00:01:54 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
00:05:31 <zzo38> (Decompose g f) will be a functor if f is, although the Functor instance for Decompose can never exist regardless of what g and f is.
00:05:51 <zzo38> Since it is not really a functor, but it has one
00:06:47 <zzo38> Although the type (Decompose g f) itself is not a functor
00:07:13 <zzo38> Yet it will make up a functor anyways if f is a functor even though (Decompose g f) cannot be.
00:07:29 <zzo38> Unless, perhaps, it is a subcategory.
00:08:09 <zzo38> I mean it is not a endofunctor on (->).
00:10:03 <zzo38> But, say you have: newtype Subcategory c f x y = Subcategory (c (f x) (f y)); data Desubcategory c f x y where { Desubcategory :: c x y -> Desubcategory c f (f x) (f y); };
00:10:56 <zzo38> soundnfury: You said you are going to post a Z80 code if someone posts a Haskell code in here? (Or, something like that)
00:12:59 <soundnfury> POP HL;PUSH HL;POP DE;INC DE;LD BC,0;LD (HL),A;LDIR
00:13:59 <zzo38> Is this Z80 code a part of some program? And, what is LDIR for?
00:14:30 <soundnfury> That Z80 code will fill all of RAM with the value in the Accumulator (A)
00:15:33 <zzo38> Now I can understand.
00:17:27 <zzo38> I have once wrote a program for GameBoy, which is not Z80 although it is similar to Z80.
00:20:46 <zzo38> The NES/Famicom has a 6502 without decimal mode.
00:21:25 <zzo38> (Actually, it has decimal mode although it doesn't do anything. You could use the decimal mode flag for your own purpose, possibly.)
00:21:47 <soundnfury> The hardest Z80 opcode to emulate is probably DAA
00:22:55 <zzo38> Is that BCD-adjust?
00:37:02 -!- Slereah has joined.
00:51:04 <zzo38> I have read something to make stereo NES. But this wouldn't seem to work with Famicom audio expansions.
00:52:09 <zzo38> But to make it sound properly with stereo you should make up the game or music file to use the stereo. (If you play a stereo file on a mono system, there won't be a problem.)
00:56:03 -!- oerjan has quit (Quit: EEEK).
00:57:07 <zzo38> Using PPMCK channels, you could have: left=A,B,G,I,K,M,P,R,T,V,X,a,b right=C,D,E,H,J,L,N,Q,S,U,W,Y center=F,O,Z I think this way is compatible with the 2A03 audio so it can be used on a real hardware if no expansion chips are used. If you use audio expansion it could be use only in emulator playing .NSF file.
01:02:00 -!- oerjan has joined.
01:18:28 -!- augur has joined.
01:19:54 -!- soundnfury has quit (Remote host closed the connection).
01:22:46 -!- zzo38 has quit (Remote host closed the connection).
01:30:34 -!- soundnfury has joined.
01:43:19 -!- Triclops200 has joined.
01:43:36 -!- Triclops200 has left ("Leaving").
01:59:48 -!- Frooxius|TabletP has quit (Ping timeout: 248 seconds).
02:00:52 -!- pikhq_ has quit (Ping timeout: 248 seconds).
02:01:01 -!- pikhq has joined.
02:43:07 -!- TeruFSX has quit (Ping timeout: 240 seconds).
02:43:28 -!- TeruFSX has joined.
02:44:20 -!- DHeadshot has joined.
03:06:40 -!- DHeadshot has quit (Read error: Connection reset by peer).
03:06:52 -!- DHeadshot has joined.
03:54:54 -!- zzo38 has joined.
04:01:59 <zzo38> The free monoid structure is a backward monoid transformer, isn't it? So the free category will be a backward category transformer, too.
04:02:52 -!- oerjan has quit (Quit: leaving).
04:03:00 -!- copumpkin has quit (Ping timeout: 248 seconds).
04:03:33 -!- copumpkin has joined.
04:17:02 <zzo38> Is anyone on today?
05:13:15 -!- azaq23 has quit (Quit: Leaving.).
05:32:30 -!- DHeadshot has quit (Ping timeout: 264 seconds).
05:45:25 <soundnfury> What kind of a fool would try to program in a language where you need to know whether "the free monoid structure is a backward monoid transformer"?
05:46:05 -!- pikhq has quit (Ping timeout: 255 seconds).
05:46:08 -!- pikhq_ has joined.
05:55:56 <coppro> soundnfury: monoids are the bestest
05:57:29 <zzo38> soundnfury: Someone who likes mathematics, I guess.
05:58:39 <zzo38> I also think the free monad is also a backward monad transformer, it is not a forward monad transformer like Edward Kmett has (he agrees it isn't, even though his article says it is)
06:28:42 <itidus21> soundnfury: these programmers are difficult to fathom. who can say what it is they aspire to
06:29:48 <soundnfury> zzo38: I recently graduated in mathematics from Cambridge. I don't like monoids.
06:30:55 <itidus21> lmao. i just saw ion in the topic
06:30:58 <zzo38> Well, there are other mathematical structures too; monoids is not the only one. There is also semigroup, group, category, functor, monad, comonad, etc
06:32:08 <itidus21> zzo38: why is sound more interesting than graphics?
06:32:31 <itidus21> this is a game of find all the fallacies in tidus' post
06:32:42 <zzo38> itidus21: I did not say a sound is more instresting than graphics!
06:33:17 <itidus21> 1) it isn't more interesting at all 2) zzo38 never said it is more interesting , hmm
06:33:28 <soundnfury> Don't get me wrong, I love mathematical structures. Groups are awesome. It's just that trying to design a programming language that reads like a proof from the abstruse end of category theory does not maintainable software make.
06:33:55 <copumpkin> soundnfury: and you've said that after a serious effort at seeing why people like that approach?
06:34:17 <itidus21> zzo38: my brain is difficult to live with at times...
06:34:25 <copumpkin> as it sounds, it just sounds like empty maxims and knee-jerk reactions :)
06:34:57 <itidus21> it is a tale told by an idiot, full of sound and fury, signifying nothing
06:34:58 <zzo38> soundnfury: Perhaps to you.
06:35:33 <soundnfury> I just think its fans are living in an epistemically closed bubble
06:35:34 <copumpkin> "I know math, and I know programming. I've never thought of trying one from the point of view of the other (and it seriously does go both ways), so it must not be worth trying"
06:35:46 <copumpkin> if I had a dollar for every time I've seen that, I'd have several dollars :)
06:36:04 <soundnfury> this is based on the fact that their observed behaviour and signalling is consistent with the standard patterns of cults
06:36:35 <soundnfury> also: don't get me wrong, I love the lambda calculus.
06:36:36 <copumpkin> nevermind, not going to try then :)
06:37:03 <soundnfury> and programming<->maths is a shining example thereof
06:37:36 <copumpkin> can you explain why people like monoids in progrmaming?
06:37:39 <soundnfury> but the Haskell world feels too much like the "formal proofs of program correctness" movement from mumble decades ago
06:37:39 <zzo38> I just think category theory is good and can make a lot of things.
06:37:50 <itidus21> soundnfury: i've said worse :P
06:37:51 <copumpkin> soundnfury: except the approach to it is completely different
06:38:53 <soundnfury> it feels like you're introducing unnecessary machinery on a huge scale (and I mean *huge*; I mean, category theory? that's like a 3000-ton dragline crane) just so you won't have to [the moral equivalent of using one little goto]
06:39:16 <soundnfury> it reminds me of everything that makes me hate Dijkstra.
06:39:22 <soundnfury> And you do not want to be like Dijkstra.
06:39:48 <monqy> are you sure you understand haskell
06:39:48 <lambdabot> monqy: You have 1 new message. '/msg lambdabot @messages' to read it.
06:39:55 <soundnfury> But this is all my subjective opinion, and I can't really make reasoned justifications of any of it. It's a gut reaction.
06:39:55 <monqy> because you're really sounding like you don't hth
06:40:17 <monqy> oh, a gut reaction
06:40:35 <soundnfury> monqy: I may be mistaken here, but I suspect that Haskell is like Quantum Mechanics: "If you think you understand it, you don't understand it."
06:40:36 * ion laughs at the topic
06:41:02 <copumpkin> soundnfury: I feel like you might need to either read more about that approach to programming or opine less about it
06:41:11 <ion> soundnfury: Err, no.
06:41:20 <soundnfury> (note that in truth the un-understandable bits of QM are all Copenhagen's fault. Yay Everett)
06:41:36 <soundnfury> but I'm feeling in an obnoxiously argumentative mood right now
06:41:48 <soundnfury> feel free to devoice me if you think it's necessary
06:42:13 <copumpkin> but I might go to sleep instead, since I must wake up for work tomorrow
06:42:17 <soundnfury> I have been up all night fixing bugs in quirc, after some wonderful person sent me about 10 bug reports
06:42:51 <soundnfury> oh, and I do mean 'wonderful', that wasn't sarcasm :)
06:49:52 <zzo38> I like mathematics, though.
06:51:36 -!- pikhq_ has quit (Ping timeout: 252 seconds).
06:51:39 -!- pikhq has joined.
06:57:05 <coppro> mathematics is wonderful
06:57:56 <fizzie> "Those who can, do; those who can't, mathematics." Isn't that the old adage?
07:01:34 <soundnfury> fizzie: I think you're thinking of "Those who can't, teach. Those who can neither do nor teach, become librarians."
07:01:57 <soundnfury> (although in fact good teachers are very talented and all too rare)
07:02:22 <coppro> copumpkin: I've figured out what you are.
07:02:29 <coppro> You're a pumpkin that shoots arrows at people.
07:03:02 <fizzie> Says an arrow-shooting ppro.
07:03:28 <soundnfury> is a copumpkin a <bra| or a |ket>? I can never remember which is the copumpkin and which is the contrapumpkin
07:05:48 <soundnfury> no, that matches copumpro and coppkin as well
07:06:36 <soundnfury> cop(umpkin|pro) is, I think, the optimal regex
07:07:00 <soundnfury> also I've just had an idea about how to go about building my regsexp engine
07:07:30 <soundnfury> unfortunately I haven't invented my regsexp language yet, so I can't start building the engine
07:08:45 <itidus21> ... in my twisted mind, i'm glad it matches copumpro and coppkin
07:10:10 <soundnfury> copumorpkin, copumumumumumr, coppppppro...
07:10:43 <soundnfury> We need to define "copumorphism" so that you can implement it in Haskell.
07:11:09 <zzo38> But what do you want to define it as?
07:11:53 <soundnfury> Something more absurd than Zygohistomorphic Prepromorphism.
07:12:28 <itidus21> i think we should drop it, it was a bad idea on my part to make a regexp out of someones nick
07:13:16 <itidus21> but, if we let it snowball nothing good can come of it
07:14:43 <fizzie> Snowball is a stemmer-making thing.
07:20:27 <itidus21> once i made a trivia bot, based on chatters... it ended up turning into an exercize in hatred
07:20:42 <itidus21> i can't let such a thing happen again =^_^=;
07:26:43 -!- Vorpal has joined.
07:26:52 -!- TeruFSX has quit (Read error: Connection reset by peer).
07:27:52 -!- calamari has quit (Quit: Leaving).
07:28:06 -!- TeruFSX has joined.
07:45:15 -!- zzo38 has quit (Remote host closed the connection).
08:03:54 -!- AnotherTest has joined.
08:08:34 <soundnfury> oh, shall I bring in my chatbot? It has terrible AI
08:08:48 <soundnfury> (the answer is no, I shall not bring it in, it's horrible)
08:08:54 <mroman> I'm going to do the turing test with him.
08:08:58 <soundnfury> unless you /like/ gibberings and ravings about brazil nuts
08:10:08 <itidus21> oops i don't think i made the bot
08:10:21 <itidus21> i took an existing bot and loaded it with new questions
08:11:06 -!- asiekierka has joined.
08:11:17 -!- Virgil has joined.
08:11:33 <Virgil> soundnfury: Difficult whatever green speak uses labor processing contention watch gurgite intercal
08:11:42 <soundnfury> I hope that prefix character wasn't already in use
08:11:48 <mroman> $What are you wearing?
08:11:54 <Virgil> mroman: Is helps long let look last efforts #bay12games alloy tetrahedrite making here akath
08:11:59 <mroman> Failed the turing test.
