←2010-03-10 2010-03-11 2010-03-12→ ↑2010 ↑all
00:00:10 -!- nooga has joined.
00:00:16 <nooga> hell-oh
00:01:50 <alise__> Hell, oh.
00:07:25 <Gregor> Hell? Ohhhh
00:08:58 <nooga> oh hell
00:15:22 -!- dev_squid has joined.
00:15:27 <dev_squid> Hey guys.
00:15:44 <dev_squid> Does anyone know anything about Lambda Calculus syntax?
00:16:28 <fax> yes
00:17:28 <Gregor> What's to know?
00:17:39 <Gregor> http://codu.org/lambda/lambda.html , btw
00:27:15 <uorygl> I know absolutely everything about lambda calculus syntax.
00:28:41 <pikhq> I've implemented it as a library in C.
00:28:46 <pikhq> >:D
00:28:52 <alise__> dev_squid: Yes.
00:30:31 <alise__> dev_squid: An expression is either λx.E where x is an identifier and E is an expression, a b where both a and b are expressions, or an identifier (preferably enclosed in a series of lambdas, one of which has this identifier as part of the head of the lambda-term, otherwise the program makes no sense).
00:30:32 <alise__> What else do you want to know?
00:30:33 <Gregor> There's about three things to know :P
00:30:51 <alise__> Oh, and parenthesisation is allowed, the E in a lambda-expression extends as far as possible, and application is right-associative.
00:31:20 <pikhq> You only need to know a few more things. The reduction rules.
00:31:22 <alise__> And an identifier is, say, any string of alphanumerics (perhaps disallowing numerics for the first character). Maybe even with some number of primes on the end, etc.
00:31:33 <alise__> pikhq: That's not syntax.
00:31:57 <pikhq> Oh, he specified syntax. Darn it.
00:33:21 <alise__> Evaluation of lambda calculus terms: eval _ = _|_.
00:33:26 <Gregor> Like my nifty lambda calculus solver btw? :P
00:33:31 <alise__> Since _|_ -> a, we can show that this is isomorphic to any other evaluation function we might define.
00:33:35 <alise__> Hooray for turing-completeness!
00:33:53 <alise__> Gregor: What syntax does it accept?
00:34:04 <alise__> Also, it doesn't even work in IE8. :-(
00:34:15 <Gregor> Lambda calculus, with the caveat that you type ^ instead of lambda.
00:34:22 <Gregor> And IE8??? WTF?
00:34:26 <Gregor> I doubt it even works on IE6.
00:34:39 <pikhq> Gregor: *Nothing* works on IE6.
00:34:44 <alise__> Er... IE8 has /superior/ support for web standards.
00:34:49 <pikhq> At least, nothing sane.
00:34:52 <alise__> Your "even" is incredibly confusingly misplaced.
00:34:52 -!- oklopol has quit (Ping timeout: 240 seconds).
00:34:57 <alise__> IE8 is actually pretty good for standards.
00:35:06 <pikhq> IE8 is finally a tolerable web browser.
00:35:21 <pikhq> Not great, but it at least *supports* standards sanely.
00:35:28 <Gregor> Oh yeah, I remember writing that note on the solver :P
00:35:53 <pikhq> alise__: The note is conditional on IE usage, not on a version or a feature.
00:35:54 <alise__> I hypothesise that perhaps your program has an error, since my estimation of IE8 being able to run sane, valid JavaScript is pretty high.
00:35:59 <pikhq> Using IE's conditional commenting thing.
00:36:23 <pikhq> <!--[if IE]> ... <![endif]-->
00:36:25 <alise__> I know, but it /literally doesn't work/ in IE8.
00:36:25 <alise__> As in, produces errors.
00:36:30 <pikhq> Oh.
00:36:51 <alise__> Length is null or not an object, line 76, char 17.
00:36:52 <Gregor> I have no IE8 to test on.
00:37:24 <alise__> IT's in iswhite
00:37:24 <alise__> *It's
00:37:43 <alise__> var toks = lex(str);
00:37:53 <alise__> ><input type="button" value="Start" onclick="handleInputPrime(document.getElementById('lambdaInput').value);"/> That should definitely work.
00:38:35 <pikhq> alise__: Uh, "var toks = lex(str);" isn't in iswhite...
00:38:35 <Gregor> Don't looka me.
00:38:48 <Gregor> pikhq: That's where 'str' in iswhite is coming from.
00:38:54 <pikhq> Ah.
00:38:56 <pikhq> Okay.
00:39:12 <nooga> http://vimeo.com/4339116 neat
00:40:47 <alise__> OK, .value works.
00:41:54 <Gregor> And yet, clearly it does not.
00:41:58 <alise__> Gregor: Can you add alert(str) to the start of handleInputPrime?
00:42:01 <alise__> It does, I checked it with alert.
00:42:14 <alise__> So it's getting fucked up between accessing and getting to the parser
00:42:26 <alise__> Does anything before iswhite mutate str?
00:42:27 <Gregor> alise__: YOU can add alert(str) there, just download it :P
00:42:55 <alise__> Too lazy.
00:43:03 <nooga> ffs use debugger instead of calling alert()
00:44:00 * nooga is thinking about booting ruby interpreter on bare metal
00:44:21 <Gregor> Ah yes, a truly worthwhile accomplishment that would be.
00:46:09 <alise__> You could shoot yourself in the foot by reopening Foot and defining its shot? method to always return true, and it'd only take years to evaluate.
00:46:57 <pikhq> Ah, Ruby.
00:47:02 <pikhq> Walking-the-tree semantics.
00:47:14 <nooga> shouldn't be so hard, interpreter depends mainly on libc
00:47:34 <pikhq> Well in that case, just link with a stubbed newlib.
00:47:48 <nooga> that's the plan
00:47:50 <Gregor> Fill in the stubbed bits with raw interfaces as you go.
00:49:07 <alise__> It is truly astonishing.
00:49:14 <nooga> %]
00:49:15 <alise__> Gregor: Stubbed bits? Is that like stubbed toes, but even worse because it's on your genitals?
00:49:15 <alise__> If so, I agree.
00:51:20 <alise__> I suggest as a resolution his immediate termination.
00:51:38 <nooga> wtf
00:55:33 -!- dev_squid has quit (Ping timeout: 260 seconds).
00:59:24 <pikhq> nooga: Oh, there's no point in arguing against it.
01:00:12 <nooga> oh hell
01:01:36 -!- alise has joined.
01:02:14 -!- dev_squid has joined.
01:02:17 -!- fax has quit (Quit: Lost terminal).
01:02:18 -!- alise__ has quit (Ping timeout: 252 seconds).
01:02:43 <alise> http://us.metamath.org/mpegif/avril1.html
01:02:48 <alise> Obviously.
01:03:22 <uorygl> I agree, that URL is very obvious.
01:03:54 <alise> [Theorem at http://us.metamath.org/mpegif/avril1.html.]
01:03:55 <alise> Obviously.
01:05:27 <uorygl> So what does that theorem state?
01:05:45 <uorygl> I don't know what conjoining with a relation means.
01:06:41 <alise> It may be helpful to realise that the whole thing, although completely valid, makes something in the vicinity of 0 units of sense.
01:07:04 <alise> And that "quidquid germanus dictum sit, altum viditur" means "everything in German sounds profound".
01:07:25 <alise> And that the First of April is the First of April.
01:07:38 <nooga> maybe tinyrb would be better choice
01:08:14 <nooga> besides that it uses boehm gc, pcre and other external deps
01:08:46 -!- oklopol has joined.
01:08:51 <uorygl> So is this theorem useful chiefly because it doesn't make much sense?
01:09:13 <alise> Who said it was useful?
01:11:39 <uorygl> I keep saying the wrong words.
01:11:42 <uorygl> s/useful/interesting/
01:14:53 <nooga> who are you two anyway?
01:18:54 -!- adu has joined.
01:19:43 -!- oklopol has quit (Read error: Connection reset by peer).
01:20:42 <uorygl> Who, me and alise?
01:21:12 <uorygl> I am e; alise is ehird.
01:21:12 <nooga> yes
01:21:14 <uorygl> s/e/me/
01:21:48 <alise> You shouldn't have told him that I'm me; maybe he thought /two/ people enjoy disparaging him to the max as a sort of idle hobby.
01:22:54 <nooga> i should check the idents first but soehow i knew that only ehird could use such gay nick while still trying to be poignant :|
01:23:13 <lament> alise is a girl
01:23:48 <nooga> named elliott
01:24:11 <Gregor> Ellietta
01:25:06 <lament> elliotta
01:25:09 <alise> Technically I'm an unnamed gender that is male in every way apart from (a) name and (b) correct pronouns.
01:25:09 <alise> Also, I wasn't attempting poignancy.
01:25:40 <alise> I merely remove all my built-in anti-extreme-silliness filters when talking about you.
01:25:44 <lament> gay is not a gender
01:25:54 <uorygl> Is lesbian a gender?
01:25:55 <alise> OR IS IT
01:26:26 <alise>
01:26:26 <alise> No, but gaylesbian is.
01:29:45 <nooga> alise: you are automatically disparage all my ideas when i share them
01:29:59 <nooga> why?
01:30:06 <nooga> WHY?
01:30:24 <nooga> WHYYYYYYYYYY?!!!
01:30:30 <alise> I need someone to drive to hopeless self-despising, and you talk about how stupid you are all the time so I figured I could do worse.
01:31:17 <nooga> okay
01:31:29 <alise> (And we're calling them ideas now? That's a step ;u
01:31:30 <nooga> sounds fair
01:31:32 <alise> *up.)
01:40:00 <nooga> gasp
01:40:17 <nooga> this won't compile ;[
01:47:25 -!- Asztal has quit (Ping timeout: 260 seconds).
01:54:15 <dev_squid> I haz a quechion.
01:54:59 <pikhq> I'm sure you haz several quechions.
01:55:26 <dev_squid> Naw, juzt 1!11!111! quechion.
01:55:50 <pikhq> I'm sorry, I don't speak stupid.
01:56:47 <nooga> wtf
01:56:57 <pikhq> Could you try speaking smart?
01:57:12 <dev_squid> Is there a name for postfix-notated statements in which the second argument in the notation (last-in argument, if you consider the statement to be a stack) is the first argument in the operation?
01:57:37 <dev_squid> I supposed #esoteric would know just as well as anywhere else.
01:57:56 <pikhq> So, this: 2 2+ ?
01:58:36 <dev_squid> Like... 23* in postfix is 2*3 in infix...
01:59:03 <pikhq> That's... Reverse Polish Notation...
01:59:04 <dev_squid> If you look at the statement as a stack, it'd make sense that it'd be written in postfix notation as 32*.
01:59:23 <dev_squid> Orly?
01:59:44 <pikhq> Unless you're asking for some sort of retarded "RPN but the arguments are flipped" thing.
01:59:52 <dev_squid> Oh, thanks! I just wanted to know if there was a name for it.
01:59:56 <pikhq> *That* is called "DEAR GOD WHY".
02:00:40 <nooga> it's Polish and Reversed, what can be worse
02:00:44 <dev_squid> Lawl.
02:00:57 <pikhq> nooga: RPN is quite a nice notation.
02:00:59 <dev_squid> So 2/3 is 32/ in RPN right?
02:01:21 <pikhq> No.
02:01:23 <nooga> 23/
02:01:34 <pikhq> What you're asking for, then, is "DEAR GOD WHY".
02:01:44 <dev_squid> ...that's just postfix notation.
02:01:47 <pikhq> ... Do you like making your parsing harder or something?
02:02:09 <pikhq> ... AKA reverse polish notation.
02:02:18 <pikhq> What you're asking for is a capital offense.
02:02:30 <nooga> it really depends on your implementation of /
02:02:56 <pikhq> Well. Yeah, you could just do (flip (/)) and voila.
