00:00:14 <AnMaster> fizzie, what does "positive and negative border" mean here?
00:02:11 -!- Oranjer has joined.
00:02:21 <augur> interesting, fizzie. sort of.
00:02:51 <AnMaster> augur, correction: enlightening
00:03:58 <fizzie> AnMaster: Hrm, well... let P be a set of patterns, S a subset of it, and define a partial order (denoted < for now) in S; now the positive border of S are such elements x of it that for all other elements y in P for which x < y, y is not in S.
00:05:10 <fizzie> AnMaster: And the negative border are the elements x in P\S such that for all elements y in P such that y < x, y is in S. Or something like that anyway.
00:05:42 <AnMaster> fizzie, okay. But err, what is that useful for?
00:05:50 <fizzie> AnMaster: It's not a very useful construction "in general", but it's useful for analysing the complexity of algorithms that try to find frequent patterns in P.
00:06:04 <fizzie> AnMaster: This was some sort of "data mining"-themed course.
00:07:15 -!- FireFly has quit (Quit: Leaving).
00:07:59 <fizzie> Actually s/patterns/sets of items/; I don't think we made very much headway into finding actual patterns, just sets of items that tend to occur often together.
00:08:39 <fizzie> (And for some reason all the examples were about supermarket sales data, and finding out that people who buy beer also buy sausages, and vice versa.)
00:08:57 -!- BeholdMyGlory has quit (Remote host closed the connection).
00:09:49 <AnMaster> fizzie, that is rather strange, more varied examples would have been nicer
00:10:14 <fizzie> "Algorithmic methods of data mining P": "The course covers general topics in data mining, such as pattern discovery, clustering of data, and approximation of probability distributions. A special emphasis is put on algorithmic techniques to analyze discrete data."
00:11:00 <fizzie> Ha, the very first slideset of the course I opened -- https://noppa.tkk.fi/noppa/kurssi/t-61.5060/luennot/T-61_5060_04_frequent_sets.pdf -- starts with "Customer 1: mustard, sausage, beer, chips" (if you discount the title page).
00:11:10 <AnMaster> fizzie, what is the "P" in the title for?
00:11:22 <fizzie> P denotes "acceptable for postgraduate students".
00:11:44 <fizzie> Yes. It's the English version of the Finnish "L", which comes from... actually, where *does* it come from?
00:12:11 <Deewiant> Acceptable for a Licenciate degree? :-P
00:12:16 <AnMaster> as in, you aren't allowed to attend such lectures otherwise?
00:12:42 <Deewiant> You can, but you won't get any credit
00:12:43 <fizzie> You can, but you can't include them in the degree.
00:12:56 <fizzie> I'll let Deewiant do the explaining, he seems to be faster.
00:16:13 -!- oerjan has joined.
00:16:41 -!- jcp has joined.
00:24:48 -!- charlls has quit (Quit: Saliendo).
00:25:56 <pikhq> Stack one is the elements before the tape head, stack two is the elements at and after the tape head.
00:26:17 <pikhq> You move forward by popping from 2 and pushing to 1, and back by popping from 1 and pushing to 2.
00:26:59 <oklopol> [0,1,2,3,4,5,6] where you're at alement 3 would be represented by ([2,1,0], 3, [4,5,6]), where [a1, a2, ...] is a stack where a1 is the top element
00:27:21 <oklopol> you can also keep the 3 in one of the stacks ofc
00:27:33 <fizzie> oklopol: When you're at alement 3, have you had three beers?
00:27:48 <pikhq> oklopol: I was presuming just a stack.
00:28:00 <pikhq> Erm. Just two stacks, and no other data structure.
00:28:34 <oklopol> right, i just thought that was the clearest way to explain the idea
00:29:02 <oklopol> the conversion isn't that complicated :)
00:32:05 <oklopol> hmm hmm alementary operations
00:32:25 <oklopol> integer matrices are pretty sexy
00:33:23 <oerjan> fizzie: What a ridiculous stoutment.
00:38:12 <augur> you're vaguely familiar with complexity, right?
00:38:33 <oerjan> slightly more than vaguely, i'd say
00:38:56 <augur> whats the relationship between complexity classes and the chomsky hierarchy
00:38:57 <oerjan> well, the basics anyway
00:39:22 <oerjan> augur: well, regular languages = finite automata = finite space
00:40:02 <augur> ok, so theres a relation between some of the complexity classes and the chomsky hierarchy
00:40:02 <oerjan> (although the finite space is not immediately obvious, needs a proof because you can still move around in the input tape of the TM)
00:40:26 <augur> so in a sense, complexity classes are a finer grain of hierarchy than the traditional chomsky hierarchy
00:40:41 <oerjan> augur: context free is within polynomial time, O(x^3)
00:40:54 <augur> ok, thats all i wanted to really know.
00:41:01 <augur> now heres another question
00:41:17 <oerjan> augur: context sensitive = linear space as i mentioned the other day
00:41:42 <augur> is there any work on translating complexity classes into formal language theoretic terms? e.g. whats a convenient way to characterize, lets say, AM, as a class of grammars
00:42:15 <oerjan> AM = arthur merlin? that's a bit beyond the basics but i've heard the name
00:42:30 <augur> i mean, it doesnt have to be AM
00:42:42 <oerjan> augur: i recall wikipedia has a list of some correspondences beyond the basic chomsky
00:43:46 <augur> i know various kinds of like
00:43:56 <augur> formalizations of certain classes of languages
00:44:06 <oerjan> that table at the end of http://en.wikipedia.org/wiki/Chomsky_hierarchy
00:44:09 <augur> so the mildly CS right
00:44:28 <augur> well, im aware of that table
00:44:44 <fax> im gonna puke my guts out in a boiling rage
00:44:53 <augur> but what im not sure of is how THAT relates to the complexity classes
00:45:03 <augur> im a grammar nerd, so i want to see grammars
00:45:11 <augur> complexity means nothing to me without interesting characterizations
00:46:55 <oerjan> augur: well some of the things listed under the automaton column are complexities: linear-bounded and finite
00:47:07 <oerjan> which i already mentioned
00:47:41 <oerjan> and Turing machine itself, although that's up in the computability part of the hierarchy
00:49:42 <oerjan> oh well, i don't know more than those listed (and not all those) anyway
00:56:03 -!- MizardX has quit (Ping timeout: 240 seconds).