08:12:52 <fizzie> fungot: What do you think of Virgil?
08:12:53 <fungot> fizzie: rewriting languages could be their stats
08:13:42 <soundnfury> it has a completely scruffy neural net
08:14:23 <Virgil> itidus21: Sense function language shall dead dinner soad recording one's message correlated
08:14:40 <fizzie> fungot: Do you even parse what is said to you at all? (No you don't.)
08:14:42 <fungot> fizzie: you can write code pretty fluently, but i suspect it sorta does matter. r5rs compliance matters.
08:14:58 <Vorpal> fungot makes more sense than Virgil
08:14:59 <fungot> Vorpal: no kidding... the basic health required for normal life without depending on others is a problem
08:15:15 <Vorpal> that first bit made perfect sense in the context XD
08:15:23 <fizzie> Sure, but it also quotes verbatim quite often.
08:15:35 <fizzie> fungot: You're such a copycat.
08:15:36 <fungot> fizzie: then i haev to er, test it, or are you just happy to see me
08:16:05 <Virgil> itidus21: , my do day : difficult noble dried deemed safest game 3 useful output physically
08:16:35 <itidus21> $Virgil, you have a lot of learning yet ahead of you.
08:16:37 <soundnfury> I'm guessing fungot is using a Markov chain model?
08:16:38 <fungot> soundnfury: there can be a valid identifier in xml is fundamentally retarded.
08:16:42 <Virgil> itidus21: Have latter much over having join few found getting fashion nist joints mad leaving
08:17:01 <soundnfury> itidus21: unfortunately much of Virgil's learning to date was in a channel whose topic is Dwarf Fortress
08:17:24 <itidus21> soundnfury: well.. im weirder than virgil
08:17:24 <soundnfury> also, xml is indeed fundamentally retarded. fungot shows good taste.
08:17:25 <fungot> soundnfury: so was fedora core 2 :d. ( which i don't
08:18:02 <Virgil> itidus21: My love colour land ii nice act result problems indent asleep she wheels generate
08:18:41 <itidus21> im entertained by the fact it doesn't have a baked-in language model
08:18:46 -!- DHeadshot has joined.
08:18:55 -!- stlangbot has joined.
08:18:57 <stlangbot> bf_cu (Brainfuck cell usage); stlang (Evaluate stlang); bf_stat (Brainfuck statistics)
08:18:57 <stlangbot> bf_in (Brainfuck inspect last active cell); df (Deadfish); undf (Undeadfish)
08:18:57 <stlangbot> dfa (Deadfish - ASCII); dfc (Compile deadfish to stlang)
08:19:07 <mroman> stlangbot: chat Can pigs fly?
08:19:12 <fizzie> soundnfury: If you want to think of it like that. It uses ngram language models, most of them trained using the VariKN toolkit from our department, but theoretically it's of course the same thing. (Though it's not quite the traditional "copy and sometimes jump" implementation but instead a "generate each word from matching ngrams" one.)
08:19:48 <fizzie> `quote research into fungot
08:19:49 <fungot> fizzie: read-file does. that's the solution there is to be able to write
08:19:49 <Virgil> all interfere their record woman aspirations paranoid man steep properties space
08:19:51 <HackEgo> 16) <fizzie after embedding some of his department research into fungot> Finally I have found some actually useful purpose for it.
08:21:27 <itidus21> internet.irc.freenode.net = bots essentially spewing random words to be read by human observers
08:21:45 <fizzie> I was going to link to the VariKN page but the PASCAL ("Pattern Analysis, Statistical Modelling and Computational Learning") page is all "PASCAL Forge Could Not Connect to Database: PASCAL Forge Could Not Connect to Database: Project Summary: PASCAL Forge Could Not Connect to Database".
08:22:08 <fizzie> Not a good summary, that.
08:22:22 <Vorpal> fizzie, is that PASCAL as in the language pascal?
08:22:24 <fizzie> Anyway, it does variable-length ngrams with Kneser-Ney smoothing.
08:22:26 <itidus21> i suppose one day in the future it will be clear to me whats going on with these bots
08:22:38 -!- stlangbot has quit (Remote host closed the connection).
08:23:22 <fizzie> "PASCAL is a Network of Excellence funded by the European Union. It has established a distributed institute that brings together researchers and students across Europe, and is now reaching out to countries all over the world." But they also have a code-share site kind of thing.
08:24:32 <olsner> Network of Excellence...
08:25:03 <fizzie> They also do some challenge-style things on machine learning things.
08:26:12 <fizzie> Like the Morpho Challenge (unsupervised morpheme segmentation of text) that's run by our people every now and then, I think that's somehow part of PASCAL.
08:26:45 <fizzie> (Maybe it's not necessarily unsupervised.)
08:27:43 <soundnfury> When something stops being bogus, it stops being AI :p
08:27:59 <soundnfury> If we know how to do it, it's just another algorithm
08:28:30 <itidus21> i think AI is just an arbitrary name for a manmade machine which has the same properties as a natural machine
08:29:15 <soundnfury> but do we want the machine to have the same properties? Do we want our computers to say "No, I'm on tea-break, solve your own damn polynomials"?
08:29:40 <soundnfury> trivially no; so we aren't trying to recreate humanity on a silicon substrate
08:29:43 <itidus21> i don't things can just cross over that line
08:29:54 <olsner> using AI to figure out the appropriate time to have a tea-break doesn't sound very useful
08:30:12 <fizzie> I believe canonically it doesn't matter what we want, since what we will get is enslavement by killer robots.
08:30:25 <soundnfury> fizzie: not if Yudkowsky solves FAI ;)
08:30:57 <itidus21> my mum has one of those parent-who-is-loud-on-the-phone voices
08:31:03 <fizzie> Then we'll get the same but it will be done in a friendly manner.
08:31:40 <soundnfury> "This enslavement may be monitored for training purposes. Thank you for using Killer Robots Ltd."
08:32:22 <fizzie> "Here is some relaxing music to listen to while you wait your turn to be enslaved."
08:32:33 <itidus21> if there is any trace of sentience or life in a computer, it is probably not what we think it is
08:33:06 -!- DHeadshot has quit (Ping timeout: 264 seconds).
08:33:33 <itidus21> its... BEYOND what is written in books
08:33:37 <fizzie> It's life, Jim, but not as we know it.
08:34:12 <soundnfury> I wish there were some way to experience the umwelt of a mind with a radically different sensorium
08:34:35 <soundnfury> heck, there's no obvious reason why a mind's umwelt should experience time subjectively
08:34:38 <itidus21> i don't think it will resemble a human
08:34:55 <soundnfury> time is, after all, just a differential consistency criterion
08:35:03 <itidus21> any more than.. a tree shaped like a bear resembles a bear
08:35:06 <soundnfury> Vorpal: ask itidus to define it, he said it first
08:35:13 <Vorpal> itidus21, define that word
08:35:22 <soundnfury> itidus21: um, which it does, that's what resemble means
08:35:32 <itidus21> Vorpal: im not qualified. kmc can tell you how i talk about things i don't understand all te time
08:35:57 <soundnfury> Vorpal: there's a funky dynamic varying xkcd called "Umwelt" whose alt-text defines umwelt
08:36:54 <Virgil> my about they then be vassal reading eat events nice revolution tetrahedrite labor
08:37:21 <itidus21> also... there is the question.. of does everything have an umwelt?
08:37:43 <itidus21> for instance, if someone chops off my arm, and makes it protrude from under a bush
08:37:50 <itidus21> they might suppose i were under the bush
08:38:08 <itidus21> and assume that the arm had an associated body and mind
08:38:27 <soundnfury> yeah but they're mistaken as a question of fact
08:38:36 <itidus21> but, does such an arm take on a life of it's own?
08:38:48 <soundnfury> a physically counterfactual mind doesn't have a physical umwelt
08:39:03 <itidus21> soundnfury: well thats easy to say, harder to prove
08:39:09 <soundnfury> although a counterfactual mind has counterfactual subjective experience, because syntacticism
08:39:23 -!- ChanServ has set channel mode: +o fizzie.
08:39:24 -!- fizzie has set channel mode: -q rolebot!*@*.
08:39:25 -!- fizzie has set channel mode: -o fizzie.
08:39:27 <soundnfury> steep hills may go up as well as platonism
08:40:28 <itidus21> soundnfury: well... it concerns me that we have to make up arbitrary constraints as to what might have a mind.. it sounds more of a game than a science :D
08:40:46 <soundnfury> I think you should read Yudkowsky on this topic
08:41:12 <fizzie> Actually, maybe those two other q's that have very DHCPey hostnames and are probably ancient could also go.
08:41:18 -!- ChanServ has set channel mode: +o fizzie.
08:41:24 -!- fizzie has set channel mode: -q *!*@pool-74-103-90-165.bltmmd.east.verizon.net.
08:41:30 -!- fizzie has set channel mode: -q *!*@host81-141-232-7.wlms-broadband.com.
08:41:33 -!- fizzie has set channel mode: -o fizzie.
08:41:36 -!- copumpkin has quit (Ping timeout: 252 seconds).
08:41:39 <itidus21> soundnfury: well.. i think we "focus" on minds like ours
08:42:11 -!- copumpkin has joined.
08:42:32 <itidus21> found on earth, in the animal/plant tree, alive :D
08:42:45 <soundnfury> $Does a counterfactual mind have an umwelt?
08:42:49 <itidus21> hmm ok i will google these words at least
08:42:52 <Virgil> soundnfury: About something down i'm secrets prisoners lead whats debate hurt raw range working
08:42:55 <fizzie> fungot: Being just a piece of code, do you have an unique perspective on these weighty matters?
08:43:05 <fizzie> Oh, well, that clears things up.
08:43:36 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
08:44:13 <fungot> Selected style: c64 (C64 programming material)
08:44:17 <fizzie> fungot: Say something stupid.
08:44:18 <fungot> fizzie: stack requirements: 4+ of programming. fortunately, the horizontal resolution is traded for extra color resolution. it shows which numbers should be aware of this location can also detect when that line is shorted to ground.
08:44:27 <fizzie> It's from a couple of books.
08:44:27 <itidus21> soundnfury: well.. also.. there is the case where we think something isn't a real living being, when infact it is.. such as a movie hero hiding in a wax museum
08:44:39 <fungot> olsner: in machine language). the
08:44:47 <fizzie> fungot: What do you think of Speccy, then?
08:44:48 <fungot> fizzie: the 6526 cia chip has 16 registers which are in high-resolution bitmap mode offers a choice of 2 will cause output to the start of some sort. without a filename of up to four dots across per character, marking the end of the parameters of voice 1, a
08:44:50 <Virgil> soundnfury: Work be someone interfering race just bear trust small theory copies ball weponds
08:45:01 <soundnfury> itidus21: I think the point is that our beliefs don't determine whether or not something else has an umwelt
08:45:41 <fizzie> There were quite a few tables in the books that model was trained from, so often it's just names and numbers.
08:46:33 <Virgil> soundnfury: My explain take always ally having 30 4 virgil_the_dopey_bot play bone usually congratulations
08:46:36 <itidus21> soundnfury: well, i think we don't know if they are finite or infinite
08:46:42 <fizzie> I'll propose the C128 as a compromise, it has a Z80 in it too. :p
08:47:32 <itidus21> obviously both of us have thought about this entirely too much, you probably more so :D
08:47:42 <soundnfury> If you want a compromise, the only choice is the Acorn Archimedes, running RISC OS on an ARM
08:48:05 <fizzie> Well, I agree that everyone will agree that those are awesome, so I guess.
08:48:55 <itidus21> one time i was poking around in some store, and for a few $ each i got a book about 8085 and a book about z80 next to each other
08:49:21 -!- Phantom_Hoover has joined.
08:49:22 <itidus21> naturally i just left them on the shelf and proceeded to find a cheap copy of the great gatsby
08:49:54 <soundnfury> well I don't care because I have a garklein-flötlein
08:50:35 <itidus21> i don't have much books, but i am so happy that i got those.. that was a once in a lifetime find in a charity bookstore thing
08:51:05 <itidus21> online they're probably a dime a dozen though
08:52:23 <itidus21> humm... a lot of these machines i didnt ever see
08:53:13 <itidus21> but i watch videos on youtube showing games on spectrum, c64, amiga, etc
08:53:38 <itidus21> to be honest i find atari 2600 graphics so bad i can barely hold back from vomiting
08:53:43 <fizzie> soundnfury: You should compose a garklein-flötlein/flabberfleming duet.