02:03:00 <dev_squid> If you really want to know, it's because I'm making a constructed (non-computer) language based on a LIFO stack where you have binary and ternary operators, which operate on the prefixed arguments in reverse order. :)
02:04:03 <nooga> bike me [rides on]
02:04:11 <nooga> how stupid is that
02:04:25 <dev_squid> In the world of languages, it make sense.
02:04:54 <dev_squid> So "X Y *" would become "Push X, Push Y, Pop A, Pop B, Ret A*B".
02:04:59 <dev_squid> Sort of.
02:05:33 <pikhq> What you are doing makes things harder.
02:05:44 <dev_squid> In terms of language, you're stating the object first, then the subject, then the action.
02:07:51 <dev_squid> So am I looking for...Polish Notation?
02:08:11 <dev_squid> Nevermind.
02:09:12 <dev_squid> In my mind, making the object closest to the operator the "subject" in the calculation (as opposed to the "object") makes more sense.
02:12:18 <nooga> how would you write your last sentence then?
02:13:47 <dev_squid> Hmm?
02:13:50 <dev_squid> Lawl.
02:13:50 <alise> Bikes I ride on; the bikes, interesting. Bikes I ride on, interesting; Bikes, interesting, I ride on.
02:13:52 <alise> English is RPN.
02:14:00 <alise> Well, "poetic" English.
02:14:00 <dev_squid> English is infix.
02:14:09 <alise> No.
02:14:13 <alise> It's multiple fixitiesl
02:14:15 <alise> *fixities.
02:14:22 <dev_squid> It CAN be postfix, but not normally.
02:14:23 <alise> You suck. = postfix
02:14:28 <alise> Suck ass. = prefix
02:14:31 <alise> You suck ass. = infix
02:14:36 <dev_squid> Okay, point taken.
02:14:43 <alise> And some are not really any in particular.
02:14:55 <alise> But "Bikes I ride on" is both valid poetic English and postfix.
02:17:38 -!- alise has quit (Quit: Page closed).
02:18:39 <dev_squid> mine mind of sense calculation "subject" in operator object closest making makes in
02:18:57 <dev_squid> It's actually easier to read that right-to-left.
02:19:11 <dev_squid> That's in RPN where the operands are reversed.
02:19:48 <nooga> please try to synthesize some practical phrase using this rule
02:20:27 * nooga pukes on tinyrb source
02:21:02 <dev_squid> Okay.
02:21:46 -!- cal153 has quit.
02:22:15 <dev_squid> "The white dog barks at the frightened man", ignoring articles, becomes "man frightened dog white barks".
02:22:26 <dev_squid> The adjectives being unary operators.
02:26:22 <dev_squid> And the verbs being binary operators.
02:27:02 -!- cal153 has joined.
02:27:44 -!- GreaseMonkey has joined.
02:29:49 <dev_squid> Besides, poetic English tends to violate the grammatical laws of spoken English.
02:30:37 <dev_squid> But, yeah, "Bikes I ride" is an example of what I'm talking about.
02:32:13 <dev_squid> A postfix notation in which the operands of binary operators are reversed, effectively creating an object-subject-verb structure.
02:32:19 <pikhq> That doesn't violate the grammatical laws of English, it merely uses one that hasn't been common for a few centuries.
02:32:29 <dev_squid> It's archaic then.
02:32:33 <pikhq> Yes.
02:32:46 <pikhq> It still parses fine, though.
02:33:01 <dev_squid> I was considering making a language based on lambda calculus.
02:33:04 <Gregor> An example, give us.
02:33:29 <dev_squid> Or maybe combinatory logic.
02:33:36 -!- augur has joined.
02:34:56 <pikhq> dev_squid: Lojban, then?
02:35:27 <dev_squid> pikhq, I'm not entirely familiar with it, but I know it's hard to to parse mentally. :)
02:35:44 <dev_squid> What's it have to do with lambda calculus.
02:35:49 <dev_squid> ?
02:37:44 <pikhq> Lojban is based on predicate logic.
02:39:33 -!- MizardX has quit (Ping timeout: 240 seconds).
02:40:04 <nooga> Lojban if pretty much f#$&*d up
02:43:39 -!- comex has joined.
02:45:32 <uorygl> I'm pretty sure Lojban has practically nothing to do with lambda calculus.
02:45:48 <uorygl> If you can find a single lambda expression in it anywhere, let me know.
02:46:23 <Sgeo> Where's alise?
02:46:47 -!- comex has changed nick to Network.
02:47:17 -!- Network has changed nick to Guest83798.
02:47:48 -!- Guest83798 has changed nick to comex.
02:47:51 <uorygl> comex is attenuating.
02:47:55 <uorygl> Never mind.
02:48:03 <Quadrescence> Sgeo: I wrote a SCHEME program today
02:48:25 <Sgeo> How difficult would it be for me to make my own Scheme?
02:48:29 <uorygl> You know, as far as I can tell, Lojban is based on predicate logic to about the same extent that English is.
02:49:11 <Gregor> OOOOOH
02:49:11 <Gregor> BURN
02:49:18 <uorygl> English has verbs, which act as predicates; nouns, which act as arguments to them; and adjectives and adverbs, which modify those.
02:49:23 <Sgeo> Quadrescence, to do what?
02:49:33 <uorygl> Lojban has brivla, which act as predicates; brivla, which act as arguments to them; and brivla, which modify those.
02:49:46 <Quadrescence> Sgeo: It pretty prints MATH
02:49:50 <Quadrescence> in ASCII
02:50:26 <uorygl> I suddenly feel an urge to capitalize all languages and language-like things.
02:51:13 <myndzi> lojban, more like loljban
02:51:14 <myndzi> har har
02:51:46 <dev_squid> Hardy har har.
02:52:10 <dev_squid> If you think about it, though, language has a lot to do with mathematical notation.
02:52:23 <comex> that's because mathematical notation is based on language
02:52:24 <myndzi> well sure, it'd be more natural to express math in something familiar
02:52:28 <uorygl> There's a reason we write math the way we do.
02:53:30 <dev_squid> IMHO, English makes very little sense as a language, as compared to other languages with more uniform notation.
02:54:26 <dev_squid> Postfix/prefix notation in math or language makes more sense. It eliminates the need to declare precedence...although declaring precedence may be easier to read.
02:56:00 <Gregor> You don't need to declare precedence to have infix notation so long as you also have parenthesis.
02:56:51 <nooga> about scheme http://vimeo.com/4339116
02:59:57 -!- coppro has quit (Remote host closed the connection).
03:04:12 <uorygl> I think the human brain simply doesn't like parentheses.
03:04:22 <Quadrescence> Um, I love parentheses
03:04:27 <Quadrescence> for lispie lisp lisp
03:04:52 <uorygl> So when you're writing a long expression from beginning to end, do you intuitively know how many parentheses to put at the end?
03:05:32 <Quadrescence> Yes
03:05:42 <Quadrescence> Especially due to how I indent
03:05:58 <uorygl> Do you know how many to put at the end when you don't indent that way?
03:06:05 <uorygl> When you use no line breaks at all?
03:06:25 <Quadrescence> I might have a slight feeling, but not definite.
03:06:43 <uorygl> I (think ((the (human brain)) (simply (doesn't (like parentheses
03:06:59 <Quadrescence> I think the human brain doesn't like clusterfucks of anything
03:07:00 <uorygl> After writing that, I had no idea how many parentheses to put at the end. Six, maybe?
03:07:09 <uorygl> Also, I once had to go back and add a parenthesis.
03:08:47 <Gregor> Oh, the human brain likes /certain/ kinds of clusterfucks ...
03:09:15 <Quadrescence> Gregor: Well yes yes
03:09:18 <Quadrescence> yes of course
03:17:26 <Sgeo> Quadrescence, are there any particular standards for what an event framework in Scheme should look like?
03:17:46 <Quadrescence> no
03:18:32 * Sgeo was thinking, until just a few seconds ago, of having, say, evt-listen take a promise as an argument
03:18:41 <Sgeo> But taking in a function makes more sense, really
03:20:59 <Sgeo> Actually, I can have delay simply be a macro written in pure Scheme, right?
03:21:16 <Quadrescence> yes
03:26:07 <Sgeo> So, my implementation might provide, say, an lle-listen, used like:
03:26:47 <Sgeo> (lle-listen (lambda (chan name id msg) (ll-owner-say msg)))
03:27:13 <Sgeo> What sort of macro would make using lle-listen simpler?
03:27:40 <Sgeo> [Also, did I use lambda correctly?]
03:27:40 -!- zeotrope_ has joined.
03:28:12 <Quadrescence> you used lambda fine
03:28:39 <Quadrescence> And it's hard to say what would make it simpler
03:29:26 -!- jcp has joined.
03:29:39 <uorygl> Aiee, a lambda expression taking arguments that don't appear in the body.
03:29:46 <nooga> i'm a goddamn bat
03:29:55 <nooga> i work @ night and sleep @ day ;[
03:30:03 -!- zeotrope has quit (Ping timeout: 240 seconds).
03:30:05 <Sgeo> But lle-listen tells the implementation to call the function with four arguments
03:31:11 <Sgeo> Also, I'd have to convince users of a crappy C-like language to actually learn Scheme >.>
03:31:27 <Sgeo> And hope that my implementation isn't too pathetically slow
03:31:48 <uorygl> What are you implementing?
03:32:12 <Sgeo> Scheme, in the crappy language known as LSL
03:32:30 <Sgeo> Ah, interesting problem here;
03:32:31 <Sgeo> :
03:32:42 <Sgeo> LSL has builtin types that act somewhat like objects
03:33:00 <Sgeo> A "vector literal" is written as <0.0, 1.0, 2.0> for example
03:33:20 <Sgeo> And if assigned to a variable my_vec, I can use my_vec.x, my_vec.y, my_vec.z
03:33:37 <Sgeo> I need to allow use of vectors in LSL-Scheme, and not have them be confused with Scheme vectors
03:34:17 <Sgeo> Similar with rotations, but once the vector problem is solved, rotations are close enough
03:34:32 <Sgeo> There's also a thing called a key, but they're pretty much simple strings, so
03:37:57 -!- comex has changed nick to toast.
03:38:42 <Sgeo> I have no clue how I'd implement closures
03:39:14 <pikhq> Represent them as a function pointer and a closed-variable pointer, if possible.
03:39:29 <pikhq> If something vaguely similar isn't feasible, good luck.
03:39:32 <Sgeo> No such thing as function pointers in LSL
03:39:44 <Sgeo> For that matter, no such thing as pointers, perioud
03:39:47 <Sgeo> *period
03:39:48 <pikhq> So, you can't reference a function?
03:39:55 <Sgeo> Correct
03:40:11 <Sgeo> Well, statically, you can call them. llSetPayPrice() works
03:40:25 <pikhq> ...
03:40:27 <Sgeo> But you can't store a function name or address or anything like that in a variable
03:40:41 <Sgeo> And there are no multidimensional lists
03:40:44 <pikhq> Good luck.
03:41:20 <Sgeo> Maybe there's a simpler language than Scheme that I can try to implement
03:41:37 <pikhq> Uh. What you are going to do is pretend LSL is an assembly language.
03:42:07 <pikhq> You are going to be using a switch-statement-esque thing to implement functions, with a manual call stack.
03:42:25 <pikhq> And you will implement data structures by treating an array as memory.
03:42:50 <pikhq> And you will do manual garbage collection.
03:42:50 <pikhq> This would be a *royal fucking pain*.
03:43:45 <Sgeo> I may completely omit call/cc
03:44:15 <Sgeo> For what it's worth, LSL arrays are heterogeneous [in a statically typed language]
03:45:34 -!- pikhq has quit (Read error: Connection reset by peer).
03:59:40 <Sgeo> Can let be written in terms of let*?
04:04:52 -!- coppro has joined.
04:08:31 -!- augur has quit (Ping timeout: 265 seconds).
04:26:43 -!- augur has joined.
04:30:41 -!- Gracenotes has joined.
