00:34:41 -!- hagb4rd has quit (Quit: hagb4rd).
00:36:22 -!- NihilistDandy has quit (Quit: leaving).
00:41:47 -!- carado has quit (Remote host closed the connection).
00:42:56 <Sgeo> Hmm, I guess REBEL's compose is effectively quasiquoting?
00:43:07 <Bike> and with a rebol yell
00:44:04 -!- carado has joined.
00:49:54 -!- Bike has quit (Ping timeout: 264 seconds).
00:50:09 -!- Bike has joined.
00:54:47 -!- SirCmpwn has quit (Excess Flood).
00:55:05 -!- SirCmpwn has joined.
01:07:33 <Sgeo> I think I really like the refinements thing
01:07:46 <Sgeo> Although it bears some similarities to keyword arguments in Racket or CL
01:07:48 <Phantom_Hoover> <cpressey> jsvine: Oh, you're good. Well -- I used to, but I'm not currently near a beach. I do enjoy long walks, though.
01:11:04 <Sgeo> "When I try to run REBOL on my Mac68K, it says I don't have enough memory. Why?"
01:13:42 -!- FreeFull has quit (Read error: Operation timed out).
01:15:44 <Sgeo> Found a bug in REBOL 3
01:15:49 <Sgeo> Unfortunately, I can't access chat
01:17:16 <Bike> what kind of uncreative question is that
01:17:27 <Sgeo> Phantom_Hoover, I didn't start writing a deadfish interpreter yet
01:17:45 <Phantom_Hoover> stop trying to learn every fucking language on the planet!
01:17:59 -!- FreeFull has joined.
01:18:15 <mnoqy> for what values of learn are we talking
01:19:21 <mnoqy> oh that biblical sense
01:19:36 <mnoqy> hm i dont think sgeo actually learns his languages that much?
01:19:51 <elliott> are there other biblical senses of learn
01:19:56 <Bike> does looking at a language constitute learning
01:20:02 <elliott> Sgeo: btw you missed the thingy
01:20:16 <mnoqy> see i had a good thing written up but then ph said a thing and i shortened it so i could respond quicker
01:20:25 <mnoqy> "eg im sure sgeos written a line of haskell but has he truly learned it? i put forward the claim that sgeo is merely practicing language tourism"
01:20:49 <Sgeo> elliott, I saw in the logs
01:20:59 <Bike> is the biblical sense of looking getting burned alive by a fire tornado?
01:23:29 <mnoqy> does anyone, truly
01:24:54 <zzo38> I have read the Bible, but it doesn't seem to help to answer your question.
01:25:19 <Phantom_Hoover> do you even know what know means in the biblical sense
01:27:33 -!- Frooxius_ has quit (Quit: ChatZilla 0.9.90-rdmsoft [XULRunner 1.9.0.17/2009122204]).
01:27:49 -!- Frooxius has joined.
01:33:15 -!- hagb4rd has joined.
01:34:53 <tswett> Coq abuse of the day: let's add axioms stating that every Type is a probability distribution!
01:38:40 <tswett> And then, uhh... no, don't do that at all.
01:38:52 <tswett> Instead, just add axioms stating that every Prop has a probability, and they obey the laws of probability theory.
01:39:18 <Sgeo> I ... think REBOL answers one of the things I dislike about Tcl.
01:40:11 <tswett> I wonder if you could do anything interesting with that. I would think that there would exist no (Prop, real number) pair, with the real number strictly between 0 and 1, such that you can prove that the probability of Prop is that number.
01:40:16 <tswett> Simply because that would be weird.
01:40:35 <tswett> Good thing I'm not talking about REBOL and Tcl, then.
01:40:46 <Sgeo> mnoqy, the compose function effectively acts as a quasiquoting mechanism
01:41:46 <Sgeo> So if you have a function that's accepting a block of literal values, you can use compose to easily dynamically make the data to be passed into it
01:42:03 <Sgeo> mnoqy, by having absolutely nothing to do with what a sane functional programmer would consider composing. The name does suck.
