00:01:24 <pikhq> In addition to Cygwin...
00:01:36 <pikhq> No. I don't know either.
00:01:49 <elliottXP> (AutoHotKey script that makes Win+K empty the recycle bin.)
00:02:23 <elliottXP> pikhq: So... got a link? I'm willing to try that out.
00:03:08 <pikhq> elliottXP: I haven't a clue how to use it.
00:03:27 <pikhq> elliottXP: All I know is they've got a shell script that bootstraps it.
00:03:39 <pikhq> elliottXP: I don't know how you're supposed to run that shell script without a shell.
00:04:24 <pikhq> elliottXP: http://overlays.gentoo.org/proj/alt/browser/trunk/prefix-overlay/scripts/bootstrap-prefix.sh?format=txt
00:04:47 <pikhq> (see private message)
00:05:15 <cheater99> let's rewrite all of minecraft in sql!!!
00:05:34 * cheater99 is content having finally come up with the dumbest idea ever.
00:06:33 <elliottXP> pikhq: Won't this just compile all that shit with Cygwin?
00:06:38 <elliottXP> pikhq: How can I tell it I want a Win32 build?
00:07:11 <elliottXP> profile="${PORTDIR}/profiles/prefix/windows/interix/${C$
00:07:14 <pikhq> elliottXP: Uh... Should be an extra argument to the script.
00:07:19 <elliottXP> profile="${PORTDIR}/profiles/prefix/windows/winnt/${CHO$
00:07:27 <elliottXP> profile="${PORTDIR}/profiles/prefix/windows/cygwin/${CH$
00:07:33 <elliottXP> pikhq: I somehow doubt this will work ...
00:10:08 <pikhq> Good God there's a lot of stuff to bootstrap Gentoo Prefix.
00:10:46 <cheater99> hey oerjan i have something for you
00:10:48 <cheater99> http://www.amazon.com/Grace-Spicy-Cock-Soup-1-7oz/dp/B002Q46EH6/
00:11:51 <oerjan> elliottXP: um cheater99 started it
00:12:10 <elliottXP> oh did i, i wasn't paying attention
00:12:18 <elliottXP> oerjan: can you kick him, that would be great, thanks
00:12:30 <cheater99> oerjan: you can kick me in the ballz
00:12:31 <oerjan> elliottXP: i'll decide after visiting that link
00:12:43 <cheater99> oerjan: i'm quite into trampling today
00:13:01 <cheater99> elliottXP: i have a hunch he'd very much enjoy that topic...
00:13:18 <Gregor> Is cock soup somehow distinct from chicken soup ...?
00:13:23 <elliottXP> oerjan: oh please, just kick him and end our national nightmare
00:13:39 <elliottXP> lol cygwin bash fails at wrapping lines
00:13:45 <cheater99> elliottXP: what does this have to do with nationality?
00:14:08 <oerjan> elliottXP: ok i find the link to be insufficient reason for kicking
00:14:29 <cheater99> oerjan: do you find elliottXP's moaning to be sufficient reason?
00:14:36 <Gregor> I am seriously intrigued :P
00:14:44 <Gregor> I want to know why cock soup is different from chicken soup.
00:14:56 <Gregor> Unless the product actually /is/ a joke.
00:15:07 <elliottXP> oerjan: well cheater99 molested me you see.
00:15:19 <cheater99> elliottXP: you have such a slender bum
00:15:22 <Gregor> cheater99: Yes, but why would that make any difference?
00:15:33 <pikhq> There is actually a "cock-a-leekie soup".
00:15:37 <elliottXP> oerjan: also, I'm going to cry every day unless you kickban cheater99. (if you kickban me, I will continue to whine every day until cheater99 is kickbanned)
00:15:38 <Gregor> Do cocks and hens really taste so different in soup form?
00:15:53 <pikhq> Gregor: Not really.
00:15:58 <elliottXP> oerjan: hey you may appreciate :list foo in ghci
00:16:05 <elliottXP> oerjan: it shows the lines that define foo in a loaded module
00:16:30 -!- cheater99 has changed nick to chelliot.
00:16:50 <chelliot> i'm a cheater trappend in an elliott's body!
00:17:02 <oerjan> elliottXP: only for interpreted modules, alas
00:17:08 -!- augur has quit (Remote host closed the connection).
00:17:16 <elliottXP> oerjan: clear impersonation, please kickban chelliot
00:17:30 <chelliot> oerjan: could compiled modules not have relevant meta data in them?
00:17:38 <elliottXP> if you give me a reason you consider adequate I can endeavour to make them come true
00:17:39 -!- amca has joined.
00:17:51 <oerjan> chelliot: _source_ metadata?
00:17:58 <elliottXP> oerjan: see, he's an idiot. kickban him.
00:18:15 <oerjan> it's not like they're not bloated already
00:18:34 <chelliot> oerjan: python modules contain source meta data after compilation
00:18:46 <chelliot> elliottXP: i'm bloated after molestating you
00:18:48 <elliottXP> chelliot just suggested that Python does anything that could even remotely be considered compilation.
00:19:12 <chelliot> elliottXP: i am michael jackson, you are macaulay culkin
00:19:28 <elliottXP> oerjan: kickban chelliot because he keeps pinging me
00:20:31 <chelliot> elliottXP: you're clearly impersonating windows xp..
00:20:36 <quintopia> elliottXP: he's an exception to the /ignoring people is more annoying than it's worth rule. no one talks to him anyway. go for it.
00:21:10 <elliottXP> quintopia: oh I do ignore him on my main client, it's just that while I can see his messages, I might as well try and get him kickbanned before resorting to ignoring him
00:21:22 <elliottXP> i think he's the only person in here that more than one person ignores
00:21:28 -!- augur has joined.
00:21:46 <chelliot> elliottXP: do your separate split personalities count as separate persons?
00:22:05 <EgoBot> 76 ++++++++++[>++++++++++>+>><<<<-]>---.+.++++++++++++++.-.++++++.-------.++.>. [189]
00:22:28 <elliottXP> oerjan: we'd buy you a cake if you got rid of him.
00:23:01 <elliottXP> quintopia: can you bake one cake, to make it three?
00:23:36 <j-invariant> !bf_txtgen aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
00:23:38 <chelliot> elliottXP: he doesn't want to get fat and depressed like a certain person
00:23:41 <EgoBot> 277 ++++++++++++++[>+++++++>+++++++>+++++++>+++++++<<<<-]>>-...>>-....<-..<..>..<..>..<..>......<.....<-.>..........>>.......<.<...>...<....>.......<...>..<.......<........>.......>..<.<...>...---------------------------------------------------------------------------------------. [898]
00:24:01 <oerjan> j-invariant: that was underwhelming :D
00:24:10 <elliottXP> oerjan: kick him for being a poophead
00:24:40 <quintopia> what the fuck egobot? someone reimplement bf_txtgen plox
00:24:45 <chelliot> that's the funny thing about it
00:25:00 -!- oerjan has quit (Quit: Kicking myself out of this madness).
00:25:47 <quintopia> next time try reasonable human created heuristics combined by learning with boosting
00:26:24 <Mathnerd314> just use a compression algorithm + implement in BF
00:26:45 -!- BeholdMyGlory has quit (Remote host closed the connection).
00:27:08 <quintopia> most of the time that'd be rubbish Mathnerd314
00:27:10 <j-invariant> Mathnerd314: it sucks that that would work better
00:27:36 <j-invariant> how many strings of text are the shortest BF programs that generate them known?
00:28:09 <j-invariant> what's the longest string for which the shortest program is known?
00:28:17 <quintopia> very few? like...the first 20 characters tops :P
00:28:30 <quintopia> after that it gets too hard to prove
00:28:49 <elliottXP> it's essentially busy beaver. i think
00:29:03 <j-invariant> I wonder how hard it would be to push that limit
00:29:14 <j-invariant> I really do want to see the first nontrivial brainfuck program to deal with
00:29:16 <quintopia> i'm pretty sure that as soon as it's long enough to implement multiplication in, we have no idea
00:41:41 -!- cheater00 has joined.
00:44:47 -!- chelliot has quit (Ping timeout: 240 seconds).
00:45:46 <elliottXP> j-invariant: want a ban? i could convince copumpkin :D
00:46:06 <j-invariant> elliottXP: well you could ban Axman and monochrom and a bunch of other peopel
00:47:04 <copumpkin> no, I am not, and several other people don't believe in it
00:47:12 <elliottXP> copumpkin: I'll have you know that #esoteric has an inalienable right and tradition of talking behind people's backs without consequences
00:47:23 -!- Mathnerd314 has quit (Ping timeout: 272 seconds).
00:47:33 <elliottXP> for instance, copumpkin is a poopy head
00:48:17 <elliottXP> 16:41:14 <j-invariant> Any tips on brainfuck termination checker
00:48:19 <elliottXP> 6: are you using Unicode symbols in your source code?
00:48:30 <elliottXP> 16:41:33 <Entroacceptor> j-invariant: it's theoretically impossible
00:48:35 <elliottXP> 16:41:41 <j-invariant> wow you guys are not that smart
00:48:36 <elliottXP> 16:41:45 <j-invariant> I thought #haskell was really clever
00:48:45 <elliottXP> j-invariant: protip: being a jerk to someone who said something entirely correct isn't a good idea
00:48:52 <elliottXP> j-invariant: you could have phrased your question better
00:49:08 <copumpkin> termination checker doesn't imply halting problem
00:49:11 <elliottXP> like "any tips on a brainfuck termination checker (that doesn't work in all cases, just a large number)?"
00:49:17 -!- augur has quit (Remote host closed the connection).
00:49:26 <copumpkin> I'm with j-invariant on this one, but I just think his attitude needs some adjustment
00:49:30 <elliottXP> j-invariant: since a ton of people don't realise you _can't_ solve the halting problem and really DO want to write a perfect termination checker
00:49:37 <elliottXP> you have to get past people's moron filters
00:50:29 <copumpkin> I honestly think he has me on ignore though
00:50:57 <elliottXP> j-invariant: you wouldn't happen to have copumpkin on ignore?
00:52:28 -!- variable has quit (Quit: Daemon escaped from pentagram).
00:53:16 <Sgeo> "The price is too damned high"
00:53:31 <Sgeo> I didn't know Jimmy McMillan was on Star Trek
00:54:36 -!- Mathnerd314 has joined.
00:54:39 <j-invariant> 01:04 -!- mode/#haskell [+o monochrom] by ChanServ
00:54:40 <j-invariant> 01:04 -!- degaus [50d448d0@gateway/web/freenode/ip.80.212.72.208] has left #haskell [requested by monochrom (degaus)]
00:55:35 <copumpkin> [08:03:14 PM] <degaus> it must be more popular
00:55:35 <copumpkin> [08:03:26 PM] <degaus> because dons never talks about anything else
00:56:16 <elliottXP> Sometimes I forget that dons actually does cool stuff and isn't just a marketroid
00:56:54 <elliottXP> I have a feeling he puts more people off Haskell than encourages :P
00:57:11 <j-invariant> I hate this "pattern match" thinking that IRC encourages
00:57:29 -!- sshc has joined.
00:58:04 -!- sshc has quit (Client Quit).
01:02:27 <Sgeo> This permanent seeming action cannot be permanent
01:02:41 <Sgeo> Why am I surprised when shows do permanent-seeming actions
01:02:57 <Sgeo> elliottXP, not a person (well, a person too)
01:03:26 <j-invariant> Sgeo: what I don't get is why permanent changes are such a problem?
01:03:46 <j-invariant> Sgeo: do they make a whole bunch of episodes in a random order then shuffle them and decide this one goes before that?
01:03:51 <Sgeo> j-invariant, because with this change there would be no show
01:04:02 <Sgeo> So clearly it's not actually occuring
01:04:06 <Sgeo> It still weirds me out
01:04:13 <elliottXP> j-invariant: DS9 subverts that by having actual plotlines
01:04:30 <elliottXP> j-invariant: the avoidance of change is mostly so that random syndication works and also so that you don't have to remember tons of canon to write an episode
01:04:51 <j-invariant> oh and so that people can join late to watch your series
01:05:19 <Sgeo> If I waited a few minute
01:05:47 -!- comex has joined.
01:07:07 <pikhq> j-invariant: It's a fairly common thing in US TV, especially as shows tend to last several seasons.
01:11:59 <j-invariant> I have this wonderful vision of a site which shows the nth brainfuck program.. and if anyone submits a proof that it terminates or does not terminate then it moves the n+1th
01:13:01 <Sgeo> http://www.ncbi.nlm.nih.gov/pubmed/8137688?dopt=Abstract
01:13:08 <elliottXP> j-invariant: why, I wrote a program to get the nth brainfuck program! ;)
01:13:39 <j-invariant> elliottXP: the point is finding the first brainfuck program that nobody knows how to decide
01:13:52 <j-invariant> I believe this program is very interesting
01:14:58 <j-invariant> I just don't know how many months I would have to work on refining and refining a termination checker to find it
01:15:43 -!- td123 has joined.
01:15:59 <td123> is there a fork bomb for brainfuck?
01:16:10 <td123> is it even possible?
01:17:11 <Sgeo> I still have trouble with the Y-combinator
01:17:42 <j-invariant> Sgeo: the Y-combinator is simple got any questions?
01:18:11 <Sgeo> bomb() -> spawn(fun bomb/1, []), spawn(fun bomb/1, []).
01:18:19 * Sgeo probably screwed that up
01:18:56 <Sgeo> What's wrong with Erlang?
01:19:16 <Sgeo> And I want to know how to do that without the named function, which is why I asked after the Y combinator
01:20:52 -!- cal153 has quit (Ping timeout: 276 seconds).
01:24:48 <j-invariant> elliottXP: Example of how stupid people are:
01:24:49 <j-invariant> 01:34 < j-invariant> Does anyone know any good software for C which tries to decide if your program terminates or not
01:24:52 <j-invariant> 01:34 < j-invariant> I need a free software
01:25:27 <j-invariant> even the people that hang around in programming channels on IRC don't know anything
01:26:19 -!- td123 has left (?).
01:29:19 <Sgeo> j-invariant, maybe CaZe was, you know, JOKING
01:29:21 <copumpkin> j-invariant: it's sophomoric computer science
01:29:47 <copumpkin> they hear "halts" and parrot "impossible"
01:29:57 <j-invariant> maybe I should use rewrite systems instead of BF
01:30:01 <Gregor> Soundfonts suck at strings X_X
01:30:55 <quintopia> also, gregor, that's because ADSR fail
01:31:03 <Sgeo> Wait, did I fall for what copumpkin said idiots fall for?
01:31:05 <j-invariant> I stoped caring about spiked math when it made that stupid mistake a while back
01:31:07 * Sgeo dies a little inside
01:31:15 <Sgeo> stupid mistake?
01:31:30 <quintopia> j-invariant: fine. but give me a hint how to do this one plox?
01:32:31 <Sgeo> j-invariant, ok, what's the thing? That you're asking for programs written in C? Or that you said "tries"?
01:33:31 -!- pumpkin has joined.
01:34:02 -!- copumpkin has quit (Ping timeout: 240 seconds).
01:36:24 <Gregor> quintopia: VSTi's do strings OK, but it's difficult to mix and match X_X
01:37:36 -!- elliott has quit (Ping timeout: 255 seconds).
01:37:41 -!- elliottXP has quit (Ping timeout: 260 seconds).
01:44:03 <quintopia> now how do i do arbitrary precision arithmetic to calculate the actual answer
01:58:17 -!- cal153 has joined.
02:07:45 -!- Adiemus has joined.
02:09:10 -!- jcp has quit (Ping timeout: 240 seconds).
02:12:03 -!- Supersonic has quit (Ping timeout: 276 seconds).
02:12:37 <j-invariant> http://spikedmath.com/358.html haha that's a good one
02:14:55 -!- jcp has joined.
02:31:31 -!- j-invariant has quit (Quit: leaving).
03:01:39 -!- pumpkin has quit (Quit: Computer has gone to sleep.).
03:02:26 <Gregor> Why must DSSI/VST/Rosegarden be 100% non-automatable X_X
03:02:37 <Gregor> I don't like anything that I can't type "make" to build
03:26:07 -!- copumpkin has joined.