04:37:33 -!- dev_squid has quit (Ping timeout: 240 seconds).
04:46:47 -!- pikhq has joined.
05:07:10 -!- coppro has quit (Read error: Connection reset by peer).
05:11:28 -!- coppro has joined.
06:08:28 <Sgeo> <Sgeo> An interpreter for any language can be written in any TC language (as long as that TC language supports access to external resources that the language being interpreted uses). Is it possible to compile any language to any TC language?
06:08:28 <Sgeo> <Sgeo> o.O trivially yes. It can just be an interpreter + the source in the language
06:08:28 <Sgeo> <Sgeo> That shouldn't count.
06:08:28 <Sgeo> <Anonycat> maybe you should specify "in polynomial time"
06:15:17 <coppro> Sgeo: you mean to rewrite it?
06:15:40 <coppro> yeah, you can do interpreter + source
06:16:11 <coppro> more direct translations are possible, but, dependending on the similarity of the languages, it may be more or less the same thing
06:36:27 -!- adu has quit (Quit: adu).
07:08:13 -!- dev_squid has joined.
07:20:26 -!- dev_squid has quit (Ping timeout: 248 seconds).
07:24:51 -!- tombom has joined.
07:26:24 -!- dev_squid has joined.
07:32:10 -!- jcp has quit (Quit: I will do anything (almost) for a new router.).
07:51:34 -!- tombom has quit (Quit: Leaving).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:06:00 -!- coppro has quit (Remote host closed the connection).
08:20:26 -!- dev_squid has quit (Ping timeout: 246 seconds).
08:23:13 -!- kar8nga has joined.
08:38:58 -!- coppro has joined.
09:00:34 -!- zeotrope has joined.
09:01:56 -!- ais523 has joined.
09:02:37 <ais523> lovers of stupidly named Turing tarpits!
09:02:40 -!- zeotrope_ has quit (Ping timeout: 245 seconds).
09:02:45 <ais523> I think I created a new one, in my head, last night
09:16:14 <fizzie> Misread that as "lovers of stupidity named Turing tarpits"; thought that a strange name for a group of people who love the stupid.
09:18:49 -!- GreaseMonkey has quit (Quit: HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it).
09:20:00 -!- Asztal has joined.
09:26:19 <ais523> anyway, the new one's called Confloddle
09:28:10 <ais523> it's based around foldl and cons, which is amazingly enough to be TC by itself, I think
09:31:22 <Gracenotes> how could I live without you, bash -.-
09:32:20 <ais523> you could use zsh and tell it to emulate bash
09:33:14 <Gracenotes> I've ended up just executing bash in csh, the default shell of a FreeBSD server my school uses.
09:34:21 <Gracenotes> hum, confloddle has functions and pattern-matching?
09:34:32 <Gracenotes> or at least unconsing
09:34:52 <ais523> no, and no
09:34:55 <ais523> it's a tarpit
09:35:16 <Gracenotes> tarpits still need syntax for expressing things
09:35:21 <ais523> it has functions to the extent that C does, but only so that you can give foldl its argument
09:35:27 <ais523> you can't actually execute them
09:35:39 <Gracenotes> yeah, I get that
09:36:31 <ais523> anyway, syntax is reverse polish; : is cons, <> is foldl (with the function to fold on going inside the angle brackets; e represents the element, and r the return value from the previous element)
09:36:45 <Gracenotes> I'm trying to think of it in terms of lambda calculus, where a simplified foldl works on Church numerals nicely.. hm.
09:36:54 <ais523> and it always uses the null list as the initial argument
09:37:01 <ais523> well, initial r
09:37:03 <ais523> for the foldl
09:39:22 <ais523> from that, you can start building up a more standard set of primitives
09:39:48 <ais523> for instance, <r> maps any list to the null list, so you can use, say, e<r> to get a null list, which is how you get started in the first place
09:40:09 <ais523> <e> returns the last element of a list, and <er:> reverses a list, so you can get the first element of a list with <er:><e>
09:45:16 <Gracenotes> what do the lists contain again?
09:57:38 <ais523> other lists
09:57:42 <ais523> there's nothing else for them to contain
09:58:23 <ais523> but you don't get an infinite regress because the null list is legal
09:58:45 <ais523> finally, to make the lang actually TC, you make the whole program a <> foldl call, and run it on an infinitely long list of null lists
09:59:10 <ais523> which allows you to get one infinite loop, which is enough
10:02:33 -!- cal153 has quit (Ping timeout: 252 seconds).
10:11:18 -!- cal153 has joined.
11:58:01 <ais523> now all I need to do with this lang is spec it, implement it, and compile some TC lang to it
12:04:08 -!- kar8nga has quit (Remote host closed the connection).
12:34:03 -!- cheater has joined.
12:39:50 <cheater> hi
12:40:07 <cheater> is there a functional version of b****fuck?
12:44:05 <pikhq> Lazy K is the closest analogue.
12:47:46 <Gregor> I would hardly call Lazy K a functional version of Brainfuck, but I suppose it's an equally evil functional language :P
12:51:39 <pikhq> Gregor: It's the closest analogue, not "a functional Brainfuck". Because of course that makes no sense otherwise. :P
12:51:50 <Gregor> True.
12:57:51 <Quadrescence> Are there any small languages that are actually enjoyable/useful to code in? Please try to understand the nature of my question before you tell me "i like brainfuck, that's enjoyable to code in"
12:58:01 <Quadrescence> I am talking about languages that actually can get stuff done.
12:58:49 <Gregor> i like brainfuck, that's enjoyable to code in
13:21:13 <fizzie> I think Glass is a bit of a get-stuff-done language, and still enjoyable. It might not be quite what you were looking for, though.
13:21:36 <fizzie> There's a pleasant Forthish feel, though I guess that's just the stack.
13:28:18 <Sgeo> "Just look! I just find manual how to login gmail account without ANY
13:28:18 <Sgeo> username!"
13:41:34 -!- MizardX has joined.
13:47:01 -!- cheater2 has quit (Ping timeout: 264 seconds).
14:01:37 -!- oklopol has joined.
14:03:57 <oklopol> o
14:03:57 <oklopol> o
14:03:57 <oklopol> o
14:03:57 <oklopol> o
14:03:58 <oklopol> o
14:03:58 <oklopol> o
14:03:58 <oklopol> o
14:03:59 <oklopol> o
14:03:59 <oklopol> o
14:04:00 <oklopol> o
14:04:02 <oklopol> ^ MY o's
14:04:33 <ais523> okokokokokokokokokokokoko
14:04:41 <oklopol> okokokokokokokokokokokokokokokokokoko
14:05:07 <ais523> ok
14:05:22 <oklopol> being on just two courses is like being on vacation
14:05:36 <ais523> Quadrescence: befunge-93 is relatively small and useful
14:06:21 <Quadrescence> ais523: meh
14:10:30 <fizzie> It's a bit hard to get anything large-scale done in such an unstructured language, though.
14:10:45 <fizzie> At least in Glass you can LEVERAGE the PRODUCTIVITY of the object-oriented ENTERPRISE paradigm.
14:11:22 <ais523> Thutu is also capable of having useful programs written in it, as long as they're the right sort of useful programs
14:11:28 <ais523> it's TC, but much better at some things than others
14:11:34 <ais523> and you'd want to use a wimpmode for things like arithmetic
14:18:12 <Sgeo> Would using PSOX be considered such a wimpmode?
14:19:52 <ais523> it's of a different nature, so I'd say no
14:20:00 <ais523> it's not a wimpmode, but a wrapper
14:20:11 <ais523> wimpmode's when you change the syntax or semantics of a lang to make it slightly less awful
14:21:54 <Quadrescence> I wonder how small one can make a decent forth compiler in C
14:52:30 -!- oklopol has quit (Ping timeout: 276 seconds).
15:34:49 -!- augur has quit (Ping timeout: 260 seconds).
15:36:33 -!- jcp has joined.
15:42:25 -!- MigoMipo has joined.
15:50:24 -!- dev_squid has joined.
16:07:04 -!- fax has joined.
16:08:48 -!- toast has changed nick to comex.
16:13:42 -!- FireFly has joined.
16:23:53 -!- augur has joined.
16:29:58 -!- coppro has quit (Ping timeout: 256 seconds).
16:34:25 -!- BeholdMyGlory has joined.
16:53:25 -!- coppro has joined.
17:01:34 -!- coppro has quit (Quit: I am leaving. You are about to explode.).
17:16:52 -!- tombom has joined.
17:25:47 -!- oerjan has joined.
17:46:33 -!- dev_squid has quit (Quit: Leaving).
17:49:11 -!- cheater2 has joined.
18:07:56 <pikhq> Quadrescence: Brainfuck is actually quite useful, so long as you don't care about file descriptors other than 0 and 1. ;)
18:08:23 <Quadrescence> Yeah sure it is.
18:08:49 <lament> Quadrescence: have you looked at io?
18:08:56 <Quadrescence> yes
18:09:07 <Quadrescence> I don't remember much from when I looked though
18:09:18 <lament> i don't know if it's enjoyable or if it gets stuff done
18:09:25 <lament> but it's definitely pretty small
18:09:52 <lament> so are forth and factor and scheme i guess
18:10:10 <lament> like r5rs scheme is actually small
18:10:23 <Quadrescence> factor is just growing and bloating
18:10:31 <Quadrescence> forth is small, scheme is small
18:11:04 <oerjan> factor should be renamed product then
18:19:06 -!- sshc has quit (Ping timeout: 248 seconds).
18:19:20 -!- tuffm has joined.
18:19:38 -!- tuffm has left (?).
18:19:51 -!- tuffm has joined.
18:20:04 -!- tuffm has left (?).
18:20:14 -!- ais523 has quit (Remote host closed the connection).
18:30:00 -!- sshc has joined.
18:40:48 -!- sshc_ has joined.
18:45:10 -!- sshc has quit (Ping timeout: 264 seconds).
19:01:19 -!- sshc_ has changed nick to sshc.
19:05:58 <uorygl> Topology: Proving your childhood dreams impossible since 1736.
19:06:30 <oerjan> what did topology do to you now
19:06:34 <uorygl> Well...
19:06:51 <lament> once a great topologist solved a famous problem by jumping off a bridge
19:06:51 <AnMaster> what was that dream?
19:07:16 <oerjan> lament: i _think_ you may have misremembered that story a tiny bit
19:07:32 <uorygl> One of my childhood dreams was to come up with a formula that could uniquely represent every line using a pair of real numbers, continuously.
19:07:59 <uorygl> Unfortunately, the topological space of lines is not equivalent to the topological space of points, so my dream is impossible.
19:08:00 <lament> one of my childhood dreams was to own a tank.
19:08:03 <AnMaster> uorygl, weird childhood dreams
19:08:37 <oerjan> hm are they even the same dimension
19:08:42 <uorygl> Yes.
19:08:50 <uorygl> The topological space of lines is the punctured real projective plane.
19:09:03 <AnMaster> uorygl, and you need more than a pair to identify a line in anything more advanced than "normal" 2D spaces
19:09:16 <oerjan> hm right most lines are defined by where they intersect the x and y axis
19:09:21 <AnMaster> you need a 3-tuple for 3D for example
19:09:38 <AnMaster> and well weird topological spaces, I have nfc?
19:09:44 <uorygl> The insight I had is that a line can usually be defined by its closest point to the origin.
19:09:51 <AnMaster> s/\?/./;s/,/?/;
19:09:53 <oerjan> i'm sure a 3-tuple is too little for 3D
19:09:53 <uorygl> All lines can, except for lines passing through the origin.
19:10:04 <AnMaster> oerjan, two of them I mean
19:10:17 <AnMaster> oerjan, I was talking about data type for the two "things"
19:10:22 <AnMaster> he said two pairs
19:10:33 <uorygl> Flip this space inside-out; the horizon becomes a missing point, and the origin becomes the horizon.