00:57:01 -!- fax has quit (Quit: Lost terminal).
01:11:50 <Ilari> Heh. cd burning program tried to burn to hard disk. Luckily IDE HDD doesn't respond to burner commands.
01:15:39 <Guest31015> ughh I'm trying to emulate a pdp-11, but I am on a mac, and I only have an emulator for DOS, so I'm emulating DOS emulating Unix 7
01:16:45 <oerjan> Guest31015: <meme age="ancient">yo dawg</meme>
01:18:03 <uorygl> Guest31015: echo $PATH
01:18:49 <uorygl> Guest31015: ls /usr/bin
01:20:06 <uorygl> Darn those keyboards that can't type all ASCII characters.
01:20:53 <uorygl> #I'm still waiting for you to ls /usr/bin.
01:21:34 -!- augur has quit (Ping timeout: 256 seconds).
01:24:26 <Guest31015> just compile yer code like this: cc "yer_code.c" -o "yer_executable"
01:25:14 <uorygl> apt-get install apt-get
01:25:52 <Guest31015> sh: apt-get : I said unix not debian ( error -14 )
01:28:44 <uorygl> afs bin boot cdrom dev emul etc home initrd.img initrd.img.old lib lib32 lib64 media mnt opt proc root sbin selinux srv sys tmp usr var vmlinuz vmlinuz.old
01:29:18 <Guest31015> void main() { printf("Hello World!\n"); }
01:29:50 <uorygl> hw.c: In function ‘main’:
01:29:50 <uorygl> hw.c:1: warning: incompatible implicit declaration of built-in function ‘printf’
01:30:37 -!- augur has joined.
01:30:50 -!- Sgeo_ has joined.
01:30:53 -!- Sgeo_ has quit (Read error: Connection reset by peer).
01:31:45 <uorygl> ftran apollo firenze ihope kaelis kyevan maxite mbiswas muffin sine sophie syrra vamp wat
01:31:54 <uorygl> aftran apollo firenze ihope kaelis kyevan maxite mbiswas muffin sine sophie syrra vamp wat
01:31:58 <oerjan> !unlambda ```.a.b`e`.c.d.e
01:32:44 <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.
01:33:25 <uorygl> su: Authentication failure
01:33:51 <Gregor> `run echo 'There is real UNIX on here, stop doing stupid fake UNIX.'
01:33:52 <HackEgo> There is real UNIX on here, stop doing stupid fake UNIX.
01:34:08 <oerjan> Gregor: which unlambda implementation is EgoBot running?
01:34:13 -!- Sgeo has quit (Ping timeout: 245 seconds).
01:34:15 <uorygl> Gregor: I'm not doing fake UNIX, I'm typing Guest31015's commands into a real command line.
01:34:20 <uorygl> After fully understanding them, of course.
01:34:21 <Gregor> oerjan: http://codu.org/projects/egobot/ :P
01:35:00 <uorygl> Gregor: so, is there some sample implementation of an ordinary Unix environment using Plash?
01:35:18 <uorygl> I.e. one that uses Plash to replicate what's already there.
01:35:37 <Gregor> uorygl: ... why would you replicate a system, when the normal system is already there?
01:36:45 <Gregor> oerjan: USED_VERSION says http://web.inf.tu-dresden.de/~bf3/unlambda/ 20060112
01:39:15 <uorygl> Gregor: as an example of how to use Plash.
01:39:30 <oerjan> Gregor: huh i hadn't seen that implementation before
01:39:41 <Gregor> oerjan: Do you have a recommendation?
01:39:46 <Gregor> uorygl: I have no idea what you mean.
01:40:04 <oerjan> Gregor: not at all, in fact it lacks the bug from the original C interpreter which i was testing :D
01:40:32 <oerjan> Guest31015: unlambda interpreter in C
01:40:38 <Gregor> !c printf("Hello, Guest31015!")
01:41:32 <Gregor> That is not a C interpreter, it's just a frontend to gcc :P
01:41:41 <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.
01:42:20 <EgoBot> /home/egobot/egobot.hg/multibot_cmds
01:42:55 <Gregor> HackEgo is a much-more-interesting implementation of the sh command, since it has a persistent filesystem.
01:43:44 <Gregor> Gee willikers, it is, innit?
01:44:00 <Guest31015> !c fprintf(stderr, "Does it print errors?");
01:45:28 <Gregor> Shhhh ... it is unaware that it's physically manifest, and we want to keep it that way.
01:45:58 <pikhq> Why would a light beep?
01:46:12 <Gregor> `run echo 'Idonno, am I?'
01:46:20 <pikhq> 1c printf("Hello, the name is HackEgo.\n");
01:46:20 <Guest31015> to show if it got dem internets and megabits and what now?
01:46:21 <pikhq> `c printf("Hello, the name is HackEgo.\n");
01:46:39 <pikhq> Oh, no c command on HackEgo? Should fix that when I'm not-lazy.
01:47:07 <Gregor> Ignore pikhq's fekked-up command.
01:47:12 <Gregor> I gave the proper command.
01:47:29 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
01:47:43 <HackEgo> bin \ cube2.base64 \ cube2.jpg \ hack_gregor \ hello.txt \ help.txt \ huh \ netcat-0.7.1 \ netcat-0.7.1.tar.gz \ paste \ poetry.txt \ quotes \ share \ test.sh \ tmpdir.21222 \ wunderbar_emporium \ wunderbar_emporium-3.tgz \ wunderbar_emporium-3.tgz.1
01:48:14 <Gregor> You can also see the remainder of various failed hacking attempts in there :P
01:48:27 <HackEgo> echo POOP! && sh ./test.sh
01:48:53 <Ilari> What, that wunderbar_emporeum isn't so wunderbar? :->
01:49:02 <Gregor> Ilari: Doesn't work in Plash.