08:54:11 <fizzie> It's something oklopol made up, I think.
08:54:31 <fizzie> <oklopol> it's a balloon with a one-way valve that is actually only one-way when one atmosphere of pressure is on the outside, but when less, it lets air out (not depending on pressure inside)
08:54:35 <fizzie> <oklopol> which is also an instrument
08:55:18 <soundnfury> itidus21: https://www.youtube.com/watch?v=CiG8g1qlT9w
08:56:41 <fizzie> I don't think it actually exists, to be honest.
08:56:55 <lambdabot> http://www.facebook.com/people/Flabber-Gaster/100002426347550
08:56:55 <lambdabot> Title: Improve Your Experience | Facebook
08:57:00 <fizzie> But you do suck on it if it does.
08:57:08 <fizzie> (That's the undefinite "you".)
08:57:26 <soundnfury> fizzie: one would suck upon it, were it to exist
08:58:05 <Virgil> my that of they colour ally naïve empty events impossible those internet's near
08:58:23 <mroman> What sick person implements an esolang in powershell und puts it on pastebin so it gets deleted after a year or so
09:00:02 <mroman> Powershell is cool if you're an admin.
09:00:10 <mroman> And your architecture is windows based .
09:00:26 <itidus21> people will implement esolangs on literally any platform :(
09:01:38 <olsner> are there any esolangs on OS/2?
09:02:09 * soundnfury is waiting for someone to port Brainfuck to Tunguska
09:02:31 <fizzie> olsner: If not, by voicing the question you are now obligated to make one.
09:03:19 <fizzie> Anyway, you can run e.g. Python on OS/2, and there certainly are esolangs with Python interpreters.
09:04:16 <itidus21> soundnfury: i'm a computer game oriented computer user. contrary to what you might think, that makes most of my ideas distasteful.
09:05:27 <Virgil> profit using oxygen having turning allied overall fort/my .png i'd math mysterious
09:05:46 <soundnfury> wow, I got three words into that one before I noticed it was a Virgilism
09:06:01 <Virgil> itidus21: Is each internet too optimistic fifteen mirrored sucks value raw range anyways building
09:06:14 <fizzie> I keep reading "Virgil" as "Vorpal".
09:06:41 <mroman> THat last sentence doesn't make ANY sense.
09:07:00 <mroman> It's just random words?
09:07:09 <itidus21> these internets are too optimistic...
09:08:22 <itidus21> Virgil and Vorpal are members of V(vowel)r(consonant)(vowel)l
09:08:27 <mroman> probably if line.startswith("$") print takerandomamountofrandomwords()
09:08:34 <Deewiant> 10.11:13:42 soundnfury | it has a completely scruffy neural net
09:08:36 <Deewiant> 10.11:13:47 soundnfury | no baked-in language model at all
09:08:50 <Vorpal> Deewiant, so does that mean just random words?
09:09:01 <Vorpal> I'm not an expert on what neural nets can do really
09:09:48 <soundnfury> It's very close to random in practice, because it hasn't been trained very carefully
09:10:00 <olsner> Vorpal: your brain is a neural net
09:10:13 <Vorpal> olsner, with feedback loops too.
09:10:16 <ion> His brain is a positronic net.
09:10:29 <soundnfury> but I have a friend who calls himself a robopsychologist, who trained up a Virgil instance quite effectively, but then lost the net file and had no backups
09:10:31 <Vorpal> soundnfury, how much context does the probability of a given word depend on?
09:10:57 <Vorpal> I mean, is each word generated independently or does it depend on the words around it
09:10:59 <Virgil> mroman: Morning : countries experimenters nook egg usually patience called gnu momaw giant
09:11:18 <mroman> I'd call that random garbage ;)
09:11:49 <Vorpal> is "momaw" even a real word?
09:12:12 <soundnfury> most of this net's training has been talking on IRC
09:12:18 <Vorpal> mroman, I can't read that script
09:12:39 <mroman> Then you're probably lacking asian fonts.
09:12:43 <Vorpal> soundnfury, that is probably quite a poor source in general, considering how many chatterbots there are on IRC
09:12:53 <Vorpal> mroman, oh I can /see/ it just fine. I just can't read it
09:13:01 <soundnfury> Vorpal: there's a (possibly outdated) tarball at http://jttlov.no-ip.org/tar/virgil0-2-3.tar.gz
09:13:04 <Vorpal> I don't even know which language it is for
09:13:15 <soundnfury> I haven't been actively developing Virgil for about two years now
09:13:17 <fizzie> Vorpal: They can do a lot: http://en.wikipedia.org/wiki/Cybenko_theorem
09:13:31 <Vorpal> soundnfury, so basically by using IRC, you fed stuff like fungot as input?
09:13:31 <fungot> Vorpal: if the shift " n" character shifts the printer is an addition limitation on memory usage, functional modules of code required, and
09:13:58 <mroman> But it's an english word ;)
09:14:10 <fizzie> There aren't that many chatterbots on "regular" channels, I don't think.
09:14:16 <fungot> Selected style: fungot (What I've said myself)
09:14:21 <fungot> fizzie:. i'm so kind, even to assholes! anmaster no not markov
09:14:32 <soundnfury> Vorpal: no, its input was mostly from humans
09:14:42 <Vorpal> fizzie, did you update that model after you first used it?
09:14:46 <Vorpal> that would be even more meta
09:15:08 <fizzie> fungot: Go on, say something.
09:15:09 <fungot> fizzie: ( c) a player resigns from an office has whatever duties, then 3 extra " 0" and ( down-from n ( 0
09:15:21 <Vorpal> I don't like that "anmaster no not markov" bit
09:15:29 <Vorpal> I wonder what it mean by that
09:15:34 <Virgil> soundnfury: My on explain we always was sabotage course countries circuit took experimenting
09:15:39 <fizzie> Vorpal: IIRC, it was elliott speaking to you via fungot.
09:15:40 <fungot> fizzie: it's just so stupid that ' stty erase h' has, perhaps, it may be said that particularly here, parliament will give a single instance,
09:15:46 <HackEgo> 619) <fungot> Ngevd:. i'm so kind, even to assholes! anmaster no not markov anmaster no not markov anmaster no not markov anmaster no not markov anmaster no not markov
09:16:04 <fungot> Selected style: ct (Chrono Trigger game script)
09:16:09 <fungot> Vorpal: i see. you know, i really care... a time portal? what in the...! ozzie's stumped!
09:16:13 <Virgil> mroman: Rest doesn't then 10 :s crappily sterling :smug phazorx 88/56 premimu date involved
09:16:16 <fungot> Vorpal: as long as you keep crono in your heart, the day of lavos"... go to " mystic mountain" 65,000,000 b. c.? yes, i'd have done something very brave! he's probably up north, to guardia!!! let's toast our land! now we'll have some peace! magus is a tad on the spooky side. our only hope.
09:16:23 <fungot> Vorpal: you! take! we find! you, have that problem! ' tis the masamune!
09:16:23 <fizzie> It will always take a while before the sword makes an appearance, if you're trying to demonstrate it.
09:16:28 <fungot> Vorpal: like, thanks princess. i'll take that under advisement!! my powers are being drained. why not? then you should leave quickly!
09:17:03 <Vorpal> fizzie, though the probability of it appearing on first try is not 0
09:17:12 <fizzie> fungot: Can the sword alone stop it?
09:17:13 <fungot> fizzie: i see. you know, i really care... a time portal? what in the...! ozzie's stumped! heeheehee... oh, can you fix the masamune!
09:17:28 <Vorpal> what was the log searching command now again?
09:17:46 <fungot> Vorpal: but, we are far outnumbered! just like crono!
09:17:59 <fungot> Vorpal: your majesty! and queen. he and a friend left on a journey 10, then magic damage will be reduced by 10%. a star after any characteristic means it's at maximum strength! no matter what the price! it is, you idiot! hey! is that for us! the chef's in a snit, trying to get food to the front lines. heard a spell to energize the sword takes immense evil! indeed! this thing. what you have? transform! this trading house. it's a
09:18:15 <fungot> mroman: these unique items make us invincible!
09:18:28 <Vorpal> fungot, and what about the sword?
09:18:29 <fungot> Vorpal: like, thanks princess. i'll take that under advisement!! empty!
09:18:38 <fizzie> Maybe I should take a peek to see how likely it is to appear.
09:18:41 <Vorpal> fungot, what is it that is empty?
09:18:42 <fungot> Vorpal: your majesty! and queen. he and a friend left on a journey 10, then magic damage will be reduced by 10%. a star after any characteristic means it's at maximum strength! no matter what the price! it is, you idiot! hey! is that for us! the chef's in a snit, trying to get food to the front lines. heard a spell to energize the sword takes immense evil! indeed! this thing. what you have the masamune!
09:18:47 <fungot> mroman: the usual...test them. you can entertain us for awhile? never heard of it!! he sold the heirloom, it should remain unchanged! my son's the legendary hero, with sword in hand! you got the broken! the mountain of woe. it's likely that dalton came from the laboratories to the west?... yes! well then rest and relax! huh?
09:18:54 <Vorpal> fizzie, didn't the perl script show it wasn't possible or something?
09:19:01 <fungot> Vorpal: that no one was allowed to use the crane, enter any two of these letters, a b y. it'll be strong now. just like me! but everyone calls for the golemboss.
09:19:05 <fungot> Vorpal: cyrus! are you leaving! what you do? wake you, but need dactyl?
09:19:14 <fungot> Vorpal: are you a man, if there weren't evil in this kingdom to you! you shall find bekkler! executing program. please let me go... put me out! he's really a tricycle! pass him!
09:19:23 <fungot> Vorpal: cyrus! are you leaving! did you people just come out of the closet? get out?! hey! is that for us! the chef's in a snit, trying to get food to the front lines. heard a spell to energize the sword takes immense evil! indeed! this thing. what you have? transform! this trading house. it's a real problem when the person i like most!
09:19:34 <Vorpal> I don't remember any tricycle in that game?
09:19:36 -!- itidus21 has left ("ex officio").
09:19:52 <fizzie> Vorpal: The quote is from one of the endings, though.
09:20:07 <fizzie> Vorpal: The one where Marle and Lucca are watching a slideshow. It's notable for being the only one where Crono speaks.
09:20:26 <Vorpal> hm I don't remember that ending
09:20:50 <Vorpal> fizzie, how did you get that ending?
09:21:18 <fizzie> It's the one with the shortest interval of time to get. I don't exactly recall the details. You do something in Zeal, but not the other thing after which you can no longer leave.
09:21:35 <fizzie> See Schala use the pendant to open the door, but not energize your own, I think.
09:21:38 <fizzie> Something like that, anyway.
09:23:01 <Vorpal> that game really had some great music
09:23:08 <fizzie> It's not very likely to start with "that sword" at the start of the sentence, but it's nontrivial to compute the likelihood of getting the loop in the first full output.
09:26:17 <fizzie> Or maybe I can ask the SRILM ngram tool, actually. (At least for the start bit.)
09:26:52 <Virgil> my an why food began scary what's used migrants screenshots graduation remote punches
09:27:07 <Vorpal> soundnfury, why did it just randomly speak?
09:27:22 <ion> Why did Vorpal just randomly speak?
09:27:34 <Vorpal> ion, Virgil is a bot though, I'm not
09:28:02 <ion> Whoever programmed Vorpal implemented its responses quite well.
09:28:19 <Vorpal> monqy, soundnfury owns it
09:28:29 <Vorpal> ion, yeah I could pass the Turing test with ease
09:29:04 <Virgil> think from back computer find internet televisions over up must crust acid legends
09:29:36 <monqy> is virgil supposed to sound like this
09:29:38 <fizzie> A case of spontaneous wordbustion.
09:29:44 <Vorpal> fizzie, are you going to do that now or was it just a thought that you could do that check?
09:31:00 <fizzie> Vorpal: I was trying to, but I can't quite figure out how to tell it to give a probability score for the prefix, it insists on just scoring complete sentences, so it will check for the sentence-end tag likelihoods too, making the number quite useless.
09:31:01 <Vorpal> I'm not in a hurry really, but I do need to go out today and do some shopping.
