 <quintopia> i like syntactic sugars, i must admit
01:24:30 <pikhq> WHO USES and FOR WHATEVER THE HELL THAT IS?
01:24:48 <alise> pikhq: x and y = y if x != 0, otherwise 0
01:24:50 <cpressey> x and x/abs(x) may be going a bit far, I agree
01:25:00 <alise> pikhq: the reason: i'm using it in my crazy golfed roguelike
01:25:02 <alise> because it's short!
01:25:05 <quintopia> pikhq: short-circuiting for fun and profit has been around since C was invented
01:25:17 <alise> quintopia: so why did it surprise you?
01:25:21 <alise> just not used to python doing it?
01:25:24 <cpressey> is there a name for "short-circuit and evaluate to" like this, though?
01:25:34 <alise> cpressey: ruby too
01:25:41 <quintopia> well, i'm not used to it behaving /precisely/ like that
01:25:47 <alise> it totally does quintopia
01:25:54 <pikhq> quintopia: Buuut "x and y" as "x ? y : 0"? What crazy crack is that?
01:25:55 <alise> y if x is true otherwise x
01:26:08 <alise> perl, ruby, python, lua
01:26:10 <alise> everything does it
01:26:12 <quintopia> pikhq: yeah i know. that's the part that surprised me
01:26:15 <alise> every dynamic scripting language, more or less
01:26:35 <cpressey> that lnguage with the same name as one of the TRON characters
01:26:45 -!- alise has quit (Quit: Leaving).
01:27:00 <Sgeo> Is the humble indie bundle server down?
01:27:02 <Sgeo> FFFFFFFFFFUUUUUUUUUUUUUUUUUUUUUu
01:27:19 <pikhq> I can totally access IPv6 Google. \o/
01:27:35 <cpressey> i think clu introduced it, or at least, claimed to have
01:27:42 <cpressey> wait, is that a TRON character?
01:28:03 <oerjan> alas it seems to require dynamic typing
01:28:16 <cpressey> b/c wp has a page "List of Tron Characters"
01:29:03 <oerjan> since either x or y can be returned, and x needs to be something treatable as a boolean
01:30:03 <cpressey> "Bit is a character from the movie Tron. Representing a bit (binary digit), it was only capable of providing yes or no answers to any question. Despite this it still managed to convey emotion and other levels of complexity."
01:30:36 <cpressey> oerjan: that doesn't require dynamic typing
01:30:54 <oerjan> well i'm sure you _could_ define a suitable haskell typeclass :D
01:30:59 <cpressey> for every type t there is a function t -> bool
01:31:23 <oerjan> but i don't think there's a common one that quite fits
01:31:41 -!- augur has quit (Ping timeout: 276 seconds).
01:31:52 <cpressey> that is kind of the problem with truthiness
01:32:02 <cpressey> i have yet to get used to the fact that in Python, [] is false
01:34:16 <oerjan> class Truthy a where truthiness :: a -> Bool
01:36:07 <oerjan> cpressey: surely that's borrowed from lisp
01:39:50 <cpressey> oerjan: possibly, but both perl and lua treat empty list(/table) as true
01:40:46 -!- augur has joined.
01:55:25 -!- sshc has quit (Ping timeout: 265 seconds).
02:00:10 -!- sshc has joined.
02:27:58 -!- augur has quit (Ping timeout: 265 seconds).
02:41:15 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
02:41:26 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
02:42:02 <cpressey> did we ever get factor onto this puppy?
02:43:22 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
02:43:27 <fungot> cpressey: i can hear you typing. i don't
02:44:10 <oerjan> cpressey: i think someone said factor was removed again for some reason
02:44:37 <cpressey> HackEgo: what is your control introducer character
02:45:36 <EgoBot> /sbin:/usr/sbin:/bin:/usr/bin
02:45:41 <EgoBot> /bin/ls: cannot access /sbin: No such file or directory
02:45:58 <oerjan> !sh ls /bin | fmt -w400 | head -1
02:46:03 <EgoBot> bash bunzip2 busybox bzcat bzcmp bzdiff bzegrep bzexe bzfgrep bzgrep bzip2 bzip2recover bzless bzmore cat chgrp chmod chown chvt cp cpio date dd df dir dmesg dnsdomainname domainname dumpkeys echo ed egrep false fgconsole fgrep fuser grep gunzip gzexe gzip hostname ip kbd_mode kill less lessecho lessfile lesskey lesspipe ln loadkeys login ls lsmod mkdir mknod mktemp more
02:48:06 <cpressey> @tell Gregor please make your bots slightly less user-hostile than OpenBSD
02:48:31 <Gregor> Doood, my bots are so not hostile at all :P
02:48:31 <lambdabot> Gregor: You have 1 new message. '/msg lambdabot @messages' to read it.
02:48:53 <Gregor> lambdabot: SHUT UP NOÖNE CARES
02:48:54 <cpressey> I was going to say "more user-friendly" but... yeah
02:49:13 <oerjan> it seems to have ignored some of cpressey's messages
02:49:52 <cpressey> see, what it could do there is say "screw you cpressey, i have no symbol table entry for this 'ping' of which you speak"
02:51:02 <EgoBot> [1,2,3] :: (Num t) => [t]
02:53:19 <EgoBot> (>>=) :: (Monad m) => m a -> (a -> m b) -> m b
02:56:06 <cpressey> what happens if bind is "overspecified" as: m a -> (a -> m a) -> m a ?
02:56:16 <cpressey> kind of like, once you pick a type, you're stuck with it?
03:01:35 <oerjan> um yeah but that wouldn't give a legal Monad instance, a and b must vary freely
03:02:08 <lambdabot> forall (m :: * -> *) a b. (Monad m) => m a -> (a -> m b) -> m b
03:04:21 -!- lament has joined.
03:04:38 <oerjan> the forall syntax is an extension for "higher-rank" types, in basic haskell it's implicitly applied in front of all type declarations
03:05:25 <oerjan> :t runST -- a function which uses a second-rank type
03:05:53 <oerjan> (for particularly magical purposes btw)
03:16:34 <oerjan> omg he's really a tree!
03:32:49 <quintopia> didn't expect much there anyway...
03:34:02 <pikhq> "320 kbps" STOP DOING THAT IT IS RAPE OF YOUR MP3 ENCODER
03:34:45 <lambdabot> cpressey said 46m 39s ago: please make your bots slightly less user-hostile than OpenBSD
03:34:58 <Gregor> Haw, no /msg for YOUUUUUUU
03:35:18 <Gregor> pikhq: #esoteric does NOT HAVE ENOUGH BOTS
03:35:33 <pikhq> Is #esoteric on the regular lambdabot join list now?
03:36:27 <oerjan> i vaguely recall a comment to effect of "no promise it will stay", so probably not
03:38:45 <cpressey> i guess i should be writin' up another, gratuitous bot to run here eh
03:39:29 <oerjan> which for obvious reasons should be called cpbot
03:48:52 -!- cpressey has changed nick to botbotbot.
03:48:58 -!- botbotbot has changed nick to cpressey.
03:51:07 -!- augur has joined.
03:58:15 <coppro> we killed commies: http://bit.ly/aP8qdB
04:03:01 -!- cpressey has changed nick to mzstorkipiwanbot.
04:03:17 -!- mzstorkipiwanbot has changed nick to cpressey.
04:03:51 -!- lambdabot has quit (Ping timeout: 272 seconds).
04:05:36 <oerjan> well i guess that means we'll soon find out if we're on the join list
04:07:27 <pikhq> oerjan: BTW: http://www.gocomics.com/calvinandhobbes/
04:08:45 -!- GreaseMonkey has joined.
04:11:21 <oerjan> huh so it actually has some net presence
04:11:23 -!- pineal_aenimal has joined.
04:11:31 <Sgeo> I always said those mutant traitors were ugly
04:12:58 <oerjan> so not quite as bad as Larson, iiuc
04:14:40 -!- aschueler has quit (Read error: Operation timed out).
04:15:03 <cpressey> the cheatsheet alise gave me for writing an ircbot isn't working
04:16:30 <cpressey> i need to have registered to join #esoteric?
04:16:55 -!- cpressey has left (?).
04:17:06 -!- carbolihy has joined.
04:17:13 -!- carbolihy has changed nick to cpressey.
04:18:52 -!- cpressey has changed nick to mzstorkipiwanbot.
04:18:58 <oerjan> pikhq: i cannot seem to find the one i mentioned before - at least "negative people" gives no hits
04:19:26 -!- mzstorkipiwanbot has changed nick to cpressey.
04:20:25 <cpressey> Gregor: I do nc to irc.freenode.net
04:20:40 <cpressey> and it tells me I'm not registered
04:20:50 <cpressey> even though I just registered too
04:21:49 <Gregor> There are steps between NICK and JOIN
04:21:53 <Gregor> Actually, before NICK IIRC.
04:22:57 -!- mzstorkipiwanbot has joined.
04:23:11 <oerjan> pikhq: i guess they have no searchable transcripts, only some tags
04:23:17 -!- mzstorkipiwanbot has quit (Client Quit).
04:23:40 <Gregor> Ahyeah, you figured that out before I did :P
04:32:33 <quintopia> how to write an IRC bot: Start with PircBot and then worry about writing the actual functionality.
04:33:42 <cpressey> but if i do it manually, it works
04:44:23 -!- wawawa has joined.
04:44:38 -!- wawawa has quit (Client Quit).
04:52:50 <cpressey> oh foo. i bet it's because stdout is buffer, eh what?
04:53:58 -!- grha has joined.
04:54:00 <oerjan> yeah i'm pretty sure you need at least line buffering
04:54:00 <grha> :oerjan: I disagree!
04:54:01 -!- grha has quit (Remote host closed the connection).
04:54:23 <oerjan> wow a run-by argument :D
04:54:59 <cpressey> It was PRIVMSGing itself ("I disagree!") in a loop so I killed it.
04:55:19 <oerjan> oh that and wawawa were your bot?
04:56:33 -!- mzstorkipiwanbot has joined.
04:56:43 <cpressey> everything up to grha was me trying to figure out why my bot wasn't working
04:56:53 <cpressey> mzstorkipiwanbot: You're my bot!
04:57:10 <cpressey> mzstorkipiwanbot: You're a very sophisticated bot.
04:57:26 -!- pineal_aenimal has quit.
04:58:32 <cpressey> also, slap me for writing it in python, but it was convenient.
05:02:17 -!- mzstorkipiwanbot has quit (Ping timeout: 276 seconds).
05:04:58 -!- mzstorkipiwanbot has joined.
05:05:41 -!- mzstorkipiwanbot has quit (Remote host closed the connection).
05:06:43 -!- mzstorkipiwanbot has joined.
05:06:48 -!- mzstorkipiwanbot has quit (Remote host closed the connection).
05:07:53 -!- mzstorkipiwanbot has joined.
05:08:39 -!- mzstorkipiwanbot has quit (Remote host closed the connection).
05:11:28 <Gregor> You might want to test that in some channel other than #esoteric 'til it works :P
05:12:17 <cpressey> Gregor: How do I send a multi-word PRIVMSG to a user?
05:12:31 <Gregor> PRIVMSG username :Stuff
05:12:47 <Gregor> That generalizes: If you start an argument with ":", that means the argument continues to the end of the line.
05:14:04 -!- mzstorkipiwanbot has joined.
05:14:30 <cpressey> mzstorkipiwanbot: Hi. I see you registered with NickServ! Good for you!
05:18:11 <cpressey> you know mzstorkipiwanbot, i'm a-gonna teach you to execute scheme. and feel pain
05:18:38 <oerjan> mzstorkipiwanbot: oh dear your creator is a megalomaniac!
05:19:15 <oerjan> well it's your problem
05:19:23 <Sgeo> mzstorkipiwanbot, I should not be crowned King of the Universe.
05:19:32 <Sgeo> mzstorkipiwanbot: I should not be crowned King of the Universe.
05:23:50 -!- mzstorkipiwanbot has quit (Ping timeout: 250 seconds).
05:24:21 <cpressey> so I need to respond to pings or something?
05:25:50 <lament> that's what it means to feel pain, for an irc bot
05:26:27 <GreaseMonkey> now say that in toki pona (or whatever it's called)
05:27:47 <pikhq> GreaseMonkey: Toki Pona.
05:28:12 * pikhq should devise a kanji mapping scheme for that
05:28:15 <GreaseMonkey> <lament> that's what it means to feel pain, for an irc bot <-- !translate english toki-pona
05:29:47 <pikhq> Thus creating an overly complex orthography for an overly simple language.
05:30:11 <pikhq> Okay, so it'd be much less complex when there's only 118 characters. Still.
05:30:53 <lament> ilo nanpa toki pi tomo toki li pilin ike kepeken ni.
05:32:15 -!- wareya_ has joined.
05:34:43 -!- wareya has quit (Ping timeout: 240 seconds).
05:37:19 -!- augur has quit (Ping timeout: 252 seconds).
05:41:28 <cpressey> i guess I could just say PONG :unknown and it accepts it
05:41:45 <cpressey> but i went through the whole rigamarole of figuring out what my hostname is anyway
05:46:37 <GreaseMonkey> i have a feeling you're supposed to just bounce the message back with one letter changed
05:48:37 <cpressey> so I should say PONG :gogol.freenode.net? that seems... ingracious
05:48:37 <GreaseMonkey> although if it doesn't accept what you bang back in whatever valid format, it's probably a screwed up server
05:48:55 <GreaseMonkey> you should take the message, change the PING to PONG, and send it back
05:49:37 <cpressey> experimentation indicates it doesn't care
05:54:48 -!- augur has joined.
06:01:22 -!- antivigilante_ has joined.
06:17:11 -!- mzstorkipiwanbot has joined.
06:18:01 -!- sftp has quit (Remote host closed the connection).
06:22:04 <pikhq> Why don't the TLDs have A records?
06:22:33 <pikhq> Rather, why don't they *all* have A records?
06:22:38 <coppro> because no website is "com"?
06:23:12 <cpressey> typing http://com/ into firefox takes me to cnet
06:23:36 <cpressey> http://www.cnet.com/ to be precise
06:23:41 <pikhq> Probably taking you to com.com.
06:23:42 <cpressey> i wonder how they finagled that
06:24:23 <pikhq> And com has no A record, so Firefox tries adding .com.
06:24:58 <pikhq> But seriously, why not just have the A record point to the site of the operator of the TLD?
06:25:35 <pikhq> (and the AAAA record, of course)
06:25:55 -!- Quadrescence has quit (Ping timeout: 240 seconds).
06:26:26 -!- oerjan has quit (Quit: Later).
06:27:06 <pikhq> Hmm. http://例え.テスト/ is a real thing.
06:29:28 <cpressey> indeed yes. running mediawiki!
06:30:41 <pikhq> Hooray, truly internationalised TLDs.
06:31:22 <cpressey> the mix of LtoR and RtoL in the URL is disconcerting
06:32:40 <pikhq> Mixed text is a bit disconcerting.
06:32:50 <pikhq> But, it's accepted practice for RtoL scripts, so...
06:33:07 <pikhq> Huh. Antarctica has a TLD.
06:38:41 <Ilari> aq. 86400 IN SOA ns1.dns.aq. noc.swizzle.co.nz. 2010100201 28000 3600 604800 86400
06:40:12 <Vorpal> I'd like a zone transfer on aq!
06:44:51 -!- Quadrescence has joined.
06:49:05 <Ilari> BE BG BIZ BR CAT CH CZ DK EDU EU FR INFO LC LI LK MUSEUM NA NU ORG PM PR PT RE SE TF TH TM UK US VC YT
06:51:54 -!- sftp has joined.
06:53:04 <Ilari> (the TLDs that have DS records in root zone, no AQ in there...) :-/
06:53:43 -!- Guest89141 has joined.
06:54:03 <Ilari> (actually, non IDN ones...)
06:54:13 <coppro> wootwoot, I proved a hard proof for marks
06:54:31 <coppro> and as I sort of expected, it's really elegant in the end :)
07:00:12 -!- antivigilante has quit (Ping timeout: 240 seconds).
07:00:50 -!- antivigilante_ has quit (Ping timeout: 272 seconds).
07:06:46 -!- tombom has joined.
07:14:27 -!- antivigilante_ has joined.
07:19:52 -!- antivigilante_ has quit (Ping timeout: 264 seconds).
07:33:28 -!- antivigilante_ has joined.
07:52:33 -!- tombom has quit (Quit: Leaving).
07:55:13 -!- lament has quit (Ping timeout: 240 seconds).
07:58:17 -!- Sgeo has quit (Ping timeout: 276 seconds).
07:59:26 -!- FireFly has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:03:06 -!- mzstorkipiwanbot has quit (Ping timeout: 252 seconds).
08:05:14 -!- cpressey has quit (Ping timeout: 260 seconds).
08:11:03 <fizzie> "Cat Head Detection - How to Effectively Exploit Shape and Texture Features" -- http://research.microsoft.com/en-us/um/people/jiansun/papers/ECCV08_CatDetection.pdf
08:11:12 <fizzie> Microsoft Research: bleeding-edge cat detection.
08:11:34 <fizzie> "Second, people love cats. A large amount of cat images have been uploaded and shared on the web. For example, 2,594,329 cat images had been manually annotated in flickr.com by users. Cat photos are among the most popular animal photos on the internet."
08:19:30 <Quadlex> I'm pretty sure felix domesticus is native to the internet
08:33:40 -!- antivigilante_ has quit (Ping timeout: 264 seconds).
08:59:53 -!- ais523 has joined.
09:08:53 -!- ais523 has quit (Read error: Connection reset by peer).
09:14:07 -!- ais523 has joined.
09:32:31 -!- antivigilante has joined.
09:32:34 -!- antivigilante_ has joined.
09:44:43 -!- Quadrescence has quit (Ping timeout: 240 seconds).
10:00:57 -!- Quadrescence has joined.
10:36:43 -!- augur has quit (Remote host closed the connection).
10:37:15 -!- augur has joined.
11:04:20 -!- GreaseMonkey has quit (Quit: Welcome honored guest. I got the key you want! would you like onderves. of Yourself).
11:23:41 <fungot> ais523: i found something which _might_ work on win98? dunno... :)
11:24:05 <fungot> ais523: lea is a dirty open source hippie commies? yes you can
11:25:36 <ais523> hmm, IRC's working fine, but the DNS seems to have gone down
11:39:36 -!- antivigilante__ has joined.
11:39:44 -!- antivigilante_ has quit (Quit: Ex-Chat).
12:14:06 -!- FireFly has quit (Quit: swatted to death).
12:18:05 -!- FireFly has joined.
12:38:26 -!- quintipod has joined.
12:48:02 -!- BeholdMyGlory has joined.
12:48:05 <ais523> any news? anything up-and-coming in the world of esolanging?
12:48:20 <ais523> I think I invented a new esolang last night, I may have to tweak it if it turns out to be sub-TC (although I think it's TC)
12:48:50 <quintipod> Esolanging? Nahhh. I am going caving and camping starting this afternoon and lasting all weekend though
12:49:12 <ais523> Slereah: a cross between http://esolangs.org/wiki/Sansism and http://esolangs.org/wiki/1L
12:56:32 <quintipod> My favorite esolang right now is Minecraft
13:00:28 -!- Slereah has quit (Ping timeout: 265 seconds).
13:05:25 -!- Slereah has joined.
13:07:33 -!- quintipod has quit (Quit: Busy busy busy).
13:12:18 <ais523> hmm, I was just looking over my old esoprograms
13:12:47 <ais523> I'm amused by the "tenloop" in Unassignable, that exists to make a single decimal digit
13:13:35 <ais523> it's implemented as a three-bit number, and a four-bit number; the three-bit number adds/subtracts 6 to the four-bit number when it overflows/underflows, the four-bit number subtracts/adds 2 to the three-bit number when it overflows/underflows
13:28:19 <ais523> hmm, who's 79.75.203.167? they wrote a P'' interp in INTERCAL, to prove it TC
13:28:23 <ais523> I'm wondering if it's someone in this channel
13:28:40 <ais523> in fact, I was wondering whether it was me to start with, but that isn't my IP
13:29:01 <ais523> also, I probably wouldn't have used CLC-INTERCAL
13:30:10 -!- alise has joined.