01:42:08 <Sgeo> http://www.rebol.com/r3/docs/functions/compose.html
01:43:18 <mnoqy> yeah i was thinking function composition
01:43:40 <Sgeo> Hmm, it's not quite adequate, though. I think. It does an automatic splicing thing, but automatic splicing skeeves me out. But disabling it leaves you with no form of splicing
01:43:49 <Bike> wow what does that have to do with composition at all
01:44:15 <Sgeo> What does composing a letter have to do with composition at all?
01:44:30 <Sgeo> (I still think the name sucks, despite my retort)
01:44:41 <Bike> it still looks weirdly used in these docs
01:45:27 <tswett> "If the result of an expression is a block, then the elements of that block are inserted into the output block:"
01:45:30 <tswett> That fact hurts a little bit.
01:45:56 <Sgeo> tswett, that can be disabled by using compose/only. But then you lose splicing altogether, I think.
01:45:56 <tswett> Okay, /only changes that. Whew.
01:48:20 <tswett> Sgeo: hey, lemme tell you about ultrafilters!
01:48:46 <tswett> Sgeo: suppose you have a set X. Then an ultrafilter on X is a collection U of subsets of X with all of the following properties:
01:49:04 <shachaf> Hey, lexande was telling me about ultrafilters the other day.
01:49:31 <tswett> The empty set is not a subset of U; given an element of U, all of its supersets are also in U; given two elements of U, their intersection is in U; and given a subset A of X, either A or the complement of A must be an element of U.
01:50:25 <tswett> Now, the obvious ultrafilters are the ones of the form { A : x is an element of A, A is a subset of X }, for elements x of X.
01:50:40 <Sgeo> All supersets of every element of U is in U? Does this include those supersets that contain elements that are not in X?
01:51:36 <Sgeo> Also "The empty set is not a subset of U" seems ... weird
01:51:45 <Sgeo> Actually, it seems impossible, come to think of it.
01:51:46 <tswett> Er, whoops. The empty set is not an *element* of U.
01:52:00 <tswett> Thinking about it for a while, it seems like these obvious ultrafilters (called "trivial ultrafilters") are probably the only ultrafilters that exist. If there are any other ultrafilters, something really weird must be going on.
01:53:08 <Sgeo> REBOL reduce also has nothing to do with functional reduce, but the name does make more sense than compose
01:53:13 <tswett> Indeed, I think it's possible to prove that it is impossible to define a non-trivial ultrafilter.
01:53:42 <tswett> However: in ZFC, every infinite set has non-trivial ultrafilters.
01:53:55 <zzo38> It says there is also free ultrafilters, according to Wikipedia, but not on finite sets.
01:54:24 <tswett> Right. Finite sets don't have non-trivial ultrafilters.
01:54:40 <tswett> ("Free" is a synonym of "non-trivial".)
01:55:58 <copumpkin> ultrafilter : ((x : A) -> Dec (P x)) -> (xs : [A]) -> ...
01:56:00 <Sgeo> REBOL has ... issues with mutability
01:56:12 <Sgeo> The copy function is essential to understand well, I think
01:57:14 <tswett> Come to think of it, maybe I can define a non-trivial ultrafilter on the integers?
01:58:13 <tswett> How about this: a set A is in U if and only if for every alternating string of 1s and 0s that ends with 1 (like "1010101"), there is an integer in A whose binary expansion ends with that string.
01:58:35 <Bike> shachaf: have you seen the Tao post
01:58:57 <tswett> Mm, no, that doesn't work. Consider the set A = {11, 1101, 110101, 11010101, ...} and the set B = {111, 11101, 1110101, 111010101, ...}. A and B are disjoint, but they are both elements of U.
01:59:35 <Bike> shachaf: http://terrytao.wordpress.com/2007/06/25/ultrafilters-nonstandard-analysis-and-epsilon-management/
01:59:57 <shachaf> The Tao that *can* be spoken of.
02:00:59 <Bike> It defines ultrafilters, and relates them to voting (and so arrow's theorem)
02:02:19 <Sgeo> do does [print "This is silly"]
02:02:34 <tswett> Hm. If you have an infinite set of voters, you can say that a bill passes if and only if the set of all yeas is an element of the ultrafilter.