03:27:10 <Gregor> Where's my goddamn shell.
03:28:27 <pikhq> I don't *hate* GUIs.
03:28:50 <pikhq> I just have yet to see one I'm actually enthusiastic about.
03:29:17 <quintopia> i made a GUI IDE I didn't hate once.
03:32:09 * Sgeo clicks the Build menu
03:32:18 * Sgeo clicks Compile and Run ( Ctrl-F5 )
03:32:34 * Sgeo completely neglects the existence of keyboard shortcuts
03:33:09 <Sgeo> Although I guess that's not as not knowing Ctrl-C and Ctrl-V
03:33:14 <Sgeo> My a key is being weird
03:33:26 <Sgeo> Maybe I shouldn't allow crumbs to get near the keybord
03:51:46 -!- augur has joined.
03:58:32 -!- zzo38 has joined.
04:00:22 <zzo38> Does Google Chrome have a maximum line length for received data?
04:04:07 <augur> j-invariant didnt understand that the halting problem isnt a problem with turing machines alone? ahahaha
04:04:30 -!- azaq23 has quit (Ping timeout: 240 seconds).
04:06:37 -!- azaq23 has joined.
04:12:47 <quintopia> you haven't read enough of the discussion then
04:13:03 <augur> i only read a tiny bit of it
04:13:09 <quintopia> summary: he is bad at asking for what he really wants
04:13:27 <augur> but what i did see was his agreeing that it was about halting behavior, but then complaining that hes not talking about turing machines
04:13:34 <augur> and why is everyone talking about turing machines
04:13:54 <quintopia> because he wasn't looking for a halting decider
04:14:11 <quintopia> he was looking for a halting predictor for specific languages
04:15:19 <pikhq> Namely, he was wanting something entirely doable: a function that accurately returns {Halts, Doesn't halt, I don't know}.
04:16:09 <Sgeo> Isn't there something along the lines of algorithmically determining that such a thing is such a thing is also impossible?
04:18:02 <quintopia> *deciding* whether an algorithm has non-trivial property X is impossible
04:18:14 <quintopia> you can make as many recognizers as you want
04:18:41 <augur> i love that its decidable that some things are undecidable
04:19:04 <quintopia> because decidability of non-trivial properties is a trivial property
04:19:07 <augur> or that you can prove that you cant prove something
04:19:12 <quintopia> all non-trivial properties are undecidable
04:19:40 <augur> like, it's possible to prove the existence of unprovable propositions
04:20:19 <augur> and there are provably unprovable propositions about that proposition as well
04:20:41 <augur> its like some power tower of provable unprovability!
04:20:58 <Sgeo> Provably unprovable propositions can be false [I hope]
04:21:44 <Sgeo> People seem to talk about provably unprovable, but true, propositions a lot
04:22:32 <augur> any provably unprovably false proposition is a negation of a provably unprovably true proposition!
04:23:11 <Sgeo> Surely it's possible that there are provably unprovable, but potentially disprovable, statements?
04:23:12 <quintopia> do you even know what you're saying?
04:23:27 <pikhq> I suspect Sgeo is word-vomiting.
04:23:38 <augur> and i can provide you with logical forms in at least two semantic formalisms.
04:23:46 <augur> Sgeo: no certainly not
04:24:01 <augur> if a statement is disprovable then its negation is provable
04:24:11 <augur> since disproof of P is proof of not P
04:24:22 <augur> making the initial statement provable
04:24:27 <Sgeo> But its negation doesn't necessarily need to be provable
04:24:30 <augur> and so its false that its provably unprovable
04:24:31 <Sgeo> Is wht I meant
04:24:47 <augur> but ofcourse that could just be a proof that leads to a contradiction!
04:24:55 <augur> so its inconsistent but complete!
04:25:14 <quintopia> well, Gdel's sentence is both true and not provable. It doesn't even make sense to talk about sentences that are false and not provable...
04:25:18 <Sgeo> If a statement is disprovable, then its negation is provable. But its negation doesn't need to be disprovable, does it?
04:25:34 <augur> we dont know if its sentence is true or not, quintopia
04:25:51 <Sgeo> quintopia, A^~A is false and, assuming consistency, not provable
04:26:33 <Sgeo> We shouldn't be able to prove that a sentence is a Godel sentence... I think
04:26:44 <Sgeo> That would be tantamount to a proof of the sentence?
04:27:01 <augur> Sgeo: disprovable(S) = provable(~S) = ~provable(S)
04:27:22 <augur> therefore disprovable(~S) = ~provable(~S) = ~disprovable(S)
04:27:44 <augur> so if S is disprovable, its negation must NOT be disprovable
04:28:02 <Sgeo> I don't get how you got provable(~S) = ~provable(S)
04:28:25 <augur> i think we can agree that disprovable(S) = provable(~S)
04:28:28 <Sgeo> I think I'm working off a different definition of "provable" than.. is correct
04:28:31 <augur> and that disprovable(S) = ~provable(S)
04:28:47 <Sgeo> By provable, I meant not proven to not be provable
04:29:01 <Sgeo> Not necessarily true
04:29:02 <quintopia> augur: yes we do. G translates to "G is not provable in this system" Assume it is false. then, G is provable in this system. therefore, G is true. Contradiction. Now, assume it is true. We can't prove it, but there is no contradiction. Hence, it must be true.
04:29:06 <augur> and thus nonsensical
04:29:18 <pikhq> augur: It does not follow that something being disprovable means that something is *not* provable.
04:29:36 <augur> but thats what disprovable means!
04:29:48 <pikhq> ... Oh, wait, dur. Sorry. I'm stupid. Carry on.
04:29:59 <Sgeo> I think augur's definition of provable relies on the provable statement being tre
04:30:05 <Sgeo> That's what was headaching me
04:30:14 <augur> it depends i suppose on whether you mean "dis-" in the metaphysical sense or the epistemological sense
04:30:42 <Sgeo> Just, forget everything I said. I think.
04:30:54 <augur> because you can say "disprovable" as in we dont know whether or not something is true, but we know for certain that we could find out SOMEHOW
04:30:59 <Sgeo> pikhq, did you fall into the same trap I did?
04:31:05 <pikhq> Sgeo: Who the hell are you?
04:31:24 <augur> but you can also say "disprovable" as in we have a way to disprove this proposition.
04:32:30 <augur> so on the one hand, "P = NP" is a provable/disprovable proposition
04:32:49 <Sgeo> "It doesn't even make sense to talk about sentences that are false and not provable..."
04:32:54 <Sgeo> This STILL makes no sense
04:32:54 <augur> we dont know if its true or false, but the truth or falsity of it is (afaik) possible to prove
04:33:16 <Sgeo> The quoted text, I mean
04:33:24 <augur> but on the other hand, "P = NP" is neither provable nor disprovable, in the sense that we dont have the ability to give you that proof right now
04:33:32 <augur> obviously this is a fact about the notion of "ability"
04:33:46 <augur> and the same semantic ambiguity relates to modal verbs like "can"
04:42:00 <Gregor> OK, so soundfonts can do woodwinds decently, and brass acceptably, VSTi's can do strings quite well but are a huge pain in the arse, and bleh.
04:43:42 * Sgeo is too tired for this shist
04:44:22 <Gregor> augur: http://en.wikipedia.org/wiki/SoundFont
04:50:41 <Gregor> augur: See http://codu.org/music/e/superturing/superturing.ogg :P
04:52:29 <augur> able to leap tall call stacks in a single bound, its: Super-Turing!
04:52:52 <Gregor> "The theme music for SuperTuring, the newest (and best) superhero. Spider-Man may have spider powers, and Superman may have super-strength, but only SuperTuring can solve the halting problem! "
04:54:38 <augur> that music sucks btw
04:55:46 <pikhq> ... Is that Prelude in a different key on the harp? Seriously, WTF?
04:56:23 <pikhq> Also, electric guitar and MIDI = OUCH
04:56:34 <Gregor> pikhq: Electric guitar and VSTi also = ouch.
04:56:48 <Gregor> And no, that's not Prelude at all, not all arpeggios are Prelude.
04:57:19 <pikhq> CLEARLY UEMATSU INVENTED THE ARPEGGIO.
04:59:01 <Gregor> augur: ITYM "That music is so awesome it changed my life forever"
04:59:22 <augur> i dont speak obscure acronym
04:59:33 <Gregor> "I Think You Mean" is not an obscure acronym
05:03:22 <zzo38> Do you have a computer which is as fast as Superman to convert Celsius/Fahrenheit?
05:03:38 <Gregor> SuperTuring transcends clock speed.
05:04:23 <Sgeo> Gregor, can SuperTuring decide whether SuperTuring will ever die?
05:04:50 <Sgeo> So, SuperTuring is SuperSuperTurinng
05:04:53 <Sgeo> How incredible
05:05:02 <zzo38> You cannot have a computer like SuperTuring, because no actually built physical computer is even quite Turing, either.
05:05:15 <augur> Gregor: i can assure you i didnt mean that.
05:05:38 <Gregor> augur: I am quite certain that you did.
05:10:55 <zzo38> What I want you do today? Make a linear-scale music, meaning that the notes are 100Hz, 200Hz, 300Hz, 400Hz, 500Hz, and so on.
05:13:17 <Gregor> pikhq: He meant "this music is so awesome it changed my life forever", right?
05:14:33 <Gregor> Well, I have two non-fans X-P
05:20:55 -!- calamari has joined.
05:21:59 <Gregor> calamari: http://codu.org/music/e/superturing/superturing.ogg
05:24:56 -!- amca has quit (Quit: Farewell).
05:28:26 * calamari solves the halting problem with ease
05:29:35 <Gregor> For all but the true SuperTuring, the effect is temporary.
05:30:49 <Sgeo> Gregor, the beginning sounds like you stole it from Falcon's Eye
05:31:08 <Gregor> Considering that I have no idea what that is, I'm gonna go with "no"
05:31:19 <Sgeo> I think it's the similar instruments
05:31:57 <Sgeo> http://users.tkk.fi/jtpelto2/nethack.html
05:43:18 <zzo38> Gregor: Can you write any John Stump style music?
06:04:17 <zzo38> What size of chapter headings should I use (from \magstep1 to \magstep5)?
06:10:18 -!- Mathnerd314 has quit (Ping timeout: 246 seconds).
06:29:24 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
06:43:29 -!- poiuy_qwert has joined.
07:02:30 <Ilari> Just for fun, looking at allocations during 2010 (IPv4 addresses in /32s, IPv6 addresses in /64s):
07:03:13 <Ilari> IANA (direct): 0 IPv4, 0 IPv6
07:06:12 <pikhq> IANA has ceased to do direct allocations.
07:11:50 <Ilari> AFRINIC: IPv4: 8 520 960, 446 677 516 288 IPv6.
07:12:11 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
07:13:37 <Ilari> LACNIC: 17 278 976 IPv4, 197 569 937 408 IPv6.
07:13:49 -!- poiuy_qwert has joined.
07:16:06 <Ilari> RIPE NCC: 56 761 248 IPv4, 8 027 328 217 088 IPv6.
07:17:00 <Ilari> ARIN: 45 240 832 IPv4, 2 508 420 284 416 IPv6.
07:18:01 <Ilari> APNIC: 120 926 208 IPv4, 13 920 020 267 008 IPv6.
07:18:09 <pikhq> APNIC now at 1.38 /8s...
07:20:48 <Ilari> All RIRs combined except APNIC: 127 802 016 IPv4, 11 179 995 955 200 IPv6.
07:22:39 <Ilari> So APNIC allocated 48.6% of the total IPv4 space allocated last year and 55.5% of the total IPv6 space allocated last year.
07:24:38 <Ilari> One of the reasons for relatively low amount of IPv6 allocations might be that LACNIC region has extensive LIR system and some of those (especially Brazil) hold amazing amounts of IPv6 address space.
07:26:28 <pikhq> Seems kinda odd that it's that low for APNIC, also.
07:27:01 <pikhq> Given that much of Asia is actually on the ball with IPv6 adoption.
07:27:06 <Ilari> Actually only brazil there holds amazing amount of IPv6 address space, but that number is really amazing: Almost the same amount as rest of the world _combined_.
07:27:40 <pikhq> For instance, end user IPv6 has been easily available in Japan for 11 years now...
07:27:44 <Ilari> Err... 13.9 billion is more than rest of the world combined...
07:27:57 <Ilari> So they are allocating IPv6 a lot.
07:28:15 <pikhq> Well, all those numbers are going to be shooting way the hell up in the next couple of years.
07:28:18 <Ilari> Oops, 13.9 trillion
07:29:21 <pikhq> Well, actually, we'll probably see much less allocation of IPv6, relatively speaking, just because even a /48 is a *gigantic* allocation for anyone not actually running an ISP.
07:29:49 <Ilari> The standard for ISP is /32. And large ISP get blocks even larger than that...
07:31:35 <Ilari> Even IPv6 /32 is more network addresses than there are total host addresses in the entiere IPv4...
07:32:11 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
07:33:32 <Ilari> That's IPv4 mentality... :-)
07:33:46 -!- poiuy_qwert has joined.
07:34:07 <pikhq> No, really, it's ::ffff:0:0:0/96. :P
07:34:08 <Ilari> Note that I said "network address" about IPv6 and "host address" about IPv4.
07:34:59 <Ilari> It has some magic properties as address in some TCP/IP stacks (at least the one in Linux)...
07:35:51 <pikhq> It's intended to be a hook for magic properties in interacting between IPv4 and IPv6.
07:39:48 <fizzie> It's ::ffff:0:0/96; with 0:0:0 you'd have 48 bits of zeroes.
07:40:01 <fizzie> Alternatively, ::ffff:0.0.0.0/96, but that's just ugly.
07:40:24 -!- azaq23 has quit (Quit: Leaving.).
07:40:43 <pikhq> No, it's actually intended to have 48 bits of zeroes. (RFC 2765)
07:41:26 <fizzie> Oh, that's the IPv4-translated block.
07:41:28 <pikhq> Though ::ffff:0:0/96 is *another* form of IPv4 onto IPv6 mapping.
07:41:53 <pikhq> Gah, the transition methods are non-trivial.
07:42:07 <fizzie> The latter is what an IPv6 socket that has accepted an IPv4 connection returns for the peer address.
07:45:09 <fizzie> The whole "IPv6 wildcard-address-bound sockets also accept IPv4" thing is very messy. On Linux the port addresses are shared, so if you bind to IN6ADDR_ANY:1234 you can't bind to INADDR_ANY:1234 later, unless you setsockopt IPV6_V6ONLY on the v6 socket; on Windows it's similar except V6ONLY is set by default; on OS X you can bind to both, and the magical map-to-the-v6-socket only happens if there is no corresponding bound v4 socket.
07:46:18 <fizzie> And I think on some BSDs they don't (by default) map v4 connections to IN6ADDR_ANY v6 sockets at all.
07:46:50 <fizzie> Solaris at least treated those things completely separately, IIRC.
07:48:12 <fizzie> So you generally have to listen-an-accept on two sockets with some sort of select/poll thing, except that on some systems binding the second socket won't work.
07:51:35 -!- cheater00 has quit (Ping timeout: 240 seconds).
07:52:10 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
07:53:49 -!- poiuy_qwert has joined.
07:54:17 -!- cheater00 has joined.
07:59:25 -!- calamari has quit (Quit: Leaving).
07:59:35 <Ilari> Isn't there way to override IPV6_V6ONLY on per-socket basis?
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:00 <fizzie> Yes, with setsockopt before binding.
08:00:12 <fizzie> But I don't think the whole option is completely portable.
08:01:32 <Ilari> Well, anything that doesn't support it is probably either 1) Totally obsolete, 2) Windows (with who knows what TCP stack) or 3) Doesn't support IPv6 anyway.
08:01:47 <fizzie> It does exist on Linux, Windows and OS X, though, so I guess you can be portable to those three by explicitly either setting it + binding both v4 and v6 separately, or unsetting it and binding just a v6 wildcard socket.
08:01:59 <zzo38> I didn't know IPv6 has magic properties. Actually I don't know a lot about IPv6 in general.
08:03:24 <zzo38> How many addresses does an end user need?