13:30:28 <alise> Dreaming is illegal
13:48:51 <alise> "Worcestershire sauce, the popular English sauce, is made from dissolved anchovies. The anchovies are soaked in vinegar until they have completely melted. The sauce contains the bones and all."
13:48:54 <alise> did not know; did not want to know
13:51:38 <Vorpal> hm, why is the heart often associated with love? I mean, biologically that is nonsense as far as I know...
13:52:43 <Vorpal> if it is just a leftover from before people knew that (which seems probable), then two new questions arise:
13:52:52 <Vorpal> 1) why would anyone think it was related to love in the first place
13:52:56 <alise> hey, you deduced the obvious and called it probable, congrats
13:53:00 <alise> Vorpal: because you think with your heart.
13:53:16 <alise> (no, you don't think with your brain, that just coordinates the body a bit, why would you think that?)
13:53:20 <Vorpal> 2) why is it still used as a symbol for love
13:53:23 <alise> (the heart is the main part of the body, you think with it)
13:53:36 <alise> 2) why do we still celebrate halloween?
13:53:54 <Vorpal> alise, well, I don't know. We don't where I live :P
13:53:56 <alise> why do we still use elements of things that we as a whole reject
13:53:56 <ais523> for an entirely different reason than we originally celebrated halloween
13:54:06 <alise> precisely, and we don't *really* think the heart loves
13:54:09 <ais523> more or less like Christmas has changed in meaning over the years
13:54:15 <alise> also, hearts don't look like <3!
13:54:24 <ais523> nowadays, it's just an indication of trick-or-treating and NetHack tournaments
13:54:43 <alise> The heart has long been used as a symbol to refer to the spiritual, emotional, moral, and in the past also intellectual core of a human being. As the heart was once widely believed to be the seat of the human mind, the word heart continues to be used poetically to refer to the soul, and stylized depictions of hearts are used as prevalent symbols representing love.
13:54:59 <Vorpal> alise, why didn't love end up getting associated with, say, the liver?
13:55:05 <alise> because you don't think with your liver
13:55:05 <Vorpal> it seems equally random
13:55:07 <alise> your liver just sits tehre
13:55:16 <alise> you can hear your heart going, it's near where you talk and stuff
13:55:34 <Vorpal> okay that seems like a plausible logic for it
13:55:52 <alise> The Roman physician Galen located the seat of the passions in the liver, the seat of reason in the brain, and considered the heart to be the seat of the emotions. While Galen's identification of the heart with emotion were proposed as a part of his theory of the circulatory system, the heart has continued to be used as a symbolic source of human emotions even after the rejection of such beliefs.[2]
13:55:56 <alise> "seat of passions in the liver"
13:56:04 <alise> Vorpal: i reject your question, it makes a false assumption!
13:56:06 <Vorpal> alise, I swear I didn't know about that one :P
13:56:10 <alise> we DID think that! or at least, something relatively close to that
13:56:12 <alise> lust not love i guess
13:56:22 <Vorpal> alise, anyone thought it was in the spleen?
13:56:39 <alise> (but no, i don't think anyone did, historically)
13:56:48 <alise> but then I didn't think anyone would place passion inside the liver, either
13:57:19 <Vorpal> I just picked a random organ in the torso above
13:58:32 <Vorpal> alise, btw, what did they think the brain was for back when they thought you used the heart to think with?
13:58:51 <alise> not much. like the liver
13:59:08 <alise> maybe it just relays stuff. if they even had a concept of signals being sent around then
13:59:13 <alise> or maybe it just regulates some random thing or another
14:00:15 <ais523> hmm, this phishing 419 spam (claiming to be from Benin, rather than Nigeria) is asking for the answers to a list of questions I don't even understand
14:00:41 <ais523> they want "Your. " Receiver, Country, City, Tel, Test question, Answer, and Passport
14:01:47 <alise> sounds Benin to me
14:02:03 <ais523> I mean, how do I send my passport via email?
14:02:08 <ais523> how do I know what the question is?
14:02:10 <Vorpal> ais523, put it in the floppy drive?
14:02:34 <ais523> maybe I'll photograph it on a wooden table, screenshot Photoshop with the photo open, paste it into Word and send that
14:02:38 <alise> This keyboard has surprisingly good tactile response for Logitech.
14:02:41 <alise> Although I'm still getting used to it...
14:02:44 <alise> And it's a bit loud :)
14:02:56 <Vorpal> ais523, as for "Receiver", they presumably want a scan of your TV receiver antenna or such?
14:02:59 <ais523> (where by "maybe" I mean "there's no chance that")
14:03:22 <ais523> presumably they're going to ignore the answers anyway, they're just looking for someone who responds to 419 scams
14:03:40 <alise> ais523: no love for "whereby"?
14:03:56 <alise> kind-of, where by, you are strange :)
14:04:07 <ais523> "whereby" means something else, doesn't it?
14:04:09 <Vorpal> ais523, because anyone _still_ doing that would have to be faking it?
14:04:17 <ais523> Vorpal: you'd be surprised
14:04:24 <alise> ais523: whereby = by which
14:04:43 <alise> definitely, whereby makes sense ehre
14:04:52 <ais523> a couple of days ago (I think when clog was down, not sure) I commented on a reddit commenter who said that when parked domains full of ads were visited, they had a clickthrough rate of above 50% as there was nothing else to click on
14:04:55 <alise> Other heads saw devolution as a whole new way of life and adopted an approach whereby the power of devolution was used to enable the school to drive the ...
14:04:56 <alise> en.wikipedia.org/wiki/wikt:whereby - Cached
14:04:59 <alise> x was used to enable y
14:05:04 <ais523> because most people didn't realise it wasn't the site they wanted to visit
14:05:04 <alise> x was used to mean y
14:05:08 <alise> maybe you're right.
14:05:18 <alise> I think "where by" is just as incorrect, though, if it is indeed incorrect
14:05:35 <alise> ais523: hasn't hit me yet
14:05:37 <ais523> (I know that a while back, when a quirk in Google made a newspaper story the top search result for Facebook, several people tried to log into Facebook via its comment form)
14:05:44 <Vorpal> <alise> en.wikipedia.org/wiki/wikt:whereby - Cached <-- wikt: ?
14:06:08 <Vorpal> isn't that a separate domain?
14:08:05 <Vorpal> ais523, how many is several? Also, I wonder how bad this is and how much is selective reporting. There has to be lots of cases where some google result ordering quirk did *not* cause similar effects.
14:08:25 <Vorpal> and presumably the majority didn't get confused
14:08:43 <Vorpal> or at least if they did, didn't try to log in using the comment system
14:09:09 <fizzie> "Several" was rather surprisingly many, if I remember the case right. Certainly not the majority, though.
14:09:20 <Vorpal> probably facebook is just large enough that on average you will get a handful of morons. And handful will be rather large.
14:09:31 <Vorpal> (large in absolute numbers I mean)
14:10:00 <Vorpal> are there any public figures on how many users facebook have?
14:10:09 -!- Guest89141 has quit (Read error: Operation timed out).
14:10:14 <alise> Vorpal: millions and millions.
14:10:28 <alise> 500 million ACTIVE users, it seems
14:10:39 <alise> "which is about one person for every fourteen in the world" --Wikipedia
14:10:42 <Vorpal> how do they count active? last 30 days? last 60?
14:10:51 <alise> but whatever it is, it's impressive
14:10:58 <fizzie> "When hundreds of clueless commenters decided mid-February that ReadWriteWeb was the place to log in to Facebook, --" so several is at least hundreds.
14:11:00 <alise> (1) thanks for the division, Wikipedia; (2) no sorry that actually is a useful statistic I just had to make that joke
14:11:53 <Vorpal> >>> (500 / 500000000.0)*100
14:11:53 <Vorpal> 9.9999999999999991e-05
14:12:17 <Vorpal> that is assuming hundreds = 500
14:12:50 <ais523> "To accomplish this, Zuckerberg hacked into the protected areas of Harvard's computer network and copied the houses' private dormitory ID images."
14:13:22 <alise> Facebook, dedicated to your privacy since day one.
14:13:39 <ais523> how the predecessor to Facebook got its info
14:13:41 <alise> Vorpal: Zuckerberg founded Facebook
14:13:48 <ais523> that site had to close down, and Facebook was founded in its place
14:13:50 -!- augur has quit (Ping timeout: 255 seconds).
14:14:02 <Vorpal> I'm not surprised it had to close down
14:14:06 <alise> ais523: Kind of scary that Harvard's network is vulnerable like that...
14:14:20 <fizzie> Vorpal: Though "hundreds" might just be the amount who actually said something. I mean, the case was that the article's commentary page had a "sign in with Facebook" thing, and people signed in, then wrote stuff to the "comments" box; presumably there are at least some who logged in and actually realized "hey, this is not facebook".
14:14:41 -!- augur has joined.
14:14:57 <Vorpal> fizzie, would "sign in with facebook" be openid or?
14:15:36 <fizzie> This was early 2010, I think Facebook's OpenID joining is a later thing? I don't really know how they do inter-a-graterion.
14:15:52 <Vorpal> also python's floating point rounding for output sucks badly
14:16:05 <ais523> Vorpal: it's Facebook Connect; vaguely like OpenID, except limited to Facebook
14:16:10 <ais523> this is, ofc, a bad idea
14:16:12 <ais523> but nobody seems to care
14:16:19 <Vorpal> ais523, hm. Does facebook provide openid as well or?
14:16:35 <Vorpal> so why roll their own system
14:16:40 <ais523> could you make Facebook Connect into an OpenID provider?
14:16:46 <alise> more people sign up for facebook to use facebook connect.
14:16:47 <ais523> or would that violate Facebook's rules?
14:16:50 <alise> more people think "facebook" more often
14:16:52 <ais523> (it's probably technically possible)
14:16:57 <alise> more people think "yes -- i am using my facebook" when using it
14:17:08 <alise> ais523: when Facebook Connect first came out, I decided I *really* hated Facebook for killing OpenID
14:17:20 <alise> you can probably make it into an openid provider
14:17:22 <Vorpal> ais523, do you use any sort of openid btw?
14:17:23 <ais523> alise: I hate Google Accounts just as much
14:17:25 <alise> but why would you want to?
14:17:32 <alise> ais523: google accounts are exposed as openIDs, at least
14:17:46 -!- jcp has joined.
14:17:49 <alise> ais523: which makes it more acceptable
14:18:08 <alise> (and you can't use google accounts on a non-App Engine site without using the OpenID solution, so that's promotion in a sense)
14:18:14 <alise> whereas facebook connect is an outright competitor to openid
14:18:29 <ais523> alise: my issue is the reverse, there's sites that allow Google accounts to log in, but not OpenID
14:18:37 <ais523> so you have to create a Google account to log in there
14:18:45 <fizzie> Twitter can be used as a Facebook Connect/OpenID -like thing, too; the ReadWriteWeb comment page currently lets you login via Facebook Connect, Twitter, or any OpenID provider.
14:18:47 <alise> ais523: well, only (1) Google sites (and this is not too surprising), and (2) sites on App Engine
14:19:01 <alise> (2) is unfortunate, but you can hardly expect Google not to provide an interface to their accounts API from one of their big services,
14:19:03 <ais523> alise: (1) isn't too surprising, but still annoying
14:19:08 <alise> and (1) is understandable if imperfect
14:19:17 <alise> it's sure as hell a lot better than facebook
14:19:30 <ais523> because if I log in to use Google Groups, I have to log out again and clear cookies to use Google search, if I want results that are relatively neutral
14:19:35 <Vorpal> openid is nice in theory, but there are a number of issues with it 1) trusting the provider you use to not abuse it and run a secure system 2) trust them not to suddenly go bankrupt
14:19:41 <ais523> rather than being attemptedly tailored to my interests
14:19:41 <Vorpal> okay you could run your own openid server
14:19:44 <Vorpal> but that is a lot of work
14:20:25 <alise> Vorpal: it is not a lot of work
14:20:29 <alise> you copy like two files and edit one file
14:20:36 <alise> there's no "server"
14:20:42 <alise> also, there are solutions to (1) and (2)
14:20:48 <alise> you can put headers on your personal web page
14:20:53 <alise> that point to an openid provider
14:20:56 <alise> then use your web page as an openid
14:20:59 <alise> swapping providers at will
14:21:12 <Vorpal> encrypted local keyring is my preferred solution, of course if you often use public computers and such that wouldn't be very convenient, but in any case you need to trust those systems not to have keyloggers installed, so they are a bad solution in any case.
14:21:25 <Vorpal> and you have to ensure backup and such of course
14:22:09 <alise> ais523: I actually wrote a thing to expose Google Accounts to non-App Engine websites in 2008. Apologies, but, in my defence, it was intended solely to implement an OpenID provider with, before Google offered OpenID.
14:22:23 <fizzie> Facebook's a sponsoring member of the OpenID Foundation, anyway, so you see, they're helping in their way.
14:22:27 <Vorpal> alise, I remember that
14:22:28 <alise> (gaccproxy.appspot.com; in my defence, that's a perfectly acceptable non-HTTP-exclusive domain name!)
14:22:29 <ais523> as long as you didn't force people to use it, I don't mind
14:22:53 <alise> i posted it on reddit and argued with people who called it the END OF SECURITY and things like that but i don't think anyone actually used it
14:22:58 <alise> also, the example site is long-dead now
14:23:03 <alise> it ran on eso-std.org
14:23:16 <alise> (just to prove it works :P)
14:23:54 <alise> the code is a little ugly iirc
14:24:05 <alise> but contains enough random numbers and verification to be fun
14:24:08 <Vorpal> ais523, don't those sites also allow creating an account and using username/password?
14:24:19 <Vorpal> ais523, like the "conventional" solution to login
14:24:27 <alise> Vorpal: instead of FB Connect?
14:24:46 <Vorpal> alise, instead of fb connect/google accounts/openid
14:24:52 <alise> <fizzie> Facebook's a sponsoring member of the OpenID Foundation, anyway, so you see, they're helping in their way.
14:25:12 <alise> I forget how ITV is structured, but the BBC at least used to partly own it
14:25:33 <alise> Vorpal: google accounts -- not really, sites that use it are on app engine, and what's the point if you have to write auth code too?
14:25:39 <alise> FB connect -- most but not all
14:25:54 <alise> openid -- most offer registration, but only because OpenID is sadly neglected
14:25:58 <Vorpal> alise, remember that thing about gmail accounts getting hacked. China was involved iirc. Now, I'm sure the risk is extremely slim for that, but not even google is 100% secure, nothing is.
14:26:01 <alise> Jyte doesn't but then it's a Jan Rain site
14:26:22 <alise> Vorpal: did anyone actually connect that to China?
14:26:30 <ais523> hmm, my Firefox was going crazy then
14:26:42 <alise> and indeed, nothing is really secure; you are far too paranoid because there are weak links far before you
14:26:46 <ais523> it wasn't responding to any input, which often happens; /but/ if I resized the window, it redrew everything accordingly
14:26:54 <alise> such as your bank, say.
14:27:11 <Vorpal> alise, the IPs were from there iirc, and the accounts belonged to people the regime didn't like.
14:27:22 <Vorpal> so um, not connect for certain
14:27:32 <alise> ais523: anyone remember when resizing Netscape used to redraw everything? 'cause I don't
14:28:09 <ais523> alise: I remember that with pre-Firefox Mozilla
14:28:10 <Vorpal> alise, I remember that
14:28:26 <Vorpal> pretty sure navigator 3 did that too?
14:30:14 <Vorpal> alise, and indeed the bank isn't completely secure... It certainly worries me.
14:30:39 <alise> Vorpal: you do realise that you're so boring, nobody would ever want to compromise your security?
14:31:31 <Vorpal> alise, stop trolling, it's just pathetic :P
14:33:26 <Vorpal> alise, still banks use security tokens and such. Reasonably secure.
14:35:12 -!- MigoMipo has joined.
14:39:44 <ais523> alise: I rarely bother with really strong levels of security, except when I'm safeguarding information for someone else
14:39:56 <ais523> in which case I expend effort to not be the weakest link in the chain
14:40:10 <alise> I still use the same password everywhere, which I *really* want to change.
14:40:16 <alise> But it's such a huge undertaking.
14:41:05 <Vorpal> alise, a good keychain program + master password? then you can stop using it for new ones and also can change old ones as you run across them?
14:41:39 <alise> I just need to figure out how to get one that works on a mobile
14:43:49 <alise> maybe. i've been meaning to replace it.
14:44:21 <alise> i have this wonderful plan for a website that can do it all seamlessly, on just about any device, including public computers, and yet, is still totally trustable
14:44:30 <alise> (i.e. you don't have to trust the website at all)
14:45:06 <alise> there's pretty much only one way I could be evil and I was planning to have a browser extension that automatically notifies you whenever the code changes and checks to see if reputable people have said it's fine before continuing
14:45:19 <ais523> alise: does it just serve client-side JS, or something?
14:45:21 -!- cpressey_ has joined.
14:45:56 <alise> ais523: I don't particularly want to reveal the whole design in public, even though nobody else would bother building it, because it's the best "mainstream" idea I've had yet, even if I can't immediately think how to make money off it
14:45:57 <fizzie> For website passwords, I've been using a Maemo PasswordSafe port on the phone; that's not too bad, I guess.
14:46:02 <alise> ais523: but basically, all the sensitive stuff is done entirely client-side
14:46:07 <alise> and the server doesn't store anything I can read
14:46:12 <alise> there's more subtlety to it than that
14:46:31 -!- mzstorkipiwanbot has joined.
14:46:40 <ais523> the way to make money off it is to get lots of users, and then sell it before the buyer notices you don't have a business model
14:46:49 <ais523> then they add ads and it collapses, but you still have the money
14:47:09 <Vorpal> alise, nice idea but limited to high end phones really.
14:47:35 <alise> Vorpal: well, it would work on any semi-modern phone
14:47:39 <ais523> low-end phones can't log into websites at all
14:47:40 <alise> probably even blackberries
14:47:48 <alise> but i don't use opera mini
14:47:54 <alise> also, it was mainly for browser usage, it just happened to work on phones too
14:48:02 <alise> Vorpal: besides, creating a java version or whatever wouldn't be too hard
14:48:05 <Vorpal> indeed, I was just disputing the claim about "just about any device"
14:48:12 <Vorpal> alise, there are lots of lower end/older phones
14:48:12 <alise> i meant, cybercafes too
14:48:17 <ais523> wow, Facebook just added/are in the process of adding an option to export data
14:48:19 <alise> yes, but cool people don't own them.
14:48:25 <alise> or rather, cool people don't internet from them :P
14:48:40 <ais523> that's... rather out-of-character for them
14:48:58 <Vorpal> alise, indeed, I wouldn't log in on anything with it. Only internet stuff I do on it is read news and check bus schedule
14:49:18 <Vorpal> and news only when waiting for bus or such
14:49:40 <Vorpal> I think the only times I'm bored is when I'm waiting for the bus...
14:49:51 <alise> ais523: what progress!
14:49:55 <alise> access to your own data
14:50:07 <ais523> I just don't see the motive here
14:50:13 <ais523> trying to dodge antitrust concerns, maybe?
14:50:17 <ais523> trying to actually become less evil?
14:50:22 <alise> maybe they've found love.
14:50:37 <cpressey_> they claim it is all secure and protected after you download it
14:50:45 <fizzie> alise: The official motivation: "It's our core belief that people should own and be able to conrol their information in Facebook," said CEO Mark Zuckerberg. "We view this as a philosophical thing."
14:50:57 <fizzie> See, it's philosophical.
14:51:04 <Vorpal> alise, we need to find a way to write a liver now. A bit tricky hm
14:51:09 <alise> fizzie: i really need a good onomatopoeium for "snrk"
14:51:22 <alise> the sort of half-nose, half-above-mouth outwards sharp release of breath
14:52:30 <cpressey_> mzstorkipiwanbot: lambdabot is gone. you know this means i'm gonna have to teach you to be our messenger service.
14:52:43 <fizzie> You have a very disagreaable bot there.
14:53:05 <fizzie> fungot: Quick, learn to do memo-passing so that these new upstarts don't steal your place as the channel's most important person!
14:53:05 <fungot> fizzie: you could also use " define" at http://www.common-lisp.net/ paste/ display/ fnord
14:53:06 <cpressey_> it's quite the contrarian little bastard, yes
14:53:24 <alise> nooo, lambdabot went
14:53:46 <alise> mzstorkipiwanbot: fort
14:53:57 <ais523> who is mzstorkipiwanbot?
