00:02:08 <CakeProphet> Igelle is the newest distro on this timeline
00:02:46 <zzo38> Is LLVM capable of compiling itself?
00:03:29 <olsner> ISTR they did self-compile clang with clang a while ago
00:04:24 <CakeProphet> well, the newest one that's also a root. There's Mageia and Fusion (on the Red Hat tree), Arch Hurd (on the Arch tree), Superb Mini Server and Imagineos (on the Slackware tree), Tiny SliTaz (on the SliTaz tree).
00:05:03 <zzo38> Because, if GCC is broken for some reason, you can use LLVM. And also other way around if LLVM becomes broken for some reason.
00:05:22 <CakeProphet> Gregor: http://upload.wikimedia.org/wikipedia/commons/8/8c/Gldt.svg
00:08:15 <CakeProphet> "Damn Vulnerable Linux" on the Knoppix looks like a winner.
00:13:05 <pikhq> I still have some fondness for it.
00:13:34 <pikhq> I got somewhat annoyed by package breaks.
00:13:36 <Gregor> Yet again, people who have forgotten the simple lesson:
00:13:51 <pikhq> Then I switched to Debian, which broke stuff fundamentally.
00:14:04 <pikhq> Gregor: Hey, build GCC. Good luck.
00:14:22 <Gregor> pikhq: Debian's choice to move those crt files was definitely a good one.
00:14:25 <CakeProphet> yeah, because "build GCC" is the first thing I do after installing a distro...
00:14:28 <Gregor> I don't know WHY they did it, but I know it was good.
00:15:03 <pikhq> CakeProphet: I do odd things. I expect the distro not to break shit behind my back when I'm doing them.
00:15:27 <pikhq> Gregor: They claim "multiarch".
00:15:45 <pikhq> Why sticking crt files elsewhere helps that is beyond me.
00:16:13 <pikhq> (seeing as the *only* compilers that are going to look in /usr/lib for those are native ones)
00:16:25 <Gregor> Because if you're compiling targeting 32-bit on a 64-bit system, but your 64-bit libraries are in /usr/lib, then /usr/lib/crt* is wrong
00:16:49 <Gregor> Multiarch != cross-compilation ... per se :P
00:16:54 <pikhq> Gregor: gcc -m32 is fundamentally broken. Any further questions?
00:17:31 <Gregor> pikhq: That's a separate observation :)
00:17:39 <pikhq> Gregor: But closely related.
00:17:58 <Gregor> CakeProphet: btw, Damn Vulnerable Linux (just looked it up) is intended to be a training tool for computer security.
00:18:08 <pikhq> Gregor: If your 64-bit libraries are in /usr/lib, a 32-bit compiler looking in there for *anything* is wrong.
00:19:30 <pikhq> CakeProphet: Which doesn't even give you the option of running 32-bit programs on a 64-bit install.
00:19:48 <Gregor> http://sttngfashion.tumblr.com/ <-- in case anybody hasn't heard of it
00:20:17 <pikhq> In conclusion, fuck everyone and everything.
00:20:36 -!- Wamanuz has quit (Ping timeout: 240 seconds).
00:20:44 -!- Wamanuz has joined.
00:25:04 <CakeProphet> pikhq: so basically I think you want Windows or OSX.
00:30:50 <elliott> i want to punch zuu hes an idiot
00:31:08 <elliott> 20:11:11: <Zuu> well, when i say turing complete i really mean LBA (ofcource)
00:31:09 <elliott> 20:12:54: <Zuu> you dont seem to understand what turing complete means then
00:31:55 -!- augur has quit (Remote host closed the connection).
00:32:49 <Sgeo> Logical Block Addressing?
00:33:12 -!- elliott has quit (Read error: Connection reset by peer).
00:33:15 <Sgeo> Oh, Linear bound automaton
00:33:18 -!- elliott has joined.
00:33:18 <CakeProphet> this modern society doesn't have TIME for your questions.
00:33:58 <Sgeo> So, basically like TC except with bounded memory?
00:34:15 <CakeProphet> elliott: so what did you do in the few seconds afforded to you by not answering that question?
00:35:10 <elliott> does anyone know if ubuntu comes with xz by default or not
00:35:12 <elliott> CakeProphet: what question
00:35:53 <elliott> i didnt go insane by answering another trivial question thats what
00:36:12 <CakeProphet> What makes a question trivial? Is this a trivial question?
00:36:34 <CakeProphet> Is answering questions maddening? I thought questions were a central component of any conversation?
00:37:06 <CakeProphet> was that last question really a question at all?
00:37:40 <zzo38> In next version of Enhanced CWEB, I plan to correct some problems with the fonts and remove the PDF stuff. And remove a few of the metamacro commands that never worked, but probably add some enhancements to the way C interpreter works.
00:39:05 <CakeProphet> I wonder why there are no commonly used record-based filesystems.
00:39:07 <zzo38> Or maybe next time I might make up LWEB for making literate programming in LLVM, including macros and stuff. So that, in addition, LLVM can have a preprocessor.
00:39:44 <zzo38> CakeProphet: Do you have some examples of record-based filesystems and in what cases they are used if not for generally common though?
00:40:02 <CakeProphet> the one I'm looking at right now is Files-11 from OpenVMS
00:40:43 <CakeProphet> I have no clue what they would be used for. But it seems like affording data types in your filesystem would improve the structure?
00:41:48 -!- oklofok has joined.
00:44:49 -!- oklopol has quit (Ping timeout: 258 seconds).
00:45:01 <CakeProphet> zzo38: also it would facillitate interprocess communication by eliminating the need to parse/deparse data into a character stream.
00:45:56 <CakeProphet> though I could also see it needlessly complicated the system.
00:46:18 <zzo38> CakeProphet: Yes it would make the complicated system.
00:46:31 <elliott> 20:17:08: <Zuu> any explanation fo C not being turing complete is a flawed explanation
00:46:39 <elliott> im going to track him down and cause him to cease existing
00:46:47 <elliott> 20:16:52: <pikhq> Zuu: Yes, attaching an infinite tape to C can model a universal Turing machine. However, this is true for many, *many* a finite state machine.
00:46:48 <elliott> 20:17:37: <Zuu> pikhq: that wouldnt be a finite state machine then
00:46:51 <elliott> 20:16:52: <pikhq> Zuu: Yes, attaching an infinite tape to C can model a universal Turing machine. However, this is true for many, *many* a finite state machine.
00:46:52 <elliott> 20:17:37: <Zuu> pikhq: that wouldnt be a finite state machine then
00:47:00 <elliott> can we all just stare at that for a while
00:48:38 <zzo38> Would C be turing complete if pointers are not interchangeable with numbers?
00:49:41 <CakeProphet> I assume that means pointer arithmetic is impossible right?
00:50:18 <CakeProphet> you could use a linked list struct for (virtually) unbounded memory.
00:50:33 <zzo38> No it would not necessarily mean that. Pointer subtraction would be impossible but not adding a number to a pointer.
00:51:24 <zzo38> But it would mean a union cannot mix pointers with non-pointers, I think.
00:52:02 <oklofok> "20:11:11: <Zuu> well, when i say turing complete i really mean LBA (ofcource)" <<<
00:52:40 <zzo38> Or maybe it would still be OK as long as pointers are considered to be stored in a different address space therefore if a union mixed pointers with non-pointers you would access them separately still.
00:53:12 <zzo38> That would allow sizeof to work as well where 1 cell in pointer memory can store unbounded pointer addresses
00:53:27 <oklofok> you computer scientists just make me sick
00:53:37 <zzo38> It would have the consequence that sizeof(int***)==1
00:53:44 <oklofok> (oh and that was not about Zuu, that was intentionally blank)
00:53:55 <zzo38> Although of course all this stuff is impossible on real computer because real computer is not turing completed!!
00:54:03 <oklofok> this is just my usual math pretentiousness
00:54:24 <oklofok> thought i'd explain that to u because you wouldn't have gotten it anyways.
00:54:56 <oklofok> zzo38: do you think we'll ever turing complete it tho?
00:55:19 <oklofok> you are right in that it clearly hasn't been turing completed yet
00:55:24 <CakeProphet> I suppose you could use a struct to emulate a tagged union.
00:55:39 <oklofok> tagged union is mean what again?
00:55:55 <zzo38> CakeProphet: Maybe you can use macros?
00:56:14 <oklofok> is it c++ then that has those
00:56:28 <CakeProphet> a cobination of a struct containing a tag and an untagged union, and a set of macros to make it not a pain in the ass would work.
00:56:35 <CakeProphet> uh, I honestly don't know all that much about C++.
00:56:53 <oklofok> because i think union is a keyword in c++ and it means... well union
00:57:19 <oklofok> two types in one, the horriblest creation ever
00:57:30 <CakeProphet> in C, union creates an untagged union, meaning there is no way to test which type the value is.
00:57:48 <oklofok> yeah you have to do that manually
00:57:55 <zzo38> oklofok: Not a bad creation, at least in C is not bad. It is very useful to make union.
00:58:10 <oklofok> zzo38: actually it's is a terrifyingly terrible and terrious creation.
00:58:27 * Sgeo wonders how Rust is coming along
00:58:34 -!- augur has joined.
00:59:18 <zzo38> oklofok: I don't know what "terrious" means but it is OK you and me can have different opinion about it.
00:59:42 <oklofok> zzo38: i think you should be able to infer it from the context
00:59:45 <zzo38> But from what I can tell, LLVM doesn't have unions.
01:01:36 <oklofok> elliott: hey linear bounded automata are totally turing complete i hear the emptiness problem for their languages is re complete i thought that's the math def lolol :SDSDSD
01:01:57 <CakeProphet> if (x.tag == SHITTY_ENUM_THING) { x.value.type1; printf("baaaah this is stupid."); }
01:02:14 <zzo38> But if you can compile a C code to LLVM code, then it must be able to make unions somehow, with it.