08:05:55 <fizzie> 18 quintillion (short-scale) addresses is what you tend to get. At least that's what 3G specs to be given to each phone, for personal-area-networking stuff like that. You know, if you happen to be carrying 18 quintillion other devices that need to share the link.
08:06:58 <zzo38> That seems too much. Even 2048 addresses should be enough for someone with many devices. There are also port numbers, too.
08:07:06 <Ilari> Or if SLAAC is to be used...
08:08:34 <fizzie> "Stateless address autoconfiguration"
08:09:28 <fizzie> It's that thing that lets hosts automagically pick their addresses. The router sends a 64-bit prefix, and the host builds the other 64 bits out of (usually) its MAC address + fffe in the middle.
08:11:11 <zzo38> Is there some commands that IPv4 programs will work with IPv6 not requiring a change?
08:12:09 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
08:13:49 -!- poiuy_qwert has joined.
08:16:11 <fizzie> I'm not sure what "commands" here refers to. There's the getaddrinfo() socket API function; if you use that to translate addresses to names (and don't explicitly specify AF_INET or AF_INET6, but use AF_UNSPEC) things will mostly work on both IPv4 and IPv6 hosts. And then there are those tricky translation mechanisms.
08:18:00 <zzo38> I mean some things like that, and some API command to tell it to make a connection with a hostname and port number, and those things.
08:18:50 <fizzie> Yes, well. getaddrinfo() takes a host and a port (string, actually; you can give it names in addition to numbers) and returns a list of address structures you can give to connect().
08:22:07 <zzo38> But IPv6 addresses can have colons. So, maybe what could be done to fix that, is have the local DNS program to check if it ends with ".ipv6" and if so, will change the dots to colons and then resolve it to the corresponding IPv6 address. Is such things exists/possible?
08:22:55 <zzo38> Programs designed for IPv4 will probably not accept colons in the address, so you would have to do something like that?
08:26:03 <fizzie> If you give "a:b:c:d:e:f:g:h" style of address to getaddrinfo, it'll return a IPv6 address back; of course usually you'd just use names.
08:27:04 <fizzie> What often breaks though is v4-oriented programs that take a "host:port" string; they might not handle numeric v6 addresses right.
08:27:40 <fizzie> Incidentally, is the "[a:b:c:d:e:f:g:h]:port" format standardized anywhere?
08:28:08 <zzo38> Yes, the "host:port" string is what I am refering to, that is why you should have some local DNS program that converts forms with dots if it ends with ".ipv6", so that IPv4 programs will continue to work with IPv6.
08:29:07 <fizzie> Well, you could do that; haven't heard of anyone doing it though.
08:32:09 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
08:32:35 <Ilari> Sometimes programs supporting IPv6 are broken with at least numeric addresses: 1) Bugs in some resolver routines, 2) Bugs in handling the [addr]:port notation.
08:33:47 -!- poiuy_qwert has joined.
08:36:41 <fizzie> That would be easier to get right if they would have made getaddrinfo support "host:port", "v4addr:port" and "[v6addr]:port" style strings internally.
08:41:23 <quintopia> what's the best way to get an image of a remote container?
08:41:40 <quintopia> i can't scp cuz that follows symbolic links
08:42:06 <quintopia> and besides, i have it set up to not allow connections as root
08:43:33 <fizzie> "ssh host tar blah-to-make-a-tarball > ball_of.tar", vaguely speaking.
08:43:55 <fizzie> As non-root it might not be so easy.
08:44:36 <fizzie> I've done tar-over-netcat too, but that's something you'd probably only do in a local, safeish network.
08:44:56 <olsner> if you have any ability to run things as root, you can "tar blah-to-make-a-tarball | ssh somewhere blah-to-save-somewhere"
08:45:19 <fizzie> Then there's tar-over-netcat-over-ssh-forwarded-port.
08:46:15 <quintopia> well, just making a tarball and then downloading it really
08:46:34 <quintopia> as long as tarring the folder you're creating the tar inside doesn't break things...
08:48:34 <fizzie> nc -l -p 1234 > ball.tar + (in another term) ssh -R 1234:localhost:1234 host + sudo + tar blah | nc localhost:1234 is one way.
08:49:51 <olsner> the trick is to not save the tar, but pipe it directly over the network
08:50:03 <fizzie> Or "nc localhost 1234" I guess. Netcat command line details depend on the variant.
08:50:07 <quintopia> but also i need to disconnect irc first
08:50:25 <olsner> why would you have to?
08:50:50 <fizzie> Pipe it (base64-encoded) here on channel, and then extract from the clog logs?-)
08:50:55 <quintopia> i was thinking because the log files might update as it is read
08:51:05 <quintopia> what the hell is that command you posted fizzie
08:51:43 <olsner> that was four commands (on two different hosts) :P
08:51:43 <zzo38> Is there anything like netcat that can receive multiple connections at the same time, send stdin to all of them, and send data received from all of them to stdout?
08:51:50 <fizzie> It's just tar-over-netcat except fed through a SSH port forwarding for safety.
08:52:08 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
08:52:45 <fizzie> zzo38: Not that I know of, but it should be called netoctocat or something if it did exist.
08:53:24 <quintopia> you are opening a netcat receiver on my end, and then creating a tunnel for that port, and then connecting to the remote serving and tarring to netcat over that tunnel
08:53:47 -!- poiuy_qwert has joined.
08:55:45 <fizzie> It depends on your netcat.
08:55:59 <quintopia> man nc says basically you can't use those options together
08:56:11 <fizzie> Some take "-l port", some "-l -p port".
08:56:24 <fizzie> NAME nc - TCP/IP swiss army knife
08:56:24 <fizzie> SYNOPSIS nc [-options] hostname port[s] [ports] ... nc -l -p port [-options] [hostname] [port]
08:59:04 <quintopia> tar without -f sends output to stdout?
08:59:23 <quintopia> it looks like it sends it to the content of the TAPE env variable...
09:01:12 <quintopia> localhost:1234: forward host lookup failed: Unknown host
09:02:05 <quintopia> does that mean i'm blocking 1234 on my end?
09:02:07 <fizzie> Yeah, that should've been "nc localhost 1234", not "nc localhost:1234".
09:04:45 -!- cheater00 has quit (Ping timeout: 240 seconds).
09:09:21 <quintopia> i just got the "error exit delayed from previous errors" message
09:12:07 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
09:13:45 -!- poiuy_qwert has joined.
09:14:08 <quintopia> it was some socket ignoreds and a file changed as we read it
09:14:16 -!- zzo38 has quit (Remote host closed the connection).
09:16:31 <quintopia> what do i need to backup in order to preserve my iptables config? isn't it like iptables-save filename?
09:17:29 <fizzie> Yes, and ip6tables-save too if you have any.
09:17:51 <fizzie> Or "iptables-save > filename", I think.
09:23:43 -!- cheater99 has joined.
09:24:49 -!- poiuy_qwert has quit (Read error: Operation timed out).
09:29:26 <quintopia> this netcatting thing is fun ... but there's no way to tell when the file has finished sending :P
09:30:15 <quintopia> anyway, i think i backed up the things i care about now...
09:33:47 -!- poiuy_qwert has joined.
09:47:53 -!- poiuy_qwert has quit (Ping timeout: 246 seconds).
10:32:52 -!- hagb4rd has joined.
10:40:16 <Ilari> Transfer rate should drop a lot after it finishes... :-)
10:40:18 -!- cheater99 has quit (Ping timeout: 240 seconds).
10:41:11 -!- cheater99 has joined.
10:58:55 -!- coppro has quit (Ping timeout: 255 seconds).
11:00:34 -!- coppro has joined.
11:02:17 -!- cheater99 has quit (Ping timeout: 240 seconds).
11:05:44 -!- poiuy_qwert has joined.
12:07:03 <Ilari> Haha... "That's good news - we need to make sure we have a /3 for both the Moon and Mars colonies. ;)".
12:07:42 <Ilari> Nah... I think RIR-class notional allocation (/7) would be enough for the Moon... :-)
12:08:37 <Ilari> There are currently two of those free...
12:11:44 <Ilari> Oops, not 2 but 10.
12:14:16 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
12:16:35 -!- poiuy_qwert has joined.
12:29:03 -!- oerjan has joined.
12:34:31 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
12:36:11 -!- poiuy_qwert has joined.
12:54:31 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
12:56:11 -!- poiuy_qwert has joined.
13:03:17 -!- Mathnerd314 has joined.
13:14:30 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
13:16:10 -!- poiuy_qwert has joined.
13:18:15 -!- Sgeo_ has joined.
13:20:37 -!- Sgeo has quit (Ping timeout: 240 seconds).
13:27:18 -!- Mathnerd314 has quit (Ping timeout: 240 seconds).
13:33:45 -!- augur has quit (Remote host closed the connection).
13:34:30 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
13:36:09 -!- poiuy_qwert has joined.
13:54:29 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
13:56:07 -!- poiuy_qwert has joined.
14:09:54 -!- poiuy_qwert has quit (Ping timeout: 246 seconds).
14:16:43 -!- Phantom_Hoover has joined.
14:19:26 -!- sebbu has quit (Read error: Connection reset by peer).
14:19:55 -!- sebbu has joined.
14:24:09 -!- j-invariant has joined.
14:25:19 -!- augur has joined.
14:28:43 -!- elliott has joined.
14:34:45 <elliott> pikhq: can you relink me that interix iso?
14:34:51 <elliott> i refound it but it's only downloading at 20 kilobytes / sec
14:35:01 <elliott> thinking i might be on a bad mirror
14:35:39 <elliott> 03:42:41 • Sgeo clicks the Build menu
14:35:40 <elliott> 03:42:50 • Sgeo clicks Compile and Run ( Ctrl-F5 )
14:35:40 <elliott> 03:43:06 • Sgeo completely neglects the existence of keyboard shortcuts
14:35:50 <elliott> I was going to something but I can't wrIte.
14:37:56 <elliott> 04:37:34 <augur> Sgeo: disprovable(S) = provable(~S) = ~provable(S)
14:38:01 <elliott> augur: excluded-middle scum
14:39:59 <Ilari> disprovable(S) is not the same as ~provable(S).
14:41:17 <copumpkin> augur: in constructive logic, I can't prove double negation elimination or LEM, but that doesn't make them false
14:43:55 <elliott> EXCLUDED MIDDLE IS FALSE ///HARDCORE CONSTRUCTIVISM///
14:44:12 <elliott> (OK, OK, so any intuitionistic logic + ~LEM is almost certainly inconsistent.)
14:44:26 <elliott> 08:56:57 <quintopia> well, just making a tarball and then downloading it really
14:44:34 <elliott> quintopia: if you did it like "ssh host tar foo >blah", you could skip the downloading
14:44:44 <elliott> by having it stream directly to your disk over the network
14:44:51 <elliott> Ilari: yeah augur is full of shit :P
14:45:02 <elliott> disprovable(S) = provable(~S)
14:45:14 <elliott> disprovable(S) also => ~provable(S)
14:45:33 <elliott> ~provable(S) does _not_ => provable(~S)
14:45:35 -!- BeholdMyGlory has joined.
14:46:25 <elliott> augur: hi, you said wrong things
14:46:31 <augur> copumpkin: im not a constructivist
14:46:35 <elliott> <augur> Sgeo: disprovable(S) = provable(~S) = ~provable(S)
14:46:38 <augur> elliott: also i didnt
14:46:52 <elliott> there are statements where ~provable(S) and ~provable(~S)
14:46:57 <elliott> which violate what you said
14:47:24 <elliott> augur: disprovable(S) = provable(~S). provable(~S) implies ~provable(S).
14:47:37 <elliott> copumpkin: tell augur why he's wrong kthx
14:47:46 <elliott> too early to deal with stupid statements
14:47:50 <augur> elliott: this depends on the meaning of provable, which we commented on afterwords
14:47:54 <augur> please read those comments
14:48:01 <elliott> augur: no, it doesn't, there's exactly one meaning of provable
14:48:11 <elliott> and it has nothing to do with whether humans have managed to prove it yet
14:48:25 <elliott> and there's exactly one meaning of disprovable, "negation is provable"
14:48:25 <augur> elliott: yes, well i was speaking in english
14:48:29 <augur> not in precise terms
14:48:36 <elliott> <augur> Sgeo: disprovable(S) = provable(~S) = ~provable(S)
14:48:38 <elliott> oh yeah that's some english
14:48:45 <elliott> augur: the precise terms were exactly what matter
14:48:47 -!- oerjan has quit (Quit: leaving).
14:48:54 <elliott> sgeo was going on about godel and all the time you were talking about ...
14:48:59 <elliott> statements that people haven't proven yet?
14:49:07 <elliott> i have no idea what you were even talking about tbh
14:49:34 <quintopia> and i even gave a specific counterexample ... ~provable(G) and ~provable(~G)
14:51:03 <elliott> i think augur just admitted he lost the argument
14:53:21 <j-invariant> 14:48 < elliott> 04:37:34 <augur> Sgeo: disprovable(S) = provable(~S) = ~provable(S)
14:53:53 <elliott> j-invariant: SORRY EXCUSE ME BY "DISPROVABLE", AUGUR MEANT "NOT YET PROVED OR DISPROVED BY HUMANS" APPARENTLY
14:53:58 <j-invariant> 14:54 < elliott> EXCLUDED MIDDLE IS FALSE ///HARDCORE CONSTRUCTIVISM///
14:54:37 <elliott> j-invariant: i also follow the axiom of LIFE
14:54:47 <elliott> because of my deep religious beliefs
14:59:44 -!- cheater99 has joined.
15:10:03 -!- Phantom__Hoover has joined.
15:10:33 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
15:12:58 <Phantom__Hoover> Oh, he admitted that he didn't have an argument and ran away because he is an intellectual coward.
15:18:38 <Phantom__Hoover> OK, so I'm being bitter, but he's done this before to me and cpressey.
15:19:42 <elliott> Hmm, I *do* need someone to blame for making cpressey leave that isn't me ...
15:20:26 <elliott> copumpkin: Yeah, I blame you.
15:20:49 <elliott> copumpkin: DRIVE AWAY THE MOST FAMOUSEST MEMBER OF THE ESOLANG COMMUNITY, WOULD YOU
15:21:14 <elliott> copumpkin: http://catseye.tc/
15:21:17 <elliott> copumpkin: inventor of Befunge
15:21:29 <Phantom__Hoover> He just said something stupid which we told him was wrong, and he basically stuck his fingers in his ears and sang "LALALALALAI'MRIGHT" until we got tired and gave up, then gloated about winning the argument.
15:22:00 <elliott> I think augur has a tendency to ... respond to questions using precise terminology by assuming they mean something else entirely.
15:22:29 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
15:22:45 <Phantom__Hoover> He said IIRC that a TM that forbade one specific program couldn't be TC, which was blatantly wrong.
15:23:24 <elliott> if you forbid a countable set of programs then maybe.
15:24:08 <Phantom__Hoover> You'd need computational equivalence to do it so that a UTM was completely impossible.
15:24:32 <elliott> Phantom__Hoover: It's arguable whether the language L = {a Brainfuck interpreter} is Turing-complete.
15:25:02 <elliott> Phantom__Hoover: L could be another language (say, Brainfuck) with a countably infinite set of programs removed (all apart from that one program).
15:25:21 <Phantom__Hoover> elliott, yes, I know. But the point stands that just adding steadily more complex nops would make it uncomputable to actually create such a TM.
15:25:26 <elliott> Specifically, it's TC by the "has a program that implements a UTM" definition, but not by the "has one program for every Turing Machine".
15:25:42 <elliott> (The latter is the original definition of a UTM generalised to arbitrary languages.)
15:25:51 <elliott> Phantom__Hoover: Eh? Erm, whatever.
15:27:13 <Phantom__Hoover> elliott, i.e. that you can't actually *prove* that an arbitrary program is equivalent to the forbidden one(s), so it's not possible to prevent TCness without making it superturing.
15:27:31 <elliott> Oh, I was assuming it was just syntactic forbiddenness.
15:27:38 <elliott> i.e., if prog in forbiddens: slkjgsdkhjl
15:28:27 <elliott> <elliott> Phantom__Hoover: It's arguable whether the language L = {a Brainfuck interpreter} is Turing-complete.