14:54:06 <alise> there's no a in it
14:54:08 <alise> it sounds like snrk
14:54:14 <alise> mzstorkipiwanbot: x
14:54:14 <alise> mzstorkipiwanbot: x
14:54:14 <alise> mzstorkipiwanbot: x
14:54:29 <alise> cheater: can I hire you as an AI consultant? :|
14:54:30 <ais523> is its only purpose to disagree with things?
14:54:33 <alise> ais523: apparently
14:54:34 <ais523> that seems kind-of pointless
14:54:41 <Vorpal> mzstorkipiwanbot, do you agree that you disagree with everything?
14:54:49 <Vorpal> mzstorkipiwanbot: do you agree that you disagree with everything?
14:54:58 <alise> cheater: happiness; the latest release of vagrant before maybe three other people
14:55:25 <Vorpal> cpressey_, you might want to add , to the list of things that goes after the nick
14:55:51 <alise> or just make it reply to it whenever its name is mentioned >:)
14:56:11 <Vorpal> alise, could get annoying, What if it happened when talking about something else? ;)
14:56:25 <alise> python really needs goto
14:56:59 <cheater> def f0001(): blah blah; f0002()
14:57:09 <cheater> def f0002(): blah blaaa; f0003()
14:57:25 <alise> cheater: that doesn't let me break out of a while loop
14:57:30 <fizzie> Vorpal: Yeah, at the very least it should reply only if it sees its nick with word boundaries around; otherwise all the words that just happen to contain mzstorkipiwanbot as a substring would get replies.
14:57:31 <alise> also it makes me write global
14:57:34 <cheater> de f0003(): if(blargh): f0001()
14:57:47 <Vorpal> cheater, but python doesn't optimise tail calls iirc?
14:57:57 <alise> Vorpal: you could write a trampoline
14:57:58 <alise> still, not the same
14:57:59 <cheater> alise: mine has less spaces.
14:58:04 <alise> cheater: yours has more bytse
14:58:10 <fizzie> Vorpal: Python also has a recursion depth limit of 1000 by default.
14:58:36 <cheater> fizzie: oh cool, didn't know that
14:58:50 <cheater> fizzie: what about nesting depth?
14:58:51 <Vorpal> so cheater's solution doesn't work
14:59:23 <Vorpal> cheater, aka: doesn't work
14:59:30 <cheater> Vorpal: it's not like we have a working model of a turing machine, so this conversation is moot
14:59:32 <fizzie> The recursion limit is runtime-configurable, though: http://docs.python.org/library/sys.html#sys.setrecursionlimit -- there's a hard limit somewhere, of course.
15:00:06 <cheater> Vorpal: because why would you talk about recursion limits outside of the interesting case of turing machines?
15:00:19 <Vorpal> alise, as for breaking out of loops, doesn't python have break [n] ?
15:00:35 <alise> it doesn't have break [n]
15:00:40 <alise> but you can't do that inside a function, duh
15:00:45 <alise> f() can't break out
15:00:48 <alise> you could raise an exception...
15:00:57 <alise> because mine sucks up the exception errors
15:01:01 <cheater> i was just going to say that
15:01:03 <alise> it's longer than my current code
15:01:12 <alise> a=(k in'lun')-(k in'hyb');b=(k in'jbn')-(k in'kyu')
15:01:13 <alise> X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y)
15:01:16 <alise> ^ these two lines are totally my favourites
15:01:33 <Vorpal> cheater, well, imagine a main loop of an httpd coded in python implementing in terms of your solution
15:01:38 <Vorpal> so, doesn't really work
15:01:52 <Vorpal> and you will always have to return up the chain in the end
15:02:00 <cheater> Vorpal: i don't care for http
15:02:02 <Vorpal> unless you exit the program
15:02:12 <Vorpal> cheater, same goes for any sort of server though
15:02:16 <Vorpal> httpd was just an example
15:02:29 * Vorpal waits for cheater to claim the concept of "server" is failed too
15:03:13 <cheater> talking to vorpal is a failed concept
15:06:35 <Vorpal> alise, btw read the annotation on iwc today
15:07:49 <ais523> claiming that http has failed is quite a bold statement...
15:08:01 <ais523> did I misinterpret you?
15:09:23 <Vorpal> ais523, long live gopher. Err.... <awkward pause> Anyone here a high level cleric?
15:10:20 <alise> cheater: so do you want to do AI or not? :P
15:12:41 <cpressey_> ais523: if its goal was to free the elephants, it has indeed failed
15:12:49 <alise> cheater: But those poor Qs.
15:13:03 <ais523> cpressey_: good thing it found a second job in serving Web pages, then
15:13:07 <ais523> it's amazing what you can repurpose some things to
15:13:15 <cheater> alise: will you keep throwing my code away?
15:13:25 <alise> cheater: only if it's really terrible
15:13:31 <fizzie> Incidentally, in the "new packages" list for my phone there's the Gophernicus Gopher server. I don't really know why someone bothered to package *that*.
15:13:32 <alise> or if I think of something even better
15:13:37 <cheater> alise: there was nothing terrible!
15:13:44 <alise> i am stating my future policy :P
15:14:13 <fizzie> Vorpal: The package homepage URL is a gopher:// thing. :p
15:14:32 <alise> undesired implications
15:14:39 <Vorpal> fizzie, well I'm not surprised
15:14:57 <cheater> not sure what "sunshine" implies
15:15:06 <alise> if r(0,14):Q('Yum! That was delicious.');L+=int(0==r(0,2)and r(5,10))
15:15:08 <alise> why do I have int() there
15:15:39 <alise> hey i can quaff indefinitely
15:15:44 <Vorpal> cheater, I think alise is better described by "messoscale thunderstorm in snow blizzard"
15:15:57 <Vorpal> rather than "sunshine"
15:15:59 <alise> IT'S FUNNY BECAUSE VORPAL DOESN'T LIKE ME
15:16:00 <cheater> alise: you can remove int()
15:16:28 <cheater> alise: i told you talking to Vorpal is a failed concept
15:17:25 <Vorpal> hm, golfing English, that might be fun
15:17:31 <Vorpal> I wonder how well it would work though
15:19:52 <alise> cheater: just getting this version a bit more polished
15:20:11 <cheater> alise: pls no heritage slurs
15:20:45 <cpressey_> a snow blizzard. as opposed to a cake blizzard.
15:21:01 * cheater wonders what it would mean to "english" a code.
15:21:22 <cpressey_> put spin on it so it bounces off funny?
15:21:23 * alise adds a NEW FEATURE
15:21:39 <cheater> alise: what feature would that be?
15:21:44 <alise> if k=='\n':D();continue
15:21:48 <alise> without costing a turn
15:22:02 <cheater> alise: you totally should set up bzr for this thing
15:22:13 <cheater> that's why you should use it
15:22:19 <cheater> so as not to hate it afterwards
15:22:26 <alise> i know i hate it for a fact.
15:22:35 <alise> anyway it's too small to version-control really. that would destroy the purity and fun.
15:22:51 <Vorpal> alise, visual sourcesafe!
15:23:09 <alise> wow, dying is broken
15:23:13 <alise> you just get informed that you die, every turn
15:23:24 <alise> and continue on with zero satiation and zero hp
15:23:33 <alise> if L<1:Q('You die...',1);D()
15:23:36 <alise> ah. that may be the issue
15:24:28 <cheater> alise: well we need to be able to work on it concurrently
15:24:39 <cheater> alise: what about a screen with vim
15:24:42 <cpressey_> pastebin and clog are your VCS for this thing
15:24:48 <alise> not really, i'm find merging in your changes when i tweak stuff
15:24:53 <cheater> alise: give me axs 2 ur shell acnt
15:25:08 <alise> i trust exactly one person in the world with my shell and it's probably ais523
15:25:24 <cheater> that's ok, ais523 trusts me
15:25:28 <ais523> heh, I like the idea that you know how many people you trust, but aren't sure who they are
15:25:40 <ais523> and no, I don't trust you, not to the extent of giving you access to someone else's trust indirectly
15:25:59 <cheater> said ais523, in fact trusting me.
15:26:13 <alise> cheater: http://pastie.org/1205374.txt?key=rxu2wgg3efutkk6cfdpryw
15:26:19 <alise> i've denoted where the AI needs to go
15:26:24 <alise> you can drop the ,w; from the global list
15:26:25 <alise> if you don't assign to w
15:28:41 -!- Sgeo has joined.
15:29:22 <alise> get Qs working first :P
15:29:56 <alise> cheater: this is why i didn't use your code :P
15:33:33 <quintopia> cpressey_: please go away and drop the _ and come back so that you'll be your normal color again! alise and ais523 are already orange!
15:33:45 <quintopia> (yay pathological cases of hashing nicks to colors...)
15:34:04 <ais523> hashing IPs, or cloaks, would be more useful, relaly
15:34:57 <quintopia> but being able to manually swap people's colors would be most useful. this script doesn't do that and i'm too lazy to add that feature...
15:35:45 <cheater> alise: what about adding shooting rays?
15:36:18 -!- cpressey has joined.
15:36:38 <cheater> (i can't really work on the code until tomorrow)
15:36:41 <alise> http://pastie.org/private/gaqwdbng4srklx1adb2o6g your code here doesn't work btw
15:36:49 <quintopia> alise: what about replacing # walls with "|" and "-" walls, so that it's easy to string them together on the fly!
15:36:51 <alise> you get floats out of the contraption...somehow
15:37:18 <cheater> alise: yes, integer floats
15:37:25 <alise> KeyError: (11.0, -10.0)
15:38:21 <quintopia> i prefer A+W root beer floats myself
15:38:57 <quintopia> someone invent a lang with that as a type
15:39:18 <alise> cheater: well either i patched it wrongly or your code is messed up :D
15:39:40 <alise> btw your use of s is unwise
15:40:15 <cheater> so i use s for something else
15:40:21 <cheater> are you getting confused by this?
15:40:30 <quintopia> alise: does cheater have a copy of the latest optimized code?
15:40:41 <alise> yes, but he can't work on it
15:40:49 <alise> cheater: no, but it also didn't work the last time I tried it
15:40:57 <alise> python scoping is fucked
15:41:08 <cheater> ok, so T gets called for every X, Y?
15:41:23 <alise> T just advances the turn
15:41:26 <alise> quaffing does it too
15:41:32 <alise> http://pastie.org/1201977.txt?key=gaqwdbng4srklx1adb2o6g ;; i swear this doesn't work
15:41:33 <cheater> why would i put ai code in T?
15:41:40 <alise> start it, space, Qs break lightspeed crazily
15:41:51 <alise> cheater: read the rest of the code and you'll see why
15:41:52 <alise> all logic goes into T
15:41:57 <alise> everything that happens once a turn
15:42:07 <alise> *such as other characters like AIs moving*
15:42:27 <quintopia> so monsters can only move when i move?
15:42:36 <quintopia> this is getting more and more drod-like...
15:42:40 <alise> or do something else that takes a turn
15:42:44 <alise> quintopia: no, nethack-like
15:42:46 <alise> ais523: back me up here
15:42:53 <alise> monsters in nethack only move when you take a turn, yes?
15:42:59 <ais523> quintopia: turn-based; your character thinks quickly, but the player might not
15:43:11 <Sgeo> Well, #esoteric isn't on lambdabot's join list
15:43:11 <ais523> so the game pauses when you're not acting to give the player time to think
15:43:18 <alise> Sgeo: it was manually joined
15:43:31 <alise> quintopia: also Crawl, Angband
15:43:56 <ais523> are there any realtime roguelikes?
15:43:57 <cheater> alise: just saying, T doesn't seem like the right placer
15:44:02 <quintopia> i'm sure they had very good technical limitations in their original implementations that made realtime monsters unwise
15:44:07 <alise> D() is called on, e.g. enter
15:44:08 <quintopia> we have come out of the dark ages now!
15:44:10 <alise> which just dismisses a message
15:44:12 <alise> without advancing the turn
15:44:13 <ais523> (I was about to say "don't say trankesbel", but that isn't this channel)
15:44:16 <alise> so of *course* it should not be in D()
15:44:24 <alise> quintopia: seriously?
15:44:26 <alise> it's called a roguelike
15:44:42 <alise> and you apparently don't have the imagination to appreciate it
15:44:54 <alise> ais523: why not; is it vapourware? i googled
15:44:58 <quintopia> i have the imagination to appreciate a rogueimprovment more than a roguelike
15:45:02 <ais523> alise: it's the PSOX of another channel
15:45:06 <ais523> fortunately, I've forgotten which
15:45:27 <alise> quintopia: dude...
15:45:30 <alise> have you ever played nethack?
15:45:34 <alise> every game after rogue is an improvement
15:45:42 <alise> making monsters realtime makes the game suck
15:45:55 <alise> removing all thinking, skill and planning
15:46:02 <alise> and just turning it into a third-person shooter with bad graphics
15:46:09 <ais523> meh, some players specialise in thinking really quickly
15:46:13 <quintopia> since there's not much in the way of puzzles in your game...
15:46:18 <ais523> but I've spent several minutes planning a move before
15:46:41 <alise> ais523: from the trankesbel guy: [[My best achievement is ascending 29 times in a row (that is, not dying between ascensions).]]
15:46:50 <ais523> alise: I know him on IRC quite well
15:46:58 <ais523> that's as far as anyone knows the current world record for that
15:46:59 <quintopia> but what makes it cool is that it has actual maps that a programmer thought about carefully!
15:47:08 <ais523> the same person /also/ holds the world record for winning NetHack in realtime
15:47:10 <alise> "Other stuff include ascending in 1 hours 42 minutes (2009 /dev/null)" fffwhat
15:47:22 <alise> http://genodeen.net/a_wins.png ;; what did /dev/null ever do to him
15:47:25 <ais523> I'm adding commentary to the recording of that run, pretty slowly
15:47:44 <alise> "Krokotiilinhammaskeittokirja (Damage calculation tool)"
15:47:48 <alise> what a hilariously terrible name
15:48:05 <alise> oh he admins pinobot?
15:48:14 <ais523> how did you find out about pinobot?
15:48:39 <alise> ais523: err, you told me to use it when playing nethack
15:48:50 <ais523> hardly anyone knows about it, you see
15:48:52 <alise> is it not well-known or something?
15:49:01 <fizzie> Krokotiilinhammaskeittokirja is Finnish for (approximately) "crocodile tooth cookbook", in case that wasn't explained there.
15:49:21 <alise> http://genodeen.net/index.clua?df_stuff ;; Hey, a way to play Dwarf Fortress in Linux nicely.
15:49:44 <ais523> so I was surprised that you did
15:49:45 <ais523> so I was surprised that you did
15:50:15 <quintopia> alise: consider what i said about walls! (i even know a way you could dynamically generate rooms without actually having to spend lines and lines of code setting them up)
15:50:32 <alise> quintopia: do tell?
15:50:38 <alise> <ais523> so I was surprised that you did
15:50:39 <alise> <ais523> so I was surprised that you did
15:50:39 <alise> <ais523> */ping ais523
15:50:50 <ais523> alise: I was having connection trouble
15:50:53 <ais523> so I was pinging myself
15:51:02 <ais523> then I tried again, so I typed the key sequence to repeat a linea
15:51:08 <ais523> and accidentally repeated the wrong line
15:51:16 <ais523> see, it's not that implausible a correction when you know the background
15:51:32 <ais523> (it's just that me pinging myself isn't sent to the channel)
15:52:37 <alise> U+=U>0;N+=1;S-=S>0;L%=301;L-=25*(L and S<1)
15:52:39 <alise> not an if in sight
15:53:07 <ais523> alise: did you figure out that k bug?
15:53:19 <alise> ais523: which j bug?
15:53:24 <alise> i almost remember what you're saying
15:53:37 <ais523> the one where you hold down j and the turncount goes negative
15:54:13 <alise> fixed by refreshing the screen each turn and getting all cargo cult about that
15:55:21 <alise> woo, actually i can make that one line shorter
15:55:29 <ais523> well, did you figure it out?
15:58:09 <cpressey> the dark ages of turn based strategy... ye-e-e-es
15:58:41 <cpressey> we need some non-turn-based interactive fiction, too
15:58:50 -!- ais523 has quit (Read error: Connection reset by peer).
16:00:03 <cpressey> the number of times i have died at zangband by moving without considering my position well enough, i cannot count
16:00:42 <alise> I wonder whether quaffing when you have no potions should take a turn.
16:02:29 -!- ais523 has joined.
16:03:59 <ais523> cpressey: are you admiring my analogy?
16:05:25 -!- Phantom_Hoover has joined.
16:06:20 <Sgeo> Is clog working?
16:06:45 <ais523> it was having trouble recently
16:06:50 <ais523> but it started working again
16:06:53 <ais523> has it stopped working again?
16:07:08 <Sgeo> It seems to be working
16:07:43 <Sgeo> I wasn't sure whether or not to bother checking logs
16:19:44 <alise> heh, my hallu is flawed
16:19:49 <alise> you can redraw as much as you want
16:19:56 <alise> which lets you determine where everything is, to high accuracy
16:20:01 <alise> because redrawing causes re-hallucination
16:20:16 -!- MigoMipo__ has joined.
16:20:53 -!- lament has joined.
16:23:29 -!- MigoMipo has quit (Ping timeout: 276 seconds).
16:28:41 <alise> "Oh, and not long after making these photos, I was stopped and interviewed by a police officer, who searched my camera bag, on grounds of suspicious activity, potentially related to terrorism. I have a carbon copy of the police report sheet to prove it." --DMM on London
16:28:47 <alise> why am I not surprised?
16:29:43 <ais523> I'm not really, either
16:30:02 <alise> ais523: the photo was of the *tower of london*
16:30:08 <alise> apparently, taking photos of it is suspicious
16:30:17 <alise> you might want to help a friend escape from there, treasoner!
16:30:20 <ais523> I know I was once moved by an armed police officer
16:30:31 <alise> you "know" it? not "remember" it?
16:30:35 <ais523> because I was accidentally standing in the way of an official car that wanted to drive into Buckingham Palace
16:30:49 <ais523> alise: well, it seems like a really absurd thing to plant a false memory of
16:30:56 -!- MigoMipo has joined.
16:31:12 <alise> ais523: I would have just omitted the "I know"
16:32:01 <cpressey> perfectly normal rhetorical device imo
16:32:13 <alise> i think i initially interpreted it as that
16:32:19 <alise> but discarded it as meaningless before finishing for some reason
16:32:33 <ais523> I think it expands to something like "well I only know that info about DMM second-hand, but here's a first-hand story..."
16:32:50 <cpressey> also, you are debugging hallucination code, so it's understandable
16:33:25 -!- MigoMipo__ has quit (Ping timeout: 250 seconds).
16:35:00 -!- MigoMipo_ has joined.
16:37:09 <alise> i don't mind it being broken
16:37:10 -!- MigoMipo has quit (Ping timeout: 260 seconds).
16:37:31 -!- MigoMipo__ has joined.
16:39:52 -!- MigoMipo_ has quit (Ping timeout: 240 seconds).
16:40:33 -!- nooga has joined.
16:47:36 -!- MigoMipo__ has quit (Remote host closed the connection).
16:47:52 -!- MigoMipo__ has joined.
16:48:52 -!- MigoMipo__ has quit (Client Quit).
16:48:56 <ais523> Sgeo: care to explain?
16:49:16 <Sgeo> What it is, or why it's memefied?
16:49:19 -!- MigoMipo has joined.
16:50:39 <ais523> I think cheater was asking the first
16:50:52 <ais523> see, you can subvert the meme this time by taking it back to its roots
16:52:13 <Sgeo> PSOX is something that sits between a program, typically written in an stdio-only esolang, and stdio
16:52:32 <Sgeo> It intercepts output, and treats it as commands to do things like open files or open sockets
16:52:35 -!- nooga has quit (Ping timeout: 276 seconds).
16:52:43 <Sgeo> And feeds the results back into the program's input
16:53:22 -!- lament has quit (Ping timeout: 250 seconds).
16:54:13 <Sgeo> Overzealous promotion, vaporware for a time, and despite having language-independence as a goal, it made many assumptions that inconvenience languages other than Brainfuck
16:54:50 <Sgeo> I rejected one of the spiritual ancestors of PSOX precisely because it assumed the concept of cells
16:55:10 <Sgeo> (And some semantic issues.. or was that something else?)