01:02:25 <Sgeo> I vaguely remmeber complaining about the bind syntax
01:02:42 <oklofok> (or nonemptiness whatever, emptiness is obviously not even RE)
01:02:48 <zzo38> Possibly just with type casting.
01:02:52 <Sgeo> CakeProphet, Rust's bind
01:02:57 <elliott> its just typecasting in a bad disguise
01:03:30 <CakeProphet> elliott: let's make our own C, with tagged unions and functional programming and parametric types. What should we call it?
01:04:55 <CakeProphet> but, seriously, a C-like language with some of the stupid things changed would be good.
01:06:32 <oklofok> nonemptiness is painfully obviously RE, as for completeness, emptiness is not RE because you can use an LBA to check inclusion in the language of any TM's valid runs in LBA style as they do in the ghetto so nonemptiness is in RE - R; we then apply the following formal theorem: everything in RE - R is RE complete unless it's some silly artificial stupidance
01:06:40 <elliott> CakeProphet: it's a long-term research project, mind you
01:06:46 <elliott> it's being designed for http://www.coyotos.org/
01:09:49 <oklofok> but more precisely, given any RE language L, and any word w, take a TM accepting it L and make an LBA for the language of its accepting runs from w, that LBA will have nonempty language iff L contains w, and thus we have reduced L to the language of LBA that have nonempty languages
01:10:12 <oklofok> perhaps i should've been more precise given that we have to keep track of multiple levels of languageness
01:12:09 <oklofok> *take a TM accepting it and
01:12:26 <oklofok> or perhaps *take a TM accepting L and
01:14:41 <elliott> 20:18:08: <Zuu> pikhq: then it coudlnt be turing complete
01:14:41 <elliott> 20:18:22: <pikhq> Zuu: AND C HAS FINITE FUCKING STATE.
01:14:41 <elliott> 20:18:24: <Zuu> either way you tuirn it you will make it false
01:14:43 <oklofok> but in fact it turns out that the emptiness of a single LBA is NOT RE complete! homework: let A be an LBA. give an algorithm that decides whether A has empty language.
01:14:47 <elliott> 20:18:30: <pikhq> Most finite state machines, *when given infinite state to work with*, are magically Turing-complete.
01:14:48 <elliott> 20:18:33: <Zuu> pikhq: no the hardware has
01:14:48 <elliott> 20:18:42: <Zuu> C is a language
01:15:01 <elliott> 20:19:45: <Zuu> so really, any language you can make up cam be said to have finite state in some spec.
01:15:11 <elliott> 20:20:00: <AnMaster> Zuu, the problem is not that the hardware is finite (of course it is). The issue is that the spec enforces any implementation to have finite state
01:15:12 <elliott> 20:20:15: <Zuu> AnMaster: and that is very much beside the point
01:15:12 <elliott> 20:20:15: <AnMaster> it is not a valid C implementation if you have infinite state
01:15:12 <elliott> 20:20:18: <AnMaster> basically
01:15:12 <elliott> 20:20:25: <AnMaster> Zuu, no it is _exactly_ the point here
01:15:13 <elliott> 20:20:26: <pikhq> Zuu: No, that IS THE POINT.
01:15:15 <elliott> 20:20:36: <Zuu> well, then you have argued for nothing
01:15:54 <oklofok> elliott: solve the problem and you'll get one
01:16:34 <elliott> 20:25:05: <Zuu> ok, give me some hardware with infinite state, and i will write a C program that will act like any universal TC but yet only access a finite set of stats
01:16:34 <elliott> 20:25:18: <AnMaster> Zuu, how.
01:16:34 <elliott> 20:25:22: <Zuu> a C program that conforms 100% to the spec i might add
01:16:34 <elliott> 20:25:35: <pikhq> Zuu: How?
01:16:34 <elliott> 20:26:16: <Zuu> By assuming that this finite state is large enough to compute whatever needs to becomputed, otherwise go into an infinite loop
01:16:35 <oklofok> show me solution, sry left that kinda hanging in implicity
01:17:24 <oklofok> you know there's a reason i switched to math
01:17:48 <elliott> 20:29:50: <Zuu> Just to be explicit, this argument ended at the sime you gius mentioned the finite state stuff form the spec
01:17:51 <zzo38> O, BitC has theorems.
01:17:53 <elliott> 20:30:08: <Zuu> it just became to rediculous by then
01:18:01 <elliott> im seriously going to yell at this guy if he ever came in again DID YOU KNOW YOU WERE AN IDIOT A YEAR AGO AND I HATE YOU
01:18:33 <elliott> 20:33:37: <Zuu> there are several other ways to achieve UTC though
01:19:27 <Sgeo> elliott, you know there's a Freenode service by the name alis?
01:19:43 <oklofok> Zuu is a horrible person, i think we all agree already, no need to continue.
01:20:17 <elliott> oklofok: but im still reading
01:20:42 <oklofok> elliott: i'm so tired i'll actually happily read some more of that shit
01:21:16 <oklofok> not enough willpower to resist the urge of enjoying being extremely annoyed
01:21:58 <elliott> oklofok: he's saying that using POSIX functions to prove C TC is ok because
01:21:59 <elliott> 20:36:51: <Zuu> remember the posic calls are mostly implemented in C aswell
01:22:59 <oklofok> is it okay if i still tell you to shut up about that old bullshit we've heard a million times even though i secretly hope you continue
01:23:47 <elliott> 20:36:59: <pikhq> By the ISO standard for C, C is a finite-state automaton.
01:23:48 <elliott> 20:37:07: <pikhq> Zuu: POSIX defines extensions *to* C.
01:23:48 <elliott> 20:37:14: <Zuu> doesnt amtter
01:23:50 <elliott> literally just punched the air
01:23:59 <elliott> 20:37:55: <Zuu> ok, i have far more insterresting stuff to do than tell about ways to use C in a UTC way
01:24:11 <elliott> like being a terrible person and fucking killing kittens
01:24:17 <elliott> * [Zuu] (~vdsvsd@77.215.149.86): uzzuu
01:25:23 <oklofok> you should probably alert the freenode staff
01:25:33 <oklofok> maybe just paste those quotes on #freenode
01:25:40 <oklofok> i mean the balls on that dude
01:26:02 <oklofok> just like hello here i am again u remember me i'm the c is an utc dude let's have a party :D
01:26:30 <oklofok> and no one understands his sentences because they are written my me so he gets away with it
01:28:06 * Sgeo watches idiots play in #wolfgame
01:28:18 <Sgeo> Even I'm not that bad I think
01:28:30 <Sgeo> Unless maybe it's... um, typical for that channel
01:28:40 <elliott> <elliott> it is impossible to use C in a UTC way, have a nice day! :)
01:28:55 <oklofok> what the fuck are you doing
01:28:57 <elliott> if you use another timezone
01:30:01 <oklofok> i use w/ all the time when i'm doing math on the blackboard
01:30:38 <oklofok> wow i'm obsessed with math today
01:38:18 -!- cheater_ has quit (Ping timeout: 255 seconds).
01:42:35 -!- CakeProphet has quit (Ping timeout: 240 seconds).
01:51:06 -!- cheater_ has joined.
01:53:26 <oklofok> although it's kind of late now, not sure i need your input anymore
01:53:43 <oklofok> except out of pure why notness
01:54:16 <augur> oklofok: what was the question now
01:55:02 <augur> oklofok: when did i say that
01:55:17 <oklofok> i was just asking you if you agreed
01:55:30 <oklofok> why exactly did you say "wat"?
01:55:36 <augur> oklofok: oh. i only got your question
01:55:39 <augur> not the line before it
01:55:57 <oklofok> but i then told you what it was about
01:56:23 <oklofok> and still u were so conf u sed
01:57:10 <oklofok> maybe i will maybe i won't
01:57:33 <quintopia> write some c code until you fall asleep
01:57:58 <oklofok> that would take so little time i would've probably been asleep all day already.
01:58:21 <oklofok> (since c is boring as well i guess?)
01:59:26 -!- quintopia has set topic: elliott using his rationale of observable sharing | Logs: http://codu.org/logs/_esoteric/ and http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
01:59:38 <quintopia> does this make more or less sense?
01:59:38 <Gregor> I decided to buy the PediaPress book on compilers.
01:59:49 <Gregor> 700 pages of Wikipedia for $40 :P
02:03:04 <Gregor> elliott: Yesssssssssssssss
02:03:32 <elliott> remember that huge book of all the featured articles
02:07:05 <Gregor> Not only do I not remember that, I'm fairly certain I never heard about it.
02:07:39 <quintopia> i bought it because my kitchen stool broke
02:07:50 <elliott> Gregor: it was just a picture of it
02:07:52 <elliott> not something actually sold
02:08:25 <quintopia> it was captioned gcc reference manual
02:14:09 <elliott> "We used the Var x variant in Kansas Lava. We use it to encode two types of variables, Var and Reg, which have different timing semantics. Thus, turning a required annotation into a useful and required syntax."
02:14:12 <elliott> stop mentioning lava u jerk
02:15:54 <elliott> http://tinyconcepts.com/invaders.html invaders in lambda calculus
02:20:36 -!- Sgeo has changed nick to Sgeo|RainboyIsWo.
02:20:39 -!- Sgeo|RainboyIsWo has changed nick to Sgeo.
02:22:21 <Sgeo> Feel free to speculate!
02:25:57 -!- augur has quit (Remote host closed the connection).
02:26:46 -!- augur has joined.
02:36:23 <zzo38> I read stuff about rulebooks in Inform 7, including about procedural rulebooks, rulebooks based on kinds, rulebooks producing values, and so on. I would like to figure out how it works I would like to see having some other programming language based on these things possibly being optimized and then compiled to LLVM which can be further optimized.
02:36:48 <zzo38> However there is some problems, one is that the "ni" program is not yet published, also procedural rulebooks are deprecated.