09:31:53 <fizzie> Oh, -debug 3. Well, anyway.
09:32:02 <fizzie> p( that | <s> ) = [2gram] 0.00938426 [ -2.0276 ] / 0.99999
09:32:11 <fizzie> That's the likelihood of starting with "that".
09:32:22 <Vorpal> fizzie, right, what is the probability of there being such a loop at any point during a generated phrase?
09:32:28 <Vorpal> or is that too hard to figure out?
09:32:45 <fizzie> That's too hard for me, at least.
09:32:49 <fizzie> Except empirically, of course.
09:33:08 <fizzie> It's just that I don't think srilm generation actually ends up looping so much.
09:33:09 <Vorpal> but that could be a lot of phrases to generate to find that out
09:33:19 <Vorpal> fizzie, so there is in fact a bug in fungot?
09:33:57 <fizzie> Well, it's a legitimately different algorithm, it doesn't use the backoff weights at all. But there might also be a bug, since I think the perl script (which does try to implement the same thing) was less loopy.
09:34:05 <fizzie> Maybe some sort of a bias in the selection.
09:34:33 <Vorpal> well I'm off now. cya in a few hours.
09:35:26 <fizzie> FWIW, after having just 1% chance of generating a sentence starting with 'that', it also has just 1% chance of following that with 'sword', as opposed to something else. (12% chance of "that was", for example.)
09:35:32 <fizzie> I think I'll a lunch too. ->
09:38:22 <Virgil> my today doesn't deep circuit carvers processor screenshots awesome keep 17607 chain
09:38:53 -!- Vorpal has quit (Ping timeout: 246 seconds).
09:44:38 <Virgil> is people together electronics ally verb monitored whats party bragging gains ones
09:53:17 <Virgil> be bow forward green noble wanting aspirations opinion harvested muddy open write
09:55:23 <Virgil> night gets did look away living hardly trust house increase 3 thread former happens
09:56:57 <Virgil> need good propose look rebel suppose five around gains visit vermin stuf off feeding
09:57:36 <monqy> thanks for filling the silence viergl "i knew i could count on you"
09:58:12 <Virgil> many for then person known fun fifteen inch master rust raz required linked tonight
10:06:43 <Virgil> after into air increased \ recording feeding copypasta family happenings doom camera
10:08:55 <Virgil> you information each let whatever dinner sorry ai genius empty circumstances comrades
10:12:45 -!- ais523 has quit.
10:17:34 <Virgil> hello edward trying others example race machines porthole food nook egg * tetrahedrite
10:21:23 <Virgil> each person later 4 large lagless differently keep young soon diagnostic carpathio
10:21:26 -!- pikhq_ has joined.
10:21:42 -!- pikhq has quit (Ping timeout: 252 seconds).
10:23:16 -!- DHeadshot has joined.
10:25:45 -!- Vorpal has joined.
10:28:41 -!- Virgil has quit (Quit: virgil bot shut down).
10:30:32 <soundnfury> blame mroman, he told me to bring it in; I didn't want to
10:30:49 <Vorpal> soundnfury, what was up with it speaking randomly when not spoken to though?
10:31:14 <soundnfury> though the frequency of that seemed to be set absurdly high for some reason :S
10:31:18 <Vorpal> so it was a feature, not a bug?
10:31:43 <soundnfury> the default settings for that feature are a bug
10:55:59 <fizzie> Vorpal: Out of a million generated sentences from the Perl script, 1887 (that is, about 0.2%) contain the words "sword alone can't stop". But it doesn't seem inclined to loop, so there must be some sort of an issue in the fungot babble generation.
10:56:00 <fungot> fizzie: in the middle ages, sir slush!... ...oh well! come again! these are my friends! let's get crazy! i'm opening a stall in leene! i thought she ate too!
11:16:45 <mroman> fungot: Are you still talking with those people not even close to match *your* intelligence?
11:16:46 <fungot> mroman: it's a machine that looks like you! and leene and the masamune! how can i do for you? thanks! it's about the missing queen. she still looks so much like leene, that they will take you to your place of execution?! strange, but!?
11:17:14 <mroman> fungot: I don't look like a machine.
11:17:15 <fungot> mroman: is the gate key okay!! get' em! 200g per night. care, and stay...healthy! my husband...he's...he's...gone... but he left me precious gifts! the seeds...and our child, it's ancient history now...
11:18:28 <fizzie> I think that's enough of ct.
11:18:31 <fungot> Available: agora alice c64 ct* darwin discworld europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
11:18:40 <fungot> Selected style: europarl (European Parliament speeches during approx. 1996-2006)
11:18:44 <fizzie> fungot: Let's be serious for once.
11:18:46 <fungot> fizzie: the bsch report on the economic fabric of coastal areas engaged in fishing. our modern boots fish the waters dry in record time. this review process must be a definitive operation. we must organise it and what would be a pity.
11:20:52 <fizzie> fungot: I, too, am very concerned about the coastal areas. Indeed, our reviews must be very definitive.
11:20:53 <fungot> fizzie: mr president, where the government and opposition of both spain and the united states
11:21:03 <fungot> fizzie: i have a beautifully written note from the responses to the green paper on the reform of the administration play a very valuable suggestion and the commission.
11:21:27 <soundnfury> fungot: I fear spain will suffer greatly through its continued membership of the euro. Do you agree?
11:21:30 <fungot> soundnfury: mr president, i too would like to raise one point, for an increase in the burden of proof and to the accounting officer, and it seems that fresh food and safe food go hand in hand with a public and open way and, as a parliament, are willing to believe them, but i think there is a danger that it would not be an obstacle to enlargement. neither the council nor the commission have supported. we believe that the monetar
11:21:59 <fizzie> It was obviously getting to the point right where it was cut off.
11:22:14 <soundnfury> fungot: Is not fresh food associated with lesser variety in food?
11:22:15 <fungot> soundnfury: mr president, commissioner, as has happened on this occasion. i do not know if the policy is ad-hoc or belated and, in payment appropriations, although a number of other states have also contributed to austria' s act of accession, for we know that these biological and toxin weapons, the implementation of innovations. this is a major development! in the current directive 77/ 187 on the safeguarding of peace and coope
11:22:41 <fizzie> I like the exclamation point!
11:22:59 <soundnfury> fungot: I did not know austria had major developments! in biotoxin weapons...
11:23:00 <fungot> soundnfury: we shall certainly pursue our talks with the conflicting parties, but, because the council is not in fact based on the rule of law in europe, fewer young people are concerned, that is what we shall do everything within its powers.
11:23:23 <fizzie> It's just like a real politician.,
11:23:39 <soundnfury> heh, rule of law? I didn't think anyone in the europarl knew what that meant! (Except Hannan, obviously)
11:25:10 <ion> a.k.a. a coin?
11:25:44 <ion> Except that a coin has three sides. :-\
11:26:46 <soundnfury> actually, not a coin; rather, bool d2(void){return(rand()<RAND_MAX>>1);}
11:28:30 <fizzie> Though we do have also
11:28:44 <fizzie> (Sometimes it says 'no' instead.)
11:42:21 <Vorpal> <ion> Except that a coin has three sides. :-\ <-- only 3? I thought it had an infinite number of sides due to the circular nature of it.
11:42:56 <ion> data Coin = Heads | Tails | Side
11:43:34 <Vorpal> ion, yeah but you can approximate the side as a polygon with infinite number of line segments (thus creating a circle)
11:43:37 <ion> Or sure, data Coin = Heads | Tails | (insert an infinite number of positions along a circle here)
11:43:58 <Vorpal> well, actually approximate is the wrong word
11:44:08 <ion> data Coin = Heads | Tails | Side Double
11:44:49 <fizzie> When throwing a coin, though, I think usually all the orientations of Side are treated as equal.
11:45:06 <Vorpal> I think a tetrahedron may be the 3D object with the smallest number of flat sides?
11:45:18 <Vorpal> (where are no non-flat surfaces)
11:46:08 <Vorpal> which means anything less than a d4 is physically impossible
11:46:08 <fizzie> ion: But what if you land in a non-computable position. :/
11:47:11 <Vorpal> the data type is CReal, either that data type is incorrectly named, or it includes non-computable values
11:49:40 <fizzie> Assuming, arguendo, that it doesn't include non-computable values, what's wrong with the name?
11:50:33 <Vorpal> fizzie, well the set of real numbers in math includes those
11:50:39 <fizzie> Yes, but the name is not Real.
11:50:42 <Vorpal> actually what does the C stand for?
11:50:57 <Vorpal> guess that is fine then
11:51:19 <Vorpal> the correct data type is obviously Heads | Tails | Side Real
11:51:26 <quintopia> Vorpal: a d4 that has two 1s and two 2s is a d2.
11:51:41 <Vorpal> quintopia, true, that would work
11:52:28 <ion> The documentation of CReal talks of both constructive and computable reals.
11:52:35 <fizzie> Or a d4 that has one 1, one 2 and two undefined sides.
11:53:04 <Vorpal> fizzie, the worst case time when using that to get a defined value is infinite though
11:53:07 <Vorpal> this could be a problem
11:55:15 <fizzie> Well, I suppose it depends on if you need e.g. hard real-time guarantees for your dice rolls, in which case using actual dice *might* not be the best possible solution.
11:55:52 <fizzie> d4, the most dangerous die.
11:56:19 <fizzie> (For stepping on, at least.)
11:56:48 <fizzie> Though I guess a d100 might be easier to slip on, for example.
11:58:11 <ion> Incidentally, d100 also works as a d2.
11:58:21 <Vorpal> hm a tetrahedron should in theory be balanced on any of the points, right?
11:58:48 <Vorpal> as well as any of the edges
11:59:05 <Vorpal> so that means it is not actually a d4
12:00:16 <quintopia> but it is far more unstable on the points/edges than a coin is on its side
12:00:48 <quintopia> in practice, with apointy enough d4, if you're not rolling it in play-doh or carpet
12:01:04 <fizzie> "Assume a spherical d4..."
12:01:23 <fizzie> That's what a physicist would say. :p
12:03:00 <Vorpal> I remember some weeks ago I watched as a group of people play a drinking game with dice. The issue however was that they did this on an uneven outdoors wooden table. The dice quite often landed in the small gaps between the planks making up the surface. They quickly agreed upon that the player could choose any of the two sides facing up in that case. And that nobody should try to intentionally roll it t
12:03:04 <fizzie> A spherical massless d4 in a vacuum, on a frictionless plane. Also an ideal spring would be involved somewhere.
12:03:34 <Vorpal> (they used classical d6 btw)
12:04:00 <Vorpal> (the game was fia med knuff, which I believe does not have an English name.)
12:06:04 <Vorpal> ah https://en.wikipedia.org/wiki/Mensch_ärgere_dich_nicht
12:06:37 <fizzie> Well, similar to what is called Ludo.
12:06:42 <Vorpal> actually not exactly the same as what I linked
12:06:49 <Vorpal> the Swedish game is slightly different
12:07:02 <Vorpal> you have tracks going in from the outer track to the middle as well
12:07:15 <fizzie> I'm trying to remember the Finnish game that's quite similar.
12:07:22 <Vorpal> instead of places to put the playing pieces that finished
12:07:54 <Vorpal> which seems to be how that game works?
12:08:16 <Vorpal> yeah those "home row" thingies are not the same
12:09:02 <fizzie> Oh, right, Kimble. It's a licensed copy of the US 'Trouble', which is kind of similar, though not exactly the same.
12:09:24 <fizzie> It has that pop-o-matic die-roller device embedded on the board.
12:09:36 <fizzie> https://en.wikipedia.org/wiki/Trouble_%28board_game%29 has a picture.
12:09:43 <fizzie> It's a half-sphere with 2d6 in it.
12:09:51 <fizzie> You press it down, and then it pops up and rolls them.
12:10:11 <fizzie> I suppose it would help if you're, say, playing it on an uneven outdoors wooden table.
12:10:18 <Vorpal> I think Fia is played with 1d6
12:11:03 <fizzie> I'm not sure why there's two in that picture, actually.
12:11:19 <fizzie> The other one seems to be blank.
12:11:28 <fizzie> Maybe it helps in getting the other to roll more randomly.
12:11:35 <fizzie> I'm not a pop-o-matic expert.