01:49:32 <Gregor> Guest31015: Did I mention "FAILED"
01:49:51 <pikhq> Gregor: What about it doesn't work? The port-opening?
01:50:01 <Gregor> pikhq: I refuse to say :P
01:50:58 <Guest31015> `run printf "#!/bin/sh \n echo Guest Hacked "
01:51:00 <HackEgo> #!/bin/sh \ echo Guest Hacked
01:51:19 <Guest31015> `run printf "#!/bin/sh \n echo Guest Hacked " > Guest_hacked
01:51:28 <Gregor> Being able to run a trivial shell script is kind of irrelevant :P
01:53:19 <pikhq> No hacking done. Just designed behavior. :)
01:53:50 <Guest31015> yeah but running a shell script is the first step :P
01:54:14 <pikhq> Not in a crazy sandbox. ;)
01:54:21 -!- Guest31015 has left (?).
02:00:03 <Gregor> Ah yes, we'll see you again, some anonymous Comcast person.
02:06:10 <coppro> Uh, I forget. Is HackEgo one of those 'please treat it nicely' bots, or one of those 'please try to break it' bots?
02:06:25 <pikhq> coppro: It's one of those "good luck breaking it" bots.
02:06:47 <pikhq> Worst case scenario, Gregor has to revert to a previous revision.
02:06:52 <oerjan> Gregor only makes those kinds of bots these days
02:06:55 <pikhq> (it's version controlled)
02:08:33 -!- songhead95 has joined.
02:08:59 <HackEgo> bin \ cube2.base64 \ cube2.jpg \ hack_gregor \ hello.txt \ help.txt \ huh \ netcat-0.7.1 \ netcat-0.7.1.tar.gz \ paste \ poetry.txt \ quotes \ share \ test.sh \ tmpdir.22425 \ wunderbar_emporium \ wunderbar_emporium-3.tgz \ wunderbar_emporium-3.tgz.1
02:09:25 <oerjan> _completely_ different number
02:10:06 <songhead95> I was the guest who pretended to be unix, than used unix, then played with the bots, then complimented the bots
02:11:00 <oerjan> you don't need to tell us. we have means to get information...
02:12:40 <HackEgo> bin \ dev \ etc \ home \ lib \ lib64 \ proc \ tmp \ usr
02:13:04 <Gregor> You can do that in private messages, btw.
02:15:24 <Gregor> Apparently "we've gathered that" is an insult
02:15:40 <songhead95> no I am just sad at my own noobishness
02:15:43 <oerjan> don't worry, we don't mind noobs
02:16:19 <oerjan> several of us may have been noobs at one time or another
02:17:22 <Gregor> Others, such as myself, are the result of a stable time loop, and as a result began our existence with full knowledge of IRC attained from the future.
02:18:04 <Gregor> No, I travel back in time, which also reverses the aging process, and asserts my existence in the year 2002.
02:19:33 <songhead95> I want to be like you, if I master IRC and time travel, then travel back in time before I started IRC to teach myself IRC and time travel, will I begin the loop?
02:19:54 <Gregor> A time loop cannot be begun, it either exists or it does not. The past is already in the past, it cannot be changed.
02:21:21 <Gregor> Sorry! I didn't invent physics!
02:27:12 -!- Gracenotes has quit (Ping timeout: 276 seconds).
02:36:37 -!- Gracenotes has joined.
02:39:38 -!- augur has quit (Ping timeout: 248 seconds).
02:56:18 -!- songhead95 has quit (Quit: songhead95).
02:56:20 -!- songhead95 has joined.
02:56:35 -!- songhead95 has quit (Client Quit).
03:10:45 <uorygl> So, I think I'd like to pretend to be UNIX again.
03:11:23 <uorygl> Gregor: well, Plash takes away the filesystem and replaces it with a fake filesystem, right? I want to see a fake Plash filesystem that acts just like a real one.
03:11:45 <uorygl> Linux thay 2.6.30-2-amd64 #1 SMP Mon Dec 7 05:21:45 UTC 2009 x86_64 GNU/Linux
03:12:20 <Gregor> uorygl: Plash doesn't "take away" the filesystem, it stands between the program and the real filesystem. Usually real parts of the filesystem are mapped directly into Plash.
03:12:54 <uorygl> Okay then, one that gives no access to the real filesystem and instead gives access to a fake one that acts like a real one.
03:13:36 <uorygl> Or I could just admit that I'm suffering from Demanding Something That Isn't What I Really Want Syndrome.
03:15:26 <uorygl> Maybe I should examine how HackBot is implemented.
03:15:51 <uorygl> And so, that thing I wanted to do to Hackiki, I might do to HackBot instead. :P
03:16:01 <uorygl> About how big is HackBot? Does it share any code with Hackiki?
03:25:12 <Gregor> HackBot shares no code with Hackiki.
03:25:25 <Gregor> It's smaller than Hackiki, IIRC.
03:26:12 <Gregor> It's ~500 lines of C and shell.
03:26:16 <Gregor> The C is purely IRC interaction.
03:26:39 <Gregor> Ohnowait, there's also C to throttle guest programs.
03:27:37 <Gregor> Basically, it's multibot (a very simple IRC bot) with some shell scripts that bind that to plash.
03:27:45 <Gregor> The shell scripts are about 175 lines.
03:27:59 <Gregor> Multibot is about 200 lines.
03:28:25 <uorygl> So does HackBot do anything with repositories?
03:29:10 <Gregor> Yeah, basic same idea has Hackiki, but all in shell.
03:31:26 <uorygl> So there's a bit of functionality overlap but no code overlap.
03:31:27 <Gregor> Wow, worst wording ever.
03:31:38 <Gregor> Yeah, basically same idea as Hackiki, but all in shell.
03:32:33 <uorygl> I guess I should look at one or the other.
03:32:52 <Gregor> Hackiki is more recent.
03:34:38 <uorygl> pikhq: Permission denied
03:37:28 <oerjan> but it discriminates against evil people!
03:41:44 <uorygl> Evil people have the same rights as non-evil people.
03:41:55 <uorygl> Just like how gay people have the same rights as straight people.