02:37:17 * Sgeo plays Mafia in #wolfgame
02:42:25 <oklofok> i had a dream that i acted in a movie i think
02:50:33 -!- augur has quit (Remote host closed the connection).
02:55:32 <zzo38> Also this new one can have property hook rules, when reading or writing a property you run the property hook rules in order of priority and the return value of one can be passed as the input of another, and so on. And also various other things. And things combining the Inform 7 style rulebooks with things similar to Magic: the Gathering rules. And more.
02:57:23 <elliott> 20:46:52: <Sgeo__> Hm, what prevents storing such a large number in a bugnum that it slows down the computer?
03:02:26 -!- augur has joined.
03:02:46 -!- BeholdMyGlory has quit (Remote host closed the connection).
03:03:41 -!- augur has quit (Remote host closed the connection).
03:13:56 -!- augur has joined.
03:17:24 -!- augur has quit (Remote host closed the connection).
03:23:52 <oklofok> storing your numbers in bugnums is usually not a very good idea.
03:27:26 -!- zzo38 has quit (Quit: Bottled water is even GST [Greenwich Sandwich Time] free.).
03:28:58 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
03:33:42 <elliott> if i sign a form with "i do not agree to this contract" do i still agree to it
03:34:01 -!- azaq231 has joined.
03:36:20 -!- azaq23 has quit (Ping timeout: 240 seconds).
03:40:12 -!- augur has joined.
03:40:18 -!- augur has quit (Remote host closed the connection).
04:00:17 <oklofok> everytime i close my eyes i'm like "so where was i... ah, the markers, let's see how to implement those"
04:01:45 <oklofok> my second proof was also kind of incomplete (actually the first one wasn't really wrong either, just even more incomplete)
04:02:32 <oklofok> i finally figured out how to do this one helper CA that kinda sparsely marks stuff, but i then realized i need to rewrite all markers already in the point
04:02:44 <oklofok> or those will get confused with stuff
04:02:58 <oklofok> of course due to entropy stuff that should be doable but it's one more pain in the ass
04:03:46 <oklofok> really this will probably blow up into a hugey when i write it up, unless it completely breaks down before i finish it
04:07:18 <oklofok> the thing is i know how to handle periodic points and i know how to handle parts of points that don't have small periods for very long anywhere, except there will be some errors occasionally because stuff happens, and i need to fix those errors and then somehow glue these two parts together
04:10:04 <elliott> I was amazed in the "Telehacks" and I've try it.. It was really cool. And I will bookmarked it. I have a question do you know how to hack? I wanted to try to hack my own laptop but i don't know where to start.. Uhm.. is there any way to hack it? I'm just a beginner/noob .
04:10:05 <elliott> p.s. sorry for my english, grammars and others.
04:12:02 <oklofok> the first step in hacking your laptop is trying to come up with your password
04:12:17 <oklofok> since you came up with it, you should be able to come up with it
04:12:25 <oklofok> if not, ask yourself what it is
04:12:33 <oklofok> and after telling yourself, type that in
04:12:40 <monqy> I looked a few seconds at the logs and wow Zuu wow
04:13:17 <oklofok> i went to a Zuu once but there were too many monkeys inside so i left quickly
04:16:43 <elliott> monqy: im seriously his enemy forever
04:16:51 <elliott> i want to find out where he lives and fly over there and open the door and say
04:16:54 <elliott> hi do you go by "Zuu" on irc
04:16:58 <elliott> and hell say yes why who are you a-
04:16:59 <monqy> is he still existing
04:19:41 <monqy> i am glad i am not in any channels he is in unless he is in good channels in which case i feel sorry for the channels :'(
04:20:14 <elliott> http://kitenet.net/~joey/blog/entry/announcing_olduse.net/
04:21:28 <elliott> http://blog.longnow.org/2011/06/17/major-update-on-the-10000-year-clock-project/ so cool
04:23:36 <oklofok> When you build this thing, please display a prominent notice, in a multitude of languages, that we do *NOT* expect the world to end when the clock overflows.
04:23:39 <Sgeo> elliott, is that olduse thing a Zuu thing, or something actually possibly decent?
04:26:56 <oklofok> what the fuck is the point of time anyway
04:27:09 <oklofok> by which i mean is the clock just gonna be like buried or will it be seeable
04:27:55 <oklofok> also i don't recall ever see
04:28:03 <elliott> http://en.wikipedia.org/wiki/Clock_of_the_Long_Now btw
04:28:40 <elliott> I want to build a clock that ticks once a year. The century hand advances once every one hundred years, and the cuckoo comes out on the millennium. I want the cuckoo to come out every millennium for the next 10,000 years. If I hurry I should finish the clock in time to see the cuckoo come out for the first time.
04:28:40 <elliott> — Danny Hillis, The Millennium Clock, Wired Scenarios, 1995
04:28:49 <elliott> "Longevity: The clock should be accurate even after 10,000 years, and must not contain valuable parts (such as jewels, expensive metals, or special alloys) that might be looted."
04:29:00 <elliott> all the talking of how to convince people not to enter
04:30:33 <monqy> clockwork more like playground
04:32:29 <oklofok> oh it required maintenance
04:33:24 <oklofok> i thought at most there'd be a population of humans that live inside the clock and keep it going without ever interfering with other humans, teaching their children to do the same
04:33:45 <oklofok> the guardians of time we'd call them
04:34:00 <oklofok> and they would have the hugest beards.
04:35:18 <elliott> it just chimes once a year
04:35:43 <oklofok> no cuckoo every mil says wp
04:35:54 <pikhq> So, my mission to make a minimal Linux system that can successfully build itself has resulted in something *somewhat* larger than I would've liked.
04:36:26 <pikhq> Though at least I was able to beat Perl into being reasonable.
04:36:40 <oklofok> does your program now contain every possible bit?
04:37:08 <oklofok> i once wrote a program whose source code contained every bit in existance
04:37:37 <pikhq> monqy: By "reasonable" I mean "I only have miniperl. As /bin/perl."
04:37:43 <oklofok> monqy: mine even had multiple copies of 0
04:38:01 <monqy> pikhq: and it works? I've never heard of it. will look it up
04:38:04 <monqy> oklofok: daaaaaaaaaaaang
04:38:15 <pikhq> monqy: Miniperl is part of the Perl build process.
04:38:21 <pikhq> You see, Perl needs Perl to build Perl.
04:38:35 <pikhq> But Miniperl just needs a reasonable C environment.
04:38:42 <pikhq> So it builds Miniperl to build Perl.
04:38:57 <pikhq> I only have it there because Linux requires Perl.
04:39:23 <pikhq> Also, though I'm reasonably confident it'll work, I haven't *tested* it yet.
04:39:31 <pikhq> I literally just got it to boot.
04:41:08 <pikhq> ... And forgot to set a root password. Which makes things somewhat difficult when I'm actually using busybox getty.
04:41:54 <pikhq> By which I mean "I forgot to make /etc/passwd".
04:42:00 <quintopia> the website says it generates a different chime sequence everyday oklofok
04:42:58 <pikhq> *Aaaaah*, that's better.
04:44:28 <pikhq> "free -m" reports 8 megs used. That feels positively amazing.
04:46:19 -!- augur has joined.
05:03:34 <pikhq> Welp, let's see what breaks.
05:18:32 <pikhq> Apparently "the toolchain"
05:19:31 <pikhq> "sh: ./a.out: not found"
05:40:38 -!- copumpkin has joined.
05:40:39 -!- copumpkin has quit (Changing host).
05:40:39 -!- copumpkin has joined.
05:45:40 <elliott> The Long Now Foundation has purchased the top of Mount Washington near Ely, Nevada which is surrounded by Great Basin National Park, for the permanent storage of the full sized clock, once it is constructed. It will be housed in a series of rooms (the slowest mechanisms visible first) in the white limestone cliffs, approximately 10,000 feet up the Snake Range. The site's dryness, remoteness, and lack of economic value should protect the clock fro
05:45:40 <elliott> m corrosion, vandalism, and development. Hillis chose this area of Nevada in part because it is home to a number of dwarf bristlecone pines, which the Foundation notes are nearly 5,000 years old. The clock will be almost entirely underground, and only accessed by foot traffic from the East once complete.
06:10:14 <pikhq> Gotta love Long Now.
06:12:11 <coppro> although the whole point of the project kind of baffles me
06:14:15 <elliott> brian eno is involved, suspend all thinking for the sake of cool please
06:14:45 <elliott> it strikes me as kind of reactionary, but i suppose their goal is better than a societal meltdown
06:15:37 <elliott> at the same time i find a lot of what they say makes sense and seems kind of obvious, but then lots of things seem obvious to me
06:16:05 <pikhq> What seems obvious to you is almost impossible to others.
06:16:34 <elliott> misanthropy: an unproductive position?
06:21:31 <pikhq> It's so very comical running GCC in a minimal changeroot.
06:21:52 <pikhq> 8m for the whole OS. 200m for the compiler.
06:22:19 <coppro> elliott: your comment about misanthropy
06:22:26 <elliott> coppro: in response to <pikhq> What seems obvious to you is almost impossible to others.
06:22:31 <pikhq> elliott: 3.0 of what?
06:22:59 <pikhq> No, though rc6 got pushed late last night.
06:23:47 <pikhq> Erm, no, it was tagged July 4th. My thinko.
06:23:58 <elliott> Length: 76737255 (73M) [application/x-bzip2]
06:24:12 <pikhq> Linux has a fuckton of shit in it.
06:25:06 -!- CakeProphet has joined.
06:25:19 -!- CakeProphet has quit (Changing host).
06:25:19 -!- CakeProphet has joined.
06:28:46 <elliott> pikhq: im going to make smaller system
06:29:04 <pikhq> elliott: Good luck.
06:29:58 -!- CakeProphet has quit (Ping timeout: 276 seconds).