19:10:40 <AnMaster> so I pointed out you need something like two 3-tuples for 3D
19:10:50 <AnMaster> and a 3-tuple is not a pair
19:11:06 <uorygl> I did mean lines in 2D Euclidean space.
19:11:32 <oerjan> AnMaster: no he said two real numbers
19:11:39 <AnMaster> uorygl, well, given two points they represent exactly one line in 2D Euclidean space
19:11:50 <oerjan> AnMaster: that is not unique
19:11:56 <AnMaster> oerjan, oh I read it as "two pairs"
19:12:08 <AnMaster> oerjan, if you mean that there is more than one way to represent the same line: true
19:12:39 <oerjan> AnMaster: and two real numbers work for almost every line, but you cannot make it work uniquely and continuously
19:12:47 <oerjan> (homeomorphically)
19:12:48 <AnMaster> anyway, you could use y=kx+m as long as the line in question is not vertical
19:13:13 <AnMaster> in which case you could use x=m
19:13:17 <uorygl> Yeah, there are lots of sets of lines that you can do this for.
19:13:25 <uorygl> You just can't do it for all of them.
19:13:26 <oerjan> AnMaster: yeah that's one option. no matter what you do you'll either leave out something or duplicate something
19:13:50 <AnMaster> oerjan, so provide two variants, one for almost all, and then a variant for the one you can't cover
19:14:05 <oerjan> AnMaster: but that doesn't become continuous
19:14:07 <AnMaster> to tell the truth, I fail to see the issue with having to use two different variants
19:15:00 <uorygl> There's no "issue" with it. It's just that the problem I posed is to do it continuously.
19:15:02 <AnMaster> oerjan, what exactly do you mean with continuous here? The functions in question are each continuous (well, I'm not sure about x=m, but I can't see why it shouldn't be, since here you have x as a function of y...)
19:15:12 <AnMaster> so it must mean something else
19:15:52 <uorygl> Intuitively, if you're representing lines as pairs of real numbers, "continuous" means that as you move the line around in a continuous manner, the real numbers change in a continuous manner, and vice versa.
19:16:00 <AnMaster> ah
19:16:10 <AnMaster> well okay, what about using point + vector then?
19:16:20 <AnMaster> and putting some restrictions on where you can place the point
19:16:50 <uorygl> Try it; there's no way you'll get it down to two points, give a unique representation for every line, and be continuous.
19:17:01 <AnMaster> such as "if it ever intersects with the y axis, that point should be the base point, otherwise the point where it intersects with the x axis should be used"
19:17:12 <AnMaster> oh wait
19:17:18 <AnMaster> use what I said, but then not a vector
19:17:26 <AnMaster> instead use the dx/dy value
19:17:53 <AnMaster> or just the angle compared to the x axis in radians
19:18:10 <AnMaster> hm, but then you need to tell if it is the x or the y axis that is the base, don't you?
19:18:15 <uorygl> Right.
19:18:16 <AnMaster> unless you can encode that in the angle
19:18:20 <AnMaster> which you can
19:18:33 <uorygl> The thing is though, you can't uniquely, continuously represent angles.
19:18:38 <AnMaster> if the angle is pi/2 then it passes the x axis
19:18:54 <AnMaster> uorygl, oh well I guess that either it isn't unique or it jumps, true
19:19:00 <uorygl> Right.
19:19:24 <AnMaster> uorygl, but what about dx/dy?
19:20:20 <uorygl> Well, how are you going to make that work for vertical lines?
19:20:36 <AnMaster> would you accept using R extended with infinity?
19:20:47 <uorygl> Nope; that's a different problem.
19:20:53 <AnMaster> hrrm
19:21:17 <AnMaster> uorygl, actually it is simple, for vertical line it reduces to a single number
19:21:37 <uorygl> Well, the problem says you have to give a pair of numbers.
19:22:14 <AnMaster> hrrm
19:22:34 <uorygl> Gee. I think the topological space of lines is a Mobius strip.
19:22:51 <oerjan> AnMaster: there are a million (understatement) ways of hacking around this. there is however no continuous, bijective mapping from R^2 to the space of lines in R^2.
19:23:56 <oerjan> (the inverse is automatically continuous so it would have to be a homeomorphism, see http://en.wikipedia.org/wiki/Invariance_of_domain)
19:23:59 <AnMaster> this is trivial if you only have positive x and y values
19:24:15 <AnMaster> and may have negative values in the pair
19:24:59 <uorygl> Technically, I never said that every pair of real numbers must represent a line, only vice versa.
19:25:08 <AnMaster> because then you can represent which axis it passes as a single number, + for y and - for x, and abs(that value) is where along said axis
19:25:09 <oerjan> ah.
19:25:29 <AnMaster> and you can select some suitable way to represent the direction
19:25:45 <AnMaster> oh wait, it wouldn't be continuous...
19:26:10 <AnMaster> <oerjan> AnMaster: there are a million (understatement) ways of hacking around this. there is however no continuous, bijective mapping from R^2 to the space of lines in R^2. <-- has this been proven?
19:26:13 <uorygl> Hmm. If you ignore all lines that don't pass through the firt quadrant, you can do it easily.
19:26:41 <oerjan> AnMaster: i assume uorygl discovered the theorem and that's what started it. i don't recall it myself.
19:26:53 <oerjan> *started this discussion.
19:27:00 <AnMaster> uorygl, well yes I gave an idea for it above, but not sure how to represent the direction of said line in a continuous way
19:27:26 <AnMaster> and, it would run into issues for lines crossing both x and y axis in said quadrant
19:27:38 <AnMaster> sure you could give one preference, but then it would jump
19:28:58 <uorygl> Okay, I think the topological space of lines can be represented more elegantly as a Mobius strip.
19:29:13 * AnMaster suspects this is possible with complex numbers btw
19:29:18 <AnMaster> but I'm not certain
19:29:30 <AnMaster> uorygl, how do you mean?
19:29:32 <uorygl> Topologically, a complex number is just a pair of real numbers.
19:30:10 <AnMaster> uorygl, well yes, but don't forget that you get 4 numbers that way
19:30:15 <uorygl> Right.
19:30:19 <uorygl> Well, your angle around the Mobius strip is the angle of the line; rotating the line by 180 degrees takes you all the way around the strip.
19:30:27 <AnMaster> wait, actually, a 3-tuple or a complex number and a real one might be enough
19:30:37 <uorygl> And then your distance above the Mobius strip's center line is your distance above the origin.
19:30:58 <uorygl> And you can s/above/to the right of/ as necessary.
19:31:21 <uorygl> And I think you'll agree that it's impossible to lie a Mobius strip flat.
19:31:23 <AnMaster> hm btw
19:31:45 <AnMaster> an infinitely long and infinitely wide Mobius strip sounds interesting
19:31:49 <uorygl> Actually proving that it's impossible to lie a Mobius strip flat is probably really difficult...
19:31:51 * oerjan cannot recall the technical name for the space of lines in R^2
19:32:09 <uorygl> AnMaster: what does it mean for a circle to be infinitely long?
19:32:23 <AnMaster> uorygl, well, I'm not a topologist, I don't know
19:32:27 <AnMaster> I leave it to them to work it out
19:32:52 <uorygl> Anyway, it doesn't need to be infinitely wide; the entire real line can be compressed continuously into an interval line (0,1).
19:33:02 <uorygl> Which is what the atan and tanh functions do.
19:33:14 <AnMaster> <uorygl> And I think you'll agree that it's impossible to lie a Mobius strip flat. <-- I disagree
19:33:23 <AnMaster> as you didn't give any surface
19:33:30 <AnMaster> it could lie flat on another Mobius strip
19:33:34 <AnMaster> with a suitable size
19:33:39 <uorygl> Mmkay, it's impossible to lie it flat on a plane.
19:34:00 <AnMaster> uorygl, well that is a completely different statement :P
19:34:47 <AnMaster> btw, I have to say I never, ever, used (sin|cos|tan)h
19:34:58 <uorygl> Wait, I don't think it's possible to lie one Mobius strip flat onto another.
19:35:02 <AnMaster> uorygl, oh?
19:35:07 * AnMaster gets some paper
19:35:33 <oerjan> uorygl: identity function
19:35:50 <uorygl> oerjan: it's impossible to place two physical objects in the same place.
19:35:57 <oerjan> oh wait hm
19:36:07 <uorygl> One strip would end up on one side, the other strip would end up on the other side.
19:36:12 <uorygl> But there's only one side, so that doesn't work.
19:36:14 <oerjan> yes you cannot do it in R^3
19:36:55 <AnMaster> uorygl, oh *physical* one?
19:37:17 <uorygl> Physical ones, yeah.
19:37:37 <AnMaster> well, then it could lie flat on some weird 4D surface probably
19:38:28 <AnMaster> or an object of suitable shape in 3D. though probably that object would need to be assembled around said Möbius strip
19:38:53 <oerjan> http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.55.9544
19:39:17 <oerjan> although it cannot be that which is the actual theorem of that paper... far too new
19:39:22 <AnMaster> uorygl, also I think Mobius is actually an incorrect name for it. Möbius seems to be what wikipedia uses, and also is what is used in Sweden for it
19:40:18 <uorygl> oerjan: the abstract sort of implies that that is indeed a new theorem.
19:40:29 <uorygl> Since it says that it has implications.
19:40:40 <oerjan> uorygl: but that is ridiculous
19:40:48 <oerjan> it's far too simple to be new
19:41:13 <oerjan> the implications could be new though
19:41:19 <uorygl> I'll go around asking my professors how long it's been known that the space of lines in R^2 is isomorphic to the Mobius strip. :P
19:41:44 <AnMaster> that isomorphy seems weird to me
19:42:13 <oerjan> i'm sure it's well-known. in fact i'm sure i've seen a name for that space but i cannot remember it
19:42:24 <uorygl> Yes, that space is called the Mobius strip.
19:42:25 * uorygl coughs.
19:42:36 * oerjan swats uorygl -----###
19:42:53 <oerjan> not just the topological space, silly
19:43:19 <uorygl> So, now I wonder about the space of planes in R^3.
19:43:55 <AnMaster> uorygl, well, a point and a normal vector to the plane works
19:44:07 <AnMaster> or a point and two vectors
19:44:17 <oerjan> uorygl: i'm sure those were just special cases of the notation i'm looking for :(
19:44:51 <uorygl> I guess the space of planes in R^3 forms a sort of analogue to the Mobius strip.
19:45:57 <uorygl> What do you call a topological space that looks locally like a product space?
19:46:20 <AnMaster> hm what about a Möbius strip but with one of the ends in the "join" rotated 180 degrees? No not in the direction where you get a "normal" circle of paper. I mean rotate with the overlapping ends laying flat on top of each other
19:46:51 <AnMaster> in one direction you get a trivial loop, but the other direction looks quite interesting
19:47:12 <uorygl> It sounds like you mean a two-twist Mobius strip.
19:47:24 <AnMaster> and if you rotate a full 360 degrees you get a two-twist one yeah
19:47:28 <AnMaster> hm
19:47:56 <AnMaster> wait...
19:47:59 <AnMaster> that can't be right
19:48:06 <oerjan> uorygl: fiber
19:48:15 <uorygl> Right, right, a fiber bundle.
19:48:41 <uorygl> A Mobius strip is a "fiber product" of a real projective line and a line segment.
19:48:51 <AnMaster> isn't something weird supposed to happen if you cut a Möbius strip along the middle iirc?
19:48:55 <uorygl> Likewise, this 3D Mobius strip would be a fiber product of the real projective plane and a line segment.
19:49:07 <AnMaster> I can't check since I can't locate any scissors atm
19:49:25 <oerjan> AnMaster: a two-twist mobius strip is topologically identical to a non-twisted one, it's just the embedding into 3D space which is different
19:49:33 <uorygl> If you do that, I think you get an ordinary, untwisted strip, except maybe entangled with itself in some strange way.
19:49:47 <uorygl> I mean, you get something topologically identical to an ordinary strip.