03:42:12 <oerjan> but it prevents evil actions!
03:42:22 <uorygl> But it doesn't prevent evil people!
03:42:30 <uorygl> So, no discrimination.
03:42:43 <oerjan> that would be like saying that gay people have the same right as anyone to have sex, as long as it is with the opposite sex
03:43:06 <oerjan> and that would _clearly_ be discrimination.
03:43:52 <uorygl> If we said gay people are only allowed to have gay sex and straight people are only allowed to have straight sex, THAT would be discrimination.
03:44:22 <oerjan> well i'm not asking for evil people to only be able to do evil actions here, am i?
03:44:33 <oerjan> after all that would make deception hard
03:45:14 -!- Oranjer has left (?).
03:57:52 * uorygl ponders what he wants in this awesome nomic thing.
03:58:12 <coppro> the answer is always Agora
03:58:21 <uorygl> Agora isn't even a codenomic!
03:59:08 -!- deschutron has joined.
03:59:10 -!- coppro has quit (Remote host closed the connection).
03:59:51 -!- coppro has joined.
04:14:45 -!- coppro has changed nick to SeanHunt.
04:14:55 -!- SeanHunt has changed nick to coppro.
04:58:35 -!- charlls has joined.
05:10:03 <uorygl> `run echo "My name is overloaded; I'll have to kick someone off" | sha256sum
05:10:04 <HackEgo> 2912b1f1f1301a0d4e14416e07d7a56bee0522e7b423cde610909f5191b9bd3e -
05:10:16 <uorygl> `run echo -n "My name is overloaded; I'll have to kick someone off" | sha256sum
05:10:17 <HackEgo> caaf3e65344624e7659aaf228f62f93faae680f71adc868a6f8da880388b12f9 -
05:17:43 -!- ssu has joined.
05:38:05 -!- deschutron has quit (Ping timeout: 245 seconds).
05:38:57 -!- ssu has quit (Quit: Ex-Chat).
05:40:00 -!- deschutron has joined.
05:59:55 <coppro> how the fuck is this guy a schoolteacher
06:07:31 <coppro> some guy on an open-source project I also work on
06:07:35 <coppro> I refuse to say we work together
06:09:47 -!- oerjan has quit (Quit: So does he give you too much homework?).
06:42:24 -!- augur has joined.
07:22:16 -!- adu has joined.
07:37:03 -!- rodgort has quit (Ping timeout: 240 seconds).
07:43:44 -!- rodgort has joined.
07:46:24 -!- pikhq has quit (Read error: Connection reset by peer).
07:59:19 -!- tombom has joined.
07:59:21 -!- tombom has quit (Changing host).
07:59:21 -!- tombom has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:06:54 -!- pikhq has joined.
08:08:38 -!- FireFly has joined.
08:27:02 -!- kar8nga has joined.
08:27:45 -!- kar8nga has quit (Remote host closed the connection).
08:42:48 -!- adu has quit (Quit: adu).
08:49:29 -!- tombom has quit (Quit: Leaving).
08:59:33 -!- MigoMipo has joined.
09:23:06 -!- jcp has quit (Quit: I will do anything (almost) for a new router.).
10:08:54 -!- lereah_ has joined.
10:22:56 * oklopol tries out coppro's patented conversational tactics
10:23:02 <oklopol> lol, i can't believe this dude!
10:23:14 <coppro> hey, that's my intellectual property there
10:23:15 <oklopol> i can't believe i just witnessed that...
10:31:48 <coppro> I just enjoy venting my frustrations, excitements, and other things into this channel
10:33:15 <coppro> also, apparently, going to be 4 hours before I wake up :/
10:33:41 -!- tehz has joined.
10:34:18 <tehz> not much going on here?
10:35:25 -!- tehz has quit (Client Quit).
10:42:04 <oklopol> venting frustrations is okay, i just like BULLYING PEOPLE.
10:42:18 <oklopol> gotta go hear this lecture thing now
10:58:27 -!- TehZ has joined.
10:58:33 -!- TehZ has quit (Client Quit).
10:58:53 -!- TehZ has joined.
10:59:07 <TehZ> If you are bored, you could take a look at lolcode
11:00:41 <TehZ> Im making a programming language with .:,;'"* as its only characters
11:00:57 <TehZ> a Object Oriented programming language
11:01:34 <TehZ> either 1: Much chatlag or 2: nobody's on
11:01:51 -!- TehZ has quit (Client Quit).
11:10:28 -!- ztirf4 has joined.
11:13:20 -!- TehZ has joined.
11:13:44 <TehZ> So, which languages have u ppl made?
11:14:16 <TehZ> Checking them out
11:14:18 -!- TehZ has quit (Client Quit).
11:14:41 -!- TehZ has joined.
11:14:51 <TehZ> Cant connect to the wiki right now
11:15:15 <TehZ> Well, i made Reversible-2D
11:15:25 -!- TehZ has quit (Client Quit).
11:15:42 -!- TehZ has joined.
11:15:50 <TehZ> right now, my browser is stupid
11:16:53 <TehZ> I didnt see that one in google chrome plugin page
11:17:53 <TehZ> Isnt it? Google searching
11:17:55 -!- TehZ has quit (Client Quit).
11:18:31 -!- TehZ has joined.
11:18:35 <TehZ> Found it, installing
11:18:36 -!- TehZ has quit (Client Quit).
11:22:06 -!- magari has joined.
11:23:25 -!- TehZ has joined.
11:23:37 <TehZ> Couldnt get mirc to work
11:23:48 <TehZ> any ideas for a programming language?
11:24:58 <lereah_> http://esolangs.org/wiki/Limp
11:25:03 -!- TehZ has quit (Client Quit).
11:25:16 -!- TehZ has joined.
11:25:19 <TehZ> Checking it out
11:25:21 -!- TehZ has quit (Client Quit).
11:26:42 -!- TehZ has joined.
11:27:01 <TehZ> i dont understand that language....
11:28:08 <lereah_> It's just a bunch of functional languages
11:28:08 <TehZ> It is nearly as unreadable as my unfinished (and unpublished) language; DotsOO
11:28:41 <TehZ> Uhm, do you live in europe or usa?