06:30:03 <coppro> elliott: that's not zepto enough
06:36:19 <elliott> pikhq: does it count if you have to use the serial port
06:36:27 -!- derrik has joined.
06:37:07 -!- cheater_ has quit (Ping timeout: 255 seconds).
06:39:10 <elliott> pikhq: that's a yes or no question
06:39:21 <pikhq> elliott: Sure, whatever.
06:39:30 <elliott> you're just upset because i'll win :(
06:39:46 <pikhq> elliott: I'm trying to make a bootstrapping system.
06:39:56 <pikhq> elliott: Good fucking luck.
06:40:00 <pikhq> Perl comes to ~50M.
06:40:11 <elliott> pikhq: i said it would be smaller than yours, not small
06:40:26 <elliott> pikhq: why aren't you using microperl
06:40:34 <pikhq> Because Linux won't build with microperl.
06:41:40 -!- CakeProphet has joined.
06:41:41 -!- CakeProphet has quit (Changing host).
06:41:41 -!- CakeProphet has joined.
06:42:03 -!- aloril has quit (Ping timeout: 260 seconds).
06:43:04 <elliott> wonder what it means if I get nothing after "Booting the kernel." :)
06:48:22 -!- cheater_ has joined.
06:54:37 -!- aloril has joined.
07:01:29 -!- CakeProphet has quit (Ping timeout: 252 seconds).
07:31:51 <Sgeo> http://www.futilitycloset.com/2011/07/10/a-logic-oddity/ yay I spotted the flaw in a decent amount of time
07:33:31 <Sgeo> Hmm, it's more interesting if you make it probabilities instead of certainties. Certainly the problem as phrased deals with probabilities
07:33:56 <pikhq> I think it pretty obvious. We're not dealing with deductive logic here.
07:34:52 <pikhq> So you can't really do modus ponens on this.
07:34:57 * Sgeo was thinking more that "If its not Reagan who wins, it will be Anderson." is in fact true >.>
07:35:08 <pikhq> Sgeo: Yes, that's 100% certain.
07:35:21 <pikhq> However, it's not 100% certain that a Republican will win.
07:35:26 <pikhq> It's merely probable.
07:35:45 <pikhq> Sorry, misread you.
07:36:15 <pikhq> Because it's not 100% certain that a Republican will win, you can't conclude that either Reagan or Anderson will win.
07:36:28 <Lymee> If a republican wins, oh shit.
07:37:29 <Sgeo> What happens if you assign the first two statements numerical probabilities
07:38:11 <Sgeo> I think my initial "flaw" thought was wrong
07:38:14 <pikhq> Then you're dealing with a form of inductive logic, and can get the probability that if Reagan doesn't win, Anderson will.
07:38:25 <pikhq> i.e. That works just fine.
07:39:09 <pikhq> (I'm not too familiar with the relevant formalisms, though, so can't say more than that, really.)
07:45:42 <coppro> the logic there makes me squirm
07:49:08 -!- Phantom_Hoover has joined.
07:49:25 <pikhq> coppro: The problem is only that the premises aren't necessarily true.
07:49:36 <pikhq> It's entirely valid modus ponens.
07:50:02 <coppro> Which part is valid modus onens?
07:50:51 <pikhq> It's of form "a->b;a;therefore b."
07:51:34 <coppro> pikhq: which part specifically, though?
07:51:42 <pikhq> ... That's the entire statement.
07:52:09 <coppro> which part of the article
07:52:19 <pikhq> ... The syllogism?
07:52:31 <pikhq> I'm restating the only thing in there that can be restated in that form?
07:53:10 <pikhq> The oddity comes about simply because a is not (entirely) true, and so we end up getting a not (entirely) true statement from perfectly valid reasoning.
08:00:45 -!- derrik has quit (Quit: no rush).
08:13:48 -!- oerjan has joined.
08:28:54 -!- monqy has quit (Quit: hello).
08:31:22 <elliott> "Hash functions are typically one-way functions (these haven't been proven to exist yet, but at this time they're essentially non-reversible)."
08:36:25 <quintopia> f(x)=1 is provably non-reversible :P
08:36:44 <oerjan> there are other conditions for a hash function duh
08:37:51 <Deewiant> He just means that you can't find preimages without brute-forcing
08:38:09 <oerjan> hm right so f(x)=1 isn't even that.
08:38:29 <quintopia> yes, i dont know what he was quoting
08:38:54 <quintopia> but my point was that non-reversibility is not the most important property of hashes
08:39:58 <quintopia> it's also pretty nice when they are bijections
08:41:23 <pikhq> And hash functions are demonstrably non-reversible as well. As are all functions that map from a space to a smaller space.
08:42:00 <pikhq> God, people, math isn't hard. :P
08:42:20 <azaq231> automorphism = isomorphism which is also an endomorphism
08:42:30 <quintopia> pikhq: i am considering the set of hashes over possible hash outputs
08:42:41 <quintopia> every hash should hash to a distinct hash
08:43:19 <quintopia> and non-reversibility without bruteforcing is meaningful in this contect
08:46:15 <elliott> <Deewiant> He just means that you can't find preimages without brute-forcing
08:46:32 <elliott> Deewiant: E directly stated that one-way functions haven't been "proven to exist yet"
08:46:45 <elliott> That might be what e /means/, but what e says is just confusing
08:47:18 <Deewiant> I thought his meaning fairly clear from the rest of the message
08:47:42 <elliott> Deewiant: Sure, you can obviously "decode" it, but that's not the point, you can decode most things, even nonsense
08:54:01 <oerjan> elliott: afaik, one-way functions _haven't_ been proven to exist. i believe they may require P != NP.
08:54:57 <pikhq> oerjan: Define "one-way".
08:55:07 <pikhq> Function for which there is not an inverse?
08:55:24 <oerjan> pikhq: no. http://en.wikipedia.org/wiki/One-way_function
08:55:40 -!- Phantom_Hoover has quit (Quit: Leaving).
08:55:49 <pikhq> Oh, sure, give a definition that makes the answer non-trivial. :P
08:55:52 <oerjan> it is _not_ the same as one-to-one/injective
09:08:24 <oerjan> although it's sort of related i think if you look at it from the point of constructive logic where objects to be proven to exist need to be constructible and if you then add the qualifier "in polynomial time"
09:09:36 <oerjan> *randomized polynomial time, if we go with the wiki article
09:09:43 -!- elliott has quit (Ping timeout: 240 seconds).
09:21:18 -!- Slereah has quit (Ping timeout: 240 seconds).
09:30:14 -!- Slereah has joined.
09:39:51 -!- Slereah_ has joined.
09:41:22 -!- Slereah has quit (Ping timeout: 252 seconds).
10:06:26 -!- sebbu2 has joined.
10:06:26 -!- sebbu2 has quit (Changing host).
10:06:26 -!- sebbu2 has joined.
10:09:40 -!- sebbu has quit (Ping timeout: 276 seconds).
10:26:08 -!- MigoMipo has joined.
11:58:58 -!- oerjan has quit (Quit: leaving).
12:00:54 -!- hagb4rd2 has quit (Quit: hagb4rd2).
12:11:29 -!- CakeProphet has joined.
12:11:29 -!- CakeProphet has quit (Changing host).
12:11:29 -!- CakeProphet has joined.
12:17:33 <CakeProphet> so is there any reason why this wouldn't work? f (c x y) ls = map ($ y) . map ((c$).($x)) $ ls
12:18:10 <CakeProphet> (I mean, other than it being unsupported in Haskell. I'm talking about a possible extension)
12:22:21 <CakeProphet> There are undecidable types. The type of the first argument is a, but what is the type of the second? a list of functions from (...?) to (...?)
12:22:37 <CakeProphet> it can't be anything, because it depends on the constructors parameters.
12:34:00 <azaq231> CakeProphet: This does in fact type check http://hpaste.org/new
12:36:10 <azaq231> err http://hpaste.org/48909
12:36:33 <azaq231> but you likely mean something different and data T a b is not what you meant
12:36:46 <Deewiant> azaq231: He meant having the T be variable
12:37:06 <Deewiant> So you could e.g. have also data Y a b = Y a b and f could handle both T and Y
12:45:13 -!- cheater_ has quit (Ping timeout: 255 seconds).
12:45:29 <CakeProphet> yeah, after thinking about it, it obviously wouldn't work with Haskell's type system.
12:46:39 <CakeProphet> it's always seemed reasonable to me to allow the tuple type to also represent a sort of anonymous data type with fields of a known type
12:47:10 <CakeProphet> thus (Int, Int) would represent any type (or subset of a type) constructed from two Int fields.
12:47:53 <CakeProphet> in that case you could type the above as (a, b) -> [b -> c] -> (c, a)
12:49:40 <CakeProphet> actually no it would be (a, b) -> [a -> c] -> (c, b)
12:50:19 <CakeProphet> I shouldn't be reasoning about Haskell extensions while heavily in need of sleep after work.
12:54:34 <CakeProphet> but I think it would reasonable to simply make tuples a special kind of anonymous constructor, whose type is all values defined by a certain number of fields with a known type for each field.
12:55:29 <CakeProphet> but then it would be tempting to simply write all of your functions to return tuples when possible, since this would generalize the function to many different possible types
12:56:12 -!- BeholdMyGlory has joined.
12:58:23 -!- cheater_ has joined.
13:40:10 -!- ais523 has joined.
13:40:17 -!- ais523 has quit (Changing host).
13:40:17 -!- ais523 has joined.
14:01:47 -!- pikhq_ has joined.
14:01:50 -!- pikhq has quit (Ping timeout: 246 seconds).
14:18:38 -!- azaq231 has quit (Quit: Leaving.).
14:28:43 -!- cheater_ has quit (Ping timeout: 255 seconds).
14:41:30 -!- cheater_ has joined.
14:50:56 -!- augur has changed nick to mhosignal.
14:51:59 -!- mhosignal has changed nick to augur.