02:02:54 <Sgeo> Static analysis of a bunch of REBOL code is probably harder than TCL
02:03:10 <tswett> Which has the curious property that no vote matters individually, under any circumstances.
02:03:39 <Sgeo> tswett, can a pair of votes matter?
02:03:52 <tswett> No finite set of votes matters.
02:04:13 <Sgeo> I guess that's easily seeable from no single vote ever mattering
02:04:34 <Sgeo> If a finite set mattered, it's possible that that finite set depends on 1 vote... unless the finite set is constrained in what it could look like
02:04:44 <Bike> i hear bachman is courting the Vitali set's vote, i wonder how that'll work out
02:05:44 <tswett> Like if you have three people, and the bill passes only if those three people all vote for it, but then it also happens that (for some reason) the three voters can't possibly all vote the same way.
02:07:49 <Sgeo> Not sure if I prefer REBOL refinements or keyword arguments. (Note that Smalltalk does not count as having keyword arguments)
02:17:20 -!- Bike_ has joined.
02:18:02 <elliott> Bike_: has anyone used "take a hike, Bike" on you yet
02:18:47 <Sgeo> I wish the whole 'so many datatypes that the parser just knows' wasn't considered such a maor selling point. Unless I can easily define my own, I'm not especially interested.
02:19:15 <shachaf> there must be 50 ways to leave your bicycle
02:19:43 <shachaf> i left my bicycle in washington :'(
02:20:30 -!- Bike has quit (Ping timeout: 264 seconds).
02:22:04 -!- Bike_ has quit (Ping timeout: 276 seconds).
02:22:33 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
02:23:05 <Sgeo> "Type secure [debug allow] or put it in your rebol.r file.
02:23:05 <Sgeo> Type secure none (but don't put it in your rebol.r file.)"
02:23:15 -!- Bike has joined.
02:23:23 <Jafet> Has anyone asked to ride Bike
02:26:44 * Sgeo wants to think through the differences between Factor and a hypothetical post-fixed REBOL
02:27:23 <Bike> but you can't, because you're ritually forbidden from thinking about computers on wednesdays
02:27:38 <shachaf> Bike: did you even see all the bicycle talk we had here while you were timing out
02:28:57 <Sgeo> Actually, I don't think there's a difference
02:30:59 <shachaf> Sgeo: when will you stop caring about languages
02:31:14 <Sgeo> shachaf, probably when I die.
02:31:31 <Sgeo> Or earlier is possible.
02:31:44 <Bike> i feel that sgeo should become a rebbe
02:32:13 <Bike> hardcore rabbi.
02:45:12 <Sgeo> Found an asymmetry: Can't have a function in REBOL return multiple values
02:47:41 <Sgeo> tswett, I may have found a use for compose without /only
02:54:10 <Sgeo> From a page on REBOL, this is valid code
02:54:11 <Sgeo> head insert tail block select list name
02:54:52 <Sgeo> :/ that's not really code to show people imo
02:56:16 -!- elieser224 has joined.
02:57:17 -!- elieser224 has left.
02:58:08 <Bike> it's just a bunch of words
03:01:26 -!- Bike_ has joined.
03:01:53 -!- Bike has quit (Disconnected by services).
03:01:58 -!- Bike_ has changed nick to Bike.
03:03:57 <Bike> hey elliott weren't you going to set up a bouncer for me or something
03:04:16 <elliott> no i just wanted to steal your password
03:11:15 <Sgeo> == $5.0000000000000000000000000
03:12:30 <Bike> floats are the future
03:12:44 * Sgeo was more concerned with the unit mangling
03:13:31 <Sgeo> "If I have $5 and see a bunch of $1 items, how many can I buy?" "FIVE DOLLARS!"
03:13:54 <Sgeo> I suck at examples
03:16:29 -!- Bike_ has joined.
03:16:54 -!- Bike_ has quit (Client Quit).
03:17:10 -!- Bike_ has joined.