19:50:40 <AnMaster> uorygl, yes you get a single strip, with a few twists it seems
19:50:53 <AnMaster> hm
19:50:56 <AnMaster> it seems one sided
19:51:11 <AnMaster> at least my thumb following it passed both sides of the joined up point
19:51:18 <oerjan> uorygl: incidentally your space of planes in R^3 etc. is obviously some kind of fibre bundle in the same way - just take the plane and _translate_ it to origo to get a map onto something projective
19:51:28 <AnMaster> wait no
19:51:38 <AnMaster> I forgot there are now two joined up points
19:51:46 <oerjan> *origin
19:52:13 <uorygl> It's definitely two-sided.
19:53:05 <AnMaster> and if you cut this one one in half you seem to get two loops that are entangled with each other
19:53:25 <AnMaster> which seems fairly interesting
19:54:00 -!- cal153 has quit.
19:54:05 <AnMaster> anyone care to explain why these things happen
19:54:06 <AnMaster> ?
19:54:20 <AnMaster> in a way that doesn't require me to learn topology first
19:54:50 <uorygl> Imagine coloring the paper strip red and blue...
19:54:57 <AnMaster> sure
19:55:02 <AnMaster> the two sided one?
19:55:08 <uorygl> Any paper strip.
19:55:11 <uorygl> So that on each side, it's red on one half and blue on the other half, and the dividing line runs down the strip.
19:55:17 <AnMaster> ah
19:55:20 <AnMaster> hm
19:55:26 -!- sshc_ has joined.
19:55:27 <uorygl> And when you flip the paper strip over, the colors don't switch places.
19:55:43 <uorygl> So no matter how you twist this strip of paper before joining it, the colors match.
19:55:59 <uorygl> And then if you cut down the dividing line, you always end up with something that's red on one side and blue on the other.
19:56:02 <AnMaster> uorygl, oh, is that physically possible?
19:56:31 <uorygl> Physically possible to color it that way?
19:56:38 <AnMaster> yeah
19:56:51 <uorygl> Sure, as long as the colors go on top of the paper, not in it.
19:56:56 <uorygl> Use paint, not dye.
19:57:01 <AnMaster> I would assume not wrt. that "<uorygl> So no matter how you twist this strip of paper before joining it, the colors match."
19:57:17 <AnMaster> because if you twist it half a turn that won't work would it?
19:57:28 <oerjan> ah
19:57:31 <uorygl> They still match if you twist it 180 degrees.
19:57:43 <oerjan> AnMaster: the paper is differently colored on each side
19:58:05 <uorygl> I'm tempted to make such a strip of paper and then make a video.
19:58:14 <AnMaster> oh yeah true
19:58:16 <AnMaster> it does work
19:58:30 -!- sshc has quit (Ping timeout: 245 seconds).
19:58:34 <oerjan> AnMaster: i was confused like you for a moment there
19:58:40 <AnMaster> well, how do we go on from there
19:58:47 <AnMaster> to the original question I asked
19:59:01 <AnMaster> oerjan, heh
19:59:28 <oerjan> well now what happens when you cut that strip along the dividing line?
20:00:13 <oerjan> now you suddenly have a strip with consistent coloring on each side - so it must be a two-sided strip
20:00:14 <AnMaster> if joined together as a möbius one, then it would end up as a two sided double twisted thingy I assume?
20:00:22 <AnMaster> not sure where the colours would end up
20:00:50 <uorygl> The colors would end up one on one side, the other on the other side. :P
20:00:59 <AnMaster> uorygl, ah
20:01:15 <AnMaster> and then what about that new one going to two intertwinned loops?
20:01:27 <AnMaster> which was what I was actually asking about
20:01:35 <AnMaster> not the one sided -> two sided
20:01:42 <AnMaster> but the next "transformation"
20:02:46 <uorygl> Colors don't help you determine whether the loops are intertwined or not.
20:03:32 <AnMaster> uorygl, well then what does
20:03:33 -!- gm|lap has joined.
20:03:51 -!- charlls has joined.
20:03:53 * oerjan assumes knot theory :D
20:04:18 <AnMaster> oerjan, which sounds like topology on steroids.
20:05:10 <oerjan> topology contains things far weirded than knots
20:05:14 <oerjan> *weirder
20:05:34 <AnMaster> what use is this btw? I mean, outside mathematics
20:05:39 <uorygl> I don't know. Stuff like knot theory is kind of difficult, because if you're inside a knot, you can't tell that you're inside a knot.
20:05:57 <uorygl> Sometimes, math, though interesting, is absolutely useless.
20:06:14 <uorygl> In the real world, you sometimes need to embed one space into another.
20:06:28 <AnMaster> uorygl, examples?
20:06:47 <uorygl> Suppose you have a circuit shaped like a Mobius strip, and you have to make it into a two-dimensional integrated circuit.
20:07:00 <oerjan> the only thing that vaguely rings a bell with knots is string theory, which doesn't exactly help
20:07:01 <uorygl> Or maybe it's shaped like a torus.
20:07:06 <uorygl> How many layers do you need?
20:07:17 -!- sshc_ has changed nick to sshc.
20:07:24 <AnMaster> hm
20:07:30 <uorygl> With a Mobius strip, you need two layers; with a torus, you need four; with a just plain strip, you only need one.
20:07:50 <oerjan> uorygl: oh right, but that's more a planar graph thing than knot theory isn't it
20:08:07 <uorygl> There's an interesting and strange isomorphism between knot diagrams and circuit diagrams.
20:08:10 <AnMaster> uorygl, I never seen anyone beginning by designing an IC circuit on anything but a flat surface, but I'm no expert in that field
20:08:40 <oerjan> or maybe those are deeply connected fields, sounds sensible actually
20:08:42 <uorygl> AnMaster: sometimes, you want to build a circuit reflecting an existing topological space, like if you want to build a circuit for cellular automata.
20:09:22 <AnMaster> hm
20:09:53 <oerjan> uorygl: why would you need four with a torus, don't you just take one layer with the upper half and one with the lower half
20:09:57 <AnMaster> however, I'm not sure things like game of life on a torus would count as a "real world application" :P
20:10:12 <uorygl> oerjan: hmm, quite right.
20:10:34 <uorygl> So, this isomorphism. Your knot diagram's edges rope off regions; the regions we're interested are the ones where to get to the outside of the knot diagram, you need to cross an odd number of edges.
20:10:39 <AnMaster> oerjan, what about the join in the other direction?
20:10:55 <AnMaster> oerjan, as in, what you said works for a tube, but I'm not sure about a torus
20:11:10 <uorygl> These regions become nodes in the circuit diagram. Wherever there's a crossing, two of these regions meet; the crossing becomes a resistor connecting the two nodes.
20:11:17 <oerjan> AnMaster: you use an actual circular arrangement?
20:11:23 <AnMaster> oerjan, of what?
20:11:26 <uorygl> If the crossing goes one way, it's a one-ohm resistor; if it goes the other way, it's a negative-one-ohm resistor.
20:11:30 <oerjan> of each half
20:11:31 <AnMaster> oh wait
20:11:35 <AnMaster> right
20:11:37 <AnMaster> that would work
20:11:46 <AnMaster> would be an awesome cpu
20:11:51 <AnMaster> if it was circular
20:12:01 <uorygl> And strangely, all the Redemeister moves, the things you can do to a knot diagram that leave it unchanged, are also things that leave the circuit essentially unchanged.
20:12:14 <fax> ??
20:12:16 <uorygl> As far as I can tell, this is simply a massive coincidence.
20:12:17 <fax> strangely?
20:12:19 <AnMaster> uorygl, !!!! resistors don't have directions
20:12:20 <AnMaster> as in
20:12:25 <AnMaster> they work the same both ways
20:12:35 <AnMaster> unlike, say, a transistor
20:12:46 <oerjan> uorygl: what do you need for a klein bottle? >:)
20:12:48 <AnMaster> (which has more than two legs)
20:12:48 <uorygl> AnMaster: let me make some ASCII art illustrating this.
20:13:42 <AnMaster> uorygl, look, I'm currently studying AC electric theory at university. Did DC a month back or so. And if there is one thing I'm certain of, it is that resistors work the same *in both directions*
20:14:01 <uorygl> I know. And crossings look the same in both directions.
20:14:31 <AnMaster> uorygl, sure, and you can't get negative resistance except by using some arrangement of transistors or such to "emulate" that
20:14:54 <uorygl> I know.
20:15:09 <uorygl> Still, negative resistors are mathematically well-defined.
20:15:29 <uorygl> The paste, if you care for it: http://pastebin.ca/1834339
20:15:49 <AnMaster> uorygl, sure, but they don't actually make any physical sense
20:16:13 <uorygl> It's math! It doesn't need to make sense!
20:16:36 <AnMaster> uorygl, and sure, if you rotate it 180 degrees, it looks the same except you now have a reverse B and a "forall" symbol
20:16:46 <AnMaster> but what about it?
20:16:49 -!- adam_d has joined.
20:17:00 <uorygl> It's a neat isomorphism.
20:17:01 <AnMaster> uorygl, where does the negative one ohm come into it
20:17:23 <uorygl> Well, if the crossing goes the other way, it's a negative resistor.
20:17:31 <AnMaster> uorygl, also how exactly is it isomorphic to a 1 ohm resistor?
20:17:58 <uorygl> It's isomorphic to a 1-ohm resistor because the isomorphism maps it onto a 1-ohm resistor.
20:18:10 <AnMaster> uorygl, and what is that isomorphism?
20:18:11 <uorygl> The bijection is an isomorphism because knot diagrams and these circuits obey the same laws.
20:18:23 <AnMaster> huh
20:18:49 <AnMaster> uorygl, so where are the poles in that ascii art?
20:19:01 <AnMaster> is it the ends of the lines? if so, why 4?
20:19:53 <uorygl> The poles are the regions A and B.
20:20:52 <AnMaster> okay...
20:21:18 <AnMaster> this seems fairly complex
20:21:47 <AnMaster> uorygl, and how do you represent a capacitor of 0.1 µF in that sort of thingy?
20:21:59 <AnMaster> (modulo typos)
20:22:20 <uorygl> By understanding this much better than I do.
20:22:25 <AnMaster> heh
20:22:29 <AnMaster> good answer I guess
20:22:32 <Wareya> hi
20:22:42 <uorygl> Hi, Wareya.
20:22:46 <AnMaster> uorygl, what about something simpler like a 1.25 Ohm resistance?
20:27:34 <AnMaster> btw, are there matrices in more than 2 dimensions? I can't see why not, but I have never seen any such
20:27:49 <oerjan> use 4 parallel copies of 5 serially connected 1 Ohm resistors. maybe.
20:28:16 <AnMaster> oerjan, hm there should be resistances that you can't reach by doing such operations
20:28:21 <AnMaster> oerjan, what about pi Ohm for example
20:28:33 <uorygl> oerjan's idea is what I would have said.
20:28:47 <oerjan> AnMaster: um do you consider an n by n matrix as 2- or n-dimensional?
20:29:14 <AnMaster> oerjan, well in this case as 2D. Since you can locate a given number in it by taking column and row numbers
20:29:22 <AnMaster> which is analogous to x and y coordinate
20:29:30 <oerjan> in that case you probably want to look at tensors
20:29:58 <uorygl> The thing is, I tried making the knot diagram corresponding to three parallel copies of three serially connected one-ohm resistors, and I'm pretty sure the result was not equivalent to simply a one-ohm resistor.
20:30:06 <AnMaster> oerjan, ah, I heard that word before. Tell me, does it extend to n dimensions for any n in R, or just to some boring fixed number n?
20:30:53 <oerjan> AnMaster: any natural number n. also you don't call it dimensions, that's for the underlying space (i.e. n in n by n)
20:30:58 <oerjan> afaik
20:31:02 <AnMaster> oerjan, ah I see
20:31:43 <AnMaster> uorygl, maybe you did the connection of those in the wrong way?