15:28:37 <elliott> That's Brainfuck minus a countable subset of programs, and I don't think it's Turing-complete.
15:29:44 <Phantom__Hoover> elliott, i.e. it's just if prog /= <specific BF interpreter>: stop else <BF intepreter>?
15:30:52 <j-invariant> or maybe Is hould do lambda calculsu instead
15:30:57 <elliott> theoretically impossible :D
15:31:05 <elliott> j-invariant: lc is probably easier to reason about in Coq, but
15:31:08 <Phantom__Hoover> j-invariant, pikhq has that super-optimising x86(-64?) compiler that he still hasn't shown me.
15:31:09 <elliott> I think harder to check termination for?
15:31:23 <elliott> j-invariant: bf seems easier to check termination
15:31:27 <quintopia> the haskell one would be a nice accomplishment
15:32:04 <quintopia> j-invariant: do it anyway, just for the challenge
15:32:21 <elliott> bf interps are not a challenge
15:32:35 <Phantom__Hoover> LC, perhaps, simply due to having to deal with alpha equivalence.
15:33:06 <quintopia> elliott: it'd be a challenge for me, who knows no haskell
15:33:23 <elliott> Phantom__Hoover: you don't have to
15:33:45 <elliott> or just substitute directly when applying
15:34:18 -!- elliottXP has joined.
15:35:13 <j-invariant> 15:44 < Phantom__Hoover> But that's trivial.
15:36:08 <Phantom__Hoover> j-invariant, it seems unrelated to his point, which was anyway poorly articulated and was never supported with anything approaching that argument.
15:36:43 <j-invariant> I can't decide whether to use brainfuck or lambda
15:37:29 <elliott> j-invariant: I have no idea how you'd write an LC termination checker
15:37:36 <elliott> brainfuck at least has several easy halting types
15:37:47 <elliott> including the extended euclidean thing for a certain type of loop
15:38:06 <elliott> j-invariant: really though
15:38:10 <elliott> j-invariant: use esotope-bfc's output
15:38:18 <elliott> j-invariant: it detects certain simple infinite loops for you
15:38:21 <elliott> j-invariant: and e.g. can figure out arithmetic
15:38:25 <elliott> j-invariant: and also gives you While loops and the like
15:48:18 <elliott> 04.11.04:14:52:59 <cpressey_> hello :)
15:48:18 <elliott> 04.11.04:14:53:31 --- nick: cpressey_ -> cpressey
15:48:18 <elliott> 04.11.04:16:16:25 <cpressey> yep
15:48:18 <elliott> 04.11.07:19:57:21 <cpressey> heatsink: re type inference: the dragon book contains a good description & algorithm
15:48:19 <elliott> 04.11.07:19:58:42 <cpressey> np
15:48:21 <elliott> 04.11.07:19:58:57 <cpressey> it still took me a loooong time to figure out exactly what was going on with it :)
15:48:24 <elliott> 04.11.07:19:59:20 <cpressey> replacing the greek letters with T1, T2, T3 and walking myself through abunch of examples seemed to help
15:48:27 <elliott> 04.11.07:20:02:47 <cpressey> no
15:48:29 <elliott> 04.11.07:20:03:44 <cpressey> word-final :)
15:48:32 <elliott> either cpressey was lagged that day, or clog has ignores
15:50:11 <elliott> 19:33:53 <jDoctor> so uh whats the norm discussion here?
15:50:11 <elliott> 19:34:06 <heatsink> nothin'
15:50:11 <elliott> 19:34:22 <heatsink> This channel isn't active most of the time
15:50:11 <elliott> 19:34:33 <slava> lets make it active!
15:50:34 <elliott> 19:35:33 <slava> i'm working on stack effect inference for postfix languages
15:50:34 <elliott> 19:35:48 * heatsink has no idea what that is
15:50:34 <elliott> 19:36:05 <heatsink> what is that?
15:50:34 <elliott> 19:36:21 <slava> eg, the stack effect of 2 2 + is [ 0 | 1 ] because it takes no values from the stack, but leaves one
15:50:34 <elliott> 19:36:38 <slava> the stack effect of dup * is [ 1 | 1 ], because it takes one value, duplicates it, multiplies the two duplicates, to yield one value
15:51:24 <elliott> 19:46:47 <heatsink> What language syntax do you use? Is it taken from an existing language?
15:51:24 <elliott> 19:47:51 <heatsink> Hey, the inventor of factor is named slava too! what a coincidence!
15:52:26 <elliott> 14:31:20 <fizzie> not related, but mooz could perhaps say something here about his "random programs" experiments.
15:52:28 <elliott> 14:31:40 <fizzie> it's esoteric enough.
15:52:33 <elliott> 14:32:14 <fizzie> apparently he keeps finding composite-number-factoring algorithms at a surprisingly high rate. :p
15:52:33 <elliott> fizzie: is that a joke...?
15:53:47 -!- copumpkin has joined.
15:55:34 <elliott> 11:01:00 <Keymaker> hi all! i'm first time here with linux
15:55:34 <elliott> 11:02:39 <Keymaker> (i hadn't linux before)
15:55:34 <elliott> 11:02:39 <mtve> client is sirc i hope?
15:55:34 <elliott> 11:02:39 <Keymaker> it's something x-chat
15:55:35 <elliott> 11:02:39 <Keymaker> but i don't know anything about this
15:56:18 <elliott> lol nooga hasn't changed since 2004
15:56:33 <elliott> 13:55:00 <lament> mmmm befunge
15:56:33 <elliott> 13:56:58 <fizzie> My girlfriend complains when I speak of scheme and lambdas in bed, maybe I should try talking about befunge.
15:56:33 <elliott> 14:04:19 <lament> Maybe you should speak of brainfuck.
15:56:33 <elliott> 14:06:00 <fizzie> I find incomprehensible rectangular blocks of befunge code very exciting, maybe she would also?
15:57:22 <elliott> 21:35:06 --- quit: ChanServ (ACK! SIGSEGV!)
16:09:49 -!- elliottXP has quit (Remote host closed the connection).
16:22:52 <elliott> WHO WANTS TO JOIN MY EPIC LOG-READING PROJECT
16:22:59 <elliott> Phantom__Hoover?! Er... clog?!
16:24:11 <Phantom__Hoover> <elliott> 14:06:00 <fizzie> I find incomprehensible rectangular blocks of befunge code very exciting, maybe she would also? ← clearly she did, assuming she's the one he married.
16:24:20 <elliott> The secret to a happy marriage.
16:42:17 -!- augur has quit (Remote host closed the connection).
16:47:33 -!- impomatic has joined.
16:50:17 -!- augur has joined.
16:50:38 <augur> elliott: two points. 1) by disprovable i mean "in principle could be proven or disproven but not necessarily already so", as in "P = NP is a (dis)provable proposition, but we haven't yet found the (dis)proof"
16:50:40 <augur> and 2) "bai" was meant as "i wont be responding because im in class"
16:51:00 <elliott> (1) that isn't what disprovable means.
16:51:06 <elliott> disprovable is a formal term meaning "negation is provable".
16:51:12 <elliott> provable means "proof exists" not "proof has been found".
16:51:23 <elliott> don't argue, because it simply /isn't/ what disprovable means in the context of logic
16:51:32 <elliott> and my definition is the one Sgeo_ meant too, as it is the only relevant one
16:51:37 <augur> elliott: i already agreed to this on grounds that theres a muddling of non-technical language
16:51:38 <elliott> in the context of goedel's incompleteness theorems
16:51:38 <Phantom__Hoover> Wait, can you just clarify here that you no longer think that provable(¬P) = ¬provable(P)?
16:54:05 <augur> elliott: actually i also think that provable(~P) -> ~provable(P)
16:54:19 <elliott> but ~provable(P) does not -> provable(~P).
16:54:26 <augur> the reverse might be true if we have a closed world assumption
16:54:31 <elliott> and therefore provable(~P) =/= ~provable(P)
16:54:33 <augur> which i think is necessarily what we dont have
16:54:37 <elliott> augur: consider the axiom of choice in ZF
16:55:45 <augur> but this also, i suppose hinges on what "dis-" means. obviously if we dont take "dis-" to be analysable in this context, fine
16:56:07 <augur> but i think its true, in common parlance at least, that "dis-" is negation
16:56:12 <j-invariant> Th being the axiomatic theory S is provable in
16:56:24 <elliott> augur: in logic, disprovable(P) means provable(~P).
16:56:31 <elliott> that's just not arguable terminology
16:56:39 <j-invariant> If you say things like "provable" or "disprovable" without respect to an axiom system you will end up writing a book like GEB one day
16:56:40 <augur> but i mean that like
16:56:49 <elliott> tr.v. dis·proved, dis·prov·ing, dis·proves
16:56:49 <elliott> To prove to be false, invalid, or in error; refute.
16:56:50 <elliott> disprovable (not comparable)
16:56:50 <elliott> Capable of being disproved.
16:57:00 <augur> if its impossible to prove that P, i think that constitutes a disproof of P
16:57:13 <j-invariant> 17:07 < elliott> augur: in logic, disprovable(P) means provable(~P).
16:57:24 <elliott> augur: in Zermelo-Fraenkel set theory, the Axiom of Choice is neither provable nor disprovable.
16:57:28 <elliott> augur: it is neither true nor false.
16:57:30 <j-invariant> In logic, disprovable_Th(P) has meaning though
16:57:32 <augur> ok, obviously this is true only in certain systems
16:57:36 <elliott> augur: both ZFC and ZF~C are valid.
16:57:39 <augur> because of completeness
16:57:44 <elliott> augur: no, it's true in no sufficiently powerful consistent system
16:57:59 <coppro> You need the completeness<->satisfiable equivalence
16:58:07 <augur> elliott: but in a consistent system, either proposition is either probably true or provably false
16:58:09 <coppro> also your axiom system must be consistent
16:58:18 <augur> so if you cant prove it's true, it must be possible to prove its false
16:58:27 <Phantom__Hoover> <j-invariant> If you say things like "provable" or "disprovable" without respect to an axiom system you will end up writing a book like GEB one day ← in that it can only be kept on reinforced steel bookcases?
16:58:30 <elliott> <augur> elliott: but in a consistent system, either proposition is either probably true or provably false
16:58:31 <coppro> augur: No, there are many consistent systems in which you can neither prove or disprove something
16:58:42 <augur> er sorry, you said consistent
16:58:47 <coppro> For instance, ZFC is a consistent system that cannot prove or disprove the continuum hypothesis
16:58:54 <elliott> in a complete system, yes.
16:59:01 <elliott> but there are no useful complete systems :)
16:59:01 <augur> i wasnt reading fully
16:59:05 <augur> and since i said complete
16:59:16 <augur> hey boolean algebra is both complete, consistent, AND useful!
16:59:26 <augur> not that its terribly complex
16:59:40 <coppro> (note that the completeness<->satisfiable equivalence, in an axiomatic logic system, follows from the principle of explosion and one other statement that I need to dig out of my notes)
17:00:18 <elliott> i hate text editors they are so bad.
17:00:24 <coppro> oh right, ((a->(~a))->~a)
17:01:00 <j-invariant> elliott: it's like teaching kids { x | P(x) } instead of { x in X | P(x) }
17:01:19 <elliott> j-invariant: yes but disprovable(P) is perfectly valid INFORMAL SHORTHAND
17:01:28 <coppro> if you have those two as theorems of your system, then you have completeness<->satisfiable which has all the nice properties you expect
17:01:36 <coppro> darn you for forcing me to unignore elliott
17:01:39 <elliott> in informal language, convenience wins over precision when the rest can be inferred
17:01:46 <elliott> j-invariant: informal discussion of formal logic
17:01:52 <elliott> in this case, the _theory_ wasn't relevant
17:02:00 <elliott> especially if left as a free variable
17:02:12 <quintopia> i agree with elliott here. it is understand that we are speaking of a particular system, namely any system where GIT applies.
17:03:10 <quintopia> but i can't prove that in this system
17:03:14 -!- elliott has set topic: githouse | http://208.78.103.223/esoteric/.
17:03:14 <augur> elliott: why do text editors suck
17:03:49 <coppro> there are lots of systems in which (~a) and (a) are provable; namely all inconsistent ones
17:04:14 <quintopia> why is vim unusable? why is emacs unusable? why is <insert other text editor you've used> unusable?
17:04:24 <elliott> because they're either vastly inefficient or nearly impossible to commit to muscle memory
17:04:48 <elliott> quintopia: With enough time I could probably turn Emacs into a usable text editor, but that's just writing a text editor myself, except with keybinding clashes.
17:04:54 <elliott> no, not pebkac, just unrealistically high standards
17:04:57 <augur> elliott: make a usable one then!
17:05:26 <coppro> I was expecting interesting conversation
17:05:27 <elliott> quintopia: in #esoteric we are generally more lenient to those who are dissatisfied with the current status of computing.
17:05:51 <elliott> being unrealistic isn't a problem if you make attempts at fixing it, which i have
17:05:57 <augur> im utterly dissatisfied with the current status of computing
17:06:07 <augur> mostly because i dont know whats new and interesting
17:06:15 <augur> coppro, elliott: whats new and interestion
17:06:16 <coppro> augur: obviously your ACM membership isn't up to date
17:06:19 <quintopia> but what's the closest *realistic* approximation to your standards?
17:06:32 <augur> coppro: i dont need an acm membership, im at a university!
17:06:37 -!- FireFly has joined.
17:06:46 <j-invariant> like I said yesterday, computer science is dead
17:06:47 <elliott> quintopia: well. acme is quite nice. yaedit has good finger feel and muscle memory.
17:06:58 <augur> coppro: what interesting stuff have you read lately
17:07:10 <coppro> augur: step 1) pick a field
17:07:11 <elliott> coppro doesn't do interesting things
17:07:18 <coppro> step 2) find the people at your university in that field
17:07:28 <augur> i dont know what the field is that im interested in
17:07:34 <augur> but its not really a field
17:07:50 <elliott> lol i was off ignore for coppro?
17:08:31 <elliott> i swear coppro gets paid by the number of times he mentions that i'm on ignore
17:08:34 <elliott> also, whining about minecraft discussion
17:08:57 <augur> constraint logic seems interesting, in principle
17:09:10 <augur> but a lot of the techniques ive seen involve just prologesque resolution
17:09:29 <coppro> you should try to solve 3-SAT in polynomial time :P
17:09:42 <augur> i cant prove p = np!
17:09:45 <coppro> but seriously; find the people, then find the research
17:09:48 <j-invariant> get interested in practical programming with dependent types
17:10:01 <j-invariant> We need an army of people interested in that in order to get things moving
17:10:20 <augur> elliott: whats your favorite dependent type language again? coq?
17:10:33 <elliott> they all suck! but you can define and prove things in Coq so that's a plus
17:10:54 <elliott> augur: come back in ten years
17:11:10 <augur> i hate you elliott :(
17:11:11 <j-invariant> since it's actually possible to automate proofs
17:11:15 <quintopia> coppro: any idea on the status of the latest try on that front?
17:11:18 -!- cheater99 has quit (Ping timeout: 240 seconds).
17:11:53 -!- cheater99 has joined.
17:12:09 <quintopia> (or rather, the latest try at showing it can't be done :P_
17:14:18 <augur> showing that what can't be done?
17:14:51 <quintopia> 12:20 < coppro> you should try to solve 3-SAT in polynomial time :P
17:17:25 <elliott> maybe i'll add haskell indentation support to yaedi
17:25:12 <Phantom__Hoover> NO, GOOGLE, NOT EVERYONE HAS AN SMS-CAPABLE DEVICE TO HAND.
17:25:47 <Phantom__Hoover> http://upload.wikimedia.org/wikipedia/commons/5/5f/Birdsniper.jpg
17:29:22 <quintopia> yaedit seems okay, but that planned "smart syntax-checking and variable coloring" thing sounds awful nice. wonder how long that will take.