03:17:22 <Bike_> ok so: sgeo does this thing use floats for currency
03:17:39 -!- Bike has quit (Disconnected by services).
03:17:44 -!- Bike_ has changed nick to Bike.
03:17:49 <Bike> does my connection use floats for currency? MAYBE
03:18:09 <Sgeo> "The money! datatype uses standard IEEE floating point numbers allowing up to 15 digits of precision including cents."
03:18:10 <shachaf> maybe it uses floats for concurrency
03:18:12 <Sgeo> Fuck you REBOL
03:18:25 <mnoqy> the money! datatype
03:18:45 <Sgeo> All datatype names are suffixed with !
03:18:56 <elliott> i wonder what comes next after rebol
03:19:01 <mnoqy> so what's the point of money!
03:19:04 -!- Bike_ has joined.
03:19:20 <shachaf> hey kmc landed a few minutes ago
03:19:52 -!- Bike has quit (Disconnected by services).
03:19:53 -!- Bike_ has changed nick to Bike.
03:22:49 <Sgeo> Oh hey Rebol3 wasn't upgraded since 2011
03:30:45 -!- Tritonio has quit (Ping timeout: 248 seconds).
03:31:50 <Sgeo> It's possible that the "we're using floating-point for money" thing is obsolete
03:32:12 <Sgeo> http://www.rebol.net/wiki/Money
03:32:49 <coppro> we totally need a bitcoin-based language
03:32:51 <Bike> money for scientific applications
03:34:39 <Sgeo> "Trembmax, unlike in R2, which you clearly used, in R3 is money! datatype BCD internally."
03:37:07 <mnoqy> what does that mean
03:38:22 <shachaf> hey mnoqy mind if i call you mnoqy!
03:38:33 <shachaf> you look like a data type to me
03:38:47 <shachaf> a data type interpreted as a proposition.......a true proposition
03:39:20 <mnoqy> who isn't, these days?
03:40:53 <Sgeo> mnoqy, binary-coded decimal, according to someone in another channel
03:41:34 <Bike> kick it like it's 1989
03:41:43 <Bike> or whenever x86 had that idk.
03:42:03 <mnoqy> binary-coded decimal. cute.
03:44:54 -!- sprocklem has joined.
03:48:09 <Sgeo> I... don't know what to think of the way apply interacts with refinements
03:48:19 <Sgeo> They're simply positional arguments
03:50:23 <Sgeo> "A scheme specifies a type of port. Schemes are defined for files, networking, sound, and many other functions. Users can define their own schemes or modify the existing schemes."
03:50:31 <tswett> So I heard once that accountants and whatnot actually need to use arbitrary-precision arithmetic.
03:50:31 <Sgeo> That last sentence makes me happy
03:51:10 <Sgeo> Ok, I misunderstood what schemes are
03:51:27 <shachaf> They're lisp dialects. There's a report for each one.
03:52:10 <tswett> So if someone has $10,000 in an account with 12% interest compounded monthly, then one year later, the balance on the account must be exactly $11,268.25030131969720661201.
03:52:59 <Sgeo> So, Rebol is still a bad idea in finance.
03:53:12 <Bike> i assume they meed computable numbers for continuously compounded interest
03:53:23 <shachaf> all numbers are computable
03:53:51 <tswett> Bike: I think computable numbers are sort of overkill there.
03:54:26 <tswett> Do they actually use continuous compounding somewhere?
03:56:24 <Sgeo> I like how this advocacy section lumps Lisp, Scheme, Haskell, and Caml together, and then proceeds to talk about Lisp
03:56:35 <Jafet> Calculus student loans.
03:56:37 <Sgeo> http://www.rebol.net/wiki/Advocacy#Lisp.2C_Scheme.2C_Haskell.2C_Caml
03:57:36 <mnoqy> well look at all those other section things though
03:58:07 -!- Bike has quit (Ping timeout: 264 seconds).
04:00:30 <Sgeo> I feel like I've mostly been mocking REBOL, despite being genuinely interested in it
04:03:25 -!- Bike has joined.
04:03:26 -!- Bike has left.