20:32:28 <oerjan> AnMaster: the term used seems to be "order"
20:32:38 <AnMaster> oerjan, silly mathematicans ;P
20:34:26 <oerjan> AnMaster: i assume you can only reach rational resistances that way, at least by combining serial and parallel separately (serial adds the resistances and parallel adds their inverses iirc)
20:34:31 <uorygl> It's pretty simple and regular.
20:34:48 <uorygl> So I don't think I make a mistake.
20:35:10 <AnMaster> oerjan, the formula is 1/R_new = 1/R_1 + 1/R_2 + ... + 1/R_n
20:35:27 <oerjan> well that's what i _said_ isn't it
20:35:32 <AnMaster> oerjan, so yes, add their inverses, *plus* then invert that
20:35:58 <AnMaster> oerjan, the sum of their inverses is 1/R_new after all
20:36:12 <oerjan> um right that may have been unclear
20:36:30 <AnMaster> oerjan, I'm tempted to say: s/unclear/forgotten/
20:36:41 <oerjan> not really
20:36:49 <AnMaster> okay
20:37:34 <uorygl> Serial adds the resistances to get the new resistance. Paralle adds their inverses to get the new inverse.
20:37:35 <fax> why is it 1/A = 1/B + 1/C?
20:38:18 <uorygl> fax: well, you can derive that using Ohm's law and some basic knowledge of how circuits work.
20:38:32 <AnMaster> oerjan, the interesting thing is that if you have a sinus formed alternating current, and write capacitances and inductances as specific complex values you can calculate as if it was all direct current
20:38:35 <fax> how do circuits work
20:38:47 <AnMaster> fax, by suitably advanced magic
20:38:53 <AnMaster> err, I mean technology of course ;P
20:39:15 <oerjan> fax: it has to do in my intuition with how resistance is the quotient of voltage and current, and serial vs. parallel adds voltage and current respectively
20:39:45 <fax> why does parallel add current?
20:39:49 <AnMaster> oerjan, the former is called Ohm's law
20:39:56 <AnMaster> fax, err what?
20:40:21 <AnMaster> oerjan, what do you mean "serial vs. parallel adds voltage and current respectively"
20:40:23 <uorygl> Well, every pole has a voltage, and every component has a current, and the currents in and out of a pole must add up to the same thing, and every component has an equation of some sort determining the relation between voltage and current.
20:40:30 <AnMaster> I'm unable to decode that in a sensible way
20:40:35 <oerjan> because for a fixed voltage across, you get the current flowing across _each_ branch for that branch
20:40:54 <uorygl> When you place two components in parallel, the voltage across the entire group is equal to the voltage across each component.
20:41:13 <uorygl> And the current across the entire group is equal to the sum of the currents across each component.
20:41:18 <fax> aha
20:41:36 <AnMaster> what uorygl said was a lot more sensible.
20:41:37 <uorygl> If you want to know how analog circuits work, this applet makes it crystal clear: http://falstad.com/circuit/
20:42:03 <AnMaster> fax, think of parallel it as opening more roads for the traffic to drive along to the same goal
20:42:08 <AnMaster> s/it//
20:42:31 <oerjan> AnMaster: well he explained why it's the case...
20:43:01 <AnMaster> oerjan, your statement seemed somewhat confused. As in what do you mean "serial adds voltage"?
20:43:02 <oerjan> also i've never learned alternating current
20:43:11 <AnMaster> what exactly does that even mean
20:43:28 <oerjan> AnMaster: the potential between the ends is the sum of the potentials across each step
20:43:29 <uorygl> AnMaster: well, do you know what he meant by "parallel adds current"?
20:43:39 <AnMaster> oerjan, oh it is fun. You just write currents and voltages on polar form, plus what I said above for the components, and calculate the same
20:43:59 <AnMaster> then you just pick the pieces back to the non-complex form for the final result
20:44:04 <uorygl> When you place two components in series, the current through the entire group is equal to the current through each component, and the voltage across the entire gruop is equal to the sum of the voltages across each component.
20:44:12 <AnMaster> <oerjan> AnMaster: the potential between the ends is the sum of the potentials across each step <-- well sure
20:44:30 <AnMaster> <uorygl> AnMaster: well, do you know what he meant by "parallel adds current"? <-- I can have a guess, but it isn't how I would express it.
20:45:15 <AnMaster> uorygl, I know how to do the math, I just don't think his way of describing the operations made a lot of sense
20:45:35 <AnMaster> to the extent I was unable to decode what operation he meant
20:46:38 * uorygl nods.
20:46:51 <oerjan> whatEVER
20:47:04 <AnMaster> uorygl, plus I'm somewhat challenged when it comes to the English terms for these things. The course is in Swedish
20:48:23 <oerjan> current = strøm, voltage/potential = spenning, if the norwegian terms help
20:48:42 <AnMaster> oerjan, that doesn't work as we have voltage = spänning potential = potential
20:49:05 <AnMaster> and current = ström I know
20:49:06 <oerjan> well i'm a little vague on the difference between those two
20:49:13 <AnMaster> oerjan, also what about impedans?
20:49:15 -!- jcp has quit (Quit: I will do anything (almost) for a new router.).
20:49:23 <oerjan> isn't that impedance
20:49:31 <uorygl> I suddenly wonder why you guys are speaking English. :P
20:49:32 * AnMaster looks on wikipedia
20:49:41 <AnMaster> uorygl, oh?
20:49:44 <oerjan> i think it's impedans in norwegian too
20:49:57 <AnMaster> oerjan, seems it is impedance from a quick look at wikipedia
20:50:08 <uorygl> It would be neat if you spoke Norwegian and Swedish and I tried to follow along.
20:50:22 <oerjan> at least i recall my dad using that word (he's an electrical engineer)
20:50:41 <AnMaster> uorygl, well we would have some issues between ourselves. While the languages are similar, they are not that close
20:50:47 <AnMaster> so some confusion would certainly arise
20:50:52 <oerjan> uorygl: Nei det ville det ikke
20:51:09 <AnMaster> and that if directly translated to Swedish is somewhat silly
20:51:22 <AnMaster> for a start what on earth does "det" refer to in there?
20:51:30 <AnMaster> the second one that is
20:51:32 <AnMaster> but also the first
20:51:52 <oerjan> AnMaster: the "It" of uorygl's sentence
20:52:19 <AnMaster> a literal translation from me would be: "no that want that not", an "idiomatic if it was Swedish" would be "no it won't it" or such.
20:52:25 <AnMaster> or rather, <grammar error>
20:52:34 <oerjan> i suppose you could add "være" at the end for better precision
20:52:49 <AnMaster> oerjan, I think your "det" can be used somewhat differently than "det" in Swedish perhaps?
20:53:36 <oerjan> AnMaster: i don't know anything about that. it corresponds to both "it" and "there (is)" in english though, though swedish was the same
20:53:39 <oerjan> *thought
20:53:42 <uorygl> Eh. Everyone, just learn Spanish. :P
20:53:44 <AnMaster> oerjan, please give an English idiomatic translation of the whole thing, because I haven't figured it out
20:53:49 <AnMaster> uorygl, agreed!
20:53:55 <oerjan> AnMaster: "No, it wouldn't be"
20:53:56 <AnMaster> oh wait
20:54:00 <uorygl> Let's see, I don't have a Swedish keyboard installed.
20:54:01 <AnMaster> I thought you said Swedish
20:54:03 <AnMaster> not Spanish
20:54:06 <AnMaster> blerh
20:54:08 <AnMaster> blergh*
20:54:13 <AnMaster> oerjan, ah...
20:54:22 <AnMaster> oerjan, "nej, det skulle det inte vara"
20:54:38 <oerjan> AnMaster: so "det" wasn't the problem actually?
20:54:40 <AnMaster> oerjan, so the issue is "ville" which means "want" but not "would"
20:55:00 <AnMaster> (well, vill is the base form rather)
20:55:05 <uorygl> börk
20:55:06 <oerjan> well i knew that about swedish, i think
20:55:27 <AnMaster> oerjan, which mean what you said didn't make a lot of sense to me. I was wondering what was wanting what
20:55:32 <AnMaster> (perhaps for a bday present?)
20:56:03 <oerjan> ville/skulle is a little awkward to translate i guess
20:56:04 <AnMaster> uorygl, you know, that is so completely un-Swedish as you can get, to a Swede that is
20:57:49 <AnMaster> oerjan, the word doesn't map 1:1 to English would for all forms, as in, the ground form "ska" maps to ~will (as in "foo will turn into bar" or whatever, not as in "last will")
20:58:32 <AnMaster> "last will" seems to make more sense if you set in the Swedish "vill", so it turns into "last want"
20:58:45 <AnMaster> perhaps it is some old form of a word still left in that phrase or such?
20:58:47 <oerjan> siste vilje
20:59:28 <oerjan> (in norwegian)
20:59:41 <AnMaster> oerjan, yep similar in Swedish. though if it is the legal document you mean it would be "testamente"
20:59:53 <oerjan> yeah
21:00:10 <uorygl> I'm tempted to speak Norwegian, but as the only words I know are "skillingsbolle" and "smultring", I can't.
21:00:11 <oerjan> "last will and testament"
21:00:22 <oerjan> uorygl: at least you won't starve
21:00:30 <AnMaster> uorygl, what is the first?
21:00:40 <uorygl> Cinnamon roll.
21:00:53 <AnMaster> the latter I can guess, if it is actually sv:"smultron"? (not sure if that is spelled with o or å though)
21:01:02 * oerjan would call that kanelbolle...
21:01:12 <AnMaster> oh kanelbulle
21:01:26 <AnMaster> oerjan, are your bullar spheres or what?
21:01:27 <AnMaster> ;P
21:01:30 <oerjan> `translatefromto sw en smultron
21:01:32 <uorygl> Don't ask me why Wikipedia calls it a skillingsbolle.
21:01:37 <oerjan> AnMaster: approximately
21:01:44 <HackEgo> smultron
21:01:45 <oerjan> uorygl: oh it's probably called that too
21:01:59 <oerjan> oh wait
21:02:05 <oerjan> `translatefromto sv en smultron
21:02:07 <HackEgo> strawberries
21:02:14 <oerjan> what the heck
21:02:15 <AnMaster> oerjan, since sv:boll = en:ball. thus "kanelbolle" sounds like a ball of kanel to me
21:02:31 <oerjan> AnMaster: _definitely_ not smultron, then (would be "jordbær")
21:02:31 <uorygl> Wiktionary says that smultringer are called "munk" in Swedish.
21:02:40 <AnMaster> oerjan, then what is "smultring" then?
21:02:42 <uorygl> Or "flottyr-ring" or "ringmunk" or something.
21:02:59 <AnMaster> oerjan, also strawberries seems wrong...
21:03:00 <uorygl> http://sv.wikipedia.org/wiki/Munk_%28bakverk%29
21:03:08 <uorygl> Those munkar look delicious.
21:03:20 <oerjan> AnMaster: plain donuts, was the conclusion of our discussion here
21:03:27 <AnMaster> oerjan, http://sv.wikipedia.org/wiki/Smultron
21:03:36 <AnMaster> which are *not* strawberries, but a related thingy
21:03:57 <AnMaster> en:strawberry = sv:jordgubbe
21:03:59 <Deewiant> They are strawberries
21:04:03 <Deewiant> Just not the same kind
21:04:15 <AnMaster> hm
21:04:17 <AnMaster> perhaps
21:04:23 <AnMaster> I'm no specialist on that sort of thing
21:04:24 <Deewiant> en:strawberry = fi:mansikka, sv:smultron = fi:metsämansikka
21:04:27 <oerjan> AnMaster: markjordbær vs. jordbær in norwegian
21:04:55 <AnMaster> oerjan, the former seems somewhat silly. "ground earth berries" and the latter "earth berries"?
21:05:24 <uorygl> Swedish Wiktionary says that smultron are Fragaria vesca, which English Wikipedia says are commonly known as woodland strawberries.