16:57:44 -!- rodgort has quit (Quit: Coyote finally caught me).
16:58:01 <alise> vapourware always :)
16:58:05 <alise> well at least for the stuff yuo said itw ould be able to do
16:58:23 <ais523> you know what would be a good model for a PSOX-alike? telnet!
16:58:36 -!- rodgort has joined.
17:00:04 <alise> ais523: hell no :P
17:00:31 -!- tombom has joined.
17:03:09 <Phantom_Hoover> So wait, what did Sgeo assert PSOX would be able to do?
17:03:31 <cheater> i thought psox was phantasy star online xbox
17:04:07 <Sgeo> alise, just because I never got around to the file stuff before abandoning it?
17:04:20 <Sgeo> And did I ever actually PROMISE a GUI domain?
17:04:21 <alise> although i may be wrong there
17:04:23 <Sgeo> The network stuff works
17:04:54 <Sgeo> pikhq wrote a wget.b
17:07:00 * Sgeo goes to watch some SG-1
17:08:01 <Ilari> Wow, for some larger operators, it is estimated that the Carrier Grade NAT logs one would have to keep would take about 2EB of space... That's A LOT.
17:08:28 -!- SgeoN1 has joined.
17:09:21 <ais523> and yet people still don't get the message about IPv6
17:09:30 <alise> 06:48:37 <fizzie> There's the Jatravartids of Viltvodle VI, who believe the Universe was sneezed out of the nose of the Great Green Arkleseizure, and who fear the Coming of the Great White Handkerchief.
17:09:31 <ais523> EB aren't out of range for modern technology, though, IIRC
17:09:35 <alise> Please, tell me you typed that from memory.
17:09:56 <alise> Ilari: carrier grade logs -- you mean legally mandated stuff?
17:10:02 <alise> i'm not so hot with networking
17:10:19 <Ilari> alise: I mean "Carrier Grade NAT".
17:10:40 <Ilari> alise: And yes, legally mandated logs. Currently one can get by with much smaller logs.
17:10:51 <alise> Ilari: obviously we will just legislate away the logs
17:10:55 <alise> after Verizon and Comcast complain
17:12:18 <Ilari> Worse yet, even those logs will not be as useful for handling abuse as what currently exist.
17:12:52 -!- augur has quit (Ping timeout: 240 seconds).
17:13:19 <Ilari> Since time skew would seriously hamper process. Yes, there's NTP, but still a lot of clocks are wrong.
17:14:46 -!- pineal_aenimal has joined.
17:16:36 -!- pineal_aenimal has left (?).
17:17:46 <alise> 15:06:43 * ais523 continues to wonder wtf expect programs (written in TCL) run, given that TCL isn't installed
17:17:51 <alise> ais523: expect actually has tcl compiled in
17:17:59 <ais523> that was explained some other time
17:18:03 <alise> it's, like, an alternative main.c or something iirc
17:18:06 <ais523> but wow, I made that statement ages ago
17:18:10 <alise> ais523: shut up i can logread as far back as I want :D
17:18:13 <alise> (only 27th August)
17:21:19 * alise examines in album's waveforms in Audacity
17:21:23 <alise> wow, atrocious clipping
17:21:31 <alise> they should remaster it properly sometime
17:23:37 <ais523> anyway, new esolang suggestion: 2D with two commands (NOP /not/ allowed, you have to use one or the other command everywhere up to the edge of the program): G rotates the IP left unless the current tape element is 0, X going left/right/up/down respectively is equivalent to BF < > + - respectively, tape is bignum, signed, and is initialised to start with 1 everywhere, IP starts going downwards at the top-left of the program
17:23:57 <ais523> I'm trying to work out if it's TC; with NOPs, it's relatively clearly possible to compile BF-minus-IO into it
17:24:22 <ais523> by separating the tape into a series of "always positive / junk / data / always positive / junk / data", etc
17:24:28 <ais523> don't have an interp yet
17:24:48 <ais523> and it's not quite clear how the lack of NOPs affects it, but I think (am unsure) it's still TC anyway
17:32:28 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>([+++++[-]>]>)*20>)*20[[[-]-]-]
17:33:50 <EgoBot> Score for ais523_tripwire2: 0.0
17:34:01 <ais523> hmm, that's suspiciously low
17:34:47 <ais523> two draws, though, so it's not like it's an autolose program
17:36:28 -!- augur has joined.
17:38:48 <Phantom_Hoover> Although I suppose it probably is in a literal sense, as well.
17:39:50 <quintopia> if one can consider meat a machine...
17:40:23 <quintopia> things that are really neat: neuroplasticity et al.
17:41:21 <augur> Phantom_Hoover: it was ALWAYS literally so.
17:41:33 -!- augur has quit (Remote host closed the connection).
17:41:58 -!- augur has joined.
17:41:59 <Phantom_Hoover> Well, the neurons themselves aren't self-replicating any more AFAIU.
17:42:03 -!- augur has quit (Remote host closed the connection).
17:42:05 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>([+++++[-]>]>)*20]>)*20[[[-]-]-]
17:42:15 <ais523> was a missing bracket, I'm amazed it actually got anywhere at all
17:42:17 <EgoBot> Score for ais523_tripwire2: 21.5
17:42:23 -!- augur has joined.
17:42:42 <ais523> looks like I just knocked GreaseMonkey off the leaderboard altogether
17:44:42 <alise> i wouldn't say the human brain can self-replicate
17:44:44 <alise> just the body as a whole
17:44:46 <alise> with the brain as the cpu
17:44:52 <ais523> neurons can't actually replicate
17:44:58 <ais523> stem cells can replicate, and transform themselves into neurons
17:45:06 <ais523> but the neurons themselves are no longer capable of replication
17:52:14 <Gregor> "SHOO!" "You could ask me to leave more politely." "I could. I choose not to."
17:53:57 <Phantom_Hoover> alise, I was referring to the various bacteria and such in my sinuses and throat.
17:54:13 -!- sebbu2 has joined.
17:54:13 -!- sebbu2 has quit (Excess Flood).
17:54:48 -!- sebbu2 has joined.
17:55:36 <EgoBot> Use: !bfjoust <program name> <program>
17:55:44 <EgoBot> Installed user interpreters: aol austro b1ff bc bct bfbignum brit brooklyn bypass_ignore bytes chaos chef chiqrsx9p choo cockney ctcp dc decisionengine drawl drome dubya echo eehird ehird fudd funetak google graph gregor he hello id jethro kraut num ook pansy pi pirate plot postmodern postmodern_aoler redneck reverse rot13 sadbf sfedeesh sffedeesh sffffedeesh sffffffffedeesh simpleacro slashes svedeesh swedish valspeak warez yodawg
17:57:42 -!- sebbu has quit (Ping timeout: 265 seconds).
17:58:12 -!- sebbu2 has changed nick to sebbu.
18:00:06 <Gregor> ais523: Still, defend7 is doing a hell of a lot better than tripwire2, so :P
18:00:35 <ais523> Gregor: heh, defend7 IIRC actually has some anti-tripwire code
18:00:44 <alise> RuntimeError: maximum recursion depth exceeded
18:00:51 <ais523> if it didn't, tripwire2 would beat it any day
18:00:55 <alise> Yes, that's right, there are so many monsters that my message-printing function crashed.
18:01:05 <ais523> http://codu.org/eso/bfjoust/report.txt
18:02:12 <Gregor> I forget the convoluted but semi-logical algorithm by which I calculated scores :P
18:02:45 <Gregor> I guess it's in report.c if I wanted to remember ...
18:03:37 <ais523> I think you even documented it
18:03:48 <ais523> yep, http://codu.org/eso/bfjoust/SCORES
18:04:38 <alise> *Somehow* this prompts "81" all the time.
18:06:19 <ais523> hmm, I wonder what evolutionary BF Joust would be like
18:11:10 <Gregor> Then a LAB will come to YOU!
18:11:27 -!- Harpyon has joined.
18:11:53 <ais523> hmm, no, defend7 doesn't have counter-tripwire code
18:11:58 <ais523> so why is tripwire2 losing to it?
18:12:27 <Gregor> Because it's more defensive than tripwire is trip-wiry?
18:13:01 <ais523> the whole point of tripwiring (one of the few tactics that worked in BF Joust 1) is to ignore the first of your opponent's decoys (or more, but that's really risky)
18:13:09 <EgoBot> Use: !bfjoust <program name> <program> . Scoreboard, programs, and a description of score calculation are at http://codu.org/eso/bfjoust/
18:13:25 <ais523> defend7 does place one extra decoy, so it does defeat the tripwire
18:13:38 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:13:48 <EgoBot> Score for ais523_tripwire2: 27.6
18:13:52 <ais523> let's try a double-tripwire; this did insanely badly last time I tried, but it may work better in today's metagame
18:14:23 <ais523> (and now beats defend7)
18:14:38 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:14:47 <EgoBot> Score for ais523_tripwire2: 33.8
18:14:58 <ais523> hmm, I wonder how many tripwires I can get away with?
18:15:06 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>>>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:15:18 <EgoBot> Score for ais523_tripwire2: 29.8
18:15:23 <ais523> looks like three is the optimal number
18:15:26 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:15:33 <EgoBot> Score for ais523_tripwire2: 33.8
18:15:37 <Gregor> I was just about to say "I'm gonna go with three" :P
18:15:44 <Gregor> But that's just the optimal number for this hill.
18:15:47 -!- nooga has joined.
18:16:17 <ais523> hmm, tripwire actually has a positive overall rating
18:16:22 <ais523> it's pretty rare to get one of those onto the hill that quickly
18:16:31 <alise> yay, my hallu now has a cure
18:16:34 <alise> (potions help alleviate it)
18:16:53 <ais523> 8 losses, one to defend7 (where a tripwire trips over the defenses altogether and lands on the other side of the flag)
18:17:33 <alise> I think it's actually possible to survive hunger if you get like 100 potions.
18:17:51 <alise> (being hungry deducts 25 HP per turn)
18:17:56 <alise> and potions give 20
18:18:12 <alise> hungry=no satiation, that is
18:18:23 <alise> so if you have enough HP, and enough potions to last you until you get to the nearby food...
18:19:01 <alise> i may make money useful, say,
18:19:06 <alise> higher max hp the more money you have
18:19:20 -!- augur has changed nick to cheesey_.
18:19:24 <alise> ehird@dinky:~/Code/vagrant$ wc -c vagrant.py
18:19:27 <alise> pretty good, I think
18:19:28 -!- cheesey_ has changed nick to augur.
18:19:46 <alise> 19.2 "standard" (80 col) lines
18:19:54 <alise> replacing newline with something else, that is
18:20:05 <alise> just need to make monsters work now :P
18:22:12 <ais523> !bfjoust tripwire2 >>>>>>>([>>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:22:21 <EgoBot> Score for ais523_tripwire2: 17.3
18:22:26 <ais523> !bfjoust tripwire2 >>>++++++++++++++<---------->>>>>>([>>>([+++++[-]>]>)*20]>)*20[[[-]-]-]
18:22:31 <ais523> OK, it seems you /do/ need decoys
18:22:34 <EgoBot> Score for ais523_tripwire2: 33.8
18:28:20 -!- augur_ has joined.
18:28:27 -!- augur has quit (Read error: Connection reset by peer).
18:30:06 <nooga> i'm looking for an idea for a complicated programs in C++ and C#
18:30:45 <Gregor> C++, it's from KHTML remember.
18:30:59 * alise wonders how he lost 5 HP
18:31:13 <Gregor> alise: Not enough entrepreneurship.
18:31:20 <alise> nooga: vagrant is my lovely little golfed roguelike
18:31:56 <Gregor> alise: Is there an actual roguelike golf out there, or are you just making it 'cuz you can?
18:31:59 <alise> 1635 bytes of Python -- well, 1634, the newline at the end is irrelevant -- and, although it doesn't have monsters that actually fight you yet, it's surprisingly feature...ful
18:32:04 <Gregor> If the latter, I have a GC in <700 lines of C :P
18:32:24 <alise> Gregor: The latter. Well, there was a 1k roguelike competition a while ago on that usenet group, but it only got like three submissions and half were in C# and Java and shit.
18:32:31 <alise> (Yes, half of three!)
18:32:36 <alise> Besides, I've broken the 1k barrier already.
18:32:46 <alise> By doing things like this:
18:32:51 <alise> w.update(q);U+=U>0;N+=1;S-=S>0;L%=301;L-=25*(L and S<1)
18:32:52 <alise> w[y,x]=32;x+=a;y+=b;w[y,x]=64;X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y);D()
18:32:53 <alise> a=(k in'lun')-(k in'hyb');b=(k in'jbn')-(k in'kyu')
18:32:55 <alise> if P and L+20<301:q=min(P,20);L+=q;U+=q*3*(U>0);P-=q;T()
18:32:57 <Gregor> So, one was in both C# and Java, another was in C, and one was in Java with JNI using C stuff?
18:33:07 <alise> q=w[v];s.addch(B-Y,A-X,choice(W)if U and 0==r(0,2)and q-32 else q)
18:33:08 <alise> s.addstr(22,0,'_'*80);C('$:%-17s T:%-17s S:%-17s HP:%-3s (%s)'%(G,N,S,L,P));global V;V();V=lambda:1;s.move(y-Y+11,x-X+40)
18:33:37 <alise> Gregor: My messaging system is awesome. Guess how it works. Actually it doesn't really work but there you go.
18:36:15 -!- Gregor has set topic: The international hub for esoterica, the occult, astrology, esoteric topics in computing and programming languages, astral projection, necromancy and scientology | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:36:44 -!- ais523 has set topic: < pikhq> Microsoft Word: the worst program to design web pages in, and this *includes* Malbolge. | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:38:03 <ais523> OK, can anyone explain this? leaning on my touchpad in a certain way causes all the pixels in the client area of my IRC client (but not the window border or tabs, etc, so it's clearly software-involved somehow)
18:38:11 <alise> <ais523> Gregor: don't
18:38:17 <alise> it's funnier that way
18:38:22 -!- Gregor has set topic: Totally NOT the international hub for esoterica, the occult, astrology, astral projection, necromancy and scientology | But actually IS the international hub for esoteric topics in computing and programming languages | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:38:44 <ais523> but it doesn't correspond to left/right/middle-clicking, or any sort of mouse movement AFAICT
18:38:47 <ais523> so what input's being sent?
18:38:55 <alise> "leaning on my touchpad in a certain way causes all the pixels in the client area of my IRC client (but not the window border or tabs, etc, so it's clearly software-involved somehow)"
18:38:58 <alise> causes all of them WHAT?
18:39:05 <cpressey> so i think my bot should be an esolang interpreter HEAR ME OUT
18:39:05 <Gregor> ais523: It just causes them.
18:39:09 <alise> "It just causes all of them!"
18:39:15 <Gregor> They wouldn't exist if not for leaning on the --- damn :P
18:39:18 <alise> except you mispinged me :P
18:39:26 <ais523> I knew I'd missed something from the sentence, just wasn't sure what
18:39:31 <cpressey> a language that just happens to overlap the irc protocol and thus work as a bot
18:39:39 <Gregor> alise: I was actually writing that before you wrote your thing, so I was telling him that :P
18:39:54 <ais523> (anyway, the main reason is that by attempting to define esoterica, there's bound to be someone who stumbles in here by mistake, disagrees with our definition, and flames us all for the rest of our lives
18:40:05 <Gregor> cpressey: Mah brain axplote.
18:40:28 <alise> i thought it was in response to <alise> causes all of them WHAT?
18:40:32 <alise> or did you write it before that too?
18:40:45 <Gregor> alise: I wrote it before that, that's what I'm saying.
18:40:55 -!- ais523 has set topic: peanut butter, the teachings of Henry XVI of Lithuania, rooftop supports, neutrinos | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:41:35 <cpressey> lithuanians sure do have funny surnames
18:41:36 -!- Gregor has set topic: George Carver did not invent peanut butter | Not that we're racist or anything | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:42:09 -!- alise has set topic: Peanut butter | The teachings of Henry XVI of Lithuania | Rooftop supports | Burma Shave | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:43:17 -!- Gregor has set topic: This channel is not about whatever you think it's about | Unless that's Lithuanian peanut-butter-based shaving creme | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:43:45 <alise> do you really mean creme there?
18:43:57 <Gregor> No, but I'm keeping it.
18:44:32 <ais523> surely it should be crème?
18:44:42 -!- trdrkia has joined.
18:44:50 <trdrkia> wow, cool, i thought nobody was as crazy as me to try it
18:45:01 <trdrkia> im from lithuania and i made some shaving cream with peantu butter...
18:45:02 -!- Gregor has set topic: This channel is not about whatever you think it's about | Unless that's Lithuanian peanut-butter-based shaving crème | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:45:07 <trdrkia> **peanut sorry my english isn't so good :))
18:45:19 <trdrkia> so what are you guys'' experiences ?
18:45:22 <ais523> alise: your IP gives you away
18:45:32 -!- Gregor has left (?).
18:45:32 <ais523> still, great performance art
18:45:33 <alise> ais523: NO MY WHOLE COUNTRY IS ON A NIC
18:45:41 -!- Gregor has joined.
18:45:41 -!- Gregor has left (?).
18:45:45 <trdrkia> it is true, alise is president of lithuania
18:45:58 <trdrkia> we have a song about it: "$*("Y!£HNO CARRIER
18:45:59 -!- trdrkia has quit (Client Quit).
18:46:03 -!- Gregor has joined.
18:46:06 <Gregor> Yes, Colloquy is certainly the worst IRC client there is.
18:46:12 <alise> Gregor: It is pretty bad.
18:46:14 <alise> Gregor: Try LimeChat.
18:46:19 <alise> http://limechat.net/mac/
18:46:21 <alise> It's got what plants crave!
18:46:52 <alise> I don't know if LimeChat has /ignore yet, but hey, what can you do.
18:46:54 <alise> (It probably does.)
18:47:09 <alise> Since 2010-05, which is... May?
18:47:10 <Gregor> /ignore is for pussies anyway.
18:47:44 <alise> Real men face their enemies!
18:47:59 <Gregor> wtf ... it won't let me install RubyCocoa ...
18:48:07 <alise> Your OS includes it.
18:48:10 <alise> Assuming you're post-Tiger.
18:48:16 <alise> http://cloud.github.com/downloads/psychs/limechat/LimeChat_2.15.tbz
18:48:18 <Gregor> Oh, I forgot what stupid name corresponds to what.
18:48:31 <Gregor> I saw "Snow Leopard" and went "know I don't have that", so skipped right to the next one.
18:48:37 <Gregor> Missing that "Leopard" was there too.
18:49:00 <alise> Gregor: You probably want to go to LimeChat -> Preferencse -> Log -> Untick "Show image links inline."
18:49:09 <alise> I think that would change http://totallygoatse/ into you know what.
18:49:20 <alise> Apparently it does it for YouTube links too.
18:49:26 <alise> (It may not even be enabled by default; I don't know.)
18:49:28 <ais523> hmm, I just noticed the zzo38 take on BF Joust
18:49:30 <alise> It's a good client though.
18:49:44 <ais523> and have been laughing for over a minute
18:49:49 <ais523> it's just so... different from how I think of it
18:49:54 <ais523> although it might nonetheless be a decent game
18:50:06 <alise> also, you laugh at *way* too little
18:50:22 <ais523> http://esolangs.org/wiki/Talk:BF_Joust
18:50:31 <ais523> well, it's not really quantity, but unexpectedness
18:50:38 <ais523> some things I find funny that other people don't, and vice versa
18:51:14 <alise> i wonder what zzo's like in real life
18:51:22 <alise> i dearly hope he speaks exactly like on IRC
18:52:04 <alise> ais523: I like how he turned it into a card betting game.
18:52:23 <ais523> it's like a mix between BF Joust and poker
18:52:24 <alise> Gregor: Notice the lesser amount of pain.
18:52:29 <alise> <alise> Gregor: You probably want to go to LimeChat -> Preferencse -> Log -> Untick "Show image links inline."
18:52:33 <ais523> you could actually do BF Joust without the BF
18:52:37 <alise> If it is indeed ticked by default.