04:09:25 -!- hagb4rd has quit (Quit: hagb4rd).
04:12:12 <tswett> Continue this sequence: 1, 1, 1, 1, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14, 14, 15, 15, 16, 16, 17, 17, 18, 18, 81, ...
04:12:17 <tswett> Or, rather, unfair problem.
04:12:24 <Sgeo> -2834198237481273409
04:12:34 <tswett> Nope. It's a strictly increasing sequence.
04:12:43 <tswett> And all of its elements are integers.
04:13:20 <tswett> I'm not actually sure that that part of the sequence is correct. Pretty sure, but not completely sure.
04:13:21 <Sgeo> ERR COUNTABLY INFINITE NUMBER OF CONTINUATIONS
04:13:43 <Sgeo> You said integers... oh, I meant for just the next number
04:14:01 <tswett> I'm pretty sure the next number is 81.
04:14:24 <Sgeo> Are there more infinite sequences of real numbers than there are real numbers?
04:14:31 <Jafet> Not strictly, strictly speaking.
04:15:34 <tswett> The numbers after that... probably 256 and 256. Then probably 1024 and 1024. Then probably 4096 and 4096. Then probably 16,384 and 16,384, then probably 78,125 and 78,125.
04:15:42 <tswett> But I'm getting less and less sure with each number I name.
04:17:28 <tswett> I could write this sequence as a Haskell function, but that would probably take hours.
04:18:41 <tswett> I could implement Conway's chained arrow notation in Haskell, and that would only take a couple of minutes.
04:19:25 <tswett> But this sequence grows faster than any sequence concisely expressible using Conway's notation. Indeed, writing any sequence that grows as quickly as this one would also take me hours.
04:21:36 <Sgeo> Is your lack of certainty reflective of your own personal lack of certainty only, or of the lack of certainty of the mathematical community in general?
04:21:54 <tswett> My personal lack of certainty. The mathematical community could probably figure it out reasonably easily.
04:22:34 <tswett> Pretty sure this sequence isn't in OEIS.
04:23:21 -!- comex has quit (Remote host closed the connection).
04:24:14 <tswett> The nth term is the largest integer that can be written as a Church numeral in the calculus of constructions using at most n+9 symbols.
04:24:14 <Jafet> (http://oeis.org/A182369)
04:26:04 <tswett> Where a Church numeral is a term of the type forall o : T, (o -> o) -> (o -> o).
04:27:36 <tswett> The number 1 can be written in 10 symbols, so the first few terms are just 1. Any number n can be written in 13 + 2n symbols, so it stutters all the way through 18.
04:28:40 <tswett> The number a^b can always be written using (13 + 2a) + (13 + 2b) + 11 symbols, which explains the following few terms.
04:28:54 -!- Bike has joined.
04:30:36 <tswett> Presumably, as n increases, the "winning expression" will become nested exponentiation, then tetration, then pentation, then the Ackermann function, then Conway's chains, then variations on those, and eventually something we haven't imagined yet.
04:30:47 <Jafet> It's kind of like a busy beaver. A... colossal coq.
04:31:00 <tswett> Yeah. It's just the busy beaver sequence for the calculus of constructions.
04:31:11 <tswett> Which, unlike the busy beaver sequence for Turing machines, is computable.
04:34:08 <tswett> I guess after a while, it seems likely that the sequence would just use the fast-growing hierarchy directly.
04:35:01 -!- carado has quit (Ping timeout: 240 seconds).
04:36:07 <tswett> We can probably figure out where the sequence falls on the fast-growing hierarchy.
04:37:27 <tswett> Perhaps it corresponds to the proof theoretic ordinal of the calculus of constructions.
04:44:58 <coppro> I have discovered the identity of the biggest pervert in the world
04:45:05 <coppro> http://csclub.uwaterloo.ca/~scshunt/pervert.png
04:48:57 <tswett> Man, f_epsilon_0 on the fast-growing hierarchy grows pretty quickly.
04:50:25 <tswett> Let's just call it fe0. fe0(1) = 2, fe0(2) = 4, and fe0(3) is too large to have an upper bound concisely expressible using Conway chains.