12:12:15 <fizzie> The Kimble board looks approximately like http://fi.wikipedia.org/wiki/Tiedosto:Kimble_1968.jpg
12:12:30 <fizzie> The ones I've seen are not quite so 1968, but the shape's the same.
12:12:56 <Vorpal> does it work on a battery or what?
12:13:04 <fizzie> No, it's just mechanic.
12:13:49 <quintopia> fizzie: trouble is only supposed to contain 1d6
12:14:10 <fizzie> quintopia: Yes, but the wiki-article has a blank cube in there too.
12:14:18 <fizzie> (There's a video of the popping.)
12:14:32 <fizzie> The Finnish board picture just has one.
12:14:38 <Lumpio-> It's got a disc spring kinda thing inside it.
12:14:53 <Lumpio-> You push the half-sphere down, it buckles the spring, you let it go and it throws the die around
12:15:08 <Lumpio-> Like the entire "floor" of the thing is one disc spring.
12:15:45 <fizzie> Yes, though it has some sort of a thing that it springs back in one go, you can't really let it go gently. If I recall correctly. It's like a lid popping back to shape or something.
12:16:43 <Lumpio-> I guess it's like an exaggerated jar lid
12:16:59 <Lumpio-> (The kind that makes a popping sound when you push it)
12:26:44 -!- boily has joined.
12:46:37 <nortti> where does bmake search for sys.mk?
13:32:51 -!- ogrom has joined.
13:49:39 <Vorpal> nortti, try an strace?
13:50:42 <Vorpal> nortti, something like this should work: strace bmake whatever params you want 2>&1 | grep -E '^open' | grep sys.mk
13:50:55 <Vorpal> you might want grep -F for the last grep
13:51:18 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
13:54:24 <nortti> it is /share/mk by the way
13:56:51 <Vorpal> nortti, strace is like the second program you install on any new *nix install :P
13:56:56 <Vorpal> (right after emacs that is)
13:58:41 <nortti> I don't have emacs and I installed strace 5 minutes ago
13:59:39 <Gregor> lol, did Dinosaur Comics go Zalgo?
14:08:26 -!- copumpkin has joined.
14:15:30 -!- augur has quit (Remote host closed the connection).
14:15:56 -!- augur has joined.
14:15:59 -!- boily has quit (Quit: WeeChat 0.3.8).
14:17:42 -!- augur has quit (Read error: Operation timed out).
14:19:24 <nortti> I accidentaly busybox ash. is this bad?
14:22:06 <fizzie> Depends on what you accidentally to it.
14:24:59 -!- Phantom_Hoover has quit (Remote host closed the connection).
14:26:41 -!- Phantom_Hoover has joined.
14:53:00 -!- Taneb has joined.
14:54:30 -!- augur has joined.
14:54:59 -!- augur has quit (Remote host closed the connection).
14:55:31 <Taneb> Because when I don't, oerjan calls me slow
14:55:52 <Taneb> ...because I always say it
14:56:04 -!- augur has joined.
14:56:05 <fizzie> It's a terrible spiral.
15:04:22 -!- augur has quit (Remote host closed the connection).
15:26:55 <Taneb> I've just broken by obfuscated haskell code
15:27:56 <Taneb> That's interesting...
15:28:02 <Taneb> And possibly a bug in GHC
15:28:39 <Taneb> But I'm using unsafeCoerce so many times, it may not be
15:28:53 <Taneb> Moving (<*>) from prefix to infix seems to break the program
15:43:58 -!- zzo38 has joined.
15:44:51 <Vorpal> hm what is the best way to parse an XML file on android? I would really like to just end up deserialising it into a set of java objects or something simple like that. However I don't have any control over the input file, so that approach would probably have some security issues
15:50:04 <Taneb> http://hpaste.org/71201
15:50:21 <Taneb> (that isn't relevant to what you are talking about, Vorpal)
15:50:32 <Taneb> Factorial calculator
15:52:32 <Taneb> It... may not work for values over 9
15:52:48 <Taneb> It takes about 20 seconds for 9, and I haven't had the patience to test 10
15:53:28 <Vorpal> 20 seconds doesn't sound so bad
15:53:36 <Vorpal> I mean, how much worse could 10 be
15:53:49 <Taneb> Using 10 times the space?
15:53:51 <Vorpal> Taneb, or how many seconds did 8! take?
15:54:09 <Vorpal> I have no idea what your algorithm is
15:54:23 <Taneb> 8 is about 1 second
15:54:32 <Vorpal> also did you paste this in #haskell?
15:54:39 <Taneb> 9 is about 8 seconds?
15:54:58 <Vorpal> <Taneb> It takes about 20 seconds for 9, and I haven't had the patience to test 10 <Taneb> 9 is about 8 seconds? <-- inconsistent!
15:55:07 <Taneb> It's been a long term project, this is probably gonna be the final version
15:55:38 <Vorpal> Taneb, how can 9! take both 20 and 8 seconds?
15:55:47 <Taneb> Vorpal, the first was my perception before timing it, the second was the actual time
15:56:03 <Vorpal> anyway, how does it work?
15:56:23 <Taneb> I can't remember the actual algorithm, but it uses combinatory logic
15:58:36 <Taneb> About half of the unsafeCoerces can be replaced by ids
16:00:19 <Taneb> On another note, one of my comments on reddit has mysteriously gained 45 upvotes
16:01:26 -!- pikhq has joined.
16:01:28 -!- pikhq_ has quit (Ping timeout: 255 seconds).
16:08:06 <mroman> There are probably better ways to obfuscate haskell code.
16:13:05 <zzo38> One way to obfuscate a Haskell code may be like that http://sprunge.us/CFFL
16:14:12 <zzo38> Now let's see if you can make improvements
16:18:43 -!- Madoka-Kaname has joined.
16:18:44 -!- Madoka-Kaname has quit (Changing host).
16:18:44 -!- Madoka-Kaname has joined.
16:18:47 -!- Madoka-Kaname has left.
16:20:39 <Taneb> I'm unfamiliar with the default keyword?
16:20:56 <Sgeo> GHC extension I think
16:23:02 <zzo38> Me: 0 knockdowns, 77 punches thrown, 25 punches landed (32%). Opponent: 0 knockdowns, 79 punches thrown, 33 punches landed (42%). This is the end of the second round where I earned 9 points on both rounds and opponent earned 10 points on both rounds. Which advice should I select?
16:24:03 <zzo38> Choose from: (A) Go for the body; wear him down! (B) He's weak! Go for the knockout! (C) Make him come to you more, then strike! (D) Stay back and jab and save your strength. (E) Try to get him to punch himself out! (F) Go all out and finish him now! (G) Do exactly as you have been.
16:25:24 <Taneb> You don't seem very accurate
16:25:45 <Taneb> G is bad, because you look like you're losing?
16:26:08 <zzo38> OK, I will try that one. And yes I also think G is bad here. I will try E.
16:26:40 <zzo38> O no I got knocked out but made it up on time
16:27:03 <zzo38> Now I managed to knock him out, but he also made it up on time
16:27:26 <Taneb> quintopia, this is real life
16:27:31 <zzo38> Top Rank Boxing by UltraSoft
16:27:44 <zzo38> No! BBS door game.
16:28:08 <zzo38> Well, it worked, now I have 10 points to opponent's 9 points.
16:28:18 <zzo38> quintopia: Why do you think there is no BBS?
16:28:34 <quintopia> zzo38: 1990 called and took all their BBSes back.
16:29:11 <zzo38> There is BBS! But now it is internet (by telnet) rather than telephone numbers. Specifically, X-BIT
16:29:55 <zzo38> AnotherTest: LLVM is one
16:33:31 -!- augur has joined.
16:41:50 <mroman> http://codepad.org/zCZaTGxw <- there you go.
16:43:06 <mroman> and more obfuscated...
16:44:15 <mroman> http://codepad.org/oRZa6L54
16:44:19 <mroman> perfectly aligned btw.
16:44:39 <mroman> that's my totaly overengineered solution to
16:44:47 <mroman> if a == 3 then 5 else a
16:45:22 <mroman> I could actually decorate it a little bit.
16:46:13 <zzo38> I meant to improve this program: http://sprunge.us/CFFL (such as by making it do some more things)
16:50:58 <mroman> Nevertheless I'm proud of my solution.
16:52:41 <Phantom_Hoover> For example, it took about a minute for speedtest.net to actually pick a server.
16:59:03 <Taneb> zzo38, I just hand de-obfuscated that, assuming minimal extenstions.
16:59:09 <Taneb> I think I missed the point slightly
17:01:31 <zzo38> Taneb: Post it anyways. And then re-obfuscate it if you want to, and/or add the check for more extensions.
17:01:46 <Taneb> I think I see how it works
17:02:02 -!- calamari has joined.
17:06:57 -!- augur has quit (Remote host closed the connection).
17:07:49 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
17:11:08 <Vorpal> why do plugs which plug into the cigarette lighter connector in cars fit so badly? After all the actual original cigarette lighter fits well into there...
17:17:33 <EgoBot> Score for quintopia_a: 55.5
17:17:37 <fizzie> It's not exactly the best-designedelectric outlet.
17:18:37 <fizzie> I suppose that's what you get with de-facto standards that just "happened".
17:18:49 <fizzie> Do they put USB charging ports in cars yet?
17:18:58 <Vorpal> maybe in high end ones?
17:19:06 <Vorpal> the car I'm driving most time is from 2001 anyway
17:19:41 <Vorpal> Also while the outlet is indeed poorly designed, the plugs are even worse. It is like they were designed to fit different sized outlets or something.
17:19:58 <Vorpal> they have much smaller diameter than the actual lighter for a start
17:20:01 <Vorpal> and springs on the sides
17:20:22 <Vorpal> actually, according to wikipedia there are different sizes!
17:20:28 <fizzie> It's possible they might.
17:20:49 <fizzie> I suppose they want to cover both without making separate plugs.
17:21:11 <EgoBot> Score for quintopia_a: 57.0
17:21:29 <fizzie> "USB standard 5-volt outlets are offered in tandem with cigar lighter outlets in newer vehicles." (From the same article.)
17:21:29 <Vorpal> fizzie, the difference is less than one millimeter though
17:22:05 <Vorpal> but I guess then you need springs to cover both, and that leads to a poor fit for any size
17:26:53 -!- Phantom_Hoover has joined.
17:27:48 <Phantom_Hoover> Moreover, my connection is over three times faster running Linux on the same computer.
17:40:20 <Taneb> My internet is faster on linux, but the connection has larger range on Windows
17:41:49 <Taneb> True, don't know why
17:49:26 -!- calamari has quit (Quit: Leaving).
17:51:39 -!- pikhq_ has joined.
17:52:18 -!- pikhq has quit (Ping timeout: 264 seconds).
17:53:10 <Taneb> I've made it slightly artier
17:53:10 <Taneb> http://hpaste.org/71206
17:53:22 <Taneb> (tip your monitor so the left side is up)
17:53:45 <Taneb> (and set it to white-on-black)
17:54:21 -!- zzo38 has quit (Remote host closed the connection).
17:54:36 -!- calamari has joined.
17:54:43 -!- boily has joined.
18:14:18 <fizzie> Regarding raneg, that could physically speaking be different transmission power (or other related) settings in the wireless drivers for Linux/Windows.
18:14:30 <fizzie> Taneb's connection raneg.
18:15:08 <Phantom_Hoover> Especially when the Windows connection was fine up until about a week ago?
18:15:22 <Vorpal> <Phantom_Hoover> Moreover, my connection is over three times faster running Linux on the same computer. <-- how many times did you switch back and forth and check? Maybe it was just a temporary issue when you ran whatever the other OS was
18:16:03 <Vorpal> <Taneb> My internet is faster on linux, but the connection has larger range on Windows <-- what do you mean with range?
18:16:12 <fizzie> Vorpal: You know, wireless?
18:16:28 <fizzie> At least that's what I assume.
18:16:39 <Phantom_Hoover> Well I've also noticed regularly having latency that's normally over 100ms and regularly over 300.
18:16:47 <Vorpal> fizzie, so he isn't talking about reaching say the US on linux, but only being able to visit websites from Japan while using windows?
18:17:03 <fizzie> Vorpal: Well, you never know. But it wouldn't be my first guess.
18:17:06 <Vorpal> (because that would have been insane)
18:17:21 <Phantom_Hoover> I was playing TF2 quite regularly when it started, too.