15:37:38 -!- derrik has joined.
15:44:15 -!- Lymee has quit (Ping timeout: 240 seconds).
15:58:06 -!- Lymee has joined.
16:27:37 -!- monqy has joined.
16:45:37 -!- MigoMipo has quit (Read error: Connection reset by peer).
16:46:09 -!- cheater__ has joined.
16:46:09 -!- MigoMipo has joined.
16:48:17 -!- cheater_ has quit (Ping timeout: 240 seconds).
17:00:17 -!- derrik has left.
17:20:36 <ais523> hmm, do we have an opinion on what to do about languages like http://esolangs.org/wiki/Crazed (it has no real information on its page, and the author has lost the specs)?
17:22:20 <quintopia> i suggest a "lost languages" category...
17:34:25 <ais523> the author suggested deletion
17:50:54 -!- iconmaster has joined.
17:59:14 <Sgeo> :( at lost language specs
17:59:37 -!- pikhq_ has quit (Ping timeout: 276 seconds).
18:00:42 <monqy> :( at brainfuck with more commands
18:03:42 <quintopia> i dunno if i agree with that sentiment...if it's just that then yes, but i think some brainfuck derivatives have been quite interesting
18:04:01 <ais523> OK, vaguely interesting BF variant: if you delete [ (and instead have all loops go back to the start of the program), is it still TC?
18:04:03 -!- oerjan has joined.
18:05:19 <quintopia> so youd have to somehow store the entire branching path through the program on the tape to do different loops?
18:06:45 <ais523> apart from . , and loops, BF is reversible
18:06:58 <ais523> so presumably you could reverse the whole program so far just after each ]
18:07:14 <quintopia> i would believe it is TC if you wrote multiply in it
18:07:19 <ais523> in order to give effectively independent bits of code
18:09:10 <oerjan> <CakeProphet> so is there any reason why this wouldn't work? f (c x y) ls = map ($ y) . map ((c$).($x)) $ ls
18:09:36 <oerjan> for one thing haskell is designed so that you cannot use constructors at all if they aren't exported from the module defining them
18:11:13 <oerjan> also the right side can be simplified to map (($y).(c$).($x)) $ ls
18:11:34 <oerjan> (in fact ghc will do so during optimization)
18:16:27 -!- Madk has joined.
18:16:51 <Madk> behold, progress
18:17:06 <Madk> <CTCP><CTCP>2YZZr'
18:17:12 <Madk> further progress has been made
18:17:38 <Madk> awesome, stuff is moving along faster now
18:18:04 <Madk> my evolutionary algorithm is trying to squeeze "Hello, world!" out of my esoteric language
18:18:33 <Madk> that most recent sample was from generation 1105
18:18:41 <Madk> and took about 10 minutes
18:19:26 <Madk> do you see that
18:19:34 <Madk> that is great progress taking place
18:19:57 <monqy> Surface, right? Why do you need such a fancy algorithm when there's a reduction from brainfuck
18:20:11 <Madk> because the instructions are wrapped around a rather small sphere
18:20:15 -!- pikhq has joined.
18:20:18 <Madk> also, because I'm lazy
18:20:31 <Madk> so instead of writing a hello, world example I wrote a program to write a hello, world example
18:21:54 <Madk> also, I intend to apply the same code to other esolangs in the future
18:22:16 <Madk> so I can make some examples for some of the more esoteric languages
18:22:47 <Madk> it goes through quite fast
18:22:54 <Madk> ~100 generations/minute
18:23:09 <Madk> thing is the complexity involved
18:23:26 <Madk> I could go larger, but I decided to try that
18:23:30 -!- Lymee has quit (Ping timeout: 240 seconds).
18:24:24 -!- Lymee has joined.
18:24:35 <olsner> maybe you could include the runtime in the fitness measure, to get faster candidates that should take less time to test
18:24:36 <monqy> is it in blitzbasic too
18:25:54 <Madk> olsner: I haven't allowed the program to use a termination command and the IP wraps, so that's a bit impractical at this point. I have set a maximum run time, though
18:26:02 <Madk> monqy: it's in blitzmax, yes
18:28:12 <olsner> sounds like you have some optimalization to do
18:28:40 <monqy> How exactly are you doing this hello world generation thing
18:29:30 <Madk> I have a pool of initially random programs, I sample 4 of them at a time and replace the program farthest from the desired output with a combination of the best two, with some opportunity for mutation
18:30:03 <Madk> the samples I've been giving you are the closest obtained output so far
18:30:16 <Madk> the esolang is 2d
18:30:22 <Madk> so not really strings, but arrays
18:31:50 <Madk> the length isn't variable, so there's that difference
18:33:36 <Madk> I just realized a mistake I made D:
18:33:43 <monqy> And how do you measure helloworldness of an output / combine two helloworldy programs
18:34:00 <Madk> I measure it by checking similarity of characters
18:34:15 <Madk> not that complicated, but it is more than a 1:1 check
18:34:39 <Madk> I combine them by using an almost 50/50 chance for each character, with a slight bias toward the program with the better score
18:35:28 <Madk> right now each character in an offspring is a 1 in 32 chance of being randomly chose
18:36:04 <Madk> program code is stored in a 32x12 array
18:36:05 <monqy> Madk: So you know exactly what parts of each program produce which characters, and are able to separate them cleanly and stick them together nicely?
18:36:25 <quintopia> that is an awfully high mutation rate
18:36:35 <Madk> monqy: I'm not really sure what you're asking
18:36:38 <quintopia> is there a way to decrease it over time
18:36:40 <Madk> I'll reduce it a bit, then
18:36:46 <Madk> I could do that
18:37:39 <quintopia> GA is a really inefficient search paradigm, you know
18:37:42 <monqy> Madk: You said you combine your programs, right? I asked how you did that, and I reaced based on your response "< Madk> I combine them by using an almost 50/50 chance for each character, with a slight bias toward the program with the better score"
18:38:04 <Madk> the distribution is random, if that's what you meant
18:38:13 <monqy> Madk: that's not what I mean
18:38:26 <monqy> Madk: I mean how do you take two programs and mash them together in a sane way
18:38:35 <Madk> it's very random
18:38:40 <Madk> not really sane there
18:38:55 <monqy> quintopia: but it's so much cooler than just translating a brainfuck hello world
18:39:03 <Madk> quintopia: if I intended to be efficient I would have written it manually
18:40:58 <quintopia> but why GA and not simulated annealing?
18:41:40 <monqy> Madk: so for combining programs you just go over every character and pick it from one of the two constituent programs? I imagine this won't get you anywhere near hello world
18:42:00 <Madk> quintopia: the end result can be more interesting
18:42:11 <Madk> monqy: what could I do that would work better?
18:42:48 <oerjan> monqy: well you'd imagine the programs that survived might be related and so have the same structure, so it starts working better after a while?
18:42:49 <monqy> I don't know. Breeding for hello world characteristics is not really a simple problem.
18:43:31 <Madk> I did a much simpler GA not long ago that operated on a much less complicated esolang
18:43:40 <Madk> though for an entirely different purpose
18:43:45 <Madk> but the GA worked wonderfully
18:44:04 <Madk> it was more similar to corewar than anything else
18:44:13 <Madk> I put a few programs in an arena
18:44:28 <Madk> the ones that had the most health (eat food, attack others) bred and replaced the one that did badly
18:44:52 <Madk> by about 100k generations I had some very complex and intriguing behaviors
18:45:30 <monqy> And I assume the language was such that good breeding was easier
18:45:41 <Madk> it was 1d and variable length
18:45:45 <Madk> so only slightly
18:45:59 <monqy> well that has little to do with it
18:46:09 <Madk> if I remember right, some of the best ones had formed good loops structures
18:46:38 <Madk> the one working currently and that one were only significantly different in that the previous was linear
18:47:07 <monqy> I still stand that that has little to do with it
18:47:45 <monqy> it's more about the structure of the language in the sense that it's easy to break it up structurally and muck with that
18:47:47 <Madk> by 1000 generations, the programs I was working with now were outputting the same character twice where 'll' would be
18:47:49 <monqy> rather than just doing character replacement
18:48:01 <oerjan> Madk: btw i was wondering how you do the sphere wrapping, since there is no entirely clean way of doing so (we had a discussion on channel previously of what surfaces have 2d grid graphs on them and they had to have the same euler characteristic as the torus, and the sphere doesn't)
18:48:03 <Madk> it's very inefficient, indeed, but it's seemed to work
18:48:14 <Madk> I have a 2d array
18:48:20 <Madk> x wraps horizontally
18:48:27 <Madk> the algorithm for y wrapping is like this:
18:48:47 <Madk> when a pole is reached, x=(x+width/2) % width
18:48:53 <Madk> and north and south are reversed
18:49:17 <oerjan> ah. so inspired by an ordinary world map, then.
18:49:45 <Madk> it's no different than latitude/longitudinal coordinates
18:50:05 <oerjan> right. this makes the poles into special points.
18:53:58 <Madk> I had to scrap that progress because I was forgetting to reset program code after going through it
18:54:11 <Madk> now's a good time to do things like change pool size, w/e
18:54:23 <Madk> I adjusted the mutation rate to decrease steadily over time
18:54:35 <Madk> I did a good deal of optimization
18:54:45 -!- Phantom_Hoover has joined.
18:55:12 <monqy> genererational algorithm to produce a hello world
18:55:21 <Madk> I'm using an evolutionary algorithm to make a hello world example in my language for me because I don't feel like doing it myself
18:55:52 <Madk> that is my alias
18:55:58 <Phantom_Hoover> Wow, I'm surprised you got anything but a steely reception here.
18:56:45 <monqy> for more information see http://esolangs.org/wiki/User:Phantom_Hoover
18:57:12 <Madk> most of my languages aren't
18:57:39 <Madk> only 3 of them are really that similar to brainfuck
18:57:44 <Madk> 1 other is kinda close
18:58:21 <Madk> anyway, I'm trying to figure the best pool size before I set it back in motion with some fixes I made
18:58:21 <oerjan> Brainfuck. Not even once.