04:52:54 -!- DHeadshot has quit (Ping timeout: 264 seconds).
04:54:07 -!- DHeadshot has joined.
04:54:14 <tswett> http://en.wikipedia.org/wiki/Fast-growing_hierarchy, and see http://en.wikipedia.org/wiki/Large_numbers#Systematically_creating_ever_faster_increasing_sequences for how to evaluate functions up to f_(omega^2).
05:03:12 <tswett> Huh. Apparently the Esperanto Wikipedia claims that "Lasero" comes from the phrase "Lum-Amplifo per Stimulata Eligo de Radioj".
05:06:54 <Jafet> Language architecture by stealing english rendition
05:09:40 -!- comex has joined.
05:13:06 <mnoqy> shachaf: btw you should smlist hth
05:13:19 <HackEgo> smlist: shachaf monqy elliott mnoqy
05:13:44 <Sgeo> Rebol is worse than Tcl in a way: Given a block of Rebol code, you can't tell where one expression ends and the next begins. With Tcl, you can do that much, even if you can't proceed to interpret code blocks inside of that block
05:15:46 <coppro> which is smlist again?
05:18:26 * Sgeo isn't entirely sure why either exists
05:18:30 <Sgeo> if/else is equivalent
05:20:10 <shachaf> mnoqy: 22:19 -!- ifeellikeabadperson whenitalkbynicking Nick change too fast. Please wait 20 seconds.
05:20:25 <mnoqy> have you tried waiting 20 seconds
05:20:41 <shachaf> where did you get that idea
05:20:52 <shachaf> are you some "kind of computer genius".........
05:23:56 -!- hagb4rd has joined.
05:45:05 -!- zzo38 has quit (Quit: zzo38).
05:53:18 -!- sprocklem has quit (Remote host closed the connection).
05:54:41 <kmc> how did it go
05:54:43 <hagb4rd> elliott: first pubic hair?
05:56:24 <elliott> it's generally considered bad to troll
05:56:59 <shachaf> kmc: welcome to california
05:57:00 <hagb4rd> you're right. i'm sorry. please go ahead
05:57:04 <shachaf> have you been to a california welcome center yet
05:57:24 <kmc> shachaf: that's the building that says "In-n-Out Burger" right
05:57:26 <kmc> in that case, yes
05:57:41 <shachaf> I suppose I can't argue with that.
05:58:10 <shachaf> Everyone said I should go there, so I went there. But there isn't much point to it when you're vegetarian.
05:58:57 <shachaf> mnoqy: are you expert in "In-n-Out Burger"
05:59:05 <mnoqy> i wouldn't call myself an expert
05:59:14 <mnoqy> but i know a few techs
05:59:21 <shachaf> expert means you went there in your life ever
05:59:54 <mnoqy> if you ask for "animal style" fries they put their goop on it, just like with their burger's
06:00:01 <hagb4rd> we say 2 experts ~ 3 opinions
06:00:05 <mnoqy> hm what other techs are there
06:01:10 <mnoqy> shachaf: that's the tech you just say "animal style" when ordering and that's the magic words
06:01:22 <shachaf> mnoqy: yes i meant teach me other techs
06:01:34 <mnoqy> i wonder if you can get an "animal style" shake that'd be hm
06:01:50 <shachaf> that doesn't sound delicious
06:02:12 <mnoqy> gosh im having trouble remembering techs
06:02:16 <mnoqy> oh i remember a tech
06:02:36 <mnoqy> i think if you ask for "grilled cheese" you get bread with cheese on it
06:02:46 <mnoqy> i've never done it myself but i have memories of witnessing it
06:03:03 <elliott> are we talking, like, untoasted bread here
06:03:15 <mnoqy> it's hamberger bun elliott
06:03:18 <mnoqy> im pretty sure they "toast" it
06:03:28 <elliott> i was really confused about california for a minute
06:03:41 <shachaf> elliott: we're talking a hamburger with cheese and without the meat
06:04:00 <shachaf> it's just a hamburger bun with cheese and maybe some vegetables
06:04:43 <fizzie> A grassburger is what you can get if you be vegetarian. It's like where you put some freshly-mowed lawn in place of the meat.