18:17:35 <fizzie> I've mentioned my old-old dialup from the company who sold also a "Finland-only Internet" connection with a cheaper per-minute price.
18:17:48 <Vorpal> Phantom_Hoover, indeed, he said it had better range on windows though
18:17:52 <Vorpal> but was faster on linux
18:18:49 <Vorpal> Phantom_Hoover, anyway /your/ internet issues, are they on the LAN/WLAN side or the WAN side?
18:19:27 <Vorpal> I personally use ethernet due to how bad most wifi routers are.
18:19:43 <Vorpal> I think to actually cover this entire house I would need two wifi repeaters
18:19:55 <nortti> I use ethernet because I don't have wifi card
18:20:04 <Vorpal> well on my laptop I meant
18:20:13 <Vorpal> on my desktop I use ethernet for the same reason as nortti
18:20:22 <nortti> I am also speaking of laptop'
18:20:28 <Vorpal> speaking of which, are wifi repeaters passive devices or do they (and possibly the router) need to be configured in some way?
18:21:35 <nortti> if I had working wifi card I wouldn't use 30 meters long ethernet cables
18:22:12 <Vorpal> if I had working wifi coverage of this room I would still use ethernet, because ethernet is faster when transferring files between the computers
18:22:14 <fizzie> At least the repeaters I know of need to be configured, at the very least with the specifics of which network they need to repeat.
18:22:29 <fizzie> Also, the wifi in this hotel is very very spotty.
18:22:40 <Vorpal> fizzie, right, but does it need any sort of support from the access point it is repeating?
18:23:24 <fizzie> I vaguely recall that perhaps it does not, if it appears as a client to the AP. But I could be completely wrong. Certainly there was something manufacturer-specific going on in there.
18:23:38 <fizzie> Haven't set up more complicated wifi topologies than a single AP ever.
18:24:09 <Vorpal> speaking of signal coverage, the GSM signal is extremely bad in parts of the living room.
18:24:24 <Vorpal> perfectly fine in the rest of the house
18:24:37 <fizzie> Anyway; there's also a single wired Ethernet hole in this room, so we're just using an ad-hoc wlan between the two laptops (that sit about 30 cm from each other) to share it. Works fine-ish.
18:25:51 <fizzie> Didn't have any cables to wire these two computers together. Except for one Ethernet cable, but both laptops have just one port, so that'd be kinda counterproductive.
18:26:23 <fizzie> And I guess I could run some sort of custom softmodem thing over the audio-in/out ports.
18:27:41 <Vorpal> fizzie, at least there is something in the socket. I have seen a setup where a wall ethernet socket was seemingly connected (the status LEDs on the port of the laptop connected to it lit up) but it could not be used to establish a connection. So far nothing too strange. However, the strange bit was (when using WLAN) the laptop only responded to ping when it was plugged into the ethernet, even though tha
18:27:41 <Vorpal> t response was on the wifi IP...
18:27:55 <Vorpal> (the laptop was running windows 7 btw)
18:28:26 <Vorpal> fizzie, you have dual-ended audio cables!?
18:28:40 <Vorpal> why would you carry those around
18:28:40 -!- ogrom has quit (Quit: Left).
18:28:44 <olsner> cross-over headset cables
18:28:46 <Vorpal> I mean, how often do you need those
18:29:03 <Vorpal> (to connect line-out on a tape player to line-in on a PC)
18:30:04 <fizzie> Vorpal: Sometimes there's a 3.5mm stereo in-plug in hotel room sound systems, to plug a personal audio player to; and there's a 3.5mm hole in the N900. So I have one cable to do that, just in case.
18:30:36 <fizzie> That'd be unidirectional, though.
18:30:53 <Vorpal> olsner, any idea what "wifi repeater" is in Swedish, trying to find what those costs, but unable to find any in any of the webshops I searched
18:31:06 <olsner> Vorpal: I'd call it wifi-repeater
18:31:12 <Vorpal> well, no luck with that
18:31:42 <Vorpal> wifi bridge? Hm that would be at a higher level in the protocol stack wouldn't it?
18:31:59 <fizzie> I also have one cable from Nokia's four-pin 3.5mm A/V connector (it's compatible with regular 3.5mm audio, but also has composite video in it) into 3x male RCA, and a 2x female RCA -> 3.5mm stereo adapter, I think that might have a chance of working as well.
18:32:41 <Vorpal> hm wifi range extenders listed under "bryggor". What a stupid translation
18:33:18 <Vorpal> oh come on, they are as expensive as access points?
18:33:23 <Vorpal> how does that make any sense
18:33:32 <fizzie> They'd have much the same hardware, presumably.
18:33:37 <olsner> because they are access points?
18:33:55 <fizzie> Incidentally, the wired Ethernet (and probably the wifi too) here NATs, so the second laptop is now in the enviable position of being behind a double NAT.
18:34:02 <Vorpal> then why not just sell them as dual function devices?
18:34:16 <fizzie> APs often do have repeater modes, actually.
18:34:39 <Vorpal> well I mostly dealt with combined ADSL-modem/switch/AP devices
18:34:42 <fizzie> At least my Linksys WAP-54G had like five operating modes, of which more than one had to do with some sort of a repeater thing.
18:34:50 <Vorpal> and those of course lacks such features
18:35:07 <fizzie> We had pure Ethernet coming out of the wall back when we got the AP, so I got a single-function device.
18:36:01 <Phantom_Hoover> <Vorpal> Phantom_Hoover, anyway /your/ internet issues, are they on the LAN/WLAN side or the WAN side?
18:36:36 <Phantom_Hoover> Well on a solid basis of uninformed conjecture I'd say it's probably on the LAN side.
18:36:47 -!- AnotherTest has quit (Quit: Leaving.).
18:36:57 <Vorpal> Phantom_Hoover, are you using ethernet, wifi, appletalk or token ring? Or possibly something else?
18:37:19 <fizzie> If token ring, check if the token has fallen out accidentally.
18:37:36 <Phantom_Hoover> Vorpal are you trying to make yourself look clever by naming as many LAN setups as you can.
18:37:39 <Vorpal> fizzie, could that actually happen?
18:37:39 <fizzie> I think that's in the BOFH excuse calendar.
18:37:42 <olsner> there should be a faint buzzing noise in the cable as the tokens swooshes past
18:37:51 <Vorpal> Phantom_Hoover, no, I'm just sleep deprived
18:38:39 <Vorpal> fizzie, hm I guess token ring would have had some sort of time out to detect if a participating device malfunctioned and didn't pass the token on?
18:38:47 <fizzie> Vorpal: I suppose it depends on how vaguely you are willing to interpret it. I mean, I suspect there could be a failure mode where one of the devices would repeatedly "drop" the token.
18:39:12 <Phantom_Hoover> Ethernet connected to a WiFi extender connected to a shitty router in the next room connected through ethernet to a modem connecting to a mysterious coax cable coming out of the floor.
18:39:43 <Vorpal> Phantom_Hoover, the modem doesn't do NAT?
18:39:49 <olsner> the way I think it works is that every NIC in the ring needs to actively forward information, so a broken device can kill the whole ring
18:40:22 <Vorpal> Phantom_Hoover, network address translation
18:40:36 <olsner> though I'm not sure ... I guess serious token rings would have several rings for backup
18:40:36 <Vorpal> Phantom_Hoover, the bit that turns a public IP into a non-public one
18:40:51 <Vorpal> Phantom_Hoover, like you have 192.168.0.1 internally and something else on the outside
18:41:03 <Vorpal> and multiple computers can share that external address
18:41:11 <Vorpal> Phantom_Hoover, yes, or 10.x.x.x or another range I forgot
18:41:25 <Phantom_Hoover> I doubt it, considering it has one ethernet port so it'd be kind of redundant.
18:41:54 <Vorpal> olsner, debugging token ring sounds a bit like fault searching a serial Christmas tree light loop thingy
18:42:00 <Vorpal> probably easier though
18:42:13 <fizzie> olsner: There's some sort of a protocol that you use to join the ring, I know that much; and one of the stations is the active monitor. I'm not sure, but I wouldn't be surprised if it were possible to de-insert a misbehaving node from the ring. (It's not physically a ring topology, after all.)
18:42:51 <Vorpal> oh? that is kind of boring
18:43:07 <Vorpal> and why would you do a ring topology if it wasn't a physical ring
18:43:20 <olsner> IIRC the connectors were made so that when you disconnect a node you automatically reconnect the previous and next nodes to the ring
18:43:25 <fizzie> It was presumably easier to do access control like that, with the token thing.
18:44:07 <olsner> from before the ethernet hippies figured out you could just let collisions happen and recover afterwards
18:44:08 <Phantom_Hoover> Vorpal, it's obviously going to be easier because messages will still get transmitted up until the broken terminal.
18:44:16 <Vorpal> the continually circulating token sounds incredibly energy inefficient btw
18:44:33 <Phantom_Hoover> With lights in serial there's no current in any of the bulbs.
18:45:43 <Vorpal> olsner, and these days we don't let collision happen any more, instead we use full-duplex connections to what is essentially a specialised computer (the switch)
18:45:48 -!- oerjan has joined.
18:46:33 <fizzie> Good old thinwire-Ethernet (10BASE2) is probably a worse thing to debug than a Token ring network, since a break anywhere will crap the whole segment.
18:46:44 <Vorpal> so today the whole collision detection stuff on ethernet is no longer needed.
18:47:03 <Vorpal> can gbit ethernet even run on a hub?
18:47:16 <Vorpal> it would kind of surprise me if it could
18:47:34 <olsner> oh, there was actually a gigabit token ring standard
18:47:36 <Vorpal> fizzie, was thinwire the one with the spikes?
18:47:37 <fizzie> We used to have thinwire network of ~20 nodes in a computer classroom at school. The cables ran across corridors under some sort of covers.
18:48:12 <fizzie> No, it's the one with the T connectors. Thickwire (10BASE5) is the one with the vampire connectors.
18:49:15 <fizzie> My "LAN" from those days also ran thinwire Ethernet using discarded networking equipment from said school, IIRC.
18:49:31 <fizzie> But that one had just two (maybe three?) computers, so isolating faults was kind of easier.
18:49:59 <Vorpal> fizzie, did faults happen though?
18:50:10 <Vorpal> not very often I presume?
18:50:17 <fizzie> I don't remember. Not often, no.
18:50:33 <fizzie> Certainly not as often as at school, where people kept trampling over the cables.
18:50:55 <fizzie> I guess in 1997 or so.
18:51:09 <fizzie> Going purely on when I was at said school.
18:51:11 <Vorpal> I can't remember ever having a problem with modern cat5e cables btw. Not with the cables themselves that is.
18:51:30 <fizzie> They went twisted-pair when they renovated the computer rooms.
18:51:55 <olsner> Vorpal: it used to be possible to confuse crossover and normal cables though
18:52:19 <Vorpal> olsner, hm I connected normal cables between computers with no issues
18:52:22 <nortti> our school still has some thinnet networks
18:52:50 <Vorpal> olsner, I did that recently to do internet sharing to a desktop for example
18:52:57 <fizzie> It's not until 1000Base-T that I think auto-MDIX became standard.
18:52:59 <Vorpal> I guess modern NICs are smarter?
18:53:05 <olsner> Vorpal: well, that's what's been recently fixed, NICs with media detection that automatically detects what cable you used and what you should've used
18:53:11 <fizzie> Vorpal: The thing that lets you use whichever type of cable.
18:53:13 <nortti> but they also still have 5" floppy disk drives so it isn't a surprise
18:53:51 <nortti> olsner: oh. and they are now moving from betamax
18:53:59 <Vorpal> olsner, the desktop was actually pretty old. Had 2 SCSI drivers in it. Actually an old server. Pentium 3 thingy. And it had 100 mbit networking.
18:54:08 <olsner> I don't think I've ever even seen non-TP ethernet
18:54:14 <Vorpal> but I guess the laptop could have handled the cable detection
18:54:17 <fizzie> Vorpal: It's enough for one end of the link to be new.
18:54:21 <Vorpal> it definitely has gbit ethernet
18:54:35 <fizzie> (And some 100Base-T cards can of course do it too, it's just not as widespread.)