18:58:41 <Phantom_Hoover> I mean, you'd hardly let a Nazi off because most of the people he killed weren't even Jews.
18:58:50 <Madk> maybe you wouldn;t
18:58:58 * oerjan hits Phantom_Hoover ===\__/
18:59:26 <Madk> this is essentailly the one thing I can't go and change without scrapping everything
18:59:51 <Madk> unless I want to inject a bunch of random crap into a developed pool
19:00:05 <Madk> no, the point being once it's started I don't want to change the pool size
19:00:19 <Madk> I can change pretty much anything else without impacting a lot
19:00:35 <Madk> it's in blitzmax
19:00:45 <Madk> the point being it saves the pool every now and then
19:00:53 <Madk> so I can pick up from where it leaves off
19:01:18 <Madk> http://esolangs.org/wiki/Surface
19:01:47 <Phantom_Hoover> You reduce it to Brainfuck in the article; how hard can it be to just rewrite the BF HW?
19:02:08 <oerjan> Phantom_Hoover: but but, that would be derivative!
19:02:11 <Madk> because the space available is too small to translate it like that
19:02:33 <monqy> oh this reminds me
19:02:36 <Madk> I started one manually and thought, "what's wrong with me? I should be having the computer do this for me."
19:02:37 <HackEgo> 478) <d1ffe7e45e interpreter> The interpreter uses an unbounded tape size, but due to technical limitations will stop being unbounded if the tap size reaches 2^63 cells.
19:02:41 <monqy> Madk: what is this supposed to mean
19:03:19 <Madk> that it's a limitation of the machine, not the language
19:03:28 <Phantom_Hoover> It's unbounded... provided you don't go over the bounds.
19:04:14 <oerjan> you have complete freedom to follow big brother
19:04:44 <Madk> I'm going on with me pool of 128
19:09:01 <Madk> stuff is maturing much faster now that it's working properly
19:09:28 <Madk> there are a few commands that toggle to another when executed
19:09:32 <Madk> was forgetting to reset them
19:09:59 -!- ais523 has quit (Remote host closed the connection).
19:10:05 <Madk> and also forgetting to keep track of a program's success instead of re-running it every time I called it into question
19:10:14 <Madk> so going generally much faster now
19:10:34 <Madk> <CTCP> //;;<CTCP>
19:11:31 <Madk> I did the unthinkable
19:11:40 <Madk> I registered an account on a furry forum just to contact you
19:11:46 <Madk> I hope you appreciate this
19:12:13 <iconmaster> Sadly, the specs for Crazed have been lost for a long time.
19:12:28 <Madk> that is too bad
19:12:47 <iconmaster> In retrospect, it was a crappy language anyways, so no big loss
19:13:03 <Madk> there are no crappy languages
19:13:06 <Madk> only crappy programmers
19:13:46 <monqy> I've seen a few crappy languages
19:14:20 <Madk> the progress being made in 5 minutes is good
19:14:33 <Madk> this'll probably happen in a matter of a couple hours
19:14:48 <iconmaster> `quote <Madk> #%%:]__t# <Madk> do you see that <Madk> that is great progress taking place
19:14:56 <monqy> FURScript is pretty crappy. Snack too.
19:15:15 <Madk> that was great progress
19:15:22 <iconmaster> `addquote <Madk> #%%:]__t# <Madk> do you see that <Madk> that is great progress taking place
19:15:24 <HackEgo> 493) <Madk> #%%:]__t# <Madk> do you see that <Madk> that is great progress taking place
19:15:24 <Madk> but my algorithm was kinda flawed
19:16:01 <Madk> how do I make it quote me
19:16:38 -!- pikhq has quit (Ping timeout: 240 seconds).
19:16:41 <Madk> right now this is the best program so far
19:16:42 -!- pikhq_ has joined.
19:16:43 <Madk> ^ /^>e.v <]xe^zo+^]+> *>++(-c )\
19:16:43 <Madk> (c)? @!-^*)?-..z+*)+o--x@^-!!+ o
19:16:43 <Madk> o)/o]c@c+z\]\ee<z+/c( <.>vz\ -\x
19:16:43 <Madk> ](!@< c(c!.z+\+++<ve -.(<.v/>/?\
19:16:43 <Madk> z<! /ov>]/x/++oo-]?c?!\/+! +x^+
19:16:44 <Madk> )v>o(*z>-@ox c\(c*-@x] c))vc>? o
19:16:44 <Madk> ^*/x.o\!<x] . *vx)/+oz*./<>@e/^>
19:16:45 <Madk> c @c\ .>^x*o. )@\-e^x\<\- ]^(<-
19:16:45 <Madk> ) *<^z!e\).?/+(e@(cv>\c ]. ece?
19:16:46 <Madk> c@z]/v+++++-!<!ec@o\<!/ v ox!-(<
19:16:46 <Madk> <c-+ ()^ocee])o@> \e@c^/e?o) \]
19:16:47 <Madk> * /cc@)\?o/o !^^-e /ec\>.!.>@*o
19:16:47 <Madk> )/ vo]zx \\o< cz.(@v/c>..*>\+(@!
19:16:48 <Madk> !^+x-xev@v)+vc\\]<+x?\>e *ee(z\<
19:16:59 <lifthrasiir> http://j.mearie.org/post/7462182919/spoken-number-to-decimal shameless advertisement for obfuscated code. :p
19:16:59 <HackEgo> 493) <Madk> #%%:]__t# <Madk> do you see that <Madk> that is great progress taking place
19:17:56 <Madk> lifthrasiir: that looks awesome
19:18:04 <Madk> what about the other way around
19:18:26 <Madk> ` quote Phantom_Hoover
19:18:42 <Madk> `quote Phantom_Hoover
19:18:43 <HackEgo> 151) * Phantom_Hoover wonders where the size of the compiled Linux kernel comes from. <cpressey> To comply with the GFDL, there's a copy of Wikipedia in there. \ 155) <CakeProphet> how does a "DNA computer" work. <CakeProphet> von neumann machines? <Phantom_Hoover> CakeProphet, that's boring in the context of DNA. <Phantom_Hoover>
19:18:44 <lifthrasiir> that'd be certainly a bit-packing challenge.
19:19:02 <lifthrasiir> I found the inverse direction is much more challenging and fun.
19:19:04 <Phantom_Hoover> Although it's almost always better to use `pastequotes.
19:19:34 <Madk> `addquote iconmaster <iconmaster> `quote iconmaster <HackEgo> No output. <iconmaster> Didnt think so.
19:19:35 <HackEgo> 494) iconmaster <iconmaster> `quote iconmaster <HackEgo> No output. <iconmaster> Didnt think so.
19:19:52 <monqy> `pastequotes Phantom_Hoover
19:19:53 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.30844
19:20:36 <HackEgo> 155) <CakeProphet> how does a "DNA computer" work. <CakeProphet> von neumann machines? <Phantom_Hoover> CakeProphet, that's boring in the context of DNA. <Phantom_Hoover> It's just stealing the universe's work and passing it off as our own.
19:21:30 <HackEgo> 400) [on Sgeo's karaoke] <not_nddrylliog> Sgeo: awesome <not_nddrylliog> sounds like a japan anime sound track \ 401) [on Sgeo's karaoke] <Phantom_Hoover> That is the thing that made me into a gay vampire.
19:22:00 <HackEgo> 477) <Phantom_Hoover> I go to clean up the shrapnel from a teabag and you're discussing the definition of god out of nowhere.
19:22:05 <monqy> I think I still have sgeo's karaoke saved
19:22:05 <HackEgo> 402) <Phantom_Hoover> Yeah, I went through a whole series of existential crises when I was 8 or so.
19:22:17 <HackEgo> 413) <ZOMGMODULES> Vorpal: it's actually called Happy <ZOMGMODULES> Vorpal: Do not use it. Use Parsec. This is the wisdom of ZOMGMODULES.
19:22:25 <HackEgo> 1) <Aftran> I used computational linguistics to kill her.
19:22:40 <Phantom_Hoover> Madk, note that the quote numbering is highly unreliable.
19:22:46 <HackEgo> 42) <oklopol> i'm my dad's unborn sister
19:22:58 <Phantom_Hoover> On account of elliott being a lazy bastard and just making it the line of the file the quote is on.
19:23:15 <HackEgo> 256) <zzo38> I have plans to make the computer and one day I will do it!! (I have access to barter some people might help with these things) It is many difference from other computer.
19:24:25 <EgoBot> Unknown command () encountered
19:24:32 <Madk> this one has a ! and an l in the right places
19:24:42 <EgoBot> c char buf[1024]; int i; fgets(buf, 1024, stdin); for (i=0;buf[i];i++)buf[i]=(buf[i]=='\n')?'\0':buf[i]; if (!strcmp(buf, "h")) printf("Hello World\n"); else printf("Unknown command (%s) encountered\n", buf);
19:24:47 <Phantom_Hoover> Madk, out of curiosity, how are you defining the fitness function?
19:25:33 <Madk> note: score=0 should be a perfect match
19:25:33 <oerjan> heh i think it's erroring on the newline
19:25:34 <Madk> Local score%=Abs(Len(output)-Len(desired))*1024
19:25:34 <Madk> For Local x%=1 To Min(Len(output),Len(desired))
19:25:34 <Madk> If Mid(desired,x,1)<>Mid(output,x,1) And stop=0 stop=x
19:25:34 <Madk> score:+Abs(Asc(Mid(output,x,1))-Asc(Mid(desired,x,1)))*(x-stop)
19:26:20 <Madk> for some reason I thought I'd also added a benefit for characters near where they belong
19:26:24 <Madk> but apparently not
19:26:56 <quintopia> it should be like exponential increase in value of a letter as it approaches the desired value
19:27:02 <oerjan> Madk: hamming distance >:)
19:27:19 <Madk> yeah, that was an intention to make it develop left-to-right but it's doing the opposite
19:27:25 <Madk> so that was my mistake
19:28:03 <Madk> I might want to fix that and start over, I'm not sure
19:28:05 <oerjan> Madk: i'd imagine left-to-right would be more robust during evolution
19:28:17 <quintopia> also are you really only breeding one new program per generation?