18:52:47 <ais523> on your turn, you can: do nothing; check if your current location is 0; adjust your current location + or -; move < or >
18:52:56 <Gregor> alise: Why does that exist, and why is it under "Log" ...
18:53:10 <alise> Gregor: "Log" is LimeChat's name for the thing you see on screen.
18:53:15 <alise> The actual displayed meat of the channel.
18:53:16 <ais523> that way you can play it as a competitive game between humans
18:53:20 <alise> Gregor: It's Japanese :P
18:53:28 <alise> As for why it exists: #goatse obvs
18:53:38 <alise> ais523: that'd be rather slow, I imagine
18:53:41 <Gregor> alise: Of course it is, it's in Ruby.
18:54:01 <ais523> alise: now I'm wondering if that channel exists
18:54:03 -!- MigoMipo_ has joined.
18:54:09 <alise> ais523: rather easy way to find out
18:54:12 -!- MigoMipo has quit (Read error: Connection reset by peer).
18:54:14 <ais523> not that I'd join it, even if it did exist
18:54:19 <alise> * You have been kicked from #goatse by ChanServ (Invite only channel)
18:54:27 <alise> It's a very exclusive anal-stretching channel.
18:54:32 <alise> You need an invite.
18:54:41 <Gregor> You have to post relevant pics to the moderator before being allowed to join.
18:54:51 <Gregor> Then they discuss techniques, skills, etc.
18:55:14 <alise> Gregor: You use a bouncer, right? Which?
18:55:34 <Gregor> It is precisely one modicum less terrible than the other ones I've used :P
18:55:42 <alise> Gregor: Ah, shame. I had some code that made LimeChat work with my bouncer's automatic scrollback feature, actually modifying the time fields and stuff.
18:55:48 <alise> So that it looked like you'd been in there all this time.
18:55:55 <alise> Gregor: Have you seen http://miau.sourceforge.net/?
18:55:58 <alise> it's a pretty good bouncer
18:56:26 <Gregor> I used bip only because it was recommended at me *shrugs*
18:56:34 <alise> I just recommended miau at you :P
18:56:43 <alise> I can't recall whether I used psyBNC or miau at the time I wrote that code though.
18:56:47 <alise> psyBNC is terrible.
18:56:50 <Gregor> Too late, bip got recommended at me earlier.
18:57:11 <fizzie> The problem with switching bouncers is that it's (potentially) yet another different log format.
18:57:32 <Gregor> fizzie: I don't log on the bouncer, I only log via xchat, and my xchat is always connected.
18:57:57 <fizzie> But that'd destroy my main reason for having a bouncer, which is to make switching actual IRC clients a lot easier.
18:58:14 <Gregor> Switching LOCATIONS is what's vital, not clients.
18:58:21 <Gregor> I'm on my laptop, at school right now.
18:58:27 <Gregor> But my home computer is still on.
18:58:44 <alise> People who use the "school" terminology post-school weird me out.
18:58:45 -!- nooga has left (?).
18:58:47 <alise> WEIRD ME OUT I say.
18:58:47 <Gregor> Also, the primary reason for me to have a bouncer is because having my hostname be codu.org is pretty damned suave.
18:58:53 -!- nooga has joined.
18:59:03 <nooga> alise: do you have something stable?
18:59:05 <Gregor> alise: Since all Americans call all forms of education "school"
18:59:19 <alise> Gregor: you're a Ph.D. student right?
18:59:34 <alise> I've never heard a Ph.D. student talk about going to school in the present tense, or anyone reference a Ph.D. student doing so :P
18:59:45 <alise> Unless they're the world's most retarded Ph.D. student, in which case maybe they go to little school too.
18:59:57 <alise> (Like theolog--*shot by the pope*)
19:00:04 <Gregor> I'm an idiot savant. I know computer science, not how to tie my shoes.
19:00:44 <fizzie> To each his own, I guess; I used to, and still do, run a permanently connected irssi that I'd ssh into for switching locations and having a zem.fi host; I just went bouncy to be able to experiment with clients without it messing logging and so.
19:02:04 <nooga> savant != i know computer science and i'm completely antisocial nerd that gets annoyed by daylight
19:02:04 <alise> Did anyone ever build an archiving system on ar before tar came along?
19:02:32 -!- augur has joined.
19:02:33 <alise> Like, an ar file that has a __DIRECTORIES file with a list of A, B and C separated by newlines,
19:02:35 -!- augur_ has quit (Read error: Connection reset by peer).
19:02:41 <alise> and that means the A.ar, B.ar, C.ar files inside are more directories
19:05:20 -!- antivigilante__ has quit (Ping timeout: 252 seconds).
19:05:20 -!- antivigilante has quit (Ping timeout: 252 seconds).
19:07:20 <Gregor> alise: I think cpio and tar came about precisely because ar was insufficient for real archiving :P
19:07:33 <alise> It's not insufficient, just handicapped! >___>
19:07:51 <alise> Gregor: Also, you forgot pax.
19:08:03 <Gregor> pax postdated by like a decade both cpio and tar
19:08:13 <Gregor> So no, pax came about because of limitations in cpio and tar :P
19:08:22 <alise> Gregor: Correct response:
19:08:26 <alise> "No, I really didn't forget pax."
19:09:12 <Gregor> I wish UNIX folks would switch to a random-access-aware compression and archival format.
19:09:34 <alise> [[Furthermore, "pax" means "peace" in Latin, so name implies it shall create peace between the tar and cpio format supporters.]]
19:09:37 <ais523> Gregor: tar and pax are random-access-aware when decompressed
19:09:49 <Gregor> "COMPRESSION AND ARCHIVAL"
19:09:52 <ais523> alise: well, it worked, when's the last time you saw a tar/cpio flamewar?
19:10:06 <Gregor> Hell, even HP-UX's bizarre .gz.tar files are better than .tar.whateverz
19:10:13 <alise> pax is what happens when the IEEE tries to design software!
19:10:21 <alise> Gregor: ...gz.tar?
19:10:28 <alise> Is every file inside gz'd or something?
19:10:29 <ais523> alise: gzip files, /then/ tar them
19:10:33 <alise> AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
19:10:42 <Gregor> alise: Gives you random access to compressed files.
19:11:08 <Gregor> Mind, HP-UX uses that format for installable packges, which have no use for random-access, but anywho :P
19:11:12 <alise> .xz.cpio *mwahahahaha*
19:11:37 <alise> Gregor: .xz is lzma btw :P
19:11:42 <alise> (.lzma is deprecated)
19:11:52 <alise> Gregor: .cab.tar.cpio.cab
19:12:17 <alise> Cab! Wonderful cab!
19:12:20 <fizzie> I mean, there's 3DES; why not 3cab.
19:12:24 <alise> Cab, cab, cab, cab, cab, cab, cab, cab...
19:12:50 <alise> Gregor: .ext3.ext2.ext.reiserfs.7z.ar
19:12:59 <Gregor> I remember I had a class that required submissions in .shar format.
19:13:00 <ais523> the Windows public-access computers in the EE computer labs here used to automatically rename .tar.gz files to .tar.tar
19:13:02 <ais523> for no apparent reason
19:13:17 <ais523> Gregor: well, you can read shars without uncompressing them
19:13:24 <ais523> they used to be common on Usenet for that reason
19:13:26 <alise> Free as in Free 30-Day Trial!
19:13:34 <pikhq> ais523: Except when they're not.
19:13:36 <alise> *.rar, for better effect
19:13:56 <pikhq> ais523: GNU shar can produce base64'd, gzip'd shars.
19:14:04 <pikhq> (not by default, mind)
19:14:05 <ais523> pikhq: the man page says that people get annoyed when you post them to Usenet
19:14:13 <Gregor> ais523: I seem to recall the submission program that's official but nobody actually uses here recompressing .tar.gz files, so you'd get a file named .tar.gz, but that was actually a .tar.gz.gz
19:14:17 <pikhq> ais523: For obvious reasons.
19:14:18 <Gregor> Which is surprisingly annoying to extract.
19:14:27 <ais523> Gregor: I can guess what you mean there
19:14:29 <pikhq> ais523: Might as well just MIME encode a tarball at that point.
19:14:33 <ais523> (why is gzip filename-sensitive anyway?)
19:14:57 <Gregor> You'd have to gunzip, rename the result, then tar zxf
19:14:58 <alise> because gzip has a badly-designed UI
19:15:10 <ais523> it's more a very specific UI
19:15:18 <ais523> good for the most common use-case, bad in other cases
19:15:18 <Gregor> And of course you'd never remember that it's doing this to you, so first you'd tar zxf, then it'll say "this shit ain't no tar"
19:15:22 -!- impomatic has joined.
19:15:37 <ais523> impomatic: EgoBot's working again, if that's what you're here for
19:15:43 <ais523> there's even been movement on the hill today
19:16:00 <Gregor> People need to PM me when it's not working :P
19:16:01 <pikhq> For uncommon usecases, you pretty much want to use it as a filter; it'll be annoying otherwise.
19:16:07 <impomatic> I joined #esolang first and wondered where everyone disappeared to :-)
19:16:13 <alise> pax -wf /dev/fd0 .
19:16:13 <alise> ATTENTION! pax archive volume change required.
19:16:13 <alise> /dev/fd0 ready for archive volume: 2
19:16:13 <alise> Load the NEXT STORAGE MEDIA (if required) and make sure it is WRITE ENABLED.
19:16:13 <alise> Type "y" to continue, "." to quit pax, or "s" to switch to new device.
19:16:14 <alise> If you cannot change storage media, type "s"
19:16:16 <alise> Is the device ready and online? >
19:16:50 <ais523> pax splits across drives?
19:17:01 <impomatic> I'm just making a BF Joust wiki page, so I wanted to test an example
19:17:04 <Gregor> Real men cat then extract.
19:17:06 <alise> with a very un-unixy UI :P
19:17:21 <ais523> impomatic: in which wiki?
19:17:28 <alise> Gregor: The best archive format is "cat directory" (works on at least Plan 9 and NetBSD!)
19:17:36 -!- antivigilante has joined.
19:17:36 <alise> Sure, it's system-dependent... but who cares, it's simple!
19:17:42 <Phantom_Hoover> What kind of thing would you use random-access compression for?
19:17:46 <alise> cat files | compress
19:17:51 -!- antivigilante__ has joined.
19:17:55 <ais523> alise: does that actually give you the files in the directory? or just the metadata?
19:18:01 <alise> ais523: just the metadata, actually
19:18:07 <pikhq> alise: The best archive format is a filesystem dump containing just the required inodes and blocks.
19:18:08 <impomatic> ais523: http://programminggames.org/BF-Joust.ashx
19:18:13 <alise> Gregor: Pretty sure it just spits out the filesystem metadata.
19:18:25 <Gregor> impomatic: .ashx ... you lose forever.
19:18:42 <pikhq> (that would actually be pretty awesome: arche2fs /dev/sda1 /home/pikhq/some_dir)
19:18:50 <alise> "ash(1) Server Pages"
19:19:34 <pikhq> impomatic: Waitwaitwait, *Almquist shell* CGI?
19:20:18 <alise> But it SHOULD be ash.
19:20:24 <alise> pikhq: Nono, ASP.NET!
19:20:25 <impomatic> Gregor: I didn't actually write any .NET, just picked a Wiki that used a flat file.
19:20:32 <pikhq> I could at least respect CGI in asp.
19:20:39 <alise> ais523: quick, link to BobTHAbsentee's tracking system
19:20:58 <pikhq> impomatic: TiddlyWiki?
19:21:01 <ais523> http://nomic.bob-space.com/agoralog.aspx
19:21:14 <alise> TiddlyWiki doesn't support server writes without evil.
19:21:20 <alise> (what ais523 said)
19:21:33 <alise> pikhq: Especially view source, look at __VIEWSTATE.
19:21:47 <alise> pikhq: All of Agora relied on this not that long ago.
19:21:54 <alise> Then he left, I think he had his, what, sixth kid?
19:21:56 <alise> And the game collapsed!
19:22:01 <pikhq> alise: Ah. Well, there must be *something* better...
19:22:12 <alise> pikhq: BEHOLD THE ASP.NET PAIN
19:22:27 <pikhq> Well, there's Wikit; the in-Tcl single-file wiki/web server.
19:22:30 <alise> proto: only infertile people can control all recordkeeping
19:23:12 <pikhq> alise: ... No, the whole wiki is a single file.
19:23:12 <alise> pikhq: Why don't you play Agora anymore btw?
19:23:18 <alise> pikhq: I meant, the wiki.tcl.tk page for it.
19:24:17 <nooga> alise: when you plan to show some bits of this golfed roguelike?
19:24:41 <alise> pikhq: /1? old page!
19:24:43 <alise> does wiki.tcl.tk run on it?
19:24:44 <Gregor> impomatic: FukYorBrane isn't there :(
19:24:44 <pikhq> alise: I don't play Agora because... Uh. I don't.
19:24:47 <alise> or did it in the past?
19:24:52 <alise> pikhq: You used to play Agora :P
19:24:54 -!- augur has quit (Remote host closed the connection).
19:24:56 <alise> nooga: I have; want the current version?
19:24:58 <pikhq> alise: It has always run on Wikit.
19:25:12 <pikhq> They believe in eating their own dog food. :)
19:25:21 -!- augur has joined.
19:25:35 <alise> vagrant.py http://pastie.org/1205917.txt?key=qvb1dnfruxbofgx4xmpog
19:25:39 <alise> debug.py http://pastie.org/1205918.txt?key=hpfmpbouu9kh8ki3kwgta
19:25:45 <alise> Run debug.py if you make any changes; it lets you see exceptions.
19:25:51 <ais523> Gregor: I still think that starting with 0, followed by defect, followed by a very very very long loop, is a breaking strategy in FYB
19:26:01 <ais523> because it causes your IP to move faster than the enemy pointer's speed of light
19:26:04 <alise> And yes, monsters do disappear if you walk into them; and yes, they do walk randomly. For now.
19:26:15 <ais523> thus, you then have unlimited time to track them down, unless they're using the same strategy
19:26:26 <ais523> there is a counter-strategy, but it sucks against anything else...
19:26:37 <Gregor> ais523: Then PROVE IT.
19:26:45 <Gregor> ais523: logicex-2 still stands as king of the hill!
19:26:51 <ais523> oh right, the hill's still up
19:29:49 <Gregor> The Haskell wiki is not written in Haskell.
19:29:53 <Gregor> WHAT DOES THAT SAY ABOUT HASKELL?
19:30:08 <Gregor> Perhaps that using it to write super-stateful things such as wikis is a form of torture?
19:30:12 <pikhq> That MediaWiki is awesome.
19:30:18 <alise> <pikhq> That MediaWiki is awesome.
19:30:28 <impomatic> Gregor: I'll add FYB at some point. Unless you want to add it?
19:30:29 <alise> Gregor: there are many wikis in haskell, but if they had written one in haskell
19:31:00 <Phantom_Hoover> Anyway, it is written in Haskell, but the code is so lazy it just calls MW.
19:31:16 <pikhq> Gregor: It's not that painful in Haskell. It's just that you should probably either abstract or use a different language.
19:32:13 <alise> <pikhq> Gregor: It's not that painful in Haskell. It's just that you should probably either abstract or use a different language.
19:32:16 <alise> Rather than write a wiki?
19:32:20 <alise> I don't understand
19:32:31 <pikhq> alise: No, I mean "for writing super-stateful things".
19:35:05 <alise> pikhq: so wikit is tclkit?
19:35:07 <alise> that seems chaeting
19:35:27 <alise> is tclkit just a tcl interpreter bundle?
19:36:07 * alise wonders what TclVfs is
19:36:22 <alise> pikhq: You know of Rebol?
19:37:03 <pikhq> alise: Tclkit is just a Tcl interpreter bundle.
19:37:18 <ais523> Gregor: I'm having trouble uploading my breaking FYB program to a pastebin, it's around a megabyte long
19:37:31 <impomatic> I've been playing with activestate Tcl.
19:37:59 <alise> pikhq: Rebol is quite cool. Similar to Tcl in a way.
19:37:59 <ais523> egobot wouldn't be able to read it, would it?
19:38:14 <alise> it's still sent in the http body
19:38:25 <ais523> impomatic: yes, but it uses ! for commands
19:38:29 <alise> i bet it ignores the attachment header
19:38:31 <pikhq> alise: TclVfs is the Tcl virtual filesystem layer.
19:38:41 <pikhq> alise: TclKit can include a filesystem image bundled with it, mount via TclVfs, and voila -- single-file distribution of your program.
19:38:44 <alise> pikhq: That I had already gathered.
19:38:58 <alise> pikhq: Does Wikit do that?
19:39:04 <alise> If so: totally cheating.
19:39:07 <alise> pikhq: You said single-file. :P
19:39:23 <pikhq> That said single file is an archive is beside the point. :P
19:39:40 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
19:39:42 <pikhq> I'm pretty sure it's only using the archive for... Archiving the wiki itself, though.
19:39:43 <alise> pikhq: So it's actually like ten files, I bet. :P
19:39:50 <EgoBot> 61 +++++++++[>+>+++++++++++++>+++++++++++><<<<-]>>-.>++.<-.+.<+. [73]
19:39:55 <ais523> impomatic: it's working
19:40:34 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957059
19:40:46 <ais523> I reduced it to 140000 bytes, should still be enough to beat existing programs
19:41:00 <ais523> (there are numerous ways to improve it, this is just a tech demo)
19:41:15 <ais523> if it loses, it's because I've screwed up the execution somewhere as I've never written FYB
19:41:47 <alise> ais523: why the reluctance to try filebin.ca?
19:42:01 <pikhq> alise: Oh, it's more than 1 file in the archive, but only because it can *also* run as a CGI script or as a Tk program for browsing said wiki.
19:42:08 <ais523> because I was already working on pastebin.ca at the time
19:42:17 <ais523> and besides, I'm not sure how efficient the FYB interp is
19:42:38 <pikhq> alise: Hey, it's pretty spiffy to be able to run the wiki on localhost and just browse it.
19:42:42 <EgoBot> Score for ais523_lightspeed1: 0.0
19:42:42 <Gregor> It shouldn't be this slow, this is weird ...
19:42:52 <ais523> and shouldn't do that badly, either
19:43:06 <ais523> perhaps I misinterpreted what [] does while defected?
19:43:09 <Gregor> You have to actually WIN
19:43:12 <ais523> the spec's rather unclear
19:43:17 <ais523> Gregor: I know, there's a loop at the end that should win
19:43:19 <alise> pikhq: BAH YOU TCL GUYS MAKING GOOD CODE
19:43:26 <ais523> by NOPing the entire enemy program, then replacing it with bombs
19:43:35 <Gregor> wtf, why isn't the report posted ...
19:43:37 <ais523> hmm, theory, it's NOPing the bombs as it lays them
19:43:39 <alise> I bet EgoBot gave up before reaching it
19:44:21 <Gregor> 17435 codu 20 0 320m 97m 6172 S 91.3 9.6 0:38.15 trac
19:44:27 <Gregor> Why are you taking 99% CPU, Trac
19:45:02 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957069
19:45:03 <Gregor> Why must Trac always be the bane of my existence?
19:45:08 <EgoBot> Score for ais523_lightspeed1: 0.0
19:46:10 <ais523> Gregor: ah, my program's just drawing with every other program
19:46:12 <ais523> due to the interp timing out
19:46:21 <ais523> it's working perfectly, just the interp doesn't handle the brilliance of my program
19:46:27 -!- Deewiant has quit (Quit: Be right back.).
19:46:29 <Gregor> Interp timing out? It gives you 1 million iterations or something.
19:46:32 <ais523> Phantom_Hoover: it just moves the IP faster than the enemy's pointer's speed of light
19:46:46 -!- Deewiant has joined.