18:55:07 <Vorpal> <olsner> I don't think I've ever even seen non-TP ethernet <-- I have, though not while in use. But I have seen a vampire spike in real life.
18:55:10 <fizzie> I'm not sure if it's mandatory in gigabit Ethernet or not, but at least it's very common.
18:55:43 <Vorpal> I think it was in some sort of historical display of technology in an internal window at university or something.
18:55:48 <fizzie> "hen two auto-MDIX ports are connected together, which is normal for modern products, the algorithm resolution time is typically < 500 ms. However, a ~1.4 second asynchronous timer is used to resolve the extremely rare case (with a probability of less than 1 in 1021) of a loop where each end keeps switching."
18:56:17 -!- pikhq has joined.
18:56:18 -!- pikhq_ has quit (Ping timeout: 250 seconds).
18:56:38 <olsner> would've been somewhat ironic if they made it so that MDIX couldn't reliably negotiate with MDIX
18:56:52 <Vorpal> fizzie, and couldn't that 1.4 second timer also end up triggering at the same time?
18:57:45 <fizzie> There's a PRNG involved for the "regular" way, that's where the 10^21 comes from I think.
18:59:13 <olsner> oh, 10^21, I thought it was 1024 minus 3 for some reason
18:59:14 <fizzie> The ~1.4 second timer resets the PRNG or something.
19:00:28 <fizzie> I suppose it's considered just completely unlikely for both ends to end up with the same random seed every time, the reset is just there in case it happens once.
19:02:08 <fizzie> I have had some autonegotiation issues with modern-ish Ethernet, though.
19:02:21 <olsner> hmm, 500ms is long enough to be noticable
19:02:26 <fizzie> Some link somewhere only worked with autonegotiation disabled.
19:03:23 <fizzie> olsner: The PRNG is clocked at 55ms, and the probability for a matching run of length 2 is around 0.1, so in 90% of cases it's about 100ms.
19:03:58 <nortti> olsner: also we still sometimes use reel-to-reel tape recorders and slide projectors. just to give you idea how moredn our school is :P
19:04:02 <fizzie> (Obviously, with a run of length 0 it's "instant", and for length 1 it'd be that 55ms.)
19:04:19 <fizzie> I've seen an overhead projector used at Aalto too.
19:04:30 <olsner> 55ms is 7MB at gigabit speeds
19:04:59 <fizzie> olsner: Sure, but consider it in the speeds of your grubby fingers plugging that cable in.
19:05:55 <Vorpal> also you can't start a transfer the instant you plug it in
19:06:08 <nortti> fizzie: those where you have to put see through sheet on the glass surface and then it projects the image?
19:06:27 <Vorpal> when I connect this laptop it takes considerably more than a second before DHCP and all that completed
19:06:28 <nortti> we use them all the time
19:06:33 <olsner> they're pretty neat - remove the mirror on top and you have an optical burning device
19:06:39 <Vorpal> olsner, so that auto negotiation is not really significant
19:06:41 <fizzie> It's got a Fresnel lens in it.
19:07:30 <fizzie> And the light output can set things on fire, that's true. Or at least make things start smoking.
19:07:49 <Vorpal> fizzie, wouldn't that make the image quality suffer somewhat? Unless it is between the lamp and the image
19:08:04 <Taneb> "That light gave me my first cigarette, and it's been downhill from there"
19:08:47 <Taneb> <fizzie> ... Or at least make things start smoking.
19:09:05 <fizzie> Vorpal: It is. The lamp is in the box, the lens forms the top of the box the slide sits on top of that, and then there's a lens-mirror combination in an arm to direct the image to the wall.
19:09:41 <Vorpal> Taneb, yes but what did it have to do with cigarettes?
19:09:57 <fizzie> The lamps fail every so often, too. And they have noisy fans.
19:10:27 <Vorpal> Taneb, yes but in the context it isn't that type of smoking. That joke was too far fetched.
19:10:39 <Taneb> THAT JOKE WAS NOT FAR-FETCHED ENOUGH
19:10:51 <olsner> Vorpal: har du aldrig sett en OH-projektor?
19:10:57 <Vorpal> fizzie, so pretty much the same issues as PC projectors have
19:11:01 <Vorpal> olsner, sure I have, what about them
19:11:10 <olsner> those are the ones we were talking about
19:11:30 <Vorpal> oh with "overhead" I imagined he meant something mounted overhead
19:11:51 <Vorpal> didn't think about what OH meant
19:12:06 <Vorpal> I never seen an OH projector with fans in it though
19:12:28 <fizzie> Pretty much all I've seen have had large fans to stop the lamp from melting.
19:12:40 <Vorpal> well, maybe the fans were just very quiet?
19:12:57 <nortti> http://ubuntusatanic.org/news/
19:12:57 <fizzie> They're generally large, so I suppose they can be quiet if the manufacturer cares.
19:13:11 <Vorpal> fizzie, also for the lamp thingy, the ones I have seen have a slider on the side that allow you to switch to a different lamp if the primary one dies.
19:13:21 <Vorpal> so you can continue the lecture
19:13:31 <fizzie> That sounds so high-end.
19:13:49 <Vorpal> also every single room at the university I attended had an OH projector
19:13:59 <Vorpal> well, every single lecture room or such
19:14:04 <Vorpal> (not the toilets obviously)
19:14:20 <ion> Your toilets didn’t have OH projectors? :-(
19:14:31 <fizzie> I don't think our lecture rooms all have them any more, they've been somewhat replaced by those document-camera dealies.
19:14:34 <fizzie> Many still do, though.
19:14:46 <Sgeo> Does Red Dwarf's continuity stay consistent?
19:14:47 <fizzie> Even if it's somewhere in a corner.
19:15:07 <Sgeo> e.g. when early episodes have things such as Lister getting married in the future, will it happen in the show?
19:15:21 <Vorpal> fizzie, I only seen those in two places. One portable that a specific teacher used, and one in the 600-person lecture auditorium
19:15:22 <Sgeo> </season-2-spoiler>
19:15:34 <olsner> Sgeo: I think it's much more important for them that things are funny than that they make sense
19:15:44 <Vorpal> <fizzie> Even if it's somewhere in a corner. <-- well yes
19:15:58 <Vorpal> most teachers seem to use portable projectors
19:16:02 <Vorpal> and those have loud fans
19:16:38 <fizzie> Vorpal: Ooooh, I remember the *best thing*. Our maths teacher in grades 7-12 (numbered linearly from start of school) had this TI-86 add-on device, it had a translucent LCD screen that you put on top of a regular overhead projector.
19:16:44 <fizzie> Then it showed the calculator screen on it.
19:16:50 <fizzie> It was the fanciest thing in the world.
19:17:09 <Vorpal> 7-12 (numbered linearly from start of school) <-- how else would you number them?
19:17:25 <olsner> some laptops have/had displays where you could remove the back of them and use them on an OH projector
19:17:37 <fizzie> They're actually 7-9 and then high-school 1-3.
19:17:38 <Vorpal> actually, we have 1-9 then three years in the gymnasium, then whatever years at university
19:18:07 <Vorpal> fizzie, you had the same teacher in primary school and high school?
19:18:10 <fizzie> We have a very similar system, but in general the school systems differ so much, I thought a linear numbering would be easier.
19:19:01 <fizzie> The school did both 7-9 and the high-school 1-3; quite many here do.
19:19:13 <fizzie> The 1-6 and 7-9 parts are generally in different schools.
19:19:39 <nortti> our school is 1-9 and 10
19:19:50 <fizzie> Not all teachers were involved in both, but they shared some infrastructure.
19:20:17 <fizzie> Also, we had this math-oriented classes-7-9 thing where we did some early high school math courses early, to get a bit of a head start.
19:20:26 <Vorpal> fizzie, I think I did 1-6 in one school, then 7-9 in another, then 1-3 of gymnasium at yet another place
19:20:26 <fizzie> So we had the same maths teacher all the way through.
19:20:30 <Vorpal> and then university of course
19:21:02 <oerjan> 16:20:39: <Taneb> I'm unfamiliar with the default keyword?
19:21:02 <oerjan> 16:20:56: <Sgeo> GHC extension I think
19:21:03 <Vorpal> actually I did 1-2 in one place but it was technically the same school, just across the road from the 3-6 one
19:21:04 <fizzie> That's also quite often the case. And I do think I had to apply for the high school bit separately. They just are in the same building, and some teachers tech in bth.
19:21:22 <fizzie> Gah, I'm drpping lttrs.
19:21:28 <oerjan> default is standard, but ghc has an extension ExtendedDefaultRules or something which makes it more flexible.
19:21:48 <Taneb> oerjan, how does it work?
19:22:04 <fizzie> Anyway, this is all beside the point of what a fancy TI-86 add-on device. It required a custom-modified calculator with a bulge where the screen cable connected to.
19:23:24 <fizzie> http://education.ti.com/educationportal/sites/US/productDetail/us_viewscreen_panel.html
19:23:48 <fizzie> Well, that's for the TI-Nspire.
19:23:55 <fizzie> But it looked rather similar.
19:25:11 <oerjan> Taneb: e.g. default (Integer, Double) makes haskell choose the first of Integer and Double which fits whenever numeric types need to be defaulted. (that's the default default, btw :P). defaulting can happen either in cases like x^2, where there is absolutely nothing to tell what type 2 should be other than an Integral, or when the monomorphism restriction hits like with x = 2 as a declaration at top level.
19:27:47 <oerjan> Taneb: the standard only allows defaulting to happen when all typeclasses involved are in the haskell report and at least one is numeric. (although i've previously found ghc doesn't include e.g. Random). ghc's extension relaxes this to allow any single type (kind * i think, so cannot be used to choose a Monad) type classes which don't need to be standard ones.
19:28:39 <oerjan> iirc ghci by default runs with the extension and default ((), Integer, Double), which means you get () as the default for e.g. Show with no numeric content
19:29:50 <oerjan> i should check that Monad thing, because i briefly had an idea for your obfuscation which might work if that isn't the case
19:31:14 <oerjan> although more likely something like default ((()->())->()->())->(()->())->()->()
19:32:33 <oerjan> http://www.haskell.org/ghc/docs/latest/html/users_guide/interactive-evaluation.html#extended-default-rules
19:32:40 <Taneb> On a different note, is it bad that I think the Env comonad makes more sense than the Reader monad?
19:48:12 <oerjan> i have never used a comonad for real, i think
19:49:48 <oerjan> oh wait ghc doesn't actually drop the numeric requirement entirely :( it just adds Show, Eq, and Ord.
19:52:01 <Taneb> I just have a vague understanding of Monads as adding context and Comonads as using context
19:52:27 <oerjan> close enough, now go write a tutorial!
19:52:43 <Taneb> Aww man, I suck at writing tutorials
19:53:01 <oerjan> ...as do most people who have written monad tutorials.
19:53:18 <lambdabot> Lemmih says: inv2004: Haskell isn't like all the other mainstream languages. You really need to read a tutorial.
19:53:25 <lambdabot> djahandarie says: I think there should be a new internet rule.... "if it exists, there is a monad tutorial using it as an analogy"
19:53:28 <Taneb> True, so I'd fit right in
19:53:35 <lambdabot> arw says: ...and a basic law of haskell is, 50% of all documentation has to be monad tutorials :)
19:53:40 <lambdabot> lispy says: monad tutorial : haskell :: conflictor representation : darcs hacking
19:53:58 <oerjan> wtf is a conflictor representation
19:54:01 <lambdabot> myname says: i prefer monat tutorials like buttiros, delicious if it comes in, but what's left is just poo
19:54:18 <lambdabot> djahandarie says: I think there should be a new internet rule.... "if it exists, there is a monad tutorial using it as an analogy"
19:56:40 <Sgeo> I am addicted to Red Dwarf
19:57:52 <Taneb> I believe it, in this case, is a TV Show which Phantom_Hoover mentioned yesterday
19:58:05 -!- zzo38 has joined.
19:59:50 <Phantom_Hoover> You'll be into the long, painful withdrawal in no time at all.
20:00:06 -!- asiekierka has quit (Remote host closed the connection).
20:00:08 <Phantom_Hoover> Especially if you don't watch anything past the second episode of the 7th series because it's all shit.
20:02:17 <oerjan> yeah, that show's got no kind of atmosphere
20:05:56 -!- kwertii has joined.