17:29:58 <elliott> quintopia: well the main thing i like about yaedit is its interface and its auto-save
17:30:18 <elliott> the interface, i.e. a spartan GUI where everything is shown at all times, accessed by the keyboard
17:30:28 <elliott> that informs you how to use the keyboard to do things when you do them with your mouse (e.g. clicking in the line dialogue)
17:30:36 <elliott> the line jumping/search is very well-done
17:30:38 <quintopia> elliott: auto-save kind of bugs me, but then, i want a text-editor that i can use as a textual scratch pad and not just programming
17:30:59 <elliott> quintopia: in leaden (my editor project), auto-save is coupled with VCS-save
17:31:09 <elliott> quintopia: that is, every change is saved to disk automatically, and Ctrl+S asks for a commit message and does a VCS commit
17:31:22 <quintopia> basically, i want to be able to use it as a multi-line input buffer that isn't associated with any file
17:31:31 <elliott> quintopia: also, undo information is stored to a file in ~/.leaden
17:31:36 <elliott> so you can undo, say, days of changes
17:31:39 <elliott> even after reopening a file
17:31:51 <elliott> the auto-save thing is great because you can test changes to a program extremely rapidly
17:31:58 <elliott> and the Ctrl+S vcs integration makes version control a lot nicer
17:32:02 <augur> Phantom__Hoover: i dont know if that word you mentioned is a word or not. that will depend on the person.
17:32:21 <quintopia> elliott: so what happens when you are editing without specifying a file?
17:32:32 <elliott> quintopia: it doesn't save it... either that or it saves it in /tmp
17:33:23 <quintopia> does leaden do all the things you like about yaedit?
17:33:50 <elliott> quintopia: also: smart language modes
17:33:58 <elliott> pervasive syntax highlighting and auto-indentation
17:34:01 <elliott> handled _programmatically_
17:34:09 <elliott> i.e., the indenter for a language is an actual program
17:34:17 <elliott> so even complex languages like Haskell can be auto-indented
17:34:53 <elliott> quintopia: also: extensive project support _without_ any project files
17:35:03 <elliott> if you have a certain directory open as a project, then you can "open in project" extremely rapidly
17:35:10 <elliott> e.g. if you have foo.h in myproj/foo/bar/baz/
17:35:29 <elliott> (assuming this is unambiguous; you select from a list)
17:36:10 <elliott> quintopia: I lost the code to the initial prototype and haven't yet got the energy to rewrite it?
17:36:46 <quintopia> so you have a binary with no source?
17:37:12 <elliott> quintopia: no, i just don't have it :)
17:37:27 <elliott> all this stuff is pretty easy to implement, but i didn't yet get around to it... if i still had the basic editor code i'd be implementing them now
17:37:35 <elliott> i had auto-indent working for python
17:38:52 <quintopia> what are the odds you'll come upon a link for your little roguelike that still has its source in it?
17:39:31 <elliott> quintopia: do you not have a copy?
17:40:39 <quintopia> i don't think so. if i did, it'd be one of the earliest ones
17:43:42 <elliott> i think cheater99 has it >__>
17:46:16 <elliott> quintopia: wiped drive by accident, then pastie.org went all fag and deleted old pastes
17:47:59 <elliott> quintopia: no, i was reinstalling linux and backed up my shit to the windows partition
17:48:03 <elliott> which i then accidentally deleted
17:48:41 <quintopia> linux is smaller than windows. basically always. i would have still tried recovering
17:50:25 <elliott> quintopia: but there was nothing of value :P
17:50:28 <elliott> other than vagrant, basically
17:52:52 <quintopia> rewrite vagrant for maximum awesomeitude
17:55:20 <elliott> quintopia: it was already max awesome
17:58:47 <elliott> quintopia: find it for me then fucker
18:10:20 <elliott> 09:37:43 <elliott> Gregor: OMG
18:10:20 <elliott> 09:37:47 <elliott> The smallest enemies, right
18:10:20 <elliott> 09:37:52 <elliott> Little tiny critters that are only a mild annoyance
18:10:20 <elliott> 09:37:55 <elliott> Gregor: Make them the favicon.
18:10:20 <elliott> 09:38:02 <elliott> The website fights back!
18:10:20 <elliott> 09:38:06 <elliott> It doesn't want you to steal its images!
18:10:22 <elliott> 09:38:14 <Gregor> ... YES. YES, YES.
18:10:24 <elliott> 09:38:29 <Gregor> Favicon goombas!
18:10:35 <elliott> I LIKE HOW THE FAVICON GOOMBAS WERE FUCKING TERRIFYING RATHER THAN A MILD ANNOYANCE
18:11:00 <Gregor> Dude ... they're goombas.
18:11:39 <quintopia> did i ever give them a death sequence? i don't remember...
18:13:35 <quintopia> then that was my next task i was trying to remember :P
18:13:54 <quintopia> also a giant flying boss with 1000 HP
18:14:34 -!- elliott_ has joined.
18:14:42 <elliott_> Gregor: They're goombas and YOU HAVE ONE HEART OF HEALTH.
18:14:51 -!- elliott has quit (Quit: Leaving).
18:14:52 <Gregor> elliott_: That's your fault.
18:17:19 <Gregor> elliott_: Also, in Mario * you have one hit worth of health until you get a powerup, and they still throw goombas at you *shrugs*
18:17:32 <elliott_> Gregor: Mario's collision detection is A LOT MORE FORGIVING :P
18:17:47 <elliott_> And it's easier to move around without slipping.
18:19:54 <quintopia> < qntm> Switching from Courier New to Consolas has made a tangible improvement to my programming happiness.
18:20:11 <quintopia> i can't believe he used courier new
18:20:29 <quintopia> it is a cool looking font, but does not look good in an editor
18:21:24 <Gregor> You crazy fontophiles.
18:23:22 <elliott_> I use Courier New in my XP VM :P
18:23:36 <elliott_> (My XP VM has font antialiasing disabled.)
18:23:42 <elliott_> Anyone know a decent Windows 7 torrent...?
18:23:54 <quintopia> i just use Droid Sans these days...
18:23:57 <elliott_> Trying to find a cracked one without any extra bullshit...
18:25:12 <quintopia> my terminal is set to use droid sans though
18:25:35 <elliott_> quintopia: Droid Sans *Mono*, more likely
18:27:42 <quintopia> cheater99: do you have um, say, like, a Vagrant, like, loitering on your system somewhere, that, uh, you'd like to, erm, share, pretty please?
18:29:29 <cheater99> didn't someone mention something about a minecraft server //__ //
18:29:51 <quintopia> i don't know about that. can't help you there.
18:30:21 <elliott_> cheater99: i demand vagrant.py
18:30:28 <quintopia> actually, i do know a few good ones...i just dont know #esoteric's
18:30:55 <elliott_> cheater99: I am not allowed to give you the server address, nor is any other member. Please give me my program.
18:32:25 <quintopia> cheater99: give it to me then? i promise i won't give that file to elliott. i'll work on it myself.
18:32:26 <elliott_> Please cut the childishness, I merely want the single latest .py file that is my program.
18:32:50 <cheater99> it's called game.py >________________>
18:33:30 <elliott_> Please just give me my program.
18:33:41 <cheater99> i feel the ratio of you being non-nice to me vs me being non-nice to you is not balanced!
18:33:49 <j-invariant> did you accidentally delete your program and this guy has a backup?
18:34:37 <elliott_> cheater99: Look, I'll be nice, just give me my program.
18:35:03 <impomatic> I'm a fan of Droid Sans Mono. I just wish it had a dotted 0
18:35:13 <cheater99> -rw-r--r-- 1 cheater cheater 2.0K 2010-10-06 01:48 game.py
18:35:20 <cheater99> elliott: is this actually genuine?
18:36:18 <elliott_> cheater99: yes. please give me my program.
18:36:28 <elliott_> j-invariant: that's the name of the program
18:37:11 <j-invariant> cheater99: you're not setting a good example
18:37:25 <elliott_> cheater99: Can you put it on sprunge/any pastebin?
18:37:47 <cheater99> i'm just trying to figure out which file is which
18:40:08 <elliott_> cheater99: Can't you just upload all of them?
18:40:21 <cheater99> http://hpaste.org/43340/vagrand_alternate
18:41:43 <elliott_> i don't know which of those is more recent. hm
18:41:49 <fizzie> VagRand, the random number generator for ladies only.
18:42:07 <elliott_> is that all the versions you have?
18:42:54 <cheater99> before that they would beeline you
18:43:05 -!- pikhq has quit (Ping timeout: 240 seconds).
18:43:07 <elliott_> both these versions are broken
18:43:11 <cheater99> now they can even go around obstacles
18:43:18 <cheater99> but they're not big bugs from what i remember
18:44:58 <elliott_> in fact they raise no exceptions
18:45:26 <cheater99> i forgot what that was... but we were able to figure it out.
18:48:53 <elliott_> cheater99: do you not have the more recent code with my rewritten ai?
18:51:44 -!- pikhq has joined.
19:00:34 <cheater99> u = int((1-2*(r(0,10)/10))*o(1,11+y-B)+11-B),int(A-40+o(1,40+x-A)*(1-2*(r(0,10)/10)))
19:00:34 <cheater99> if u in w and w[u] in (0,32,33,36,37) and u not in s:
19:00:48 <cheater99> (that's with one extra indent for the try/except)
19:02:51 <cheater99> in fact you can take off those int()'s
19:03:03 <elliott_> that's your lame ai version though
19:03:12 <elliott_> what's the bug in the other :P
19:03:23 <elliott_> hm i suppose i could backport my ai to the first one
19:03:42 <elliott_> does anyone know a good X server for windows?
19:05:40 <Gregor> Are there even any options besides the one that comes with Cygwin and XMing?
19:06:20 <elliott_> Gregor: I don't think so... but the thing with Xming is that the author thinks that only offering a version from 2007 for free and then asking for a "donation" (read purchase) to get something recent is a good strategy.
19:06:41 <elliott_> Great way to pretend to be an "open source community program" while charging for it ..
19:07:39 <elliott_> Gregor: We evidently have different definitions of cool.
19:08:02 <elliott_> Gregor: If it was "Newer versions cost money and are released for free after four years", fine.
19:08:11 <elliott_> But it's worded in a slimy way.
19:08:33 <Gregor> I <3 both "Donor Password" and the incorrect use of "Public Domain"
19:08:39 <Gregor> It's still under the X11 license, right?
19:08:50 <elliott_> Gregor: Only the PUBLIC DOMAIN versions.
19:08:59 <elliott_> The cost versions are proprietary, pay-for software, plain and simple.
19:09:09 <elliott_> Almost makes you wanna use the GPL :
19:09:11 <Gregor> Well that's retardalicious.
19:09:22 <Gregor> Just makes me want to not use Windows *shrugs*
19:09:27 <elliott_> Gregor: Hell, I'd prefer a restricted free version rather than just an OBSOLETE free version.
19:09:38 <elliott_> Gregor: The problem here is a moron, not Windows :P
19:10:02 <Gregor> And yet, I'm running a recent version of X11 ;)
19:10:17 <Gregor> (Again, where's my goddamn TROLL FACE codepoint)
19:10:49 <elliott_> Gregor: Cygwin/X exists ... and requires Cygwin, which is the worst piece of software ever created :P
19:11:20 <elliott_> Gregor: I wonder if your ELF loader thing would work with Interix.
19:11:31 <Gregor> I can't imagine why not.
19:11:45 <Gregor> I mean, it might require some minor tweaking to get the image offset right, but otherwise.
19:14:09 <quintopia> Gregor: there is that one trollface emoticon that's <funkyunicode>_<funkyunicode>
19:14:23 <Gregor> 'snot trollfacey enough :(
19:16:12 <Gregor> Y'know, when the electric guitar isn't shredding MIDIly, it's actually not that bad ;P
19:16:56 * Gregor had no luck finding any VSTi's that do electric guitars any better ... particularly at the end, where it's really vital.
19:17:41 <elliott_> "crumbs", he said, "i'm all out of crumbs"
19:17:59 <Gregor> cheater99: End of http://codu.org/music/e/superturing/superturing.ogg
19:18:04 <elliott_> Gregor: Play it on a real guitar.
19:18:13 <Gregor> elliott_: I don't play the geetar
19:18:16 <elliott_> Gregor: In fact, play all of it.
19:18:25 <elliott_> Gregor: Yeah, but neither does SuperTuring's theme tune :P
19:18:41 <Gregor> cheater99: ... more reverb? :P
19:19:01 <Gregor> I really think elliott_ is exaggerating how bad the guitar is, all things considered it's pretty darn good :P
19:19:10 <elliott_> Gregor: It's just not hardcore enough.
19:19:19 <Gregor> elliott_: Well that's definitely true.
19:20:08 <Gregor> cheater99: Oh it does not.
19:20:46 <cheater99> you totally need to work on your mix
19:21:46 <cheater99> well, the fact that everything shares the same reverb is not helping at all
19:22:04 <Gregor> But I'm not an audiophile.
19:22:11 <cheater99> and that you're not using any eq on anything
19:22:28 <cheater99> and that all velocities are set to 100 or some other constant number
19:22:35 <cheater99> elliott_: it sounds like an mpu 401
19:22:39 <elliott_> No they're not... at least I doubt they are.
19:22:41 <cheater99> if that's what you want, it sounds awesome
19:22:46 <Gregor> cheater99: No, they're not.
19:22:59 <cheater99> but, i wager a bet that wasn't gregor's intention
19:23:06 <elliott_> cheater99: It's meant to be a cheesy theme tune.
19:23:14 <elliott_> So, yes, it's meant to sound like that.
19:23:17 <Gregor> It is most assuredly meant to be cheesy, yes.
19:23:25 <Gregor> It's not supposed to be dramatic, it's supposed to be ridiculous.
19:23:40 <Gregor> Or rather, ridicoulous in a way that makes it sound like it's trying to be dramatic but failing hard.
19:23:56 <cheater99> are you recording it all with vst synths?
19:24:24 <Gregor> 's all soundfonts ... my experience with VSTs is somewhat limited and I find them very annoying.
19:24:44 <elliott_> I decided to pretend VSTs existed when I found out that they all have custom skins.
19:24:47 <cheater99> do you have a physical midi interface?
19:24:50 <elliott_> Custom, terrible skins meant to look like REAL AUDIO HARDWARE MAN.
19:25:07 <cheater99> here's a trick you can use to make it more cheesy
19:25:11 <Gregor> The problem I have with VST's is that I can't "make" and go from source -> sound.
19:25:19 <cheater99> dump the output as a midi file through the interface
19:26:27 <Gregor> Uhhh, awesome, but that's not really the way in which it's supposed to be cheesy :P
19:26:38 <Gregor> Well, not necessarily ... maybe.
19:28:45 <cheater99> also another thing you can do is play it back through cheesy speakers and a cheesy microphone, and mix that with the original
19:29:03 <cheater99> especially when lots of bass comes in and the tiny speaker starts farting
19:29:24 <elliott_> I don't think cheater99 has a very good conception of cheesy.
19:33:41 <Gregor> The real problem here is that I'm a classically-trained musician, and electronic music ain't my thing ... I'm used to composition and performance being well-separated, and a lot of the aspects of tonality being related to environment rather than human fine-tuning. And I hate that fine-tuning, and I'm not an audiophile :P
19:34:01 <cheater99> well... the first half a minute isn't tonal at all
19:34:37 <Gregor> Sorry, I meant character, not tonality >_>
19:34:46 <Gregor> As in, character of tone.
19:35:40 <Gregor> That is to say, things like reverb are out of the hands of both the composer and the performer.
19:36:16 <Gregor> Even making my hand-played piano works sound as good as I'd like is annoying for me because I am not an acoustical engineer :P
19:36:38 -!- miekko has joined.
19:37:42 <Gregor> Anyway, I'm learning, so shut your facehole.
19:37:46 <Gregor> (Which you already have :P )
19:39:51 <elliott_> who is this miekko guy, i blame fizzie
19:39:55 <elliott_> looks suspiciously fizzie-related
19:40:45 <cheater99> Gregor: you have no idea what an awesome amount of work goes into fine-tuning the reverberation of a concert area!
19:40:58 <Gregor> cheater99: Fine-tuning BY ACOUSTICAL ENGINEERS, not the performers.
19:41:01 <Gregor> cheater99: That's my goddamn point.