11:28:52 <TehZ> cauz this chatlag is irritating
11:28:58 <TehZ> im in europe too
11:29:34 <TehZ> Either the server is slow, the server is in USA, or both.
11:30:19 <TehZ> Do you feel like there is chatlag too?
11:31:13 <lereah_> How do you know there's chatlag?
11:31:26 <TehZ> it takes a long time to get an answer
11:31:42 <lereah_> It may be that I don't anszer right away
11:32:06 <TehZ> lets test it: I write Test, and you answer with Test, k?
11:32:40 <magari> irc clients have ping functions for a reason
11:32:55 <TehZ> im a little new to IRC
11:33:41 <TehZ> you ping by doing what?
11:34:21 <magari> 04:57 [ctcp!TehZ] PING 1268737056 202884
11:34:21 <magari> 04:57 CTCP PING reply from TehZ: 0.514 seconds
11:34:54 <magari> my terminal is in afghanistan
11:35:42 <TehZ> Europe -> afghanistan -> Usa -> afghanistan -> Europe?
11:37:15 <TehZ> maybe i should continue working on DotsOO
11:37:52 <magari> depends on the freenode server your connected to
11:38:13 -!- TehZ has quit (Quit: Page closed).
11:40:54 -!- magari has quit (Quit: leaving).
11:42:18 -!- FireFly has quit (Quit: Leaving).
11:44:45 -!- comex has quit (Ping timeout: 245 seconds).
11:45:00 -!- puzzlet has quit (Changing host).
11:45:00 -!- puzzlet has joined.
11:48:57 -!- ztirf4 has quit (Quit: Nettalk6 - www.ntalk.de).
12:35:15 -!- ais523 has joined.
12:59:02 -!- MizardX has joined.
14:03:20 -!- sebbu has quit (Quit: reboot).
14:32:24 -!- sebbu has joined.
14:35:37 -!- charlls has quit (Ping timeout: 264 seconds).
14:46:37 -!- oerjan has joined.
14:52:35 -!- deschutron has left (?).
15:21:10 -!- puzzlet has quit (Quit: leaving).
15:32:11 -!- BeholdMyGlory has joined.
15:46:48 -!- coppro has quit (Quit: I am leaving. You are about to explode.).
15:58:45 <AnMaster> fizzie, another panorama soon, for this one I balanced the mobile phone on an armrest and rotated it around that point, result is that it is a fairly "boring" 180° panorama of a corridor.
16:00:07 * AnMaster is resuming the laptop with the photos and hugin project on from sleep atm
16:00:18 -!- oerjan has quit (Quit: Later).
16:21:08 -!- charlls has joined.
16:21:49 -!- comex has joined.
16:31:14 -!- FireFly has joined.
17:16:35 <uorygl> Hang it by a thread from the ceiling!
17:32:37 <AnMaster> uorygl, the images are low quality anyway
17:33:14 <AnMaster> anyway, doing photometric optimisation atm
17:42:02 -!- oklopol has quit (Ping timeout: 248 seconds).
17:57:42 -!- lereah_ has quit (Remote host closed the connection).
17:59:37 <augur> someone who knows about compsci!
17:59:41 <augur> i need your assisstance!
18:01:20 -!- fax has joined.
18:02:20 <augur> i need test graphs that have known maximum cliques
18:02:30 <augur> so i can test a maximum clique algorithm
18:02:39 <ais523> hmm, that's not the area of compsci that I work in, unfortunately
18:02:46 <ais523> does the web not help?
18:02:57 <fax> are they hard to generate?
18:03:10 <augur> fax: finding the max clique of a graph is NP complete
18:03:38 <ais523> hmm, if a problem is NP, its reverse is normally P
18:03:44 * fax doesn't see how you got that
18:03:50 <ais523> but it rather depends on what "reverse" means
18:03:51 <augur> NP complete is hard!
18:04:07 <ais523> besides, even exponential algorithms can run quickly on small enough data sets
18:04:38 <augur> true, but i need it to run on large graphs :)
18:05:11 <augur> with upwards of 5000 words
18:05:18 <augur> maybe even 10000! or 100000!
18:05:26 <augur> not factorial, mind you.
18:05:40 <fax> though nurikabe is NP and generating puzzles for that is hard
18:05:47 <fax> and making sokoban puzzles is an art
18:05:58 <augur> i'd make _your_ sokoban puzzle
18:06:01 <augur> if you know what i mean
18:06:48 <ais523> fax: generating random Sokoban puzzles is something I'm interested in myself
18:06:49 <augur> i need test graphs tho
18:07:05 <augur> so satisfy to myself that i havent just proven that P = NP
18:07:10 <ais523> not necessarily particularly good ones, but they can't be trivial
18:07:20 <ais523> augur: do you have a P-time algo, or think you do?
18:07:34 <ais523> it's probably easier to poke holes in the algo by looking at it than by trying to create test graphs
18:07:49 <augur> its P on complete graphs up to 300 nodes, or so it looks
18:07:59 <augur> i _dont_ know if its P on non-complete graphs.
18:08:04 <ais523> maximum clique for complete graphs is easy
18:08:15 <fax> ais523, I make a game called 'multiban', which is sokoban with n-players (over the internet)
18:08:19 <augur> its especially easy if you know its a complete graph!
18:08:25 <augur> but i dont assume that it is.
18:08:32 <augur> the question is what its runtime is on non-complete graphs
18:08:55 <fax> it's great fun because the game really is about collective thought -- but you run out of levels quickly :(
18:09:01 <augur> and since theres a LOT of graphs with n nodes, this is not an easy thing for me to check
18:09:17 <augur> infact, theres like
18:09:26 <augur> what 2^(n^2) graphs with n nodes?
18:09:44 <augur> adjacency matrix is n^2, and each of the cells can be filled or not, so 2^(n^2)
18:10:14 <augur> or 2^(0.5 n^2) if you require that the graph is undirected
18:10:53 <ais523> anyway, here's a test case from Wikipedia: abcdaedfecfgbfega
18:11:03 <ais523> (I take it you're familiar with Eodermdrome graph notation?)