20:11:28 -!- sclv has quit (Read error: Connection reset by peer).
20:17:24 -!- Taneb has quit (Quit: Leaving).
20:17:33 -!- zzo38 has quit (Remote host closed the connection).
20:18:43 <pikhq> Phantom_Hoover: It's easy to stave off for a bit.
20:18:54 <pikhq> By watching more British TV.
20:19:03 <pikhq> Unfortunately, this only *delays* the problem.
20:19:12 <Phantom_Hoover> Sgeo, have you watched Spaced, Spaced is even harder to cope with because there are only 14 episodes and coming down again renders you unable to laugh at lesser comedy.
20:19:48 <Vorpal> Phantom_Hoover, so recommend that he doesn't watch it then?
20:20:11 <Phantom_Hoover> pikhq, dunno what you're watching, but sitcoms are so short you can easily burn through the entire run in a day or two if you're binging.
20:20:21 <Vorpal> pikhq, if there is eventually enough British TV it can delay it for the rest of your life?
20:20:53 <pikhq> Phantom_Hoover: Yes, but you can move on to other sitcoms. This delays the problem for a few days.
20:21:01 <Vorpal> Phantom_Hoover, indeed, I'm not saying it is currently possible, TV as a medium hasn't been around long enough
20:21:09 <pikhq> Alas, you quickly come across the problem that 99% of everything is shit.
20:21:14 <Vorpal> but your great-great-grandchildren might be able to do so
20:21:38 <Phantom_Hoover> Your average sitcom has maybe 9 hours of footage total, more for real long-runners.
20:21:53 <Vorpal> okay, great-great-great-great-great-great-grandchildren then
20:22:50 <pikhq> I'd imagine there's enough British TV you could spend a fairly large portion of your life watching all of it. However, there's not enough British TV *worth watching* for that.
20:23:36 <Phantom_Hoover> You could get a lot of mileage out of classic Doctor Who, but it's a) often bad and b) often lost forever.
20:24:31 <pikhq> Kinda wish there was just a list of decent classic Doctor Who...
20:24:50 <pikhq> Perhaps it'd be short enough you could watch all of it without being highly obsessed.
20:29:14 <Phantom_Hoover> Well, the BBC does have that comprehensive episode guide with rough commentary.
20:30:26 <Phantom_Hoover> (It's here if you want to look: http://www.bbc.co.uk/doctorwho/classic/episodeguide/ )
20:31:33 <Phantom_Hoover> Oh wow they still have the same early 2000s media player and everything,
20:33:56 <Phantom_Hoover> (Note especially the bottom line and analysis sections in the "In Detail" page.)
20:35:25 -!- calamari has left ("Leaving").
20:39:43 <Sgeo> Phantom_Hoover, I've heard of Red Dwarf before yesterday
20:39:52 <Sgeo> Haven't heard of Spaced until you mentioned it earlier
20:40:33 <Sgeo> Heard of Shaun of the Dead. I think because of the Basic Instructions strip
20:41:03 <Sgeo> Now starting Season 3 of Red Dwarf
20:41:36 <olsner> I liked Hyperdrive because it features Miranda Hart: http://akas.imdb.com/title/tt0481449/
20:41:42 <Phantom_Hoover> Well if you'd seen either I'd say it's that but in sitcom form and also even better, but otherwise it's really hard to describe.
20:42:45 <Phantom_Hoover> Because TbH I could just watch Miranda, considering it has a higher Hart:everyone else ratio, and also it has that lady from Mitchell and Webb.
20:43:09 <olsner> well, Miranda is also better
20:43:27 <Sgeo> Phantom_Hoover, is Red Dwarf still ongoing, the Facebook page said something about "Series X coming later this year!"
20:44:37 <Phantom_Hoover> It was cancelled after going to shit, but I think Dave bought the rights and are trying to resurrect it.
20:45:01 <Phantom_Hoover> They did a 3-part miniseries in 2009 which I haven't seen, although the one clip I have watched was funny
20:48:03 <Phantom_Hoover> tell me sgeo did you have a humorous misunderstanding at the use of 'dave' there
20:48:51 <Sgeo> Of course I thought the character, but I assume an actual... person
20:48:58 <Sgeo> *thought the character at first
20:49:06 -!- tswett has quit (Changing host).
20:49:06 -!- tswett has joined.
20:50:14 <Phantom_Hoover> that's funny because i was actually talking about the tv channel
20:51:10 <Sgeo> I did not know there was a TV channel called Dave.
20:53:12 -!- boily has quit (Quit: WeeChat 0.3.8).
20:54:52 <Sgeo> What's with the different H?
20:56:43 <Phantom_Hoover> Are you at series 3? They changed a lot of stuff round for series 3.
20:57:13 <Phantom_Hoover> They changed everything again for 6, and again during 8 (or so I hear).
20:57:46 <Vorpal> what does that stand for
20:57:52 <Sgeo> Quite Interesting
20:58:24 <Phantom_Hoover> What it stands for is irrelevant; it's a comedy quiz show.
20:58:41 <olsner> you could've googled using just QI, you know
21:00:55 <Vorpal> olsner, that gave me stuff about taoism in all the top ten hits
21:01:18 <Vorpal> actually wait. the tenth is for some sort of software thingy
21:01:53 <olsner> maybe you used swedish derp-google?
21:02:31 <Vorpal> says google.com and everything is in English
21:03:06 <olsner> for me, hits related to the tv show are number 1, 2, 5, 6 plus three video hits
21:03:07 <Vorpal> I am logged into gmail, but since I don't usually google taoism or religion I'm not sure why it would customise it like that
21:03:32 <olsner> based on context you should be able to figure out that it was a tv show, and there are no other tv shows with that name afaik
21:04:08 <olsner> otoh, maybe google customized the hits for me, since I happen to be logged in on gmail
21:05:13 <oerjan> i have no google account and i got the tv show on norwegian google
21:08:34 <Phantom_Hoover> Google doesn't work that way, and I'm astonished you don't know that.
21:11:40 <oerjan> stormåse, gråmåse, svartbak...
21:12:04 <oerjan> oops, stormåse = svartbak
21:12:41 <olsner> a bird is a bird is a bird
21:12:53 <oerjan> wait, actually stormåse can mean _both_ the others
21:17:22 <Phantom_Hoover> I'm not even sure what the seagulls common in Edinburgh are.
21:17:33 * oerjan looks at the swedish wikipedia and wonders what the difference between mås and trut is...
21:17:42 <olsner> oerjan: they're both birds
21:18:04 <olsner> trut is also slang for mouth
21:18:05 <Phantom_Hoover> The closest I can find on Wikipedia is the herring gull, but ISTR hearing or reading that it's closely related but not the same.
21:18:15 <oerjan> olsner: i said _difference_. they're both må[sk]e in norwegian
21:18:54 <olsner> why not move all bird questions to #esoteric-birds? :)
21:19:18 <Phantom_Hoover> (One of the comedians in the Fringe once joked that Edinburgh seagulls are abnormally big and I've been trying to confirm this ever since.)
21:19:55 <Vorpal> olsner, what about fiskmås?
21:20:40 <olsner> mostly it's that I don't know anything about birds :)
21:20:43 <Vorpal> hm that is "common gull"
21:21:23 <Vorpal> and common gull is apparently also a name for a type of butterfly.
21:22:15 <oerjan> "Arterna inom denna kategori anlägger adult dräkt efter drygt tre år efter det att dom kläcks, dvs på deras fjärde levnadsår. Inom denna kategori finns alla trutarna."
21:23:09 <Vorpal> oerjan, hm I think that must be a typo
21:23:16 <Vorpal> because that makes no sense
21:23:42 <olsner> is that from the swedish wikipedia?
21:23:49 <oerjan> http://sv.wikipedia.org/wiki/M%C3%A5sar_och_trutar
21:24:50 <Vorpal> oerjan, actually it is accurate, första året : 0-1, andra året, 1-2, ..., fjärde året: 3-4
21:25:03 <oerjan> it's the only mention i can find distinguishing them... they're very mixed up in the genus classification :P
21:25:40 <olsner> oh, apparently adult is an actual ornithological term, and not just a very bad translation from english
21:25:54 <oerjan> oh wait "Generellt kan man säga att de mindre arterna kallas måsar och de större trutar"
21:26:05 <oerjan> olsner: i was wondering about that too :P
21:26:08 <Vorpal> oerjan, with a few exceptions?
21:26:18 <nortti> someone give me programming project. I'm bore
21:26:39 <Vorpal> nortti, implement a JIT compiler for befunge-98 using LLVM before fizzie gets back working on his again
21:27:08 -!- pikhq_ has joined.
21:27:08 <Vorpal> nortti, what about that?
21:27:16 -!- pikhq has quit (Ping timeout: 248 seconds).
21:27:19 <nortti> umh. something not involving LLVM would be nice
21:27:36 <Vorpal> nortti, implement a JIT compiler for befunge-98 using by generating your own machine code before fizzie gets back working on his again
21:28:03 <Vorpal> nortti, does that sound good to you?
21:28:16 <nortti> that sounds better. what architecture?
21:28:30 <Vorpal> nortti, which ones do you have access to?
21:28:41 <Vorpal> nortti, 32-bit x86 I presume, anything else?
21:28:53 <Vorpal> I don't have access to the other ones
21:29:10 <Vorpal> nortti, 32-bit, running under linux
21:29:22 <Vorpal> nortti, as a user space program
21:29:47 <Vorpal> nortti, okay, if it runs with minimal modifications on multiple *nix for 32-bit x86
21:29:49 <nortti> why not as self hosted OS?
21:30:04 <Vorpal> I mean, the differences aren't that large, are they?
21:30:19 <nortti> yes. and netbsd supports linux abi
21:30:46 <Vorpal> nortti, because... hm. Because it is more useful to a larger group of people as an user space program
21:31:04 <Vorpal> I would assume a tracing JIT would be the way to go
21:31:10 <Vorpal> that is what fizzie did at least
21:43:51 <nortti> ok. currently me thinks that compiling 1 row column at a time would be easier
21:47:00 -!- azaq23 has joined.
21:47:08 -!- azaq23 has quit (Max SendQ exceeded).
21:47:48 -!- Vorpal has quit (Ping timeout: 276 seconds).
21:48:12 -!- azaq23 has joined.
21:58:53 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
22:06:37 <Sgeo> Phantom_Hoover, is Holly completely gone?
22:09:52 <Phantom_Hoover> I think he comes back in series 8 but see every other time I've mentioned series 8.
22:11:15 -!- zzo38 has joined.
22:21:56 <Sgeo> "Holograms don't produce heat and neither do androids"
22:22:12 <Sgeo> I know it's roughly the opposite of hard science fiction but that still bothers me
22:22:19 -!- copumpkin has joined.
22:22:56 <nortti> only shell better than busybox ash
22:24:18 <nortti> too much pain to configure
22:26:17 <nortti> well it seemed kinda like emacs. it does multiple things and is pain to configure to work the way you want it to
22:27:34 <nortti> heirloom shell is bit limeted but otherwise nice
22:28:06 <nortti> (I kinda like tab completition)
22:32:09 <nortti> I know that there are people who love zsh but there are also people who love emacs so yeah
22:32:54 <nortti> oh and sash is otherwise nice but lacks tab completition
22:36:36 -!- kallisti has quit (Ping timeout: 248 seconds).
22:39:55 <nortti> sash -aqp $USER@$HOST"$ " makes trying to unfuck your system so much more pleasant
22:40:07 <Sgeo> Phantom_Hoover, accidentally saw a series 8 ending spoiler
22:40:36 <Phantom_Hoover> Sgeo you should have picked up by now that the show has minimal continuity.
22:44:37 -!- kallisti has joined.
22:44:38 -!- kallisti has quit (Changing host).
22:44:38 -!- kallisti has joined.
22:51:12 <ion> Which show?
23:03:08 <zzo38> Is that the show you mean?
23:11:18 -!- pikhq has joined.
23:11:29 -!- pikhq_ has quit (Ping timeout: 244 seconds).
23:39:26 -!- kwertii2 has joined.
23:41:58 -!- kwertii has quit (Ping timeout: 244 seconds).
23:41:58 -!- kwertii2 has changed nick to kwertii.
23:54:31 -!- kwertii has quit (Read error: Connection reset by peer).
23:54:54 -!- kwertii has joined.