19:41:08 -!- elliott_ has changed nick to elliott.
19:41:11 -!- elliott has quit (Changing host).
19:41:11 -!- elliott has joined.
19:41:11 <cheater99> you're the acoustical engineer here
19:41:18 <cheater99> since you're building from ground up
19:41:22 <Gregor> THAT - IS - MY - POINT
19:41:27 <cheater99> alternatively, you can get altiverb
19:41:50 <elliott> "Altiverb 6 is a convolution reverb plug-in for Mac OS X and Windows XP."
19:41:52 <elliott> "Mac OS X and Windows XP."
19:42:05 <cheater99> or use one of those philharmonic ensemble samplers that also have a reverb/space placement solution
19:42:20 <elliott> I suppose they have the same stunning Linux support too.
19:42:34 <cheater99> elliott: i wish audio software would come to linux :(
19:42:42 <Gregor> So basically, by dumping hundreds of dollars into something I make no money on, I can make something good enough to make cheater99 say "this is tolerable"
19:42:57 <elliott> cheater99: My 100% disbelief that Gregor will stop using Linux for his music especially when he's stated that he doesn't want any solution that he can't use from the command-line for the fact that it already sounds fine.
19:43:06 <Gregor> cheater99: I don't pirate stuff.
19:43:11 <cheater99> elliott: well, you can use windows vst under linux
19:43:18 <elliott> Welp, cheater99's going back on ignore now that I have what I want.
19:43:48 <Gregor> cheater99: The only VSTs I've found that enable my absolute requirement of "make -> music" are DSK's (quite awesome) VST's of brass and strings. The brass I haven't tried integrating into this yet.
19:44:40 <Gregor> Or rather, I tried integrating it's trumpet and it's not as good (IMHO) as the relatively-good soundfont I have. Its trombone MUST be better than the one I have since the one I have has shitty audible looping X_X
19:45:11 <Gregor> But controlling the dynamics mix when you have both soundfonts and VSTis in a make-able way is a PITA *sigh*
19:45:15 <miekko> elliott: I am a Finn, I study CS and am writing my bachelor's thesis on complexity theory, no association to any fizzie of any kind
19:45:26 <elliott> miekko: Aren't all you Finns related?
19:45:37 <Gregor> Related by Finnish EVIL.
19:45:39 <elliott> Are you sure fizzie isn't your uncle or something?
19:45:59 <cheater99> Gregor: there are soundfont vst's.
19:46:05 <Gregor> cheater99: Terrible ones.
19:46:31 <Gregor> cheater99: FluidSynth is still, by a wide, wide margin, the best SoundFont renderer for any platform available legally for free.
19:47:07 <miekko> elliott: maybe, but if that is the case, his identity as fizzie is unknown to me
19:47:10 <Gregor> (I haven't tried all of them, blah blah blah disclaimer)
19:47:26 <miekko> I was told to come here by slereah
19:50:43 <fizzie> I'm not in abo.fi at all.
19:50:52 <fizzie> I'd blame the oko instead.
19:51:57 <fizzie> Anyway, not everything that ends in .fi is related to me.
19:53:00 <cheater99> in finland everyone is a second cousin
19:53:28 <miekko> uhm, I am probably not a second cousin to everyone else, because my village is so badly inbred that it's ...
19:53:36 <miekko> well, let's say, all my second cousins are my third cousins
19:54:00 <cheater99> how many ancestors 20 generations ago would that be?
19:55:08 <miekko> of course it grows exponentially, but you can also make exponential-sized cuts in it
19:55:54 <miekko> if we're going to have it grow exponentially, we're assuming every grandparent only is present in one place
19:56:04 <miekko> generations may be of different length
19:56:22 <miekko> so your paternal whateverwhateverwhatever may be half-sibling with your maternal whateverwhateverwhateverwhatever
19:56:39 <miekko> I think we should avoid counting anyone twice?
19:56:53 <elliott> miekko: you are not helping the "finns are not crazy" case here
19:57:05 <miekko> well, there's undoubtedly been multiple bottlenecks throughout biological history
19:57:15 <Gregor> The whole human race is related by incest :P
19:57:41 <elliott> I never thought my life would turn out like this ... updating a Gentoo installation that's running on Windows...
19:57:58 <miekko> hey, how's gentoo these days?
19:58:13 <miekko> I abandoned it a few years ago
19:58:44 * Gregor begins screaming "Debian" over and over again and foaming at the mouth
19:58:55 <elliott> miekko: I hate Gentoo, it's just that nothing else is maintained for Interix...
19:59:06 <elliott> miekko: (POSIX kernel running on top of Windows NT.)
19:59:11 <elliott> It's like Cygwin, except not terrible :P
19:59:12 <Gregor> elliott: Why do you have Interix? :P
19:59:24 <elliott> Gregor: Because if I didn't, I'd have Cygwin, and Cygwin is the worst piece of software ever invente.
19:59:28 <Gregor> elliott: Why in the sense of "by what totes-legal means did you obtain it"
19:59:40 <elliott> Gregor: By downloading it from microsoft.com, it's free :P
20:00:00 <elliott> Gregor: You almost certainly want to download the Gentoo Prefix ISO instead.
20:00:04 <elliott> That gets you things such as "a gcc newer than 3.3".
20:00:06 <miekko> hm, I guess gentoo isn't worth getting back to then
20:00:09 <elliott> Gregor: And it installs Interix for you.
20:00:16 <Gregor> INTERIX COMES WITH GCC?!?!
20:00:27 <elliott> Gregor: Yes, and Gentoo Prefix comes with gcc 4.2.
20:00:30 <elliott> You can of course update that.
20:00:41 <Gregor> But ... it's a Microsoft product!
20:00:47 <elliott> Gregor: In fact, you can use Gentoo Prefix to build native Windows applications with GNU make...
20:00:56 <elliott> Gregor: It also comes with ksh... Gentoo Prefix gets you bash :P
20:01:28 <elliott> Gregor: The only Interix available for XP is 2004-vintage but it works fine. If you have Windows 7 Ultimate (Ultimate only... well, also Enterprise) it comes with Interix.
20:01:40 <elliott> If you have a Windows Server OS (why?!?!!) it comes with it.
20:01:44 <elliott> If you have some other Windows 7 version sucks to be you :P
20:01:57 <elliott> But yeah, Interix and Gentoo Prefix are actually pretty polished...
20:02:00 <Gregor> I have Windows No Edition
20:02:05 <elliott> For instance fork() is instant because it isn't implemented in a retarded way on top of Win32.
20:02:09 <elliott> Gregor: That's the BEST edition!
20:02:45 -!- pikhq has quit (Ping timeout: 265 seconds).
20:03:09 <elliott> My god, emerge is slow to synchronise.
20:06:21 -!- impomatic has left (?).
20:07:55 <fizzie> elliott: My only Windows here is Windows Server 2003; I suppose that'd work. (Except I somewhat fail to see the point.)
20:08:18 <elliott> fizzie: I *think* you need the standalone for that, only Vista's server onwards have it. *Think*.
20:08:24 <elliott> fizzie: Also, the point is FUN!
20:08:42 <elliott> fizzie: You could probably build mcmap with the existing Makefile and Gentoo Prefix :P
20:09:15 <fizzie> Heh. Would it then work on non-POSIX-subsystemy Windowses?
20:09:45 <elliott> fizzie: Yes, you set it up using the I-forget-the-name Gentoo thing so that "cc" is just a wrapper around Visual Studio.
20:09:57 <elliott> fizzie: Admittedly it would be easier just to use MinGW with the Gentoo Prefix's Makefile.
20:10:22 <elliott> fizzie: One thing that Cygwin can do that Interix can't is build a hybrid Win32/POSIX program.
20:10:28 <elliott> e.g. a Win32 program that uses ptys.
20:12:37 <elliott> Is "emerge --sync" meant to take five years?
20:13:17 -!- oerjan has joined.
20:13:18 -!- pikhq has joined.
20:13:28 <elliott> pikhq: How long is $ emerge --sync meant to take.
20:15:37 <Vorpal> elliott, first sync or later on?
20:15:48 <Vorpal> elliott, since it is rsync the time will vary
20:15:52 <elliott> Vorpal: First sync after install of what I think is a stage1.
20:16:07 <elliott> I think it might be synchronising everything...
20:16:09 <Vorpal> elliott, not stage3? I thought stage1 wasn't supported any more
20:16:18 <elliott> Vorpal: Well, maybe. This is Gentoo Prefix.
20:16:23 <Vorpal> elliott, well yes first time it will fetch the portage tree
20:16:34 <elliott> Except it says it's deleting things, so I bet I already had one.
20:16:40 <elliott> Vorpal: It has gcc 4.2 and stuff, so maybe it is stage3.
20:16:47 <Vorpal> elliott, well probably
20:16:55 <Vorpal> elliott, maybe a very out of date tree?
20:17:06 <elliott> Vorpal: 2010-03; so only about 7 months out of date.
20:17:13 <elliott> I guess that's reasonably close to "very" for computers.
20:17:45 <Vorpal> elliott, for the portage tree: extremely
20:20:07 <elliott> I wish a better distribution was maintained for Interix :P
20:20:27 <elliott> Hm, Funtoo was started by the same guy as Gentoo?
20:20:39 <elliott> I seem to recall Funtoo's website having something crazy on it...
20:22:26 <elliott> So, er, is it wise to try and install a new package with Portage before updating the system?
20:28:24 <Vorpal> elliott, it might not complain though
20:28:29 <elliott> I don't want to devote the next five years to updating the system :P
20:28:32 <Vorpal> elliott, initial sync can take like 20-30 minutes
20:28:38 <Vorpal> elliott, as far as I remember
20:28:43 <Vorpal> elliott, you might be using a bad mirror
20:28:48 <Vorpal> did you select a UK mirror
20:28:53 <elliott> Vorpal: I didn't select any mirror.
20:28:57 <Vorpal> elliott, or did you go with the default US one
20:29:00 <elliott> Vorpal: Note that I doubt every mirror has Gentoo Prefix.
20:29:01 <Vorpal> elliott, that's the issue then
20:29:14 <elliott> So this is probably one of, like, 3 I could pick :P
20:29:28 <elliott> Vorpal: Anyway I don't mean update, I mean upgrade.
20:29:33 <Vorpal> elliott, iirc the default US mirror is rate limited if you don't run an official mirror since it is the master sync server for them
20:29:38 <elliott> Vorpal: Can I install a package without upgrading the world? (Isn't that the TERMINOLOGY?)
20:29:39 <Vorpal> or something like that
20:30:19 <Vorpal> elliott, you can but the issue with doing this while it is syncing could be that 1) ebuild is changed while building 2) deps might be messed up possibly
20:30:56 <elliott> So do the binary packages have different names or whatever? I just want the basic X programs i.e. xlogo, xterm, etc. and if there are binary versions that is preferable.
20:31:28 <Vorpal> I'm starting to forget them
20:31:39 <elliott> Vorpal: Won't be long until he does, he's a Debianer now
20:33:44 <Vorpal> elliott, so why windowas
20:34:09 <elliott> Vorpal: Because this is entertaining.
20:34:22 <elliott> It disturbs me slightly that the Windows in a VM is performing better than its Ubuntu host.
20:34:44 <Vorpal> elliott, is it xp 64-bit?
20:35:00 <elliott> It's also allocated only 768 megs of ram and a single core.
20:35:02 <Vorpal> elliott, then I'm utterly surprised
20:35:16 <elliott> Vorpal: To be fair, I did disable almost every service... but still.
20:35:19 <Vorpal> elliott, 64-bit XP is the best windows version
20:35:32 <elliott> I dunno about that, very little supports 64-bit on XP :P
20:35:43 <elliott> XP is pretty good though if you configure it.
20:35:55 <elliott> Although Windows 7 has several advantages...
20:36:13 <Vorpal> elliott, xp 64 bit was based on 2003 server iirc
20:36:16 <elliott> Phantom__Hoover: Yes. And don't ask for a comprehensive list of why.
20:36:18 <Vorpal> which explains why it is snappy
20:36:40 <elliott> Vorpal: NanoXP is based on XP SP3 corporate.
20:36:43 <elliott> (Corporate = doesn't bug you about activation. Ever.)
20:36:55 <elliott> Vorpal: And it's the snappiest Windows around! :P
20:37:47 <elliott> The problem with the server ones is that they're usually configured to give priority to background services, not applications.
20:37:51 <elliott> Which is the opposite of what you want.
20:37:57 <Vorpal> elliott, do nano-2003!
20:37:58 <elliott> And if you change that, well, they're essentially the consumer edition :P
20:38:14 <elliott> Vorpal: Maaaaybe... 2003 is very un-desktop-suitable by default.
20:38:48 <elliott> Vorpal: That will never happen, since XP/64-bit is incredibly unsupported by everything :P
20:39:06 <elliott> Windows 7's start menu is better than XP's at least... because you can just type in a program name and hit enter :P
20:40:41 <Vorpal> elliott, hey I was able to play an old windows 9x game on xp 64
20:40:52 <Vorpal> elliott, planescape iirc
20:41:18 <elliott> Major advantage of NanoXP: It uses the Windows 2000 installer interface.
20:41:22 <elliott> Which is waaaay less gaudy :P
20:41:32 <elliott> ...and doesn't even _install_ the Fisher Price theme.
20:41:59 <fizzie> Is that the XP default blue thing?
20:42:09 <elliott> Vorpal: You know, Fisher PRice.
20:42:29 <Sgeo_> Vorpal, go eat shit fuckers
20:42:38 <Sgeo_> [Possibly not what elliott was referring to]
20:42:45 <elliott> NanoXP comes with two themes: Windows Standard, and Windows Classic. Both are Windows Classic themes, the latter looks like Windows 95 :P
20:44:18 <elliott> Vorpal: Have I mentioned that the resulting C:\WINDOWS is 200 or so megs?
20:44:40 <Sgeo_> slava was in #esoteric ?
20:44:43 <elliott> MORE BENEFITS: C:\Programs instead of C:\Program Files! C:\Users instead of C:\Documents and Settings!
20:45:04 <Sgeo_> elliott, I was being an idiot
20:45:11 <Sgeo_> For a proof to exist, a statement must be true
20:45:19 <elliott> Sgeo_: NOT IF YOU'RE INCONSISTENT
20:45:23 <elliott> Vorpal: Pestov, creator of jEdit, Factor.
20:45:46 <elliott> Oh, wait, I forgot his greatest achievement, Flying Shits 2000.
20:46:11 <elliott> http://factorcode.org/slava/FlyingShits.html
20:46:36 <Sgeo_> So, you have to catch elliott?
20:46:38 <elliott> EMERGE STILL HASN'T FINISHED SYNCING
20:46:40 <Vorpal> "You need a Java compatible browser to run this applet!"
20:46:43 <Vorpal> elliott, will check later
20:46:49 <Vorpal> since I can't run more than w3m atm
20:46:50 <elliott> Vorpal: It's quite possibly not worth it :P
20:47:01 <Vorpal> elliott, summary of it then?
20:47:15 <elliott> It's a game where you catch feces.
20:47:24 <elliott> Evil shits are coming to Earth from Uranus! The only way to stop them is to catch them with your Super-Loo.
20:47:36 <elliott> It was, as the page said, made when he was 13 :P
20:47:44 <Sgeo_> I seem to be having some trouble with it
20:48:05 <Sgeo_> Clicking does nothing
20:48:23 <Sgeo_> Also, some of the text is cut off
20:48:44 <elliott> "The Mac OS X cat program is not actually interrupt-safe"
20:49:21 <elliott> Ohh, {^Raven^} is jonripley.
20:50:08 <Vorpal> Phantom__Hoover, I bet gnu could anywhere
20:50:17 <Vorpal> interrupted system call... why the heck really
20:50:22 <Vorpal> EINTR is the stupidest thing ever
20:50:28 <Phantom__Hoover> They have a unique talent for messing up the simplest programs.
20:50:37 <elliott> is the basic transcript on os x
20:50:45 <elliott> Vorpal: EINTR is basically the PC luser problem, isn't it?
20:50:50 <elliott> As outlined in Worse is Better...