19:46:54 <ais523> by using a [] loop with the current element set to 0
19:48:50 * ais523 replaces the kill stuff at the end with the simple bomber example from the README
19:49:05 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957070
19:49:45 <EgoBot> Score for ais523_lightspeed1: 0.0
19:50:25 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957070
19:50:29 <EgoBot> Score for ais523_lightspeed1: 0.0
19:50:30 <ais523> I think EgoBot just doesn't handle the program
19:51:12 <ais523> still, this doesn't mean that the /principle/ is broken
19:51:26 <Gregor> There's a super-secret program length limit apparently :P
19:52:15 <Gregor> Maaaan I was a shitty coder back in '05 X-D
19:52:30 <Gregor> Make it 32k or less and see what happens :P
19:52:55 <Gregor> (I should rewrite this ... )
19:54:07 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957076
19:54:11 <EgoBot> Score for ais523_lightspeed1: 0.0
19:54:27 <ais523> report.txt blank again
19:54:28 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957076
19:54:32 <EgoBot> Score for ais523_lightspeed1: 0.0
19:54:41 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957076
19:54:47 <EgoBot> Score for ais523_lightspeed1: 0.0
19:54:59 <ais523> and again, I give up trying to figure out what's going on here
19:55:07 <ais523> that's only 31 and a bit K...
19:55:25 <Gregor> That's not the issue any more.
19:55:31 <Gregor> Wait for the report to finish generating.
19:56:21 <Gregor> It's not actually doing anything it seems :P
19:56:26 <Gregor> Crufty piece o' garbage
19:57:23 <EgoBot> Score for Gregor_nothing: 23.1
19:57:23 <Gregor> I don't understand why the BFJoust hill is so stable and FYB isn't, when they're basically the same software ...
19:57:36 <ais523> maybe it's the interp itself
19:57:43 <alise> Gregor: Did you rewrite the FYB hill?
19:57:46 <Gregor> 3 | 0 + 0 + 0 + - - + - 0 - - + 0 | 25.3 | 0 | ais523_lightspeed1.fyb
19:57:48 <alise> Or is the Joust one based on the FYB one?
19:58:17 <ais523> Gregor: clearly 32k isn't enough for complete immunity from everything
19:58:21 <Gregor> alise: The Joust one is based on the FYB one. Probably improved. If anybody cared I'd fix it though :P
19:58:30 <ais523> given that I used a /very/ slow but relatively sure method of bombing the opponent
19:58:33 <Gregor> ais523: Good excuse there, Mr. My-Strategy-Doesn't-Work-Waaaah :P
19:58:48 <alise> ais523: why not run it locally?
19:58:54 <ais523> Gregor: the strategy can be mathematically proven to work, given a long enough size advantage over the opponent
19:59:00 <ais523> alise: because then Gregor would never believe me
19:59:29 <Gregor> I just don't want FYB to be broken X-P
20:01:06 <ais523> btw, a long string of NOPs at the start of the program works just as well, possibly even more effectively
20:01:30 <ais523> because unless the opponent starts with an equally long string of >s, you outspeed them by moving at lightspeed while the opponent has to do some sort of logic
20:03:01 * Gregor is presently trying to figure out why logicex-2 wins.
20:03:19 <alise> my satiation counter wraps around
20:03:22 <alise> $:909 T:808 S:0 HP:100 (345)
20:03:24 <alise> after i ate some food
20:03:30 <alise> good thing i have all those potions
20:03:40 <ais523> is there a potion counter?
20:03:42 <alise> (that's 345 hp i can add on to mine, 20 per turn)
20:03:47 <Vorpal> how can anyone possibly manage worse manual translation to English than the automated translation of English with google translate...
20:03:47 <alise> ais523: yes, after the HP count in ()s
20:03:53 <alise> q moves 20 from that to the hp
20:04:10 <ais523> exactly the same notation as NetHack, entirely different meaning
20:04:11 <alise> didn't take a potion first turn
20:04:23 <Phantom_Hoover> "<ais523> Gregor: the strategy can be mathematically proven to work, given a long enough size advantage over the opponent"
20:04:32 <alise> ais523: it'll probably be HP:now/max (potion) if i add a max hp
20:04:36 <Phantom_Hoover> "Moshe Sipper invented non-uniform cellular automata, which are cellular automata in which the CA-rules are local to each cell and can be copied onto neighbouring cells. Non-uniform CA rules can be designed to model conservation of mass."
20:04:38 <alise> if L<1:Q('You die...',1);D();endwin();print'You survived for '+N+' turns and had $'+G+' when you died.';exit()
20:04:38 <alise> TypeError: cannot concatenate 'str' and 'int' objects
20:05:00 <ais523> you can do that in Java...
20:05:15 <ais523> and if even /Java/ lets you do something, it must be pretty nonesoteric
20:05:17 * Vorpal invents a silly addition function.
20:06:12 <fizzie> Is the Python % operator still the way to do formatted output, or did they do something different there? (I mean, I know it works, but if it's somehow unclean.)
20:06:24 <alise> if L<1:Q('You die...',1);D();endwin();print'You survived for %s turns and had $%%s when you died.'%N%G;exit()
20:06:29 <alise> It's shorter than %(N,G)!
20:07:24 <Sgeo> Golfing in Python?
20:07:46 <alise> Sgeo: i will now give you a heart attack:
20:07:50 <alise> W=[32]*1000+[36,81]*5+[37]*3+[35]*50+[33]
20:07:52 <alise> q=w[v];s.addch(B-Y,A-X,choice(W)if U and 0==r(0,2)and q-32 else q)
20:07:53 <alise> s.addstr(22,0,'_'*80);C('$:%-17s T:%-17s S:%-17s HP:%-3s (%s)'%(G,N,S,L,P));global V;V();V=lambda:1;s.move(y-Y+11,x-X+40)
20:08:18 <alise> w.update(q);U+=U>0;N+=1;S-=S>0;L%=301;L-=25*(L and S<1)
20:08:22 <alise> w[y,x]=32;x+=a;y+=b;w[y,x]=64;X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y);D()
20:08:28 <alise> a=(k in'lun')-(k in'hyb');b=(k in'jbn')-(k in'kyu')
20:08:30 <Sgeo> Oh, fun fact: I posted a link to something. Warrigal posted on Reddit. Warrigal got almost 1k karma from it
20:08:33 <alise> if P and L+20<301:q=min(P,20);L+=q;U+=q*3*(U>0);P-=q;T()
20:08:50 <alise> Sgeo: let me guess, you're angry at him now.
20:08:57 <alise> and yes, it's that obvious
20:09:01 <ais523> I'd be happy for him if that happened
20:09:10 <Sgeo> Not _actually_ angry. Joking angry
20:09:11 <ais523> alise: where's the redraw in that?
20:09:26 <Sgeo> And angry at myself for not thinking of posting it myself
20:09:33 <alise> ais523: that's not the whole code, just particularly abhorrent snippets
20:09:38 <Vorpal> Render the addition as an image, the usual way humans write such downs (numbers above each other and such). Font should be Comic Sans if that is available on the system. Now do OCR on the image for each column, writing the result to the relevant place below that column. If you get a carry, update the image above the next column as usual. Continue until done. Then read the result line back using OCR.
20:09:41 <yorick> alise: that's horrible :P
20:09:48 <alise> yorick: that's awesome.
20:10:03 <Vorpal> alise, cool code above :P
20:10:08 <alise> Sgeo: you mean the kind of joking angry that means actually angry? :p
20:10:30 <Vorpal> alise, is there any IOPCC?
20:10:31 <ais523> I like using nonalphabetic variables in Perl so you don't need to use a space between them and a keyword
20:11:11 <yorick> wait...that's actually python?
20:11:27 * Sgeo misread ais523 as zzo38
20:11:30 <Vorpal> alise, why does this need to be two lines:
20:11:34 <Vorpal> <alise> w[y,x]=32;x+=a;y+=b;w[y,x]=64;X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y);D()
20:11:34 <Vorpal> <alise> a=(k in'lun')-(k in'hyb');b=(k in'jbn')-(k in'kyu')
20:11:47 <Sgeo> Nick colors, I think
20:11:48 <ais523> Vorpal: they aren't consecutive, is my guess
20:12:15 <alise> yorick: yup, python
20:12:19 <Vorpal> alise, ah. What remains to be written?
20:12:26 <ais523> I still think you should use thin-spaces to golf it even further
20:12:38 <alise> Vorpal: it has the need to eat, turn count, money, HP, potions, walls, full movement...
20:12:40 <Vorpal> ais523, I doubt python will parse that?
20:12:48 <alise> Vorpal: to actually be fun in a sense just requires a little bit of dumb AI code
20:13:03 * yorick should create an irssi script that will color the nicks the way I want
20:13:04 <Vorpal> alise, and the TARDIS?
20:13:11 <alise> even now it's fun to play and see if the RNG hates you (i.e. see how long you can survive; food doesn't last long)
20:13:13 <alise> (and rack up cash)
20:13:25 <alise> Vorpal: not at that stage yet
20:13:39 <alise> Vorpal: you can have the current code if you want
20:13:45 <cpressey> yorick: you can script this thing? bitchen
20:13:46 <alise> oh yeah it has hallucination too
20:13:49 <Vorpal> alise, also about the food, surely if the world is infinite it shouldn't be impossible to find more?
20:13:50 <alise> 1/15 chance of bad food
20:13:53 <Gregor> laaaaaaaaaaaaaaaaaaawl
20:13:58 <Gregor> ais523: You know there is no '-', right?
20:14:04 <alise> Vorpal: you have limited satiation, though
20:14:12 <yorick> cpressey: I probably could, but I hate perl
20:14:14 <alise> Vorpal: so can you survive long enough to get to it?
20:14:17 <alise> (yes, almost always)
20:14:20 <ais523> should have been [+], I wasn't thinking
20:14:22 <alise> (but still, it's easy to lose track)
20:14:24 <ais523> (it doesn't matter for that program)
20:14:29 <alise> Vorpal: max. 300 turns before you lose 50 hp/turn
20:14:35 <alise> (maximum satiation is 300)
20:14:42 <Gregor> ais523: Uhh, yes it does, you'll be in a tight worthless loop.
20:14:42 <alise> and food doesn't always replenish it fully
20:14:45 <Vorpal> alise, ouch, that is a bit too nasty
20:14:56 <ais523> Gregor: I mean, it doesn't matter whether it's + or -
20:14:57 <yorick> (if you convert a random wav file to text, you get executable perl)
20:15:01 <ais523> (nor how tight the loop is, using that strategy)
20:15:03 <alise> Vorpal: that's why you keep your HP up so that once you lose 50 you have enough time to get to food :)
20:15:12 <Gregor> ais523: Right, so fix it :P
20:15:24 <Vorpal> alise, can't you carry food?
20:15:26 <alise> guh, my HP wrapped arround to 4
20:15:36 <alise> Vorpal: nope, i avoid state like the plague :D
20:15:41 <alise> Vorpal: most things are done directly on the grid
20:15:45 <alise> for instance there is no model behind the object grid
20:15:50 <alise> walking onto something makes it disappear
20:16:00 <alise> and monsters are moved by actually looking at the visible screen and moving them around
20:16:03 <Vorpal> alise, it isn't just 'EM or 'ELM that is needed, it is 'EILM
20:16:16 <alise> slav, Super Lotsa Additions Vagrant
20:16:30 <cpressey> Phantom_Hoover: So non-uniform cellular automata are... neighbouring state machines.
20:16:35 <Vorpal> alise, how does hallu work then?
20:16:42 <Vorpal> alise, it permanently changes the grid?
20:16:44 <alise> "L%=301" fun fact, this does not impose a max health!
20:17:05 <cpressey> just make the playfield layout non-uniform too, and it's... a big state machine.
20:17:06 <alise> no, it makes you drop health rapidly if you get too much :P
20:17:14 <alise> i now have 4 hp due to no fault of my own
20:17:18 <cpressey> Phantom_Hoover: feels like a decay of theory
20:17:46 <Vorpal> alise, evil, surely you can make L=min(L,300) shorter somehow?
20:17:53 <cpressey> Phantom_Hoover: part of what makes them special is their uniform structure
20:18:04 <alise> Vorpal: i'm just not doing it, i'm sure i can check L<301 elsewhere
20:18:13 <alise> Vorpal: L-=25*(L and S<1)
20:18:14 <Phantom_Hoover> cpressey, well, you could simulate it with loads and loads of states.
20:18:19 <Vorpal> alise, as for the playfield, does it remember previous screens?
20:18:20 <alise> that is, "if L>0 and S==0, decrease L by 25"
20:18:30 <Vorpal> so the playfield grows over time
20:18:31 <alise> if we are hallucinating, increase the hallucination counter
20:18:37 <alise> explore too much and you run out of memory
20:18:42 <Phantom_Hoover> Incidentally, mechanical self-replicators are the coolest things ever.
20:18:53 <alise> If we have any satiation, lose one point of it.
20:19:09 <alise> "X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y)" <-- figure out what this does (not having been there when we worked it out) and get a prize
20:19:38 <alise> it scrolls the playfield
20:19:40 <Vorpal> <alise> "S-=S>0" <--- err
20:19:57 <alise> Vorpal: (S>0) = True equiv-to 1
20:20:01 <alise> (S>0) = False equiv-to 0
20:20:04 <alise> S -= 1 decreases S
20:20:05 <cpressey> and i ruined Phantom_Hoover's chance!
20:20:06 <alise> S -= 0 does nothing
20:20:28 <alise> my hallu is so pitiful, you can just keep telling the game to redraw and it works :D
20:20:36 <alise> now what would be cool is hallu that actually permanently fixed the map!
20:20:40 <Vorpal> <Phantom_Hoover> Incidentally, mechanical self-replicators are the coolest things ever. <-- what about biological ones?
20:20:43 <cpressey> alise: that's how hallu works in real life too
20:20:54 <cpressey> you just keep redrawing and you can tell what's real
20:20:59 <alise> without taking turns
20:21:01 <alise> and it re-hallucinates
20:21:07 <alise> meaning if you hold down enter it's easy to see what's what
20:21:17 <alise> <Vorpal> <Phantom_Hoover> Incidentally, mechanical self-replicators are the coolest things ever. <-- what about biological ones?
20:21:25 <alise> well i don't know if mechanical self-replicators enjoy replicating as much...
20:21:25 <Phantom_Hoover> Vorpal, http://www.cogs.susx.ac.uk/users/ctf20/dphil_2005/Thesis/Chapter1/Chapter1Figs/breivik2001.pdf
20:21:41 <alise> Oh yeah baby, feed me that molten metal so that I can assemble a copy of myself.
20:21:53 <alise> Yes, you totally assemble computers with MOLTEN METAL
20:21:57 <Vorpal> Phantom_Hoover, has any been built btw?
20:22:01 <alise> Also I suppose that's a bit of a role reversal there.
20:22:15 <Vorpal> Phantom_Hoover, give me server ip!
20:22:20 <alise> hee, i think i'm a wall
20:22:56 <cpressey> so a mechanical replicator would need to be "smart" enough to assemble a copy of itself -- so it would need a relatively powerful computer -- so it would need to build integrated circuits
20:23:05 <cpressey> which is a bit fucking nontrivial
20:23:08 <Vorpal> Phantom_Hoover, host presumably
20:23:27 <cpressey> i could see it not needing a trult integrated circuit
20:23:43 <Vorpal> lets hope it isn't using vhosts
20:24:05 <Phantom_Hoover> It works with lots of little plastic nucleotides, though, not true self-replication.
20:24:14 <ais523> cpressey: relay logic might work
20:24:24 <ais523> several telephone exchanges work on that basis
20:24:40 <Vorpal> cpressey, it depends on what you define as the raw material
20:24:58 <Vorpal> cpressey, imagine an universe where stuff like ARM cpus occur naturally :P
20:25:28 <Phantom_Hoover> In this case, the raw material is elements designed to stick together magnetically in an appropriate way.
20:25:31 <Vorpal> oh and li-ion batteries are found in certain geological formations
20:26:58 <cpressey> SAND. all you have to start is SAND.
20:27:06 <Phantom_Hoover> Penrose did a similar but less sophisticated system with 2 elements in 1 dimension.
20:27:45 <cpressey> Hard to make motors out of just sand.
20:27:50 <Phantom_Hoover> Given an infinite series of A B... a single cluster will tend to self-replicate.
20:28:17 -!- impomatic has left (?).
20:28:37 <alise> cpressey: MAKE IRON OUT OF SAND
20:29:31 <alise> uh oh, i just ate rotten food when already hallucinating
20:29:42 <alise> although in some cases that can actually cure you.
20:29:45 <cpressey> alise: RADIOACTIVE MOBILE CHIP FACTORY
20:29:53 <alise> in fact, in all cases.
20:30:32 <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957097
20:30:37 <EgoBot> Score for ais523_lightspeed1: 25.3
20:30:39 <ais523> sorry, closed the tab rather than submitting it for some reason
20:30:57 <ais523> Gregor: how do you get the report to come up?
20:31:05 <ais523> cpressey: that's the same score it got last time, I think
20:31:07 <ais523> the scoring thing is broken
20:31:21 <Gregor> ais523: Idonno, I just submitted something else then it worked that time :P
20:31:28 <EgoBot> Score for Gregor_nothing: 23.1
20:31:34 <Gregor> ais523: I'll have to debug that tonight.
20:32:19 <alise> quick, what's the range of printable ascii?
20:32:28 <Vorpal> <ais523> !fyb lightspeed1 http://pastebin.ca/raw/1957097 <-- what was % now again?
20:32:51 <alise> q=w[v];s.addch(B-Y,A-X,choice(W)if U and 0==r(0,2)and q-32 else q)
20:32:57 <Gregor> ais523: Ohhhh, it seems to get confused if you use a name that's already in use X_X
20:33:01 <ais523> Vorpal: defect (start analysing your own program)
20:33:03 <alise> q=w[v];s.addch(B-Y,A-X,r(32,126)if U and 0==r(0,2)and q-32 else q)
20:33:05 <alise> That's more like it.
20:33:08 <alise> ais523: is 126 valid too?
20:33:13 <ais523> !fyb lightspeed1a http://pastebin.ca/raw/1957097
20:33:22 <alise> ais523: valid as in printable. righ
20:33:24 <ais523> 127 is a control char, though
20:33:46 <Gregor> ais523: I'll have to look into this weirdness :P
20:33:47 <EgoBot> Score for ais523_lightspeed1a: 21.1
20:34:11 <cpressey> mzstorkipiwanbot: So eventually you'll support four variable scopes: message scope, nick scope, channel scope, and global variables.
20:34:14 <Vorpal> ais523, why resubmit the same thing?
20:34:22 <cpressey> I guess I could even have a server scope, but...
20:34:38 <fizzie> Lightspeedia, Wikipeedia.
20:34:55 <Gregor> ais523: I'll fix the borkitude on Saturdayish ...
20:35:03 <Vorpal> ais523, scanned last screensful, saw nothing
20:35:07 <cpressey> Lightspeedia is an encyclopedia made entirely out of light.
20:35:35 <ais523> Vorpal: <Gregor> ais523: Ohhhh, it seems to get confused if you use a name that's already in use X_X
20:37:51 <cpressey> Phantom_Hoover: since this morning (my time)
20:38:33 <cpressey> and its nick should really be "mzstorkipiwanbotbotbot" but that's too long
20:40:06 * alise has been hallu for the past N minutes
20:40:25 <alise> cpressey: Wait, are you actually doing that IRC-language thing?
20:40:59 <Gregor> Wait, why am I back in crappy Colloquy anyway ...
20:41:02 <alise> $:2689 T:2307 S:252 HP:300 (380)
20:41:38 <Gregor> cpressey: Will it have MEMORY SAFETY WITHOUT GC?
20:41:56 <cpressey> i need to rewrite what i have so far of the bot in some language less unappealing to me than python
20:41:59 <alise> Memory safety without GC: No free(), all memory leaks
20:42:26 <cpressey> Gregor: no. it will be unsafe in the sense that anyone could come along and change your global variables anyway.
20:42:35 <Gregor> alise: Apparently you missed cpressey's and my big argument about this, resulting in Eightebed which does fit my requirements but is so terrible :P
20:42:38 <alise> HOW AM I LOSING HP
20:42:47 <alise> Gregor: I did not miss it
20:42:55 <alise> Gregor: I named Eightebed
20:43:12 <Vorpal> cpressey, what irc language thingy?
20:43:18 <Vorpal> hm dns seems to work again
20:43:21 <alise> Gregor: http://catseye.tc/projects/eightebed/doc/website_eightebed.html; /Legal Issues/
20:43:40 <alise> "To have a literate specification written in SUPER ITALIAN, thus giving all programs the power of UNMATCHED PROPHETIC SNEEZING."
20:44:21 <Sgeo> Does reference counting count as GC?
20:45:00 <Gregor> Sgeo: It doesn't fit the requirements I set out because it leaks.