06:05:25 <shachaf> are you sure you don't put drugz instead
06:05:34 <kmc> shachaf: I'm in a hotel in MV
06:05:35 <mnoqy> i've seen lots of vegetarian burger alternatives
06:05:46 <mnoqy> veggy burger, soy berger
06:05:47 <kmc> shachaf: in-n-out Animal Style fries are good
06:05:49 <kmc> as are the shakes
06:05:56 <kmc> and they will make a 'grilled cheese' which is like a burger minus meat
06:06:11 <elliott> 07:02:36 <mnoqy> i think if you ask for "grilled cheese" you get bread with cheese on it
06:06:19 <elliott> 06:59:54 <mnoqy> if you ask for "animal style" fries they put their goop on it, just like with their burger's
06:06:23 <kmc> i got scoop'd
06:06:23 <elliott> way behind kmc, way behind
06:08:40 <kmc> so how did the interview go
06:09:55 <mnoqy> oh, that interview
06:10:00 <mnoqy> yeah it went swimmingly
06:10:06 <shachaf> kmc has some interviews this weekend.......
06:10:08 <mnoqy> he even left us a gift
06:10:13 <shachaf> or maybe tomorrow "who knows"
06:11:40 <mnoqy> but is it for wsj!!
06:12:14 <shachaf> no its for the csc(CASTRO STREET COMPANIES)
06:13:40 <shachaf> kmc: Were we going to meet in SF on Fri or something like that?
06:14:27 <kmc> shachaf: sounds good to me
06:14:52 -!- `^_^v has quit (Ping timeout: 256 seconds).
06:19:27 <hagb4rd> can somebody pick me please explain in few words whats that reporter-interview-wsj thing is about? i somehow missed the kickoff and it seems to keep you all excited these days
06:20:45 <kmc> shachaf: help how do i prepare for job interview
06:20:53 <fizzie> hagb4rd: http://catseye.tc/wsj.html <- good link
06:21:04 <fizzie> (If slightly outdated.)
06:21:04 <elliott> this answer was personalised to you
06:21:13 <kmc> elliott: thx
06:21:23 <fizzie> Was the extent of the personalization that "z"?
06:21:28 <kmc> elliott: they do say memory is state dependent so maybe if i have some bong hits I will remember some of the CS i learned while stoned
06:21:42 <elliott> they say memory is state dependent b/c we rely on the govt for everything
06:22:32 <shachaf> kmc: teach me when you find out plz thx
06:22:43 <elliott> kmc: who are you interview'sin'g with first
06:22:54 <shachaf> elliott: hey that was goin g to be my question
06:28:14 -!- comex has changed nick to iamthewolf.
06:31:33 -!- iamthewolf has changed nick to comex.
06:32:40 <shachaf> you could blackmail them or kidnap their children/pets
06:32:40 <kmc> flew into SFO today. got as far as Milbrae station before I smelled someone smoking pot and heard someone use the word "hella"
06:32:43 <kmc> god bless SF
06:36:20 <kmc> Sgeo, shachaf: I finished _The Tangled Web_
06:36:26 <kmc> good stuff
06:36:51 <shachaf> two exclamation marks? ok i m in
06:37:36 <kmc> i like that you can convince the user to manually change the character encoding of a page, which also changes the encoding used for a hidden iframe of an unrelated site, defeating escaping mechanisms on that site
06:39:14 <shachaf> kmc: What are your plans on Fri if any?
06:39:55 <kmc> no plans, you?
06:40:32 <shachaf> Ideally I'll get there sometime before 13:00 and try to do the passport thing. I might also meet $people but I'm not sure.
06:40:54 <kmc> also there's discussion in that book of a mechanism that's kind of like the NX bit for the web, and a workaround exploit strategy that's like ROP for the web
06:41:29 <kmc> some HTTP header that says which URLs can be used for <script> tags
06:41:31 <kmc> forgot the name
06:42:52 <shachaf> I guess I should read the book.