21:05:30 <uorygl> s/Wiktionary/Wikipedia/
21:05:37 <AnMaster> oerjan, why not "ground earth zero-altitude berries" next
21:05:52 <oerjan> AnMaster: well "mark" here sounds me like it means "in the wild" in this case
21:06:03 <AnMaster> oerjan, aha
21:06:24 <AnMaster> oerjan, there are no such connotations of that word in Swedish
21:06:29 <uorygl> Lowly ground earth dirt berries!
21:06:35 <oerjan> while the other kind is a farmed variety, which iirc is a hybrid of markjordbær with an american relative...
21:06:42 <AnMaster> probably it would be called "vilda &" or "vild&" or such
21:06:50 <AnMaster> (where & as in sed)
21:06:59 <oerjan> (i think i heard that mentioned on television recently)
21:07:13 <oerjan> which must mean christmas, since that was last i watched tv
21:07:37 <AnMaster> <uorygl> Wiktionary says that smultringer are called "munk" in Swedish. <-- sv:munk is *also* en:monk
21:07:39 <AnMaster> just to confuse things
21:07:51 <uorygl> Yum, monks.
21:08:11 <AnMaster> so if you google translate a Swedish article about either and happen to get some of the other things you now know the cause
21:09:00 <AnMaster> http://sv.wikipedia.org/wiki/Munk_%28olika_betydelser%29 <-- apparently it can also be something related to hydrology
21:09:05 <AnMaster> I have no idea what
21:10:14 <uorygl> `translate Munk är inom hydrologin en regleranordning för att kunna reglera inloppet eller utloppet till en damm.
21:10:18 <HackEgo> Monk is in hydrology a regulated regime to regulate the inlet or outlet to a pond.
21:10:26 <oerjan> oh wait wikipedia disagrees with me
21:10:35 <uorygl> A regulated regime.
21:10:40 <oerjan> "The garden strawberry was first bred in Brittany, France in 1740 via a cross of Fragaria virginiana from eastern North America , which was noted for its flavor, and Fragaria chiloensis from Chile and Argentina brought by Amédée-François Frézier, which was noted for its large size."
21:10:44 <AnMaster> uorygl, mistranslation
21:10:55 <uorygl> `translate Ursprungligen betecknar ordet munk ett vertikalt rör, där vatten kan strömma in över kanten när vattenytan i dammen överstiger kantens höjd.
21:11:00 <HackEgo> Originally, the word denotes a monk, a vertical pipe, where water can flow onto the top when the water in the pond exceeds the edge height.
21:11:08 <AnMaster> uorygl, it would be "regulation apparatus"
21:11:13 <AnMaster> or some such
21:11:13 <oerjan> or with my recollection, rather. anyway neither ancestor was actually markjordbær then
21:12:41 <oerjan> strangely afaik in norwegian "munk" means only the same as english "monk"
21:12:51 <AnMaster> heh
21:12:57 <oerjan> never heard any of the other meanings
21:13:37 <AnMaster> oerjan, flottyr-ring works in Swedish too (for the non-monk-or-hydrology type of munk that is)
21:13:43 <AnMaster> well
21:13:47 <AnMaster> drop that -
21:14:03 <AnMaster> it seems out of place
21:14:35 <AnMaster> uorygl, what is your native language then?
21:14:48 <oerjan> heh sw:munk as bakery is named for the monk hairstyle :D
21:14:57 <AnMaster> hah
21:14:59 <AnMaster> that explains it
21:15:04 <AnMaster> oerjan, also: sv not sw
21:15:12 <oerjan> yeah yeah
21:15:17 <uorygl> English.
21:15:45 <AnMaster> uorygl, then I feel sorry that you can't almost freely concatenate words to form new longer ones!
21:15:51 <uorygl> :)
21:15:56 <AnMaster> (with a well defined meaning of course)
21:16:00 <uorygl> We can do that; it's just that we still put spaces in between.
21:16:15 <AnMaster> uorygl, then it isn't true concatenation
21:16:18 <uorygl> So we say "Grand Valley State University Honors College application" instead of "Grandvalleystateuniversityhonorscollegeapplication".
21:16:19 <AnMaster> so it doesn't count
21:16:42 <AnMaster> uorygl, we need to add a few binding letters in there in Swedish between some of them
21:16:54 <AnMaster> it will take a few seconds to translate that
21:17:22 <oerjan> ordsammensetningsumulighetssorg
21:17:49 <AnMaster> oerjan, "umulighet"?
21:17:55 <uorygl> `translatefromto en sv grand valley state university honors college application
21:17:57 <HackEgo> Grand Valley State University utmärkelser högskola ansökan
21:18:04 <oerjan> omöjlighet?
21:18:09 <AnMaster> oerjan, ah
21:18:15 <uorygl> Apparently Google knows what Grand Valley State University is.
21:18:27 <AnMaster> uorygl, I'm unable to translate "collage"
21:18:34 <AnMaster> to anything except "universitet"
21:18:38 <AnMaster> which we already have
21:18:47 <uorygl> In this case, a college is a sub-unit of a university.
21:18:53 <AnMaster> Stordalgångslänsuniversitetsheders<insert collage translation here>ansökan
21:18:56 <AnMaster> and well
21:19:02 <AnMaster> state had to be translated to län
21:19:07 <AnMaster> which is the not exactly
21:19:14 <AnMaster> but we don't have a real word for an US state
21:19:31 <AnMaster> except, stat, which is also a synonym to land (country)
21:19:47 <uorygl> Well, "state university" has a special meaning.
21:19:49 <AnMaster> a Swedish län is *way* smaller than a US state
21:19:56 <AnMaster> uorygl, oh? well then I have no clue
21:20:16 <uorygl> A state university is a public university funded by an individual state.
21:20:29 <AnMaster> <uorygl> Apparently Google knows what Grand Valley State University is. <-- no it doesn't since it didn't translate it
21:20:35 <AnMaster> or if it is a name then I guess it does
21:21:05 <AnMaster> uorygl, oh and it's "högskola" for "college" is not a sub-unit of a university
21:21:15 <AnMaster> it is closer to what in UK I think used to be called a polytechnic
21:21:21 -!- pikhq has quit (Read error: Connection reset by peer).
21:21:40 <uorygl> It did translate it. It translated it from "grand valley state university" to "Grand Valley State University".
21:21:52 <AnMaster> uorygl, which makes no sense in Swedish
21:21:58 <AnMaster> since you don't do caps in the names like that
21:22:05 <uorygl> Well, it's an English name.
21:22:11 <AnMaster> uorygl, we don't have title case you see
21:22:17 <uorygl> "Grand Valley State University (GVSU) er et universitet med hovedbase i Allendale i Michigan i Amerikas forente stater."
21:22:31 <AnMaster> uorygl, that was norwegian
21:22:35 <uorygl> True.
21:22:36 <AnMaster> or perhaps Danish
21:22:38 <AnMaster> can't really tell
21:22:40 <uorygl> Norwegian.
21:22:57 <AnMaster> mhm
21:22:59 <oerjan> i think danish would be "forenede"
21:23:03 <oerjan> or something like that
21:23:05 <AnMaster> förenta in Swedish
21:23:13 <AnMaster> and "er et" would be "är ett"
21:23:23 <AnMaster> and hovedbase would be "huvudbas"
21:23:30 <AnMaster> but that would sound awkward in that context
21:23:40 <AnMaster> it sounds like some military main base or such
21:23:44 <oerjan> i'm not sure there would be any other differences between danish and norwegian there
21:24:23 * AnMaster would write it in a different way to get idiomatic Swedish
21:24:46 <oerjan> danish wikipedia has no article on GVSU
21:24:51 <AnMaster> probably instead of "with main base in" it would be like "located in"
21:25:37 <oerjan> um main base would imply there could be other bases elsewhere
21:25:50 <oerjan> it seems to me
21:26:05 <AnMaster> oerjan, perhaps, well I would check if there was, and then write something that fitted that
21:26:21 <AnMaster> also I guess the more idiomatic English term would be HQ
21:26:34 <oerjan> of course the english article uses "located" :D
21:26:40 <AnMaster> which reminds me of RTS
21:27:09 <uorygl> GVSU does indeed have multiple campuses.
21:27:26 <uorygl> The main one is in Allendale, there's another one in downtown Grand Rapids, and there are a couple of tiny ones in other places.
21:27:39 -!- pikhq has joined.
21:28:45 <AnMaster> uorygl, "with campus in Allendale, foo, bar and so on", with the Swedish translation for "and so on" sounding a lot better in the context than it does in English
21:29:04 <AnMaster> "med flera" (literally: with more)
21:29:09 <AnMaster> well no that isn't a literal translation
21:29:24 <uorygl> I think we would tend to use "and others".
21:29:27 <AnMaster> oerjan, English doesn't have different words for mer and fler does it?
21:29:36 <AnMaster> uorygl, closer to the Swedish yeah
21:29:37 <uorygl> With campuses in Allendale, Grand Rapids, Holland and others.
21:29:52 <AnMaster> uorygl, shows I'm not a native speaker, not thinking of that English way
21:30:04 <uorygl> The Holland campus is really inconvenient because it's in a different continent.
21:30:07 <oerjan> AnMaster: i think not, it has "less" vs. "fewer" but not the other way iirc
21:30:13 <AnMaster> yeah
21:30:31 <AnMaster> which is rather unsymmetrical when you think about it
21:30:38 <uorygl> And they speak Dutch on that campus.
21:30:40 <AnMaster> uorygl, don't you agree?
21:30:47 <uorygl> I really have no idea why GVSU has a campus there.
21:30:55 <uorygl> Yeah.
21:30:56 <AnMaster> uorygl, I was just about to ask you heh
21:31:20 <uorygl> Of course, I'm entirely kidding, and Holland is actually a city in Michigan.
21:31:31 <AnMaster> uorygl, err? really?
21:31:44 <AnMaster> now that is confusing
21:31:45 <uorygl> There is a city called Holland, Michigan, yes.
21:31:51 <oerjan> Paris, Texas
21:31:58 <AnMaster> should anyone in Michigan actually *want* to travel to the real Holland
21:32:08 <AnMaster> like, when ordering tickets
21:32:11 <uorygl> Once, before I knew about it, my mom said, "Bye! I'm going to Holland!", and I thought, "Wait, what?"
21:32:16 <AnMaster> at least if the city also have an airfield
21:32:46 <fizzie> Isn't it quite a common thing for US people to do to name their cities and such after "real" places?
21:33:03 <AnMaster> fizzie, why did you have to put quotes around real there?
21:33:07 <AnMaster> it ruined the thing
21:33:38 <fizzie> I notice it whenever I go to maps.google.com and type in a city without bothering to scroll away from the default US-centered start view; then it zooms to somewhere in Texas or whatever.
21:33:46 <uorygl> Azeroth, California
21:33:53 -!- gm|lap has quit (Quit: HydraIRC is a child molester -> http://silverex.net/news <- i couldn't change my quit message).
21:34:03 <AnMaster> Azeroth?
21:34:10 <AnMaster> that sounds somewhat familar
21:34:15 <uorygl> World of Warcraft.
21:34:23 <AnMaster> uorygl, I don't play that
21:34:24 <Deewiant> Or just Warcraft.
21:34:25 <AnMaster> so can't be
21:34:28 <AnMaster> nor that
21:34:47 <uorygl> Wikipedia says that Holland, Michigan has two airports but no commercial flights.
21:34:58 <AnMaster> two!?
21:35:06 <AnMaster> what kind of metropol is that?
21:35:25 <uorygl> Population 35,000.
21:35:32 <AnMaster> okay that is just weird
21:35:40 <uorygl> I think they're the sort of airport that consists of a hanger and a runway.
21:35:44 <uorygl> s/e/a/
21:35:46 <AnMaster> even so
21:35:54 <AnMaster> a runway takes a lot of space
21:36:02 <uorygl> A road also takes a lot of space.