20:51:03 <Vorpal> elliott, I prefer correctness over simplicity
20:51:32 <elliott> Vorpal: I prefer reducing the problem until making it correct requires no thought :P
20:51:41 <elliott> For instance, the obvious thing to do here is to not have signals.
20:51:47 <Vorpal> elliott, well sure that would be nice....
20:51:49 <Gregor> [20:58] eazyigz: can't i manually gc? I mean like in C/C++?
20:51:49 <Gregor> [20:58] bradleymeck: you can use delete
20:51:49 <Gregor> [20:58] brianc: maybe "delete"
20:51:50 <Gregor> [20:58] Gregor bashes all three of you in the head.<CTCP>
20:51:52 <Gregor> [20:59] Gregor: "Manually GC" is meaningless, you cannot manually allocate/deallocate, and delete in JavaScript is not the same as delete in C++.
20:51:53 <elliott> Ha, it wants me to update portage.
20:51:54 <Gregor> [20:59] eazyigz: Gregor: but v8 is c++ based, not JS based
20:52:14 <elliott> Gregor: I suggest you /part that channel forever
20:52:15 <Vorpal> <elliott> Ha, it wants me to update portage. <-- yeah if it wants that you should
20:52:19 -!- pumpkin has joined.
20:52:32 <elliott> * IMPORTANT: 2 news items need reading for repository 'gentoo_prefix'
20:53:13 -!- Behold has joined.
20:53:21 <elliott> Sweet, cmd.exe fails at Unicode :P
20:53:39 <elliott> eselect is ridiculously slow ...
20:53:52 <elliott> Is it written in bash or something? :P
20:55:19 -!- copumpkin has quit (Ping timeout: 255 seconds).
20:56:19 <Vorpal> elliott, hm... eselect is iirc written in bash yes
20:56:27 <Vorpal> elliott, presumably fork() still sucks
20:56:46 -!- BeholdMyGlory has quit (Ping timeout: 276 seconds).
20:56:47 <elliott> Vorpal: Interix is implemented at the same layer as Win32; i.e. directly on top of NT.
20:56:55 <elliott> Vorpal: NT has code in it specifically to support fork().
20:57:02 <elliott> So, no, Interix fork() is actually pretty fast.
20:58:34 <elliott> Vorpal: Remember that Interix is actually used in ENTERPRISEY environments.
20:58:43 <elliott> Vorpal: Unlike Cygwin, which is a creaky, half-broken toy.
20:59:40 <elliott> "/opt/gentoo/usr/lib/portage/bin/chpathtool"
20:59:46 <elliott> You might think that this is the most ridiculous path ever.
20:59:56 <elliott> Allow me to tell you what its path is in the Win32 subsystem.
21:00:08 <elliott> C:\SFU\opt\gentoo\usr\lib\portage\bin\chpathtool
21:00:11 <elliott> THAT is the most ridiculous path ever.
21:00:36 <Vorpal> elliott, what happens if you put a literal \ in a filename using this
21:00:58 <Vorpal> elliott, after that: newline in filename (yay)
21:01:05 <elliott> Newlines in filenames are a bug :P
21:01:17 <elliott> They're never useful and make it harder to use things like find etc. safely.
21:01:28 <elliott> Because you need a separate print-with-\0s function... which is useless for console output.
21:01:35 <elliott> So the whole Unix "pass around text" stuff kinda falls apart.
21:01:47 <elliott> Has anyone ever put an \n into a filename intentionally? If yes: Without being a moron?
21:02:17 <elliott> Vorpal: I get a "no such file or directory" when I do $ touch 'a\b'.
21:02:50 <Vorpal> elliott, what about the other ones that are forbidden. Such as :
21:03:12 <elliott> That works too. Remember that that restriction is at the Win32 level, and Interix is accessing NTFS directly through NT.
21:03:29 <Vorpal> elliott, but \ is on NTFS level?
21:03:32 <elliott> They show as squares in explorer for me.
21:03:48 <Vorpal> elliott, how does c:\ and d:\ and so on show up from interix
21:03:59 <elliott> Vorpal: /dev/fs/C, /dev/fs/D
21:04:08 <elliott> Not the prettiest solution but you can always symlink e.g. /c: there.
21:04:14 <Vorpal> elliott, under /dev? *cringe*
21:04:14 <elliott> Vorpal: / itself is C:\SFU.
21:04:26 <Vorpal> elliott, well yes. But I would have expected under /mnt
21:04:41 <elliott> But they're not mountpoints :P
21:04:53 <elliott> * IMPORTANT: 2 config files in '//opt/gentoo/etc/' need updating.
21:05:17 <Vorpal> elliott, you can set it to use colordiff if you install it
21:05:39 <Vorpal> elliott, maybe it tries /etc
21:05:39 <elliott> Well. It paused for a second then exited.
21:05:56 <Vorpal> elliott, there is another tool for it too
21:06:25 <elliott> Scans, and then says "nothing left to do, exiting"
21:06:32 <elliott> Maybe they didn't really need updating.
21:06:40 <elliott> OK, pikhq: what's the binary package for the basic X executables?!
21:06:44 <Vorpal> elliott, iirc portage does that already
21:06:54 <Vorpal> elliott, find /opt/gentoo/etc -iname '.*'
21:06:59 <elliott> (SAD OF THE DAY: I need the Cygwin DLL or something somewhere on the system so I can use mintty.)
21:07:00 <Vorpal> elliott, look for any strange ones
21:07:07 <Vorpal> elliott, just to make sure they *are* merged
21:07:14 <Vorpal> elliott, probably ._something
21:07:22 <elliott> .keep_app-admin_eselect-python-0
21:07:31 <elliott> Those are the only strange ones.
21:08:30 <Vorpal> elliott, hm they are fine
21:08:37 <Vorpal> elliott, they are like .keep for CVS kind of
21:08:44 <Vorpal> portage doesn't track dirs. Just files
21:12:07 <Sgeo_> My dad spent $150 to let me have mobile Internet access while preventing my step-mom from knowing about it?
21:12:27 <elliott> I stopped watching Zero Punctuation and pressed alt-tab to reply, but I can't think of anything to say.
21:14:20 <elliott> Vorpal: So, that masking thing... wtf is it?
21:14:22 <Phantom__Hoover> Sgeo_, I... your stepmother must be the most awful person ever
21:14:26 <elliott> Apparently xeyes is MASKED.
21:14:33 <elliott> So I'm just imagining Zorro here.
21:15:28 <elliott> Phantom__Hoover: What kitten thing.
21:16:20 <Phantom__Hoover> You know, the one that adds a kitten which follows your mouse around.
21:17:31 <Vorpal> elliott, more specific
21:17:42 <elliott> Vorpal: xeyes is apparently Masked.
21:17:49 <elliott> I'd rather have plain old unmasked eyes on my desktop rather than Zorro.
21:18:09 <Vorpal> elliott, well masked by what
21:18:12 <elliott> At least I seem to recall you can't install things that are masked because Gentoo loves to make everything complicated.
21:18:17 <elliott> Vorpal: * x11-apps/xeyes [ Masked ]
21:18:23 <Vorpal> elliott, uh. No more details?
21:18:25 <elliott> Maybe X11 is masked or something?
21:18:42 <elliott> Do I need to run something specific to ask for more details?
21:18:42 <Vorpal> elliott, what was -s now again
21:18:50 <Vorpal> elliott, try emerge -pv xeyes
21:18:54 <Vorpal> elliott, -p for pretend
21:19:12 <elliott> Masked by missing keyword.
21:19:30 <Vorpal> elliott, what is the interix arch called
21:19:47 <elliott> x86-interix, I think... or something. Let me check.
21:19:54 <Vorpal> elliott, the ebuild is missing that
21:20:23 <elliott> Vorpal: Is there a standard way to look at it, or should I just fish in directories?
21:20:37 <Vorpal> elliott, which version number?
21:20:48 <elliott> 1.1.1. but there's also 1.1.0.
21:20:51 <elliott> I assume you mean of xeyes.
21:21:10 <Vorpal> elliott, less /whatever/your/prefix/is/usr/portage/x11-apps/xeyes/xeyes-1.1.1.ebuild
21:21:18 <Sgeo_> The reason my step-mom would be mad is because she'd see it as a waste of money
21:21:43 <Vorpal> elliott, in other words: the package category/name + version gives ebuild path and name
21:21:53 <elliott> Vorpal: KEYWORDS includes ~x86-linux...
21:22:00 <elliott> does that mean that regular Gentoo users can't install xeyes? :-D
21:22:04 <Vorpal> elliott, they changed it to read ~x86-linux?
21:22:07 <elliott> Vorpal: Maybe 1.1.1 is TESTING xeyes.
21:22:17 <Vorpal> elliott, is 1.1.0 stable?
21:22:20 <elliott> Nope, 1.1.0 has ~x86-linux.
21:22:24 <elliott> Vorpal: x86 is in there but also ~x86-linux.
21:22:35 <Vorpal> elliott, I never heard of ~x86-linux before
21:22:38 <elliott> Vorpal: x86-linux is Gentoo Prefix.
21:22:50 <Vorpal> elliott, well x86 is plain linux then
21:23:02 <elliott> OK, so, how can I tell portage that I don't give a shit about no damn maskin' and I want it anyway? Isn't it editing some conf file?
21:23:02 <Vorpal> elliott, ask pikhq for how to work around this. I forgot
21:23:16 <Vorpal> elliott, oh wait wasn't it /etc/portage/something.keywords and add a specific line
21:23:22 <Vorpal> I don't remember details
21:23:43 <elliott> Vorpal: I have a serious question. Do Gentoo users think that every distribution is this over-complicated?
21:23:54 <elliott> I'm just trying to understand the mind of a person who would subject themselves to this.
21:24:04 -!- Tritonio has joined.
21:24:20 <elliott> package.keywords it seems.
21:24:35 <Vorpal> elliott, well having several versions available of a package at one time *is* actually useful sometimes
21:24:40 <Vorpal> elliott, apart from that: no clue
21:24:51 <elliott> *I mean the masking stuff :P
21:25:07 <Vorpal> elliott, when I began with it I wanted rolling release. I didn't know about arch then. Heck was arch around in 2004 even?
21:25:23 <elliott> I'm not exactly a noob but just about every Gentoo document confuses me... so much random terminology.
21:25:33 <elliott> Vorpal: arch started in 2002
21:25:43 <Vorpal> elliott, well I didn't know about it anyway
21:25:49 <elliott> Maybe Kitten will run on Interix. :-)
21:26:11 <elliott> Supported platforms: x86/Linux; x86/NT
21:26:46 <elliott> x11-apps/xeyes x86-interix
21:27:04 <Vorpal> elliott, iirc you need
21:27:18 <Vorpal> elliott, you need to list a keyword that it has
21:27:19 <elliott> I need the full version too it seems
21:27:33 <elliott> Vorpal: "For example, if we're not on x86 but would like to install quake3-demo anyway
21:27:37 <elliott> games-fps/quake3-demo x86"
21:27:54 <Vorpal> elliott, well it might :P
21:28:09 <elliott> Vorpal: Well it *should*, there's a compiled xfce-terminal on the DVD image.
21:28:22 <elliott> USE: elibc_Interix kernel_Interix prefix userland_GNU x86-interix
21:28:28 <elliott> FEATURES: nostrip preserve-libs
21:28:29 -!- Behold has quit (Read error: Connection reset by peer).
21:28:49 <Vorpal> elliott, also they put random-non-user-options in USE
21:28:51 <elliott> Phantom__Hoover: I'd have to fight with Mac hardware to develop it on here.
21:28:56 -!- Behold has joined.
21:28:58 <elliott> When I get a desktop work will continue.
21:29:03 <elliott> Vorpal: Yeah, I know USE is crazy.
21:29:04 <Vorpal> elliott, which was really an abuse (pun not originally intended) of USE
21:29:20 <elliott> Phantom__Hoover: More trouble than it's worth, it's not like it has rigorously-supported hardware either.
21:29:26 <elliott> A desktop will really be the most convenient platform to do it on.
21:29:51 <Vorpal> elliott, a thinkpad would work well too
21:29:54 <elliott> I would just like to say that Kitten's package manager will be rigorously sane.
21:30:09 <elliott> Vorpal: Yeah, but only because ThinkPad gets preferential treatment by kernel devs... the hardware is just as nonstandard as any laptop :P
21:30:23 <Phantom__Hoover> So how long until ais or Keymaker vapes the ClearBF page on the wiki?
21:30:25 <Vorpal> elliott, I wonder why kernel devs like thinkpads?
21:30:41 -!- Mathnerd314 has joined.
21:30:44 <Vorpal> elliott, maybe because it works well mostly. Circular preference
21:31:08 <Vorpal> Phantom__Hoover, "vapes"?
21:31:30 <elliott> Vorpal: Because they were decent when laptops were terrible :P
21:34:55 <elliott> Vorpal: Why does xeyes depend on libxslt.
21:36:17 <Vorpal> elliott, emerge -pvt xeyes
21:36:21 <Vorpal> elliott, this list it as a tree
21:36:25 <elliott> Vorpal: I have no idea, I did "emerge xeyes" and it's installing libxslt, so I don't want to question it :P
21:36:28 <elliott> Can I do that while it installs?
21:36:44 <Vorpal> elliott, it won't list already installed ones
21:36:45 <olsner> maybe you found xcb or something in the depends
21:36:56 <Vorpal> elliott, anyway you can do multiple emerges side by side
21:36:56 <elliott> olsner: oh, right, I blame xcb.
21:36:57 <olsner> (iirc that involves XML)
21:37:25 <elliott> All Xlib programs depend on XCB depend on libxslt, wooooo :P
21:37:31 -!- oerjan has quit (Remote host closed the connection).
21:37:46 <olsner> indeed, that's PROGRESS right there
21:40:35 <elliott> Wow, source-based distros are so tedious ... and emerge is so noisy.
21:40:39 <elliott> Why do people put up with this stuff?
21:40:57 <olsner> for the speed of course
21:41:18 <olsner> it takes a while to compile, but it's SO FAST afterwards, like SEVERAL PERCENT faster
21:41:34 <elliott> olsner: Several percent? I doubt that very much ... :P
21:41:43 <elliott> A *single* percent, maximum, unless the comparison is _really_ biased.
21:44:15 <Vorpal> elliott, so use debian interix
21:44:23 <elliott> Vorpal: Not even vaguely maintained.
21:45:14 <Vorpal> elliott, is gentoo/interix maintained?
21:45:39 <elliott> Vorpal: Gentoo/Interix is just Yet Another Gentoo Prefix port.
21:45:45 <elliott> Gentoo Prefix's main targets are OS X, Solaris, other Linuxes.
21:46:04 <elliott> Vorpal: Interix is POSIX, and it has gcc, so porting Gentoo Prefix to it is honestly not that hard.
21:46:07 <olsner> it has "other" as a main target?
21:46:16 <elliott> olsner: Well, Linux, yes. :p
21:46:19 <elliott> Vorpal: It only has one main developer, but hey, it has a swanky installer.
21:46:26 <olsner> this port is *mainly* for everything, but mostly everything else
21:46:49 <elliott> Vorpal: It just uses the main Gentoo Prefix tree (I don't know whether that's the main ports tree, I don't think so since --sync kept printing out a figlet "Gentoo Prefix" from the ports server).
21:46:59 <elliott> Might be maintained as a diff to the ports tree adding the target keywords or whatever.
21:47:03 <Vorpal> olsner, no it is so you can do Gentoo/Debian Linux
21:47:12 <elliott> Vorpal: But really, with a source distro it's quite easy to do ports like this :P
21:47:27 <elliott> But yeah, there's one developer of the port itself, but many more of Gentoo Prefix.
21:47:27 -!- Behold has quit (Read error: Connection reset by peer).
21:48:07 <elliott> Emerging 6 of 23... boy oh boy this is fun.
21:48:15 <elliott> I can almost taste the wasted CPU cycles.
21:48:52 <olsner> hmm, I shouldn't have closed that mplayer window, I'm never going to be able to watch this episode without rewatching something I've already seen
21:49:14 -!- Behold has joined.