20:45:00 * Sgeo goes to try Samorost
20:45:16 <Gregor> Regardless of whether it counts as GC or not.
20:45:32 <cpressey> Gregor: unless your language forbids circular structures.
20:45:37 <Gregor> Alas, poor yorick, whoTF is this guy.
20:45:45 <Gregor> cpressey: Then your language is made of FAIL.
20:45:54 <fizzie> Perl folks call what they have a GC, and it's reference-counting. (With a "real" GC sweep on interpreter termination.)
20:45:59 <alise> $:3151 T:2859 S:291 HP:286 (324)
20:46:00 <cpressey> Gregor: Then all functional languages are made of FAIL.
20:46:08 <alise> I'll die now, since the game is insufferably boring.
20:46:22 <cpressey> fizzie: Wait, what? Why GC at the end? The OS will reclaim it all anyway!
20:46:23 <Sgeo> alise, the game you made?
20:46:28 <Gregor> cpressey: Yes, they are X-P
20:46:35 <alise> but i agree, circular structures are iffy
20:46:50 <alise> i read some stuff about a language lacking them, was interesting.
20:47:12 <fizzie> cpressey: The two official motivations I've seen are (a) making sure all finalizers get called at least once, and (b) for the case where the interpreter is embedded in some other process that's not going to terminate that point, and might in fact restart it again.
20:47:14 <cpressey> alise: OK, maybe not *all* functional languages, depending on implementation details etc, but you catch my meaning
20:47:15 <yorick> omg Gregor doesn't know me!
20:47:48 <cpressey> fizzie: ok, that's fair i suppose
20:48:18 <cpressey> yorick: nor do i, except that you've been here the past few days
20:48:27 <yorick> which is about all you could know about me
20:48:34 <yorick> I've been here since saturday
20:49:23 <cpressey> yorick: so what's your story, Dark Stranger who Wandered into Town Recently?
20:49:24 -!- tombom_ has joined.
20:49:28 <fizzie> cpressey: Oh, and also because Perl's "ithreads" mean each thread gets a completely separate copy of the whole interpreter, and they want those not to leak either.
20:49:35 -!- tombom_ has quit (Changing host).
20:49:35 -!- tombom_ has joined.
20:49:52 <yorick> cpressey: I knew about this channel for some time, but never bothered to join...and then I got an assignment that got brainfuck
20:49:56 <fizzie> "A more complete garbage collection strategy will be implemented at a future date."
20:50:03 <Gregor> What's really incredible about refcounting is that it's usually slower than even a naive mark-and-sweep :P
20:50:07 <yorick> and then I saw the beauty if this channel (saw beyond Vorpal)
20:50:36 <yorick> and now I'm idling here
20:50:36 <Gregor> alise: Looks like you've got a Vorpal-hatred buddy X-P
20:50:37 <cpressey> Gregor: but responsiveness is more predictable (go the arguments)
20:51:03 <Gregor> cpressey: Eh, fair enough.
20:51:27 <yorick> Vorpal: you scare new people away
20:51:46 -!- Wamanuz2 has joined.
20:52:22 <Vorpal> <alise> i read some stuff about a language lacking them, was interesting. <-- um, lots of functional languages lack circular structures
20:52:55 <cpressey> Gregor: except it's not really true. i unlink a structure -- i have to check if it's an orphan -- i have to traverse it -- oh wait, how big is this thing?
20:53:02 <cpressey> excpet i think there might be tricks around that
20:53:19 <alise> "(With a "real" GC sweep on interpreter termination.)" ...why? For embedding?
20:53:21 <yorick> Gregor: especially against Vorpal eh
20:53:33 <Vorpal> alise, scheme without setf! and similar wouldn't allow circular structures I think?
20:53:40 <Gregor> yorick: No, I feed on hatred in general.
20:53:42 <cpressey> isomorphic to needing a sweep phase, i guess
20:54:01 <pikhq> Vorpal: And no optional laziness.
20:54:02 <Vorpal> <alise> "(With a "real" GC sweep on interpreter termination.)" ...why? For embedding? <-- um look above
20:54:12 <cpressey> alise: yes, he said embedding. and finalizing.
20:54:31 <Vorpal> pikhq, I'm pretty sure you can't get circular structures in erlang btw
20:55:20 <Vorpal> cpressey, I'm not familiar with that one. One of yours?
20:55:39 <cpressey> Vorpal: it's Scheme minus Scheme
20:55:49 <Vorpal> cpressey, the null set?
20:56:13 <cpressey> Vorpal: no, more like Garfield minus Garfield
20:56:30 <cpressey> Vorpal: YOU SEE HOW THE ONE WORD CAN HAVE DIFFERENT MEANINGS
20:56:40 <olsner> hmm, there's an interesting paradigm, every input is an incorrect program
20:57:04 <Vorpal> cpressey, indeed. I admit I'm not 100% sure how to interpret the meanings in this case though
20:57:36 <olsner> which would be the effect if the set of correct programs is the null set
20:57:54 * pikhq found his in-C SKI interpreter
20:57:57 <alise> ehird@dinky:~/Code/vagrant$ python debug.py
20:57:57 <alise> You survived for 3176 turns and had $3505 when you died.
20:57:58 <pikhq> Good God that's crazy.
20:58:28 <Vorpal> olsner, depends on what you do on "incorrect" data. For brainfuck-brainfuck every program would be a nop
20:58:29 <alise> More money than turns!
20:58:33 <pikhq> I had forgotten how mad this code was.
20:58:52 <Vorpal> olsner, unless the definition of what you do on incorrect data is part of the language that got substracted away
20:58:55 <olsner> Vorpal: well, the interpreter should print an error and abort
20:59:10 <pikhq> Especially where I copy a thunk before it's evaluated so that I can do an infinite list of input.
20:59:15 <Vorpal> olsner, in which case all programming languages minus themselves are the same programming language
20:59:16 <olsner> supposedly any computation would have to be in controlling the kind of error you get
20:59:37 <Vorpal> olsner, "how do we make this TC" ;)
21:00:02 <olsner> how to find TC-ness or making something TC is all I ever think about
21:00:41 <olsner> pikhq: sounds awesome... is this on the internet?
21:01:18 -!- alise_ has joined.
21:01:48 <pikhq> olsner: Lemme tar it up.
21:01:55 <olsner> for instance: do type-level calculation, cause a type mismatch, the "expected foo, found <...>" prints the output you've produced
21:02:16 <olsner> but that would be a slightly boring way to do it
21:02:30 <alise_> 12:46:22 <cpressey> fizzie: Wait, what? Why GC at the end? The OS will reclaim it all anyway!
21:02:35 <alise_> holy shit i missed this and everything after
21:02:38 <olsner> (oh, and the point would be that you have to make the type system so that there are no correct programs)
21:02:55 <olsner> I don't quite know how to do that
21:03:15 <alise_> 12:50:36 <Gregor> alise: Looks like you've got a Vorpal-hatred buddy X-P
21:03:15 <pikhq> I really, really need to get a webhost.
21:03:28 <olsner> pikhq: just put it on github?
21:03:37 <olsner> it works for pages and code anyway
21:04:36 <alise_> <alise> min(L+min(P,20),300)
21:04:36 <alise_> <alise> min(L+P,L+20,300)
21:04:36 <alise_> <alise> q=min(P,20);L=min(L+q,300)
21:04:36 <alise_> <alise> if P and L<301:U=min(U+q*3*(U>0),300);P-=q;T()
21:04:38 <alise_> <alise> Prizes for golfing this.
21:04:40 <alise_> <alise> Since when is Dwarf Fortress available for Linux?
21:04:57 <olsner> speaking of git: has anything been done in the field of esoteric version control?
21:05:23 <alise_> 13:00:10 <Vorpal> alise, debug.py?
21:05:27 <alise_> vagrant's debug launcher
21:05:39 <alise_> olsner: i think ais523 had some thoughts
21:05:43 <alise_> i have some thoughts too
21:05:47 <alise_> olsner: as for existing examples
21:05:58 <pikhq> olsner: No. Wants host.
21:06:00 <ais523> alise_: I have, but haven't got much futher
21:06:08 <ais523> also, it was a surprisingly non-eso project
21:06:22 <alise_> olsner: http://www.gnu.org/software/gnu-arch/tutorial/index.html
21:06:30 <alise_> or http://www.gnu.org/software/gnu-arch/tutorial-old/arch.html
21:06:37 <alise_> just http://www.gnu.org/software/gnu-arch/tutorial-old/arch.html
21:06:40 <alise_> is the complete, crazy one
21:06:47 <ais523> but I started with the concept of "version control system which can manage any command in any other common DVCS as a special case of one or more of its commands + a UI" and "only has five commands"
21:06:53 <ais523> and then tried to make things fit from there
21:07:01 <alise_> ais523: that's how git is designed
21:07:08 <alise_> except it's more like 10 commands i think
21:07:11 <ais523> yes, but git fails at it
21:07:14 <ais523> the 5 is very important
21:07:20 <alise_> that *exact* number? :P
21:07:25 <ais523> also, it fails at doing some things that, say, darcs does trivially
21:07:28 <ais523> and yes, that *exact* number
21:08:30 <olsner> minimalistic version control, I like the sound of it
21:08:37 -!- tombom has quit (*.net *.split).
21:08:38 -!- alise has quit (*.net *.split).
21:08:38 -!- Wamanuz has quit (*.net *.split).
21:08:53 <ais523> look, you can set arbitrary goals for an esolang, right?
21:08:57 <ais523> why can't you set them for other programs?
21:09:11 <olsner> maybe a place to start is to figure out the counterpart of turing complete for version control
21:09:27 <yorick> esoteric version control!
21:09:45 <ais523> alise_: I do have an idea of the model to use, though, and a name
21:09:50 <ais523> the name is... scapegoat
21:10:26 <ais523> alise_: <tswett> Who's ə? Me?
21:10:34 <ais523> ^ what is wrong with Agoran nicknames
21:10:35 <Gregor> Somebody needs to make a 2D competitive programming languages.
21:10:51 <ais523> someone was talking about BeYourFunge, or something like that
21:11:03 <Gregor> Awww, they're naming it after mine.
21:11:40 <alise_> Gregor: I did BeYourFunge, I think
21:11:43 <alise_> BeYorFunge it was I think
21:12:07 <Vorpal> alise_, ah ok (wrt debug.py)
21:12:19 <Gregor> Actually, somebody needs to make a 3D competitive programming ENVIRONMENT in which the programs can be written in any arbitrary language, and there's some overarching laws of physics to dictate how one might find another :P
21:12:25 <alise_> Vorpal: quick, if i give you the latest version will you see how long you can survive?!:!?!?!?!
21:12:29 <Gregor> In other words, we need a space-flight simulator where people write captain programs.
21:13:05 <alise_> > > > I lower ehird's position on the list by 1, for a fee.
21:13:42 <alise_> ais523: please tell me your model has deduplicative storage as a separate, non-main part
21:13:52 <alise_> (deduplicative = store diffs, not the entire tree, basically, for every revision)
21:13:56 <ais523> alise_: it's an implementation detail
21:14:11 <alise_> well, i'm talking details
21:14:12 <ais523> but the model is more easily implemented by storing diffs, than by storing trees
21:14:25 <ais523> doing it via trees would be awfully complex, except for caching
21:17:05 <alise_> how does it work, then?
21:17:26 <ais523> OK, rough explanation because a) this isn't written down anywhere, and b) I worked it out entirely in my head
21:17:37 <ais523> but basically, each line ever added to anything has its own unique ID
21:17:38 -!- alise_ has changed nick to alise.
21:17:44 <alise> ais523: line? how ...
21:18:14 <ais523> also, indentation increase/decrease counts as a special token on a line of its own, but that's more incidental to the way everything works
21:18:38 <ais523> and it's not the literal text of the line that matters, more its platonic existence
21:18:58 <alise> no, but, being line-based
21:19:07 <ais523> well, it's meant to work like diff
21:19:16 <ais523> lines give a good approximation to how people think of programs
21:19:21 <ais523> you could do it with characters, it'd work just as well
21:19:29 <Gregor> Each byte added has a 128-bit ID.
21:19:30 <ais523> but probably confuse people because it didn't do merges how they wanted
21:20:15 <ais523> now, what's stored in the version control system are diffs, vaguely darcs-like, except that the diffs have a complete context of where they go
21:20:33 <ais523> e.g. "line ab239ca0 added between line 30adb123 and line 8923ad01"
21:21:09 <ais523> they don't need any context (as in unidiff context) other than that, because each line number refers to the line itself, plus its complete history, because every line belongs to the diff where it was added with complete context of where it comes from, etc.
21:22:22 <alise> wait, you have a concept of individual object history?
21:22:26 <alise> not versioning trees?
21:22:26 <ais523> in order to figure out the current state of the version control system, you effectively start with nothing and replay all the diffs until you end up with the files (this is likely to be optimizable, but as-if rule; that's how it works /conceptually/)
21:22:31 <alise> congrats -- you've invented either RCS or CVS
21:22:33 <alise> and I'm not sure which
21:22:44 <ais523> alise: that's the bottom level of abstraction
21:22:46 <alise> ais523: that's how darcs does it
21:22:51 -!- augur has quit (Ping timeout: 240 seconds).
21:22:55 <ais523> yep, darcs was a major inspiration
21:23:01 <Gregor> alise: CVS is just some centralization and network crap on top of RCS.
21:23:03 <ais523> although I didn't do everything exactly the same way
21:23:21 <ais523> anyway, above this layer of abstraction, you can also group sets of changes into a single change that represents them all
21:23:26 <alise> Gregor: And ... multiple file support, dude.
21:23:38 <ais523> that's the svn/darcs layer of abstraction if you group once (and also does multiple files)
21:23:46 <ais523> and the tag, etc, layer of abstraction if you do it twice
21:23:53 <alise> Gregor: Well, RCS works entirely on one file with no special directory for versioning.
21:23:56 <alise> So more than kinda :P
21:24:03 <ais523> thus, individual object history, repo versions, and tags are all the same concept, really
21:24:09 <ais523> helping to keep the command count down
21:24:56 <Gregor> alise: Yeah, but it didn't actually change anything intrinsic to do that, it just has per-file data, separated properly, and maintained in "modules" which are basically directories. At the actual version-control level, it's still one-file.
21:25:31 <Gregor> SVN has actual multi-file version control, and even that is just "give all the files the same number so people don't go batshit crazy trying to find a revision"
21:31:14 <olsner> alise: I don't think ais' thing is anything like RCS or CVS
21:33:06 <olsner> in particular, both RCS and CVS work on trees of files that contain history
21:33:42 <olsner> or each file contains a tree of versions of that file
21:36:04 <alise> Thanks entirely to quintopi, I have shortened Vagrant further!
21:36:07 <alise> X+=17<abs(X-x)and cmp(x,X);Y+=5<abs(Y-y)and cmp(y,Y)
21:36:12 <alise> X-=(X-x)//17;Y-=(Y-y)//5
21:36:32 <olsner> and CVS is mostly just networked centralized storage for all those RCS files instead of storing them locally
21:39:10 -!- GreaseMonkey has joined.
21:39:36 <olsner> ... looks like Gregor already said that though
21:40:36 <alise> / instead of // works
21:40:51 <alise> ehird@dinky:~/Code/vagrant$ wc -c vagrant.py
21:43:38 <alise> GreaseMonkey: what is?
21:44:22 <alise> actually in this case it ends up as floats, it just so happens that the way i use them makes this irrelevant
21:44:28 <alise> and jesus, it's python! slow already
21:44:35 <alise> / vs // won't help that
21:44:44 <alise> microoptimising python is an amazingly silly idea
21:44:52 <Sgeo> Isn't Python faster than Ruby or something?
21:45:09 <GreaseMonkey> alise: it's fast enough to play a .mod file on a 1.44GHz 32-bit single-core AMD machine at 32000Hz in stereo
21:45:11 <Sgeo> Also, this is #esoteric. Everything that happens here is silly
21:45:12 <alise> hmm wait I guess it isn't returning a float
21:45:23 <alise> GreaseMonkey: what a useless dump of statistics
21:45:39 <Sgeo> We should microoptimize the sample of BancSTAR code that we have
21:45:46 -!- augur has joined.
21:45:51 <GreaseMonkey> source: http://pubacc.wilcox-tech.com/~greaser/stuff/pymod.py.txt
21:45:52 <ais523> how, if you don't know what it does?
21:46:07 <alise> GreaseMonkey: 32000 Hz = .000032 GHz, your computer could sleep for a bloody year and it'd still play that
21:46:12 <Sgeo> There are people in the world who DO know what it does
21:46:14 <ais523> its author claimed it wasn't leaking commercially sensitive information because nobody knew what it did or what the commands were
21:46:16 <olsner> bancstar code starts out slow, but damn does it microoptimize well
21:46:45 <Sgeo> There's no place where we can find out more about it?
21:47:03 <GreaseMonkey> alise: it has to mix 4 channels at different frequencies, it's not a .wav file silly
21:47:15 <Sgeo> We should purchase a license from... whoever it is
21:47:41 <Sgeo> Although the latter might be tricky
21:47:45 <GreaseMonkey> and it's using about 54% CPU playing cd_orbit.mod
21:51:56 -!- augur has quit (Ping timeout: 272 seconds).
21:55:08 <olsner> hmm, bancstar is obviously a list of opcode,arg1,arg2,arg3
21:55:44 <olsner> Control structures available include the 3000 ("conditional"), 3001 ("block conditional"), and 3101 ("reverse block conditional"), as well as 8500 ("GOTO") and 8550 ("combination GOTO").
21:55:57 <olsner> (the opcode goes first btw)
21:56:06 <olsner> which I already said ... d'uh
21:56:58 <olsner> a bunch of "block conditional" in that one sample
21:58:18 <olsner> http://www.oocities.com/connorbd/tarpit/bancstar.html
21:59:02 <Sgeo> Does Broadway & Seymour still exist?
21:59:49 -!- MigoMipo_ has quit (Read error: Connection reset by peer).
22:03:34 -!- augur has joined.
22:03:35 <olsner> I think that given documentation, this bancstar thing wouldn't even be very esoteric... it's just a poor source language
22:04:21 <olsner> in fact, why the heck didn't these people just write their own source language and compile it to bancstar?
22:05:43 <Phantom_Hoover> But the source language was crappy, so the developers just coded everything in BANCStar.
22:05:59 <olsner> ok, I should rephrase that... since the real source language was so crappy, why didn't they replace the source language instead of coding in the bytecode?
22:07:02 <olsner> but they obviously *could* be bothered writing this awful bytecode crap :)
22:07:26 <Sgeo> http://charlotte.bizjournals.com/charlotte/stories/2001/07/16/daily41.html
22:07:35 * Sgeo attempts to follow the trail
22:07:42 <Phantom_Hoover> http://www.youtube.com/watch?v=2YPYYvZOGlU Does anyone else think this is cool?
22:09:24 <Sgeo> http://www.linkedin.com/companies/webtone-technologies
22:09:31 <olsner> there's also this: http://www.allbusiness.com/banking-finance/banking-lending-credit-services/7277371-1.html
22:10:43 <Sgeo> Searching for WinPrism gets some book management thing
22:11:37 <Sgeo> Fidelity Information Services still exists!
22:13:28 <Sgeo> Searching the site for WinPrism reveals nothing
22:13:44 <olsner> wealth management solutions!
22:15:12 <cpressey> < Gregor> In other words, we need a space-flight simulator where people write captain programs.
22:15:23 <cpressey> So, C-Robots, in 3D, and not just C.
22:15:38 <Gregor> If that's what C-Robots is, then YES! :P
22:15:55 <cpressey> < pikhq> I had forgotten how mad this code was. <-- to what do you refer?
22:17:08 <cpressey> < alise_> "Like Feather!" <-- FeatherSCM! woo!
22:17:10 <pikhq> cpressey: SKI interpreter using lambda in C.
22:17:22 <cpressey> it'd be the only one i'd use. if it existed
22:18:32 <cpressey> < alise> microoptimising python is an amazingly silly idea <-- happens all the time here :/
22:20:08 -!- augur has quit (Ping timeout: 250 seconds).
22:20:11 <cpressey> pikhq: i think i remember you writing that. or i;m confused.
22:20:26 <pikhq> cpressey: And it's fucking nuts.