18:11:33 <ais523> but this is #esoteric!
18:11:38 <fax> OH MY GOD THIS IS AWESOME
18:11:41 <ais523> http://esolangs.org/wiki/eodermdrome
18:11:42 <fax> http://esolangs.org/wiki/Eodermdrome
18:11:45 <augur> yes im there now :P
18:11:58 -!- tombom has joined.
18:11:58 <augur> you're so narcisistic, ais523 :P
18:12:15 <ais523> I'm not entirely sure if I translated it properly from the Wikipedia article, but it's so much easier to send over IRC than trying to send a graph normally
18:12:26 <augur> can you give this graph to me in non-eodermdrome notation? :P
18:12:36 * fax how to pronounce Eodermdrome
18:12:59 <augur> or is it just that adjacency in the string = adjacency in the graph?
18:13:05 * ais523 pronounces it "E. O. derm drome"
18:13:22 <ais523> I've only ever seen the word written, though
18:13:49 <ais523> it was originally coined in a wordplay book, to try to create a minimal-length word that lead to a nonplanar graph
18:18:01 <augur> ais523: i presume this graph is not directed, yes?
18:20:28 <augur> btw, which graph is this that you found? like, where is it on the wiki page
18:20:51 <ais523> on http://en.wikipedia.org/wiki/Special:Search?go=Go&search=Clique_problem
18:20:57 <augur> oh ok, the bluedots one
18:21:08 <augur> wheres it say this is known to be hard?
18:21:19 <ais523> the graph? or the problem?
18:21:57 <ais523> it isn't known to be hard; it's just known to have a maximum clique of 4
18:22:08 <ais523> and that's what you asked for
18:22:15 <ais523> graphs with known maximum cliques
18:22:19 <augur> oh. no i need lots of test cases tho :p
18:22:57 <ais523> generate them at random? by definition, you can test if an answer to an NP-complete problem is correct in P-time
18:23:00 <augur> not just _one_ but lots of them, with lots nodes, so i can test the algorithm
18:23:27 <augur> i dont know how to generate these things at random.
18:23:36 <fax> augur, why don't you just prove the algorithm correct?
18:23:38 <ais523> graphs? that's trivial
18:23:41 <augur> i dont know how to, fax!
18:23:52 <augur> ais523: well, i could generate _graphs_ at random
18:24:03 <fax> did you just implement a random algorithm them realize it solves an NP problem in P? :)))
18:24:05 <augur> but i need to make sure that the algorithm isnt just hitting easy graphs
18:24:29 <AnMaster> fizzie, there are a few seems, but way less than in the one yesterday. Some image data at the edges is cut, and a few tiny (5x10 pixels or so) holes along some edges was filled in in gimp) http://omploader.org/vM3VvNA
18:24:31 <augur> no, i implemented the algorithm intended to solve one problem, and then a prof at the school pointed out that it seems to be solving the maximum clique problem
18:24:54 <augur> so its not a _random_ algorithm
18:25:26 <ais523> augur: why don't you just show us the algorithm, and we can find flaws in it / show it isn't P-time?
18:25:46 <augur> but then you'll steal my discovery! D:
18:25:48 <ais523> if it really did solve the maximum clique problem in P-time, your prof would have claimed the money for himself already
18:25:51 <AnMaster> fizzie, irritatingly enblend placed the seam over that door so a blurry image of it was included. There is another sharper image of it..
18:25:59 <augur> i didnt show him the algorithm
18:26:03 <augur> just the description of the original problem
18:26:10 <augur> i havent mentioned that i think its PTIME.
18:26:14 <augur> im not _that_ stupid
18:26:21 <augur> "lol i proved P = NP 8D"
18:26:23 * fax decided that
18:27:03 <augur> ill show you guys the algo when i get back. i have to go pick up some cookies.
18:27:11 <augur> be back in an hour or two
18:27:25 <AnMaster> strange, getting to the cookie jar taking that long
18:27:42 <augur> no, these cookies were delivered
18:27:44 <fax> augur is in no rush
18:27:57 <augur> and are sitting in a UPS depot in Laurel, MD
18:28:12 <augur> http://www.cherylandco.com/ << this kind. gooood cookies.
18:28:13 <fax> wow augur, it must be hard to get cookies where you are
18:28:25 <uorygl> They weren't delivered very well.
18:28:58 <augur> shut your face you ruined my joke
18:29:05 <ais523> PSPACE-complete is the bounded equivalent of TC, isn't it?
18:29:38 <uorygl> There was a paper titled "Sokoban is PSPACE-complete" that essentially showed that you can make a Sokoban computer.
18:29:58 <uorygl> Yay, my computer is PSPACE-complete. :P
18:30:15 <ais523> uorygl: strangely, someone linked me to that a few days ago
18:31:26 <uorygl> Speaking of linking to stuff: http://www.jaapsch.net/puzzles/
18:31:59 <uorygl> Jaap's Puzzle Page, the most important page ever.
18:33:23 <uorygl> Well, it isn't really.
18:34:14 <ais523> augur: anyway, write a P-time algorithm for checking if an alleged clique is in fact maximal
18:34:54 <ais523> (that's easy: check all subsets of the graph that are one element larger than the alleged clique to see if they're cliqued; that's P-time with a rather horrible exponent)
18:35:30 <fax> 7"the most important page ever" lol
18:38:46 <ais523> hmm, what complexity class is finding a maximal cycle in a graph?
18:38:55 <uorygl> Little did fax know that eir subconscious had begun counting down to eir death. Eir next typo will involve the number 6.
18:41:33 * fax terabytes oowriggle
18:46:50 -!- ais523 has quit (Remote host closed the connection).
19:07:57 -!- hiato has joined.
19:08:44 <hiato> Hey all, got a new esolang, and I'm struggling to write a quine in it. Does a language have to be TC to have quines?
19:09:17 <hiato> But that is a language feature. Ok, non-trivial quines
19:09:42 <Deewiant> Probably not, but I can't prove it.
19:09:57 <fax> hiato new esao lang?