19:28:19 <Madk> that was the intention, I wasn't thinking clearly
19:28:28 <Madk> all initial programs are generation 1
19:28:38 <Madk> a new program is the greater generation of its 2 parents +1
19:29:11 <quintopia> then what actually does "100 generations per minute" mean?
19:29:36 <quintopia> also, you are caching scores for programs that are kept alive yes?
19:29:37 <Madk> in one minute the average generation increases by about 100
19:29:46 <Madk> that's faster now
19:29:55 <Madk> because I wasn't caching when that was the case
19:30:19 * oerjan detects some kind of basic
19:30:38 <Madk> blitzmax has sort of basic syntax
19:30:47 <Madk> as in, {} are replaced with words
19:30:53 <Madk> other than that it's not really basic
19:31:00 <Madk> it just looks prettier in that respect
19:31:57 <Madk> it is advertised as being basic-like but that's very misleading
19:32:18 <Madk> it's OO and just happens to have some stuff that's like basic in addition to it
19:36:54 <oerjan> ah it's descended from basic, wikipedia implies
19:36:58 -!- cheater__ has quit (Ping timeout: 255 seconds).
19:37:25 <Madk> yes, the guy who made it made a few basic languages before blitzmax
19:37:30 <Madk> so its predecessors were basic
19:39:28 <Madk> ok, how does this look for my fitness function
19:39:29 <Madk> Local score%=Abs(Len(output)-Len(desired))*1024
19:39:29 <Madk> For Local x%=1 To Min(Len(output),Len(desired))
19:39:29 <Madk> score:+Abs(Asc(Mid(output,x,1))-Asc(Mid(desired,x,1)))*(Max(Len(output),Len(desired))-Min(x,Len(desired))+1)
19:39:44 <Madk> again, the lower the score the more accurate the output
19:46:31 <quintopia> a parabolic curve towards the desired character/~?
19:49:42 -!- cheater__ has joined.
19:51:58 <olsner> 372) <Phantom_Hoover> DON'T MOCK ME WITH YOUR ABILITY TO DIVIDE BY TEN
19:58:41 -!- choochter has quit (Quit: lang may yer lum reek..).
20:15:45 -!- zzo38 has joined.
20:43:28 -!- Madk has quit (Quit: Make it idiot proof and someone will make a better idiot.).
20:50:00 -!- zzo38 has quit (Remote host closed the connection).
20:59:01 -!- aloril has quit (Ping timeout: 276 seconds).
21:14:29 -!- aloril has joined.
21:26:38 -!- MigoMipo has quit (Read error: Connection reset by peer).
22:07:46 * Sgeo is not happy about his phone being shattered
22:08:27 <coppro> that's what you get for walking around near red mages holding an artifact
22:08:53 <Sgeo> I have no idea what that's a reference to
22:12:11 <coppro> http://gatherer.wizards.com/Pages/Card/Details.aspx?multiverseid=194147
22:17:36 <Phantom_Hoover> "LC is a really nice language. I wasn't able to find a modern, untyped, lazy language."
22:17:45 <oklofok> i was like o_O and it was like O_o
22:20:36 <coppro> Phantom_Hoover: locale?
22:21:38 <Sgeo> Lambda Calculus?
22:31:02 -!- DocHerrings has joined.
22:37:27 <Phantom_Hoover> I assume that by now you have given up on the Eodermdrome interpreter.
22:37:38 <Phantom_Hoover> oklofok, are the rumours of you writing an incredibly inefficient one true?
22:38:04 <oklofok> yeah, also it was asymptotically fast enough
22:38:08 <DocHerrings> Quite. There was no way to implement it nicely, so I decided not to implement it at all. :)
22:38:14 -!- elliott has joined.
22:38:25 -!- zzo38 has joined.
22:39:08 <oklofok> i think i did some small optimizations before abandoning it, there were no programs so i didn't quite feel the need to make it better.
22:39:28 <oklofok> anyway it's just a graph rewriter, i've written lots of those
22:39:51 <pikhq_> I'm really kinda surprised. I don't think anyone else has bothered setting up a Linux system which is just barely capable of building itself.
22:40:01 <Phantom_Hoover> <DocHerrings> Quite. There was no way to implement it nicely, so I decided not to implement it at all. :)
22:40:12 <oklofok> pikhq_: i think pikhq did that as well
22:40:29 <pikhq_> And no, it didn't bootstrap until just now.
22:40:29 <Phantom_Hoover> I think it's possible, but my incredible lethargy forbids me from working out further details.
22:40:37 <pikhq_> Linux also needs GNU sed.
22:40:48 <oklofok> pikhq_: i don't actually recall if you were underscoreless last time although i have a vague feeling you were
22:41:18 <DocHerrings> Personally, I made a function that converted my tag system into a generalized structure of nil values, then attempted to compare them. But it was terribly inefficient.
22:41:55 <oklofok> and graph rewriting is np-complete since as a special case you need to solve the isomorphic subgraph problem, so you can't really completely solve the problem
22:42:06 <oklofok> but i'm sure you can solve all cases that actually occur in programming
22:42:24 <Phantom_Hoover> oklofok, there's an algorithm which allows it to be done in polynomial time for fixed match subgraphs.
22:43:01 <oklofok> well yeah, given a single program, it will not be np complete to run it
22:43:04 <Phantom_Hoover> <DocHerrings> Personally, I made a function that converted my tag system into a generalized structure of nil values, then attempted to compare them. But it was terribly inefficient.
22:43:23 <oklofok> but that's kind of beside the point since you need to run more than a fixed set of programs, it becomes relevant that the polynomial grows very fast
22:44:11 <zzo38> Are there hardware description languages without arithmetic?
22:44:26 <oklofok> "<Phantom_Hoover> oklofok, there's an algorithm which allows it to be done in polynomial time for fixed match subgraphs." <<< from each node, do a depth n search where n is the number of nodes in the needle graph (as opposed to the haystack graph), obviously that's polynomial
22:44:50 <oklofok> because it's x^n where x is the size of the haystack
22:45:35 <DocHerrings> Phantom_Hoover: But how would you reliably compare all the gensyms?
22:45:40 <oklofok> i'm going to guess there's also an algorithm that works in x^(sqrt n), based on nothing really.
22:46:45 <Phantom_Hoover> DocHerrings, you'd need to use a subgraph isomorphism algorithm, which is where the inefficiency comes in.
22:46:50 -!- iconmaster has quit (Quit: Pardon me, but I have to go die in NetHack again.).
22:48:10 <DocHerrings> Phantom_Hoover: Seems to me that we have just agreed that the tag system is arbitrary. :)
22:48:17 <lambdabot> elliott: You have 3 new messages. '/msg lambdabot @messages' to read them.
22:48:28 <oklofok> so you'd call O(n) and O(n^2) polynomial time but O(n^1.5) not?
22:49:25 <oklofok> n is usually the changing thing
22:49:35 <oklofok> but not in what i just said
22:50:00 <oklofok> in x^n x is the haystack and n is the needle, and the needle is fixed
22:50:28 <Phantom_Hoover> I suppose you could have O(n^1/2) if you were doing something to a matrix.
22:50:28 <oklofok> O(x^(sqrt n)) is still polynomial, since it is in particular in O(x^ceil(sqrt n))
22:51:31 <oklofok> you have O(n^1/2) in many things
22:51:48 <oklofok> well okay not O(n^1/2) so much since usually you have to read the input
22:52:24 * oklofok tries to come up with a natural reason for that to occur...
22:52:48 <oklofok> obviously if there's some kind of squares being built that might happen
22:53:36 <oklofok> (might happen and often happens, just means you are linear w.r.t. height)
22:54:12 <Phantom_Hoover> Like, you have n integers and arrange them into a square matrix and then sum along one of the columns.
22:54:52 <oklofok> well n^3/2 occurs if you have to, for each column, do something in every cell, what you just described is O(n) if i understand correctly
22:54:55 <Phantom_Hoover> Or wait, actually, I think just arranging n integers into a square matrix is O(sqrt(n))
22:55:13 <oklofok> you have to go through the integers
22:55:34 <oklofok> if you have n integers, and you touch all of them, that's at least O(n).
22:55:48 <Phantom_Hoover> Assuming you have a square matrix as a C-style array of pointers to arrays.
22:56:12 <oklofok> well if it's already given like that then summing along say the first column would be n^0.5
22:56:42 <oklofok> you have n objects, and while summing you touch roughly every squareth of them
22:56:53 <Phantom_Hoover> You work out sqrt(n), take its ceiling m, then go along each m places in the array of ints you have and sticking the pointer to it in the matrix array.
22:57:18 <oklofok> how do you work out sqrt(n)?
22:57:32 <oklofok> are you also given n in binary?
22:58:51 <oklofok> in that case the working out takes a polynomial of log n time and is just not significant, and not spacewise either. then if you have random access and other real life computery stuff at hand, true, n^0.5 to make a matrix out of the array.
23:00:10 <oklofok> "<Phantom_Hoover> Assuming n has an upper bound, it doesn't matter." <<< that's a dangerous way of looking at things though, said the mathematician
23:00:49 <oklofok> just say there's an additional log n space wasted but it is hidden by the invisible constant in the linear term
23:01:35 <oklofok> what thing did you not understand?
23:01:50 <oklofok> do you know what O(f(n)) means by the way? i usually assume cs people have a firm intuitive grip at least
23:02:24 <Phantom_Hoover> It's not actually that I don't understand, more that I'm tired and there are too many words for me to bother with.