22:20:59 <cpressey> Gregor: http://en.wikipedia.org/wiki/Crobots
22:21:27 <olsner> Sgeo: one can only hope that FIS have gotten rid of this crap... then again they are still marketing a MUMPS-based thingy
22:21:41 <cpressey> ais523: if i were to design a SCM it would only work on S-Expressions
22:21:51 <Sgeo> They wouldn't have it in some storage facility somewhere?
22:22:18 <alise> lol, Sgeo is suddenly worried about bancstar being lost to the world
22:22:24 -!- Harpyon has quit (Quit: Harpyon).
22:22:35 <Phantom_Hoover> <pikhq> cpressey: SKI interpreter using lambda in C. ← Sophisticated enough to run Lazy K?
22:22:50 <cpressey> < olsner> but they obviously *could* be bothered writing this awful bytecode crap :) <-- THAT'S HOW IT WORKS IN DE BIZNESS WURLD
22:23:29 <cpressey> the source language was probably *terrible* and they didn't have the resources/justification to make a better one
22:23:47 <Sgeo> alise, hey, nostalgia isn't a factor here
22:23:52 <cpressey> (the first * is the correction *)
22:24:29 <alise> Sgeo: but you're still worried about it
22:24:45 <cpressey> terrible, like, programming language implemented by someone with no familiarity with parsing or interpretation, theory or practice
22:24:48 <alise> cpressey: i think that it was some gui tool, i may be mistaken
22:25:01 <pikhq> Phantom_Hoover: It currently implements a subset of Lazy K.
22:25:10 <pikhq> Phantom_Hoover: Namely, the SKI subset.
22:25:16 <ais523> it was the bytecode used by a GUI code-gen tool
22:25:17 <Sgeo> I want to program in BancSTAR!
22:25:24 <ais523> but people used it directly because the code-gen tool wasn't very good
22:27:17 <pikhq> ... Wait, why the heck is this segfaulting?
22:27:38 <cpressey> well, if it was built by "gui programmers" that pretty much guarantees it was AWESOME.
22:27:44 <olsner> Sgeo: they have things called BancLine and BancPac, note the spelling of 'banc'
22:28:11 <alise> "Not a game, but rather a lawsuit" --TV Tropes
22:28:15 <cpressey> they've probably switched to Plain English now
22:28:18 -!- GreaseMonkey has quit (Quit: Welcome honored guest. I got the key you want! would you like onderves. of Yourself).
22:28:33 <Sgeo> Would it be possible to ask if there's any relation to BancStar, particularly the bytecodes?
22:28:37 <alise> olsner: probably just a brand name
22:28:45 <alise> Sgeo: they probably don't know it ever existed. or deny it
22:29:45 <Phantom_Hoover> Sgeo, FFS if you want it so hard, just write something that looks identical.
22:29:57 <olsner> and this TellerPlus that looks related to those BancLine/BancPac things is apparently *customizable* (in bancstar? :D)
22:30:05 <Sgeo> But it wouldn't be genuine BancSTAR
22:30:15 <Sgeo> Just something that fits what little we know of it
22:30:36 <cpressey> Use one of those fractal interpolators to infer the rest of BancSTAR
22:31:58 <ais523> alise: which game? I can't visit TV Tropes because I'm at work
22:32:02 <ais523> and thus, can't search it
22:32:21 <alise> ais523: Capcom vs. X series
22:32:26 <alise> some street fighter ripoff
22:32:29 <alise> the quote is the only amusing part
22:33:11 <Sgeo> Numbers below 3000, or maybe 2000, might be variables
22:33:27 <Sgeo> Wait, there are NEGATIVES?
22:33:48 <olsner> Sgeo: the text mentions a limit of 2000 variables
22:34:01 <ais523> also, that the language doesn't have constants
22:34:09 <ais523> so you need a spare variable to store constants in
22:34:17 <Sgeo> Are any numbers between 2000 and 3000?
22:34:30 <Sgeo> If not, maybe those would be meaningless
22:34:32 <olsner> maybe constants are just in the code, but surrounded by if's
22:34:51 <Sgeo> How do you store constants if you can't write constants?
22:35:11 <alise> "I really wish they made a sequel to The Matrix. It would have been awesome!" --reddit
22:35:12 <ais523> presumably variables can be initialized statically, or something
22:35:17 <alise> Phantom_Hoover: it's on my todo list!
22:35:29 <ais523> Phantom_Hoover: are you sure you can use a diaeresis in that context?
22:35:37 <alise> ais523: it is correct
22:35:42 <alise> it isn't noooooooooone
22:35:50 <alise> start of a new syllable
22:35:54 <ais523> alise: according to normal pronunciation rules, yes
22:36:01 <ais523> that doesn't mean the word's actually spelled like that, though
22:36:07 <alise> Phantom_Hoover that video doesn't load for me
22:36:10 <alise> ais523: err, it's orthography
22:36:14 <alise> ais523: diareses are valid everywhere
22:36:14 <ais523> words tend not to be spelled phonetically in English
22:36:18 <alise> *no " " at the start
22:36:20 <ais523> alise: and in reverse, too?
22:36:25 <ais523> I wouldn't call Zo-e a common name
22:36:27 <olsner> oh, based on this presentation I found, I get the impression that BancLine really could be the continuation of bancstar
22:36:28 <alise> ais523: it's an accent mark
22:36:42 <olsner> it looks like the kind of forms that would be built by something like that
22:36:42 <ais523> alise: and thus part of the spelling of a word
22:36:44 <alise> "noöne" is definitely correct.
22:37:00 <ais523> I mean, I accept, say, "preëmptive" because that's actually how you spell the word
22:37:16 <alise> hey, the diaeresis mark has a name
22:37:57 <ais523> but a "you can always replace a hyphen between syllables with adjacent vowels with a diaeresis" rule doesn't seem to be correct
22:38:09 <alise> ais523: doesn't even have to be between syllables
22:38:18 <alise> although doing it everywhere it can go is annoying, it's not incorrect
22:38:22 <ais523> well, you have to pronounce the two vowels separately
22:38:28 <ais523> that's what a diaeresis means
22:38:37 <ais523> and I find it hard to see how you could do that within a single syllable
22:38:45 <alise> you don't need a hyphen
22:38:46 <cpressey> I think if you were a Greek scholar you'd spell it... n/m
22:39:11 <alise> "The designated driver concept was developed in Scandinavia over several decades beginning in the 1920s"
22:39:17 <alise> Scandinavia, always inventing newer, more modern ways to be drunk.
22:39:26 <alise> cpressey: just use TeX codes for it!
22:39:35 <alise> romanise greek text :P
22:40:56 <alise> Phantom_Hoover: it does now
22:41:52 <olsner> Sgeo: you know what you must do, you must aquire bancline and tellerplus and take it apart
22:42:13 <Sgeo> Do we in fact know that those are descended from BancSTAR?
22:43:07 <olsner> maybe you could just ask FIS if they know what happened to the code?
22:43:09 <Sgeo> At some point, I'm going to write an email to someone at FIS
22:43:25 <Ilari> Bancstar (what is known about it) looks like good base for an esolang...
22:45:43 <olsner> it's only esoteric because so little is known about it
22:46:35 <ais523> doesn't that also apply to Schrödilang?
22:47:06 <Sgeo> a.k.a. Feather?
22:47:08 <alise> cpressey: PC-BSD; pontificate
22:47:22 <ais523> olsner: a joke language, I think it's on the wiki
22:47:38 <ais523> the joke is that there's only one known description on a floppy disk, with a powerful magnet to erase it at random
22:47:49 <ais523> so nobody knows if there are any surviving descriptions or not
22:51:50 <Sgeo> Should I email World Headquarters?
22:52:09 -!- tombom_ has quit (Quit: Leaving).
22:52:16 <Sgeo> moreinformation@fisglobal.com
22:52:28 <alise> cpressey: didn't think so :)
22:52:33 <cpressey> alise: i was wondering if you could efficiently do occurs checks by hashing the terms somehow
22:52:34 <alise> Detroit; pontificate
22:52:50 <cpressey> Detroit is not nearly as cool as Windsor
22:53:03 <alise> Windsor; monoglicks?
22:53:16 <cpressey> Windsor: a Canadian city where you look north to see a major US city
22:53:27 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
22:54:01 <alise> no, Windsor: it's a place in Berkshire
22:54:15 <alise> cpressey: please -- stop me creating my own distro
22:54:17 <pikhq> cpressey: Which Windsor? There's 3 in Canada.
22:54:18 <alise> before all hell breaks loose
22:54:28 <pikhq> alise: 4 in the UK.
22:54:31 <cpressey> pikhq: the cool one where the SALT is packages
22:54:41 <alise> pikhq: the one with the castle in it
22:54:55 <cpressey> alise: ... there are cooler things to do with your time
22:55:08 <alise> yeah but they're harder
22:55:19 <cpressey> alise: that's why there are so many damned distros
22:55:20 <alise> they don't solve my day-to-day irritation
22:55:30 <pikhq> cpressey: So, Windsor, Ontario.
22:55:33 <alise> cpressey: trust me, mine is nothing like others :P (I've had it planned a while)
22:55:40 <alise> no, really, nothing.
22:56:08 <alise> ais523: I thought of what a FeatherSCM would look like
22:56:25 <cpressey> in windsor, ON, if you look north, across the detroit river, you see detroit.
22:56:47 <pikhq> Y'know, Detroit-Windsor is itself a very good argument for opening up the US-Canada border.
22:56:51 <cpressey> YakLinux ships with FeatherSCM. Film at 11
22:56:56 <alise> ais523: every change *retroactively* changes all the previous ones, because changes are in reverse: that is, each revision is an entire diff that turns the revision after it into the empty string. so you modify the previous diff so it turns the latest revision into the empty string, then the one before it to compensate, so on, forever
22:57:13 <ais523> that's not all that Feather-like, really, just ridiculous
22:57:18 <pikhq> (Detroit & Windsor are effectively a single city, seperated by a damned border crossing on the bridge between the two halves)
22:57:19 <alise> ais523: the key is that you collapse these
22:57:21 <alise> so instead of turning into ""
22:57:24 <alise> they turn into the revision before it
22:57:41 <alise> ais523: so to revert you just run a diff
22:57:48 <alise> ais523: it is feather, it's retroactive :P
22:58:06 <alise> is Windsor any good?
22:58:41 <pikhq> alise: Unlike Detroit, it has more to its economy than car manufacturing.
22:58:47 <alise> what i'm saying is
22:58:52 <alise> merging Detroit and Windsor would make Canada worse
22:58:58 <alise> for the benefit of *Detroit!*
22:59:01 -!- SgeoN1 has quit (Quit: Bye).
22:59:43 <pikhq> alise: The border there is a bit like this: http://upload.wikimedia.org/wikipedia/commons/d/da/Baarle-Nassau_fronti%C3%A8re_caf%C3%A9.jpg
22:59:49 <cpressey> This is how the rest of Canada knows Windsor: http://www.canadiandesignresource.ca/officialgallery/wp-content/uploads/2006/05/windsor_salt_canadian_design.jpg
23:00:01 <pikhq> ... Except with a damned customs checkpoint.
23:00:26 <cpressey> alise: Windsor kind of sucks too, but not anywhere near as bad as Detroit. It has some cool at the centre of its suckage.
23:00:44 <pikhq> cpressey: Like I said, Windsor has more to its economy than car manufacturing.
23:01:10 <alise> q=min(P,20);L=min(L+q,300)
23:01:10 <alise> if P and L<301:U=min(U+q*3*(U>0),300);P-=q;T()
23:01:14 <alise> still offering prizes for golfing this
23:01:30 -!- cheater99 has quit (Read error: Operation timed out).
23:01:33 <alise> pikhq: the river is a bit wider than that i think :)
23:01:42 <cpressey> alise: Also, one of my esolangs was designed and written up there.
23:02:12 <pikhq> alise: Yes, but still.
23:02:16 <alise> i think canada is my kind of country
23:02:20 <pikhq> alise: http://en.wikipedia.org/wiki/File:AmbassadorBridgesunsetting1.jpg That's the actual bridge.
23:02:36 <alise> pikhq: 's a pretty big bridge :P
23:02:45 <alise> is that ice in the water?
23:02:56 <alise> i can't see clearly
23:03:06 <pikhq> Yes; Detroit is fucking cold in February.
23:04:08 <pikhq> alise: There's also a tunnel between the two, and a couple more bridges being built.
23:04:38 <alise> cpressey: have you ever been to winnipeg?
23:04:57 <pikhq> Because for some reason some 25% of trade between Canada & the US goes through Detroit-Windsor.
23:04:58 <cpressey> alise: Yes. In fact, I was born there.
23:05:16 <alise> cpressey: Is it, in fact, a, and I quote, "frozen shithole"?
23:05:21 <cpressey> and lived there for the first, er, >2/3 of my life.
23:05:53 <cpressey> alise: i don't know that i've ever heard that *exact* term, but it's not wholly inaccurate
23:06:14 <alise> cpressey: Is it, in fact, a "dogshit dildo"? Is it "fucking over"? Is it Steven Stapleton's armpit? Is it a boiling pot of cranberries?
23:06:33 <cpressey> alise: i will stick to the first term you proposed
23:06:48 <alise> PSHT NOBODY CAN APPRECIATE MY REFERENCES
23:07:05 <alise> cpressey: (http://en.wikipedia.org/wiki/Winnipeg_Is_a_Frozen_Shithole)
23:07:09 <cpressey> yeah well no one's going to ever figure out mzstorkipiwanbotbotbot either.
23:07:42 <cpressey> possibly best if i wait til i'm home to follow that link
23:08:11 <alise> what a radical idea
23:10:30 <ais523> mzstorkipiwanbot: you still disagree with everything anyone says to you, right?
23:11:03 <ais523> fungot: do you disagree with mzstorkipiwanbot?
23:11:04 <fungot> ais523: i think now you're working on an os ( even ' kind of' proposed mean? ;p it might make sense
23:11:26 <ais523> oh well, I should go home
23:11:32 -!- ais523 has quit (Remote host closed the connection).
23:12:51 <pikhq> http://upload.wikimedia.org/wikipedia/commons/2/2b/Peacearch-usside.jpg The irony is you need to go through customs if you intend to go more than 50 feet past that "gate that will never close".
23:13:12 <alise> pikhq: is that the border?
23:13:17 <nooga> mother of common children mait
23:13:22 <alise> "Children of a common mother" is just wat
23:13:35 <alise> America, I think you need to read up on Canadian history
23:14:04 <cpressey> hey, another border crossing i've been through multiple times
23:14:16 <pikhq> As a special exception, you are permitted to go past the border crossing *there* without clearing customs.
23:15:07 <pikhq> However, you can't go past the nearby customs gates.
23:15:11 <alise> what format is /dev/dsp again?
23:15:17 <Sgeo> I _will_ play Aquaria
23:15:26 <pikhq> 8-bit unsigned PCM
23:17:10 <olsner> Sgeo: you _will_ find bancstar
23:17:18 <pikhq> At, uh. 8 kHz mono.
23:17:52 <pikhq> /dev/audio is 8-bit mu-Law at 8 kHz, BTW.
23:18:25 <alise> perl -e'print pack "c*",map {$_/=100;10*sin ($_*$_)} (0..25000)' >/dev/dsp
23:18:27 <Sgeo> olsner, yes, I will. Later
23:18:32 <alise> A swooping sci-fish effect.
23:18:47 <pikhq> alise: Well, that's fun.
23:18:53 <pikhq> If a bit shockingly loud.
23:19:43 <pikhq> 1.5 is *not* shockingly loud.
23:20:38 <alise> pikhq: Now make a song out of it.
23:23:22 <alise> #!/usr/bin/env perl
23:23:22 <alise> sub p { print pack 'c*', @_ }
23:23:22 <alise> close STDOUT; open STDOUT, '>', '/dev/dsp';
23:23:25 <alise> There's a header for you.
23:24:33 <Sgeo> Aquaria download is 209MB
23:24:36 <Sgeo> I have 199MB free
23:24:40 <Sgeo> After deleting SL
23:25:52 <Sgeo> Now I seem to have plenty of space
23:25:57 <cpressey> mzstorkipiwanbot: ~@f={/a>100?~/a=~/a-1&"PRIVMSG mzstorkipiwanbot :@f} -- or similar
23:26:58 <cpressey> alise: it'll look something like that
23:27:26 <cpressey> maybe #SELF can expand to the name of the bot or something
23:31:18 <pikhq> Taiwan's relationship with the US is so very fucked up.
23:31:29 <pikhq> “The act provides for Taiwan to be treated under U.S. laws the same as "foreign countries, nations, states, governments, or similar entities".”
23:31:43 <pikhq> Really. We say that they are an independent nation in all but name.
23:32:18 <alise> pikhq: please debug my perl >__>
23:32:35 <alise> zeroes should be silence to /dev/dsp, right?
23:32:50 <pikhq> alise: Not with unsigned PCM.
23:33:17 <alise> interestingly, my sin program works just as well piped to /dev/audio
23:33:26 <alise> is mu-Law really that similar to PCM?
23:33:59 <alise> a lot of my bytes seem to be getting dropped
23:34:04 <alise> sub z { my $x = shift; p (15 x $x) }
23:34:04 <alise> sub tck { p int(rand(256)) foreach 0..500 }
23:34:08 <alise> this is the same as tck;tck
23:34:25 <alise> i guess it isn't silence
23:34:38 <pikhq> alise: /dev/dsp is like http://upload.wikimedia.org/wikipedia/commons/b/bf/Pcm.svg but with more bits.
23:34:57 <alise> So... the midpoint.
23:35:09 <alise> Now tell me why it doesn't work.
23:35:27 <alise> so the packing is destroying it somehow ffff
23:35:46 <pikhq> And mu-Law is very close to PCM. It's PCM with a logarithmic scale instead of linear.
23:36:51 <Sgeo> You don't want to hear me song?
23:37:54 <Sgeo> Torture device: Me singing the Stargate Infinity theme
23:39:07 <cpressey> also make me watch ST:V at the same time
23:39:31 <Sgeo> ST:V isn't bad
23:40:09 -!- augur has joined.
23:41:08 <alise> pikhq: pretty hard to get a sharp sound here
23:41:32 <pikhq> alise: Well, yes. It's 8kHz.
23:42:01 <alise> pikhq: yeah but there's gotta be some sort of cymbally noise
23:42:08 <alise> not everything has to be sludge!
23:42:21 -!- BeholdMyGlory has quit (Remote host closed the connection).
23:42:30 <pikhq> Could you do it on a phone?
23:44:42 <alise> Say "TSSSS" into the phone with a ... hiss, not a proper s.
23:48:41 <Sgeo> How bad an idea would it be to put an installer on a remote machine (my school's machine) and run it locally?
23:49:16 <cpressey> Sgeo: i'm sure your school wouldn't mind!
23:49:33 <alise> pikhq: http://pastie.org/1206539.txt?key=rtv18vjbsp2ihxz2blygq My attempt at laying down the phat beats.
23:49:45 <Sgeo> cpressey, sarcasm?
23:52:39 <cpressey> Sgeo: what's on the schoolmachine?
23:54:09 <Sgeo> Some variety of Linux
23:54:52 * Sgeo creates an sdf account
23:55:20 <alise> pikhq: BTW, you know how you were talking about university financial aid sucking in the US?
23:55:31 <alise> Sgeo: the sdf admin is a huge jackass
23:56:51 <alise> Sgeo: when someone said something about the prices for registration or something he ranted about them, uhh, i don't remember, something idiotic and insulting about McDonalds
23:57:05 <alise> worse than Theo de Raat level insults :)
23:58:59 <pikhq> alise: BTW, that's just one of *many* things about US education that sucks.
23:59:01 <alise> pikhq: I looked it up out of curiosity and that situation definitely seems to be non-universal. At least, MIT was the first one I found giving clear info about it--
23:59:25 <pikhq> alise: MIT and the Ivy Leagues are some of the *few* schools that offer decent financial aid.
23:59:27 <alise> Statistics: 58% of undergrads awarded scholarship, average $26,800
23:59:38 <pikhq> That is coming *from MIT*.
23:59:57 <cpressey> Sgeo: I guess as long as you don't have the root password you can't fuck anything up TOO badly. for some reason i was assuming it was windows