19:09:57 <uorygl> Take any language with a quine. Figure out how much memory it uses. Restrict it to that much memory.
19:10:03 <hiato> Ok, cool, then I guess it's programmer error here :P
19:10:16 <hiato> uorygl: Right, but that requires a quine to reduce the memory
19:10:23 <hiato> fax: yep, I'll upload the spec now :)
19:10:24 <Deewiant> Well, that doesn't imply that your language has quines
19:10:37 <pikhq> No, but I seem to recall a proof that a TC language with output capabilities sufficient to output itself must have a quine.
19:10:56 <fax> it's pretty much one quine that you can rewrite into whatever language
19:11:12 <uorygl> hiato: well, the language with the memory restricted, which is not TC, still has a quine.
19:11:15 <hiato> there was something about fix point thereom, but I can't remember what it meant or why it was important
19:11:30 <uorygl> So, a non-TC language with a non-trivial quine.
19:11:49 <hiato> uorygl: well, memory isn't really the issue here, it's string stuff and delimeters, actually, only delimeters
19:12:00 <pikhq> C has quines, and C is not TC.
19:12:31 <hiato> pikhq: right, but it can easily operate on strings, and it has the augmented print function %c, %i etc
19:12:42 <fax> C is not TC?
19:12:56 <pikhq> fax: Mandated finite memory.
19:13:08 * hiato thinks it's probably best to keep quite about this now, sign, C again
19:13:28 <hiato> fax: there was that "discussion" under bit-bit-jump
19:15:34 <pikhq> fax: Pointers must be a multiple-of-char size. A char must have a bound, and that bound must be storable in a char. Thus, C has finite memory. QED.
19:18:25 <hiato> http://dpaste.com/172601/
19:18:30 <fax> http://www.jaapsch.net/puzzles/morph.htm
19:18:48 <hiato> sorry, the explanation is terrible, but the source should clear it up
19:21:39 <hiato> the problem is that # (as a string, so as to avoid execution) needs to be escaped viz., :# Then, : needs to be escaped and it's a horrible route from there on in
19:24:02 <hiato> unless I'm going about it in the wrong manner, and perhaps the stack should hold individual tokens instead of strings
19:24:06 -!- TehZ has joined.
19:25:27 <pikhq> Switching from Japanese manga to English chatroom = possible confusion.
19:25:41 <fax> uorygl uragirl?
19:25:57 <fax> I have to find some puzzles for you
19:25:58 <augur> ok who wants to see my little algo now?
19:26:16 <uorygl> What, people manage to make that mistake even with this spelling of my name? :P
19:27:00 -!- TehZ has quit (Client Quit).
19:27:01 <fizzie> AnMaster: That's a very long-looking corridor.
19:29:08 * hiato cannot believe someone would have the time to do this http://en.wikipedia.org/wiki/Tupper%27s_self-referential_formula
19:29:30 <AnMaster> fizzie, to the right it continues for another, wider, "section" like the one I took the photo in.
19:31:25 <hiato> well, it seems that Quine -> TC may not hold, but TC -> quine does : "By the fixed point theorems in logic, such a program exists in any Turing-complete language in which printing..."
19:31:43 <AnMaster> * hiato cannot believe someone would have the time to do this http://en.wikipedia.org/wiki/Tupper%27s_self-referential_formula <--- .... wow...
19:32:10 <hiato> AnMaster: Yep, I'm still picking my jaw up. I mean, I knew you could do relations, but encoding bitmaps?
19:32:27 <AnMaster> hiato, well, plotting relations is common.
19:32:33 <fax> uorygl, http://www.youtube.com/watch?v=ONYN8yAEfpA http://www.youtube.com/watch?v=uzgnpf2URhQ http://upload.wikimedia.org/wikipedia/en/4/42/Haberdasher-anm-01.gif
19:32:34 <AnMaster> well, not as common as functions
19:35:48 <uorygl> I'll use the fact that I'm unimpressed by Tupper's self-referential formula as a status symbol. :P
19:36:09 <fax> how can you not be impressed by that? Do you know how to make it?
19:36:58 <fax> i guess it's basically a less elegant and 2D version of godels beta function...
19:38:42 <fax> oh! "The formula was first published in his 2001 SIGGRAPH paper that discusses methods related to the GrafEq formula-graphing program he developed."
19:38:48 <fax> I learned taylor series from this program
19:41:41 <fax> uorygl did you see the links I gave you
19:47:29 <augur> should i use haskellish notation for my algorithm, even tho its written impertively?
19:55:14 <hiato> augur, I'm not sure that makes terribly that much sense to me
19:55:32 * hiato ok, that is how not to write a sentence
19:56:28 <hiato> and augur, what is exactly is your algorithm?
19:57:06 <augur> a relativized maximal clique algo
19:57:36 <hiato> right. sure it is. does it make coffee? otheriwse I would say your deltas are off a tad
20:00:48 <augur> that is, the largest clique that some node is in
20:00:58 <augur> not the largest clique in the graph but just the largest clique a given node is in
20:01:11 <uorygl> Couldn't you find that by dropping the node and all the nodes it's not connected to?
20:03:45 <fax> uorygl?!??!?!?!?
20:04:12 <fax> geez I went and looked it up in my book for you and you're just ignoring me
20:08:44 <fax> uorygl fuck you
20:08:55 <augur> uorygl: no, because that doesnt mean that they form a clique
20:10:36 <augur> uorygl: http://wellnowwhat.net/linguistics/node-relative%20maximal%20clique.pdf
20:10:37 -!- jcp has joined.
20:12:29 <fax> don't youknow it's rude to ignore people
20:13:19 <augur> i need to go buy some spraypaint. bbiaf guys <3
20:13:29 <uorygl> I mean by doing that and then finding the maximal clique of the result.
20:13:58 <augur> oh, you mean taking the subgraph of G that contains the node and all its neighbors
20:14:11 <augur> (and thus all the edges that only include those neighbors)
20:14:28 <augur> and then finding the maximum clique in that?
20:14:35 <uorygl> The subgraph that contains the node's neighbors only, not the node itself.