21:49:37 <Vorpal> elliott, minecraft weirdness: the fastest tool to break a wooden pressure plate is a pickaxe
21:49:48 <elliott> But is it a WOODEN pickaxe?
21:50:07 <Vorpal> elliott, diamond. Because everyone know diamond is so good for cutting wood
21:50:08 <olsner> amber tamblyn sucks in house :/ but I liked her in joan of arcadia
21:50:31 <fizzie> Bleh. Figured out how to make a ray-transparent (it's what Blender calls actually raytracingly transparent stuff, as opposed to just z-sort + blending) object cast shadows based on the material (so the gridwork I have in windows casts grid-shaped shadows), but it's horreebly slow to render that. (I keep dabbling with that house-render.)
21:50:47 <olsner> she's incidentally the daughter of one of the actors in twin peaks
21:50:50 <elliott> We are terrible at this, aren't we.
21:51:15 <fizzie> Also I first thought you were criticizing it for being not minecraft-related enough.
21:51:53 -!- Tritonio has quit (Quit: Leaving).
21:52:09 <olsner> hmm, taub's wife looks like she could be taub's sister
21:53:12 <olsner> Phantom__Hoover: yeah, I just realized they look a lot like each other
21:53:21 <elliott> olsner: well duh that's because of genetics!
21:53:32 <olsner> right, they mix the genes when they have intercourse
21:54:00 <olsner> luckily this means infidelity is not that big a deal since you get your genes in there anyway
21:54:43 <olsner> (there is one of those yahoo answers thingies on this topic, it's kind of fun)
21:56:35 <olsner> was that gentoo installing stuff?
21:59:34 <elliott> olsner: It has a fancy installer and everything.
21:59:39 <elliott> olsner: tl;dr it's the only maintained Interix distro :P
21:59:56 <elliott> olsner: But hey, it has a fancy graphical installer that lets you set up accounts and everything, and installs Interix for you.
22:00:20 <olsner> gentoo/winnt includes interix? this sounds crazy convenient
22:00:34 <olsner> if I had an nt kernel running anywhere, that is
22:00:39 <elliott> olsner: yep, and all the hotfixes
22:00:47 <elliott> olsner: also, it uses the system interix on later OSes
22:00:52 <elliott> olsner: downside is it's a 2 gigabyte ISO :P
22:00:56 <elliott> but yeah, it is really convenient
22:01:02 <elliott> you basically click next a bunch of times and you're done
22:02:00 -!- Behold has quit (Read error: Connection reset by peer).
22:02:13 <olsner> what is the nature of its non-integration with win32? can it start windows programs at all?
22:02:15 -!- Behold has joined.
22:04:56 <elliott> olsner: Well, /dev/fs/C/WINDOWS/notepad.exe works.
22:05:13 <elliott> olsner: The "non-integration" is simply that you can't have a program that makes both Interix and Win32 calls (like e.g. mintty does).
22:06:26 -!- Sgeo_ has changed nick to Sgeo.
22:06:47 <elliott> olsner: Also it's a lot faster than Cygwin. :p
22:08:52 <olsner> made out of win and awesome, eh?
22:10:03 <olsner> Win32some, Lose32some?
22:10:14 <olsner> incidentally 32 is the ascii for space
22:10:43 <elliott> olsner: Downside: at least Cygwin has binary packages :P
22:11:00 <elliott> olsner: You can even run Gentoo Prefix on Cygwin, if you want the true slow-compilation experience.
22:12:02 <olsner> hmm, i should have proper whisky glasses
22:12:46 <elliott> olsner: you should be like me, use this channel as a substitute for drinking hard liquor while looking grumpy and pissed off
22:13:02 <elliott> ok, i don't exactly have first-hand experience, but I'm pretty sure it's equivalent
22:13:35 <olsner> sure, being grumpy and pissed off doesn't require alcohol (or anything at all, really)
22:13:50 <elliott> olsner: oh sure, but I do it in a really classy way.
22:14:00 <elliott> Interix might be faster than Cygwin but ./configure still ain't quick :P
22:16:01 -!- Behold has quit (Read error: Connection reset by peer).
22:20:59 -!- Behold has joined.
22:21:38 <Gregor> PANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORAPANDORA
22:22:43 <olsner> Gregor: http://www.youtube.com/watch?v=UZcqV7pLC0A
22:23:41 <quintopia> someone help replace the speechbox? this one's shortcircuited
22:24:15 -!- oerjan has joined.
22:25:02 -!- augur has quit (Remote host closed the connection).
22:25:17 <elliott> `addquote * quintopia sits on gregor
22:25:29 <elliott> 17 of 23 ... holy fuck this is slow.
22:25:37 <HackEgo> 282) * quintopia sits on gregor
22:25:39 <elliott> Vorpal: is rolling release really worth this crap
22:26:50 <olsner> "rolling release" is where packages get updated continuously?
22:28:40 <elliott> "Putting Gentoo users on the same level as ricers is a bit unfair; the Gentoo users don't annoy the general public and cause traffic hazards."
22:28:41 <olsner> ah, that sucks... but the alternative also sucks :D
22:28:54 <elliott> Gentoo users don't annoy the general public?
22:29:01 <elliott> well maybe not but certainly the Linux public...
22:29:14 <elliott> olsner: why does rolling release suck? sure everything is hideously unmatched with each other but you get that anyway
22:29:18 <olsner> the general public is luckily complete unaware of their even existing
22:29:20 <elliott> since no fucking release cycles are synched up
22:29:41 <elliott> they do cause the analogy of traffic hazards too... well, for developers
22:29:43 <Sgeo> I asked my game programming professor if there was an equiv. of Haskell's Maybe types in C#
22:29:50 <Sgeo> He doesn't know Haskell, so
22:29:52 <elliott> "game programming professor"
22:30:04 <elliott> quintopia: this is where we pull out our standard line about Sgeo's college
22:30:06 <Sgeo> I think I'll write some code that does the same
22:30:10 <olsner> hmm, "their even existing", but I think that's how you say it?
22:30:29 <elliott> Sgeo: useless, C# already has null
22:30:34 <elliott> so Maybe is completely unsafe, basically
22:30:57 <j-invariant> 22:40 < elliott> "game programming professor"
22:31:04 <Sgeo> Can you put null into a bool?
22:31:13 <olsner> null is like Maybe with implicit fromJust
22:31:19 <elliott> Sgeo: you can certainly put _|_, but then you can in Haskell too :-)
22:31:35 <elliott> Vorpal: from a thread about funroll-loops.org on the Gentoo forums, circa 2004: "One week later: this is on the Arch forums. One month later: the Arch people get the joke."
22:32:13 <Vorpal> elliott, that *might* be biased though :P
22:32:41 <olsner> aw fuck, pandora is now making schlager music
22:33:00 <elliott> here's Kitten's CRAZY policy: I upgrade packages, except when that would break shit!
22:33:23 <elliott> also, I upgrade gcc every two major releases! :-P
22:33:36 <elliott> (Well... I'm not going to ship the latest gcc, considering its stability record...)
22:33:51 <oerjan> olsner: vikingarnarna?
22:34:14 <elliott> "* Gentoo does save time because it is easier to administer -- emerging does not require user intervention and its CPU usage can be tuned"
22:34:15 <elliott> This person has never used apt.
22:36:33 -!- FireFly has quit (Quit: swatted to death).
22:40:50 <elliott> "That's the power of Gentoo, being able to choose whether or not you want to install all of KDE just to get arts, or whether you want to enable gnome support in Abiword. No RPM-based distro can do that."
22:41:07 <elliott> (Reading anti-Gentoo material helps me forget that I'm running emerge in the background.)
22:43:34 <Sgeo> How does Gentoo deal with self-hosted compilers other than gcc?
22:44:01 <elliott> Same way it deals with gcc?
22:46:39 <Sgeo> So, it just allows binaries
22:46:41 <Sgeo> Just like that
22:48:10 <elliott> Sgeo: Um, no, it uses the binary to compile the source.
22:48:21 <elliott> Vorpal: http://bugs.gentoo.org/show_bug.cgi?id=35890
22:49:01 -!- Phantom__Hoover has quit (Remote host closed the connection).
22:49:23 <Sgeo> So, if I write a self-hosted SgeoLang compiler, I can just get the binary to be downloaded, even if it's only temporary before it gets compiled on the user's machine?
22:49:44 <elliott> Sgeo: Are you assuming that Gentoo has some kind of... religious objection to binaries?
22:52:31 <Sgeo> I think I thought that part of the reason behind it using source instead of binaries mostly is ... safety. Somehow. Although what sane person actually reads all of the source code for everything that they use?
22:52:49 <Sgeo> I'm sure zzo38 could do it
22:55:04 <Vorpal> <elliott> Vorpal: http://bugs.gentoo.org/show_bug.cgi?id=35890 <-- what
22:55:37 <elliott> Vorpal: Lead GAIM developer gets fed up of Gentoo users, hilarity ensues.
22:56:11 <Vorpal> elliott, his request is a bit stupid though
22:56:32 <Vorpal> elliott, but stupid as well
23:03:07 -!- hagb4rd has quit (Ping timeout: 240 seconds).
23:04:16 <elliott> Did I mention that Kitten is made of happiness and yay?
23:06:19 <elliott> Vorpal: wow! emerge crashed!
23:06:37 <elliott> "error: sys/poll.h: No such file or directory"
23:09:11 <Sgeo> "*** Bug 124595 has been marked as a duplicate of this bug. ***
23:11:09 <elliott> Sgeo: Yeah, that's the "get rid of Quod Libet god damn" bug.
23:13:26 <elliott> Vorpal: will libX11 even _work_ on a platform without poll?
23:14:07 <elliott> it's just in poll.h, not sys/poll.h
23:15:17 <Sgeo> What's wrong with Quot Libet?
23:15:40 <elliott> Sgeo: The Quod Libet developer asked for the package's removal, as the package was unmaintained and broken.
23:15:51 <oerjan> latin lesson: quod = what, quot = how many
23:15:53 <elliott> And because it was causing many, many incorrect bug reports for the developer.
23:16:04 <elliott> (Due to the package being broken.)
23:16:07 <elliott> The Gentoo developers said "no".
23:17:18 <oerjan> maybe he should send them a bill for his wasted time
23:17:46 <elliott> oerjan: hmm, that should be possible
23:19:01 <elliott> oerjan: consider that things like Bitcoin essentially use time as a currency :)
23:19:10 -!- azaq23 has joined.
23:19:16 <elliott> because a bitcoin is a proof that you've done at least this much computing work, which takes about this much time
23:23:49 <elliott> Vorpal: oh wow: http://pastebin.com/7VRZpfmy
23:24:04 <elliott> In which Gentoo tells everybody who installs libX11 to bother the X.Org developers.
23:27:24 <pikhq> elliott: Implicit declarations can have really surprising results sometimes.
23:27:38 <elliott> pikhq: Yeees... but a distro shouldn't be telling EVERYONE who installs a package to bug the developers.
23:27:44 <elliott> Because that way lies FIVE THOUSAND DUPLICATE BUG REPORTS.
23:28:06 <pikhq> elliott: I'm not entirely sure why emerge outputs those QA notices by default.
23:28:11 <elliott> pikhq: Better would be for the *damn package maintainers* to be able to run it and get those notices to report.
23:28:16 <pikhq> elliott: Cause it's kinda intended for the maintainers.
23:28:34 <elliott> Right ... but the wording doesn't say "maintainers, report this" it says "report this" :P
23:28:50 <elliott> pikhq: Anyway, the Gentoo Prefix installer Just Works.
23:29:11 <elliott> pikhq: And I'm on the slow, slow path to installing xeyes with only one thing I had to fix (link /usr/include/sys/poll.h to /usr/include/poll.h).
23:30:19 <elliott> "Then there are the people who compile everything with "-O3 -fomit-frame-pointers -ffast-math -fguess-at-hard-math -mmmx -msse2 -fexpensive-optimizations -mcpu=Pentium4 -march=AthlonXP -finline-functions" and don't understand why things crash from time to time." --Gentoo forums
23:30:26 <elliott> Vorpal: You should integrate -fguess-at-hard-math into the cfunge build process
23:32:13 <elliott> pikhq: Have I mentioned that in Kitten, the package manager automatically updates all packages before installing a new one? :p
23:32:24 <elliott> That's my INNOVATIVE solution to the "packages don't like each other" problem.
23:33:23 <elliott> pikhq: Anyway, I just realised something. You know how you said that dynamically-linked stuff will fail if it's built for prefix / and you put it in e.g. /arch/x86?
23:35:24 <elliott> pikhq: That's not a problem if you don't dynamically link!
23:35:45 <elliott> pikhq: Wait, I just realised something.
23:36:00 <elliott> pikhq: The only directories that actually differ depending on the architecture are /lib and /bin, right?
23:36:25 <elliott> Most distros have /lib -> /lib64 or vice versa and then use /lib32. Well... why not just have /bin64 and /bin32? Except by /bin64 I mean /bin-x86-64, of course.
23:42:04 <pikhq> elliott: A handful of packages will actually install different headers as well.
23:42:36 <elliott> pikhq: I am about to run xeyes on Gentoo on Interix on Windows NT, shown on an X11 server running on Win32 running on Windows NT.
23:43:59 <lambdabot> foldl f z (x:xs) = foldl f (f z x) xs
23:44:10 <lambdabot> foldr f z (x:xs) = f x (foldr f z xs)
23:44:27 <elliott> pikhq: How does one take a screenshot in Windows if one has no Print Screen key, do you know? :p
23:45:12 <elliott> Apparently you have to use the onscreen keyboard :P
23:45:25 <quintopia> just because it is defined recursively using itself doesn't mean its functionality is absolutely necessary as a primary syntactical element... is there a way to define it with respect to other functions only?
23:45:46 <elliott> quintopia: no. haskell has no recursive primitives
23:45:52 <elliott> everything is implemented in terms of recursion
23:46:00 <elliott> quintopia: you can implement the Y combinator with a data type
23:46:11 <elliott> and you can use that to implement fix
23:46:17 <elliott> so technically you don't need to recurse
23:46:43 <elliott> pikhq: Jesus though, is installing xeyes meant to take hours?
23:46:55 <pikhq> elliott: Uh, from scratch?
23:47:02 <pikhq> elliott: Yeah, X has a giant dependency tree.
23:47:06 <elliott> pikhq: Well, installing libxcb upwards and all, yeah.
23:47:13 <elliott> pikhq: Faster than Cygwin :P
23:51:14 <fungot> <CTCP>.. !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ...
23:51:32 <fungot> <CTCP>.. !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ...
23:52:39 <fungot> . "$&(*,.02468:<>@BDFHJLNPRTVXZ\^`bdfhjlnprtvxz|~
23:53:51 <quintopia> fungot strips a lot of special chars by replacing them with .
23:53:52 <fungot> quintopia: that'll never grow the stack indefinitely. i assume irc has limits to the language you choose pales in comparison to its peers, perl, etc.
23:54:32 <oerjan> quintopia: i see only a few .'s
23:55:01 <elliott> pikhq: for some reason xeyes causes 64% cpu usage in the posix process...
23:55:55 <quintopia> so what happened to ^B? that makes bold right? did the ^C override it?
23:55:56 <oerjan> quintopia: however ^B-^H are probably stripped by the channel +c mode
23:56:33 <oerjan> ^O, ^V and ^W also are missing
23:56:51 -!- cheater00 has joined.
23:57:52 <oerjan> does anyone else see the 128 and up a while characters as double inverted?
23:58:06 <elliott> oerjan: that's ctrl+foo, I think
23:58:09 <elliott> which irssi shows as inverted foo
23:58:29 <elliott> I think ^@ is 127 or something
23:58:36 <elliott> and something = offset of H from @
23:58:40 <quintopia> after which are undisplayable characters in this font...
23:58:42 <elliott> and that's shown as inverted H in irssi
23:59:11 <oerjan> elliott: except i don't see them as a single char but as two equal copies of the x-128 one, which is itself x-64 inverted i think
23:59:29 <oerjan> that is, i see char 128 as two inverted @'s
23:59:53 -!- cheater99 has quit (Ping timeout: 240 seconds).
23:59:58 <oerjan> ^bf ++++++++++++++++[>++++++++<-]>.