23:02:42 <oklofok> oh well, i'm not saying anything important.
23:03:05 <oklofok> my little toe has lost feeling again
23:03:43 <Phantom_Hoover> Also you can't be dying it's colder here than it is in Finland in summer TRUE FACT,
23:04:10 <oklofok> it is actually reeeeeeeally hot i wish i had a fan
23:04:45 <oklofok> you mean merrytime climate amirite
23:04:52 <Phantom_Hoover> I actually took my jumper off for significant periods of time.
23:05:10 <elliott> This is the summer you idiot.
23:07:20 <Phantom_Hoover> Also there are no hills; how do you not go insane it is a mystery to me.
23:07:56 <oklofok> we have upside down hills but they're full of water :(
23:08:36 <oklofok> a pair of gills and you're ready for some serious mountaineering
23:08:45 <Phantom_Hoover> Perhaps I can lend some hills to the poor folk of England.
23:09:24 <Phantom_Hoover> Oh wait no the hills have already spread to where elliott is.
23:11:25 * oklofok mutters something inappropriate about elliott's sanity and Phantom_Hoover's comment earlier
23:12:17 <Phantom_Hoover> oklofok, come now, how can elliott be insane with all those hills.
23:13:48 <olsner> oklofok: how can you want a theremin and complain about other people's sanity at the same time?
23:14:30 <oklofok> yeah theremin has got elliott, that's what reminded me
23:18:18 <olsner> hmm, elliott has the theremin and the theremin has got him, are they in a relation of mutual ownership?
23:19:14 <cheater__> sounds like a degenerated cocktagon
23:23:33 <Phantom_Hoover> I like the way the cats, when confronted with a new thing, try to eat it.
23:23:51 <Phantom_Hoover> This is essentially what a world run by cats would be like,
23:24:11 <elliott> It's a good way to handle problems.
23:24:13 <olsner> oh, he owns both cats and a theremin, that must make some complicated ownership graphs
23:24:34 <monqy> invent massage and then eat it
23:24:35 <olsner> unless it just becomes a complete graph where everyone/thing owns everyone/thing else
23:24:43 <Phantom_Hoover> oklofok, please inform olsner that you cannot have very complicated graphs with only three nodes.
23:24:55 <olsner> Phantom_Hoover: did you count the cats?
23:24:55 <elliott> you can if the connections get knotted
23:25:18 <olsner> you can make at least up to K5 based only on that video, assuming an unshown human owner
23:25:45 <olsner> there's also *another cat* added since the video was filmed
23:25:48 <oklofok> THERE ARE ONLY TWO DIFFERENT CONNECTED UNDIRECTED GRAPHS WITH 3 NODES MISTER OLSNER
23:25:56 <oklofok> one of them is extremely complicated
23:26:16 <olsner> oklofok: the graph is directed though
23:26:25 <Phantom_Hoover> oklofok, it's almost as mysterious as cDonald's Theorem.
23:27:19 <Phantom_Hoover> Perhaps the classification of these graphs could provide crucial insights into it.
23:27:54 <elliott> connecting back into itself in a manner bordering on the circuitous
23:27:59 <DocHerrings> Ho hum. Making a efficient implementation of /// is lisp is kinda fun.
23:28:04 <elliott> a sort of... one-element tangle.
23:28:57 -!- oerjan has quit (Quit: Good night).
23:29:14 <Phantom_Hoover> Wait elliott we need to make sure that oklofok understands cDonald's theorem.
23:29:17 <olsner> hmm, K5 is an undirected complete graph of 5 nodes, I wonder what the standard name is for the corresponding directed graph
23:29:30 <olsner> (that's the one I actually meant)
23:29:48 <oklofok> i dunno what cdonald's theorem is :\
23:29:58 <oklofok> but i can check if i understand it
23:30:24 <oklofok> cats are sooooooooo cute :D
23:32:20 -!- cheater__ has quit (Ping timeout: 255 seconds).
23:33:03 <oklofok> 45000000000 => 45000000001 BING
23:33:37 <olsner> oklofok: not proven to exist and be a number yet, obviously
23:37:21 <oklofok> oh cdonald's theorem is therefrom
23:37:28 <oklofok> perhaps i should've guessed.
23:37:37 -!- Phantom_Hoover has quit (Quit: Leaving).
23:40:22 <olsner> oh my ... cats that play the theremin :/ irresistable
23:44:19 -!- cheater__ has joined.
23:45:47 <oklofok> cheater__: aren't cats cute?
23:48:17 <olsner> I think cats are not cute, they just *look* cute as an adaptation to a world where looking cute brings them food
23:48:24 <cheater__> oklofok, cat is a substantial part of my GNU/Linux kernel quine
23:48:49 <oklofok> so i have my energy drink in a can
23:49:15 <oklofok> and i opened it just enough for stuff to slowly spill out through the invisible hole
23:49:27 <cheater__> oh yah? cute? say that to satan cat http://www.inquisitr.com/26619/satan-cat/
23:49:57 <oklofok> the problem with drinks is they are homogenous, why not have like compartments in the can
23:50:39 <oklofok> well you know, many drinks in one
23:50:44 <elliott> 12:45:29: <CakeProphet> yeah, after thinking about it, it obviously wouldn't work with Haskell's type system.
23:50:50 <oklofok> separated by these weird little bubbly compartments
23:50:51 <elliott> CakeProphet: see Data.Data, Data.Typeable
23:50:52 <cheater__> yeah, mix the drink and put it in the can
23:51:05 <cheater__> alternatively use fluids that don't mix
23:51:23 <cheater__> i'm sure you'd enjoy that wouldn't you WOULDN'T YOU
23:51:40 <elliott> 18:04:01: <ais523> OK, vaguely interesting BF variant: if you delete [ (and instead have all loops go back to the start of the program), is it still TC?
23:51:44 <elliott> no, all loops terminate immediately
23:52:18 <oklofok> cheater__: mixing drinks together is like taking all of mozart's works and listening to all of them at once
23:52:42 <olsner> oklofok: ... which is why drinks are usually sold in individual containers
23:52:52 <oklofok> and in general a drink - and most foods - are kind like taking a song and then just taking the average of the notes and listening to it for three minutes.
23:52:57 <olsner> what is the point of making containers with subcontainers in them?
23:52:57 <cheater__> oklofok, how did you know my hobby
23:53:26 <cheater__> simultaneous parallel audition is the best
23:53:30 <oklofok> olsner: the point is you don't have to be the composer yourself
23:53:40 <oklofok> not everyone knows what sequences of drinks taste the best
23:54:47 <elliott> 18:55:42: <Phantom_Hoover> Madk?
23:54:47 <elliott> 18:55:52: <Madk> that is my alias
23:54:47 <elliott> 18:55:58: <Phantom_Hoover> Wow, I'm surprised you got anything but a steely reception here.
23:54:47 <elliott> this is just going to be painful
23:54:50 <elliott> prediction of the next few lines
23:54:54 <elliott> <PH> YOU SEE WE ALL HATE YOU ESPECIALLY ME
23:55:53 <elliott> 18:58:41: <Phantom_Hoover> I mean, you'd hardly let a Nazi off because most of the people he killed weren't even Jews.
23:55:53 <elliott> 18:58:50: <Madk> maybe you wouldn;t
23:55:54 <oklofok> drinks should come in long long pipes that drip liquid at varying speeds, and you shouldn't just casually taste to them, you should really try to understand what the artist (the canposer?) was trying to convey when making the drinkdrink
23:56:09 <elliott> oklofok: do you have a newslatter
23:56:34 <elliott> oklofok: also when is the National Museum of Modern Drinks opening
23:56:42 <olsner> oklofok: ok, this idea is making a lot of sense to me
23:57:17 <olsner> but I would be too lazy to make it work
23:57:48 <olsner> I don't care about canposers conveying ideas, I care about drinking something that tastes good
23:57:54 <elliott> 19:02:37: <HackEgo> 478) <d1ffe7e45e interpreter> The interpreter uses an unbounded tape size, but due to technical limitations will stop being unbounded if the tap size reaches 2^63 cells.
23:58:01 <oklofok> olsner: well you know i'm a genius. anyway i like how food works tho, because it has both the element of composing and choice, it's kind of like an rpg really
23:58:03 <elliott> olsner: everything is art, but everything isn't art by default obviously
23:58:07 <elliott> you have to make things art
23:58:24 <olsner> elliott: woah, you have to make art art? yourself?
23:58:34 <elliott> olsner: no. oklofok does it for you
23:59:27 <zzo38> What are the best Free-software ARM emulators?
23:59:28 <elliott> `addquote <oklofok> mixing drinks together is like taking all of mozart's works and listening to all of them at once <oklofok> and in general a drink - and most foods - are kind like taking a song and then just taking the average of the notes and listening to it for three minutes. <oklofok> olsner: the point is you don't have to be the composer yourself <oklofok> not everyone knows what sequences of drinks taste the best
23:59:29 <HackEgo> 495) <oklofok> mixing drinks together is like taking all of mozart's works and listening to all of them at once <oklofok> and in general a drink - and most foods - are kind like taking a song and then just taking the average of the notes and listening to it for three minutes. <oklofok> olsner: the point is you don't have to
23:59:36 <elliott> `addquote <oklofok> drinks should come in long long pipes that drip liquid at varying speeds, and you shouldn't just casually taste to them, you should really try to understand what the artist (the canposer?) was trying to convey when making the drink <oklofok> olsner: well you know i'm a genius. anyway i like how food works tho, because it has both the element of composing and choice, it's kind of like an rpg really
23:59:36 <HackEgo> 496) <oklofok> drinks should come in long long pipes that drip liquid at varying speeds, and you shouldn't just casually taste to them, you should really try to understand what the artist (the canposer?) was trying to convey when making the drink <oklofok> olsner: well you know i'm a genius. anyway i like how food works tho,