20:14:43 <augur> that still doesnt guarantee it, because the biggest clique in that might not include the node itself
20:14:58 * fax wonders how uorygl leaned to be such a bastard
20:15:05 <augur> no but it must, right
20:15:10 <augur> sure then i guess that would work
20:15:18 <augur> if you included the node, crucially
20:15:20 <uorygl> Surely any clique in that can be extended to include the node itself.
20:15:34 <augur> anyway, check out the algo while im off doing stuff
20:20:01 -!- augur has quit (Ping timeout: 264 seconds).
20:20:37 -!- Slereah has quit (Ping timeout: 264 seconds).
20:24:20 -!- comex has changed nick to comex_.
20:24:21 -!- comex_ has changed nick to comex.
20:24:33 -!- comex has changed nick to comex_.
20:24:35 -!- comex_ has changed nick to comex.
20:26:12 -!- Slereah has joined.
20:32:16 <AnMaster> fizzie, any such corridors at your university btw?
20:32:32 -!- Gracenotes has quit (Quit: Leaving).
20:33:42 <fizzie> I guess the main building corridors look a bit like that, though not quite that.. how should I put it, open.
20:34:32 <AnMaster> fizzie, I think the eqvirectangular (?) projection is partly to blame for that
20:35:42 <AnMaster> fizzie, not that cylindrical projection looks much better, and rectilinear is just not feasible with that fov
20:38:37 <AnMaster> fizzie, according to hugin I have ~185° fov
20:39:24 <AnMaster> fizzie, actually more (214), but not all was usable
20:41:29 <AnMaster> (not usable, as in photo ends up looking not very nice due to corridor not being in "center")
20:41:49 <AnMaster> (oh and all there was to see was the power cable to my laptop lying over a table basically
20:42:32 <AnMaster> some day I should try to take a 360° panorama
20:42:43 <AnMaster> also I wonder if that really is a degree sign, or if it is a ^0...
20:43:30 <AnMaster> hm... °⁰ well degree sign I guess
20:44:09 <AnMaster> hey, am I holding a monologue or something?
20:44:38 <fizzie> I'm sort of a bit away here.
20:44:55 <fizzie> I'd photograph places at the university, but I don't want to do it when there are people there.
20:45:13 <AnMaster> fizzie, btw fish eye looks funny
20:45:38 <AnMaster> stereographic actually seems to kind of work for this...
20:57:47 -!- augur has joined.
21:01:43 -!- Phantom_Hoover has joined.
21:01:52 -!- Phantom_Hoover has quit (Client Quit).
21:11:06 <fax> uorygl is a twerp
21:11:23 <fax> he can go stick it up himself
21:13:33 * uorygl disassembles a pen and screws a TRS size adapter onto one of the parts.
21:14:31 <uorygl> I imagine that in my younger days, I would have expected to be able to do that, plug it into an amplifier, and sing into the tip of the pen, and hear my voice coming through the amplifier.
21:14:53 <fax> uorygl I hope yu get electrocuted
21:29:19 -!- Ilari_ has joined.
21:30:54 -!- Ilari has quit (Quit: "Address/window change").
21:30:54 -!- Ilari_ has changed nick to Ilari.
21:47:35 <augur> well uorygl? is my algorithm awesome or not? 8D
21:51:05 -!- Phantom_Hoover has joined.
21:59:16 -!- hiato has quit (Quit: underflow).
22:20:32 -!- johsam1 has joined.
22:20:59 -!- johsam1 has left (?).
22:28:49 -!- Phantom_Hoover has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100214235838]).
22:32:02 <AnMaster> * uorygl disassembles a pen and screws a TRS size adapter onto one of the parts. <-- TRS?
22:32:39 <AnMaster> <fax> uorygl I hope yu get electrocuted <-- why?
22:35:03 <fax> because he ignored me
22:35:29 <fax> he linked to some puzzles site so I tried to show him my favorite puzzles but he just completely ignored me, why would someone do that? It's so horrible
22:51:04 -!- Ilari has quit (Quit: Reconnecting).
22:51:12 -!- Ilari has joined.
22:57:04 <uorygl> AnMaster: tip-ring-sleeve, consumer audio connectors.
22:57:53 -!- comex has left (?).
23:05:04 <augur> uorygl: ouch. im doing some tests on random graphs, and the time complexity seems to be exponential :(
23:05:12 <augur> its _fast_ but its exponential
23:06:13 -!- werdan7 has quit (Ping timeout: 619 seconds).
23:07:27 <augur> and i had thought i'd done it! D:
23:13:51 -!- werdan7 has joined.
23:17:44 -!- comex has joined.
23:29:50 <AnMaster> augur, I doubt P=NP is the case
23:30:00 <AnMaster> there seems to be nothing that points in that direction
23:30:36 <AnMaster> augur, what I find more likely is than P=NP is that P!=NP but that is unprovable. (Of course it could be unprovable if it is unprovable or not...)
23:32:13 -!- fax has quit (Ping timeout: 260 seconds).
23:32:52 -!- BeholdMyGlory has quit (Read error: Connection reset by peer).
23:33:20 <lament> it's provable.. it's just very, very hard to prove.
23:33:29 <augur> oh yeah? prove it.
23:34:15 <augur> actually, this is interesting, the graphs that seem to be taking the longest to calculate are the ones that are very nearly complete, but those are _easy_ when just run by themselves. so i might be actually hitting computational randomness in the calculation times
23:34:16 -!- coppro has joined.
23:38:15 <augur> yeah, that seems to have been what it is; after averaging multiple compute times for each graph, the number of graphs that are extremely long to compute for has dropped
23:40:32 -!- tombom has quit (Quit: Leaving).
23:52:17 -!- MigoMipo has quit (Read error: Connection reset by peer).
23:58:38 <augur> damnit, i keep doing this
23:58:47 <augur> the algorithm ive been testing? wrong fucking algorithm.
23:58:57 <augur> i have two methods on my graph class that solve this problem
23:59:07 <augur> both with the same name, except for a version number
23:59:21 <augur> e.g. relmaxclique and relmaxclique2
23:59:26 <augur> and the first i know to be a shitty algorithm