21:36:07 <fizzie> Incidentally, now that I went to test that maps.google.com, it seems to jump to the actual places I was actually looking for. Maybe they've changed that a bit.
21:36:13 <AnMaster> usually 1 km x 20 m or so at least
21:36:14 <uorygl> Michigan has plenty of rural space to spare.
21:36:52 <uorygl> Is that 20 meters for the runway itself or also for the surrounding cleared area?
21:37:38 -!- pikhq has quit (Read error: Connection reset by peer).
21:37:41 <uorygl> I guess we have some roads that are about 20 meters wide.
21:37:47 <AnMaster> uorygl, of it itself, but this varies widely with the intended use of the runway
21:37:50 <AnMaster> some can be much wider
21:37:52 <fizzie> The local "small" airport here in Helsinki seems to have runways of approximately that width.
21:37:53 <AnMaster> and much longer
21:38:06 <fizzie> (Based on Google's satellite imagery.)
21:38:35 <AnMaster> consider KEDW, that has some *very* wide runways
21:38:41 <AnMaster> (don't know exact values)
21:39:01 <fizzie> The "main" (Helsinki-Vantaa) airport runway width, eyeballing from that satellite photo, seems to be around 50 m.
21:39:11 <AnMaster> well http://www.airnav.com/airport/KEDW gives 4R/22L as 91 meters wide
21:39:17 <AnMaster> which is very wide
21:39:34 <AnMaster> also it is over 4.5 km long that one
21:39:52 <fizzie> Aw, I can't check my eyeball measurements, that site seems rather US-centric.
21:40:09 <AnMaster> fizzie, what is the ICAO code for it?
21:40:39 <AnMaster> fizzie, well?
21:40:57 <AnMaster> fizzie, I could probably look it up with relative ease if I can just get the ICAO code for it..
21:41:17 <AnMaster> it would start on E (north Europe) I know that
21:41:25 -!- pikhq has joined.
21:41:30 <AnMaster> I don't know what the second letter is for finland
21:41:45 <fizzie> EFHK.
21:42:08 <fizzie> It's F for Finland. (Or maybe it's just a happy accident and not by design.)
21:42:31 <uorygl> Does K mean United States?
21:42:49 <AnMaster> uorygl, yes, you have 3 letters for inside the US
21:43:00 <AnMaster> iirc france has a single letter prefix too
21:43:04 <AnMaster> and a few other big countries
21:43:23 <uorygl> I just realized that I don't know what France calls itself.
21:43:41 <AnMaster> fizzie, I forgot how to extract this from the flightgear airport data...
21:44:06 <uorygl> They call themselves France. That was easy enough.
21:44:22 <oerjan> _la_ France
21:44:22 <AnMaster> fizzie, what is your equiv of luftfartsverket?
21:44:54 <AnMaster> they should have some sheets of info on the airport
21:44:59 <AnMaster> on their website
21:45:27 <fizzie> AnMaster: I guess it's called Finavia nowadays. And yes, I guess they'd have that.
21:45:28 <AnMaster> with stuff like VOR freqs and any ILS glideslopes and landing patterns and what not
21:45:35 <oerjan> AnMaster: hey you're not supposed to spell that exactly the same as in norwegian!
21:46:27 <AnMaster> fizzie, "Finavia" sounds like some company flying cheap flights to tourist filled up islands or such
21:46:32 <oerjan> hm actually it's no longer called that in norwegian
21:46:47 <oerjan> Avinor now
21:46:56 <AnMaster> ooh that sounds as silly...
21:46:58 <AnMaster> as the Finnish one
21:47:11 <oerjan> fizzie: shall we beat him up?
21:47:38 <fizzie> oerjan: Yes. Do you want to be the one doing the beating, or the holding?
21:47:47 <AnMaster> hah
21:47:49 <oerjan> ouch, choices
21:48:01 <AnMaster> oerjan, also from which direction to attack?
21:48:08 <AnMaster> and how to represent it?
21:48:24 <oerjan> stealth attack! -----###
21:48:28 <AnMaster> as a vector? If so, what would the base (is that the correct English term?) be
21:48:34 <AnMaster> oerjan, no
21:48:41 <AnMaster> oerjan, you got that wrong i think
21:48:52 <oerjan> hm?
21:48:55 <AnMaster> oerjan, it seems to be one - short
21:49:13 <oerjan> no
21:49:23 <AnMaster> fizzie, btw http://www.finavia.fi/files/finavia2/Taulukot/Airport_Facts_Chart_Jan09.pdf
21:49:37 <uorygl> Five hyphens and three pounds, right?
21:49:43 <AnMaster> fizzie, your measurement is way off, those are all about 60 m wide
21:50:07 <AnMaster> the one I gave was for a rather small airport
21:50:15 <oerjan> uorygl: as it should be!
21:50:22 <fizzie> AnMaster: It still rounds to 50 if you quantize things with large enough granularity.
21:50:28 <AnMaster> also, just CAT II ILS heh
21:50:37 <AnMaster> is there a larger airport near there?
21:51:03 <AnMaster> if not I'm really surprised at the main Helsingfors airport not having a CAT III ILS
21:51:21 <fizzie> There's nothing larger than Helsinki-Vantaa in Finland, no.
21:51:27 <AnMaster> fizzie, heh
21:51:48 <fizzie> But the one I approximated at 20 m width was another place.
21:52:38 <AnMaster> fizzie, oh? you said EFHK when I asked
21:52:53 <fizzie> Yes, that was for the 50 m figure.
21:53:02 <fizzie> <fizzie> The "main" (Helsinki-Vantaa) airport runway width, eyeballing from that satellite photo, seems to be around 50 m.
21:53:11 <AnMaster> fizzie, I was asking about the 20 meter one, I must have missed that line
21:53:34 <fizzie> That one; it's EFHF for the tiny almost-in-the-city-centre airport.
21:54:28 <fizzie> It's not listed in that facts chart either; it's possible that it's still administered by the (apparently still existing as a subset of another government thing) Finnish Civil Aviation Authority thing, and not by the Finavia state-owned-enterprise nonsense.
21:55:26 <AnMaster> ffs, lfv.se has redesigned, I'm unable to find the data on Swedish airports
21:55:32 <AnMaster> flight weather yes...
21:55:36 -!- Gracenotes has quit (Quit: Leaving).
21:55:41 <AnMaster> but not the landing patterns and so on
21:55:45 <AnMaster> I know I have read it before
21:55:49 <fizzie> If I read https://ais.fi/ais/eaip/aipcharts/efhf/hfad.pdf right that 20 m is actually 30 m. (At least in the N x M numbers the N matches the listed runway lengths.)
21:56:23 <AnMaster> fizzie, charts like that plus a lot more I remember seeing on lfv.se ...
21:56:23 <fizzie> I like it how they have that other runway exactly 2^10 metres.
21:56:25 <oerjan> AnMaster: clearly it's to confuse the terrorists
21:56:46 <AnMaster> oerjan, -_-
21:57:23 <AnMaster> fizzie, yes seems to be 30 meters wide asphalt
21:57:53 <AnMaster> fizzie, also note the listing with the four runways down in the corner
21:58:06 <AnMaster> fizzie, do you know where the two "missing" ones are ;P
21:58:35 <AnMaster> what about you oerjan?
21:59:26 <fizzie> Are those just the same things in both directions? It sure looks that way.
21:59:32 <AnMaster> fizzie, indeed
22:00:01 <AnMaster> fizzie, I expect oerjan would have answered before you if the directions had been given as radians instead of magnetic heading
22:01:07 <AnMaster> what is the google thingy to search for things that links to a given url?
22:01:20 <fizzie> Anyhow, https://ais.fi/ais/eaip/en/ seems to be the site for that sort of detailed information; it's Finavia-operated, but couldn't find any links to it from the mostly-PR-stuff finavia.fi site. At least Google helped.
22:01:36 <uorygl> I suddenly wish .523 were a TLD.
22:01:43 <AnMaster> uorygl, why?
22:02:50 <fizzie> AnMaster: Do you mean the link: operator?
22:02:57 <AnMaster> fizzie, ah thanks
22:02:58 <fizzie> (http://www.google.com/intl/en/help/operators.html)
22:03:01 <AnMaster> well, that didn't solve it
22:03:51 <fizzie> Doing "link:" searches seems to work rather randomly nowadays, anyway.
22:04:26 <AnMaster> aha http://www.lfv.se/sv/LFV/Flygtrafiktjansten/FPC/IAIP/AD-0-4/AD-2/
22:04:28 <AnMaster> there we go
22:04:42 <AnMaster> wow that is a lot of separate pdfs: http://www.lfv.se/sv/LFV/Flygtrafiktjansten/FPC/IAIP/AD-0-4/AD-2/S/
22:05:18 <oerjan> AnMaster: i might have answered before fizzie if i had either (1) been looking at your page (2) not been browsing reddit
22:05:32 <AnMaster> oerjan, suuuure ;P
22:06:33 <AnMaster> fizzie, oh and it seems ESSA at least have CAT III ILS :P
22:06:47 -!- werdan7 has quit (Ping timeout: 615 seconds).
22:09:34 <uorygl> AnMaster: because then we could have ais.523.
22:14:18 -!- werdan7 has joined.
22:16:01 <AnMaster> uorygl, -_-
22:18:32 -!- MigoMipo has quit (Read error: Connection reset by peer).
22:29:11 <lament> Quadrescence: thermo says STAY OUT of #nm
22:29:29 <lament> so you should probably join just to piss him off
22:32:25 <AnMaster> lament, what is #nm about?
22:32:31 <AnMaster> nanometers?
22:32:37 <AnMaster> nautical miles?
22:32:52 <oerjan> naughty midgets
22:33:33 <AnMaster> no mathematicians
22:34:00 <AnMaster> or perhaps naughty mathematicians
22:34:23 <AnMaster> would that be like not writing => when you should?
22:35:59 <uorygl> What's the irssi command to swap the current window with the given window
22:36:05 <uorygl> ^H?
22:36:08 <olsner> oerjan: hmm, what was it again - two contumacious cubits per naughty midget?
22:36:20 <AnMaster> olsner, ???
22:36:38 -!- MizardX has quit (Read error: Connection reset by peer).
22:36:44 <olsner> AnMaster: !!!
22:36:45 <AnMaster> why cubits?
22:37:06 <oerjan> olsner: itym cupids
22:37:07 <AnMaster> oh wait, not qubits?
22:37:26 <AnMaster> "A cubit is the first recorded unit of length and was one of many different standards of measurement used through history."
22:37:28 <AnMaster> says wikipedia
22:37:29 <olsner> AnMaster: for their contumacity, of course
22:37:46 <AnMaster> olsner, and itym qubits
22:37:59 <olsner> oerjan: not that I did, but I could have meant it
22:38:53 -!- MizardX has joined.
22:39:20 <oerjan> `define cubit
22:39:23 <HackEgo> * an ancient unit of length based on the length of the forearm \ [19]wordnetweb.princeton.edu/perl/webwn \ * CUBIT is an open source multi-touch system designed by Stefan Hechenberger and Addie Wagenknecht for NOR_/D. It was developed to "demystify multitouch" technology through making its software and hardware
22:40:45 -!- tombom has quit (Quit: Leaving).
22:41:36 -!- charlls has quit (Read error: Connection reset by peer).
22:45:48 -!- zeotrope has quit (Read error: Operation timed out).
22:48:29 -!- zeotrope has joined.
22:53:01 -!- nooga has quit (Ping timeout: 264 seconds).
23:00:00 -!- nooga has joined.
23:14:26 -!- adam_d has quit (Quit: Leaving).
23:30:18 -!- BeholdMyGlory has quit (Remote host closed the connection).
23:54:37 -!- fax has quit (Quit: Lost terminal).
23:57:18 -!- coppro has joined.
23:57:53 -!- FireFly has quit (Quit: Leaving).
←2010-03-10 2010-03-11 2010-03-12→ ↑2010 ↑all