06:44:40 -!- Taneb has joined.
06:47:02 <kmc> also i read like a third of _Stories of Your Life and Others_
06:47:05 <kmc> good stuff
06:47:40 <shachaf> airplanes are nice for reading sometimes
06:48:00 <kmc> i have an e-reader now; i don't use it much at home but it's nice for traveling
06:48:04 <elliott> more like ehhhroplanes because i don't have any strong opinions on them
06:48:06 <kmc> i might use it more at home, now!
06:48:10 <kmc> elliott: thx for sharing
06:49:10 <shachaf> kmc: I wish I could enjoy carbonated things.
06:57:54 <Jafet> They built a new ehhkranoplan, but nobody cares.
06:58:23 -!- BillyZane has quit (Read error: Operation timed out).
07:04:45 <Jafet> “In order to prevent the occurrence of CFIT accidents, manufacturers and safety regulators developed terrain awareness and warning systems (TAWS). [...] In the case of Mount Salak Sukhoi Superjet 100 crash, the TAWS was working but the pilot intentionally turned it off.[9]”
07:08:19 -!- Bike has quit (Ping timeout: 264 seconds).
07:10:47 <kmc> BWOOP BWOOP PULL UP PULL UP
07:12:29 -!- BillyZane has joined.
07:16:28 * kmc → sleep, maybe
07:23:29 <hagb4rd> kmc: http://www.youtube.com/watch?v=SE0fhflzoHM
07:27:56 <hagb4rd> make sure to master the company song
07:46:37 <hagb4rd> this is amazing! i never realized zzo38 published THAT much work for the esolang community..our sneaky humble rascal of zzo38
07:48:31 -!- conehead has quit (Quit: Computer has gone to sleep.).
07:51:31 -!- epicmonkey has quit (Ping timeout: 264 seconds).
07:59:53 -!- ion has quit (Quit: Moving the server).
08:10:20 <fizzie> If you want to compliment a professional musician, can you say "don't quit your day job", or does it always sound like an insult?
08:11:58 <mnoqy> i think the "good sounding" form is "keep it up" or "keep up the good work" or "attaboy"
08:12:58 <Jafet> Keep it up, she said
08:13:25 <shachaf> mnoqy: are you "an attaboy kinda person"
08:13:31 <mnoqy> shachaf: no i was making a funny
08:13:37 <mnoqy> Jafet: who's this?
08:14:07 <mnoqy> maybe it has attagirl
08:14:13 <shachaf> (sry wn but u r u barely no english..................)
08:14:24 <lambdabot> *** "wordnet" wn "WordNet (r) 3.0 (2006)"
08:14:24 <lambdabot> n 1: any of the machine-readable lexical databases modeled after
08:14:24 <lambdabot> 2: a machine-readable lexical database organized by meanings;
08:14:26 <lambdabot> developed at Princeton University [syn: {WordNet}, {Princeton
08:14:50 <Jafet> mnoqy what is it with you and the tough questions
08:14:54 <fizzie> http://sprunge.us/XgOA compensationary WordNet atta's.
08:15:19 <Jafet> @wn compensationary
08:16:05 <fizzie> The best thing about WordNet are the hierarchies.
08:16:38 <fizzie> Aircraft carrier is a warship is a ship is a vessel is a craft is a vehicle is a conveyance is an instrumentality is an artifact is a whole is an object is a physical entity is an entity.
08:17:23 <fizzie> Some people do semantic similarity measures by counting path lengths in WordNet.
08:17:27 <Jafet> @wn instrumentality
08:17:28 <lambdabot> *** "instrumentality" wn "WordNet (r) 3.0 (2006)"
08:17:28 <lambdabot> n 1: a subsidiary organ of government created for a special
08:17:28 <lambdabot> purpose; "are the judicial instrumentalities of local
08:17:28 <lambdabot> governments adequate?"; "he studied the French
08:17:39 <lambdabot> instrumentalities for law enforcement"
08:17:39 <lambdabot> 2: the quality of being instrumental for some purpose
08:17:39 <lambdabot> 3: an artifact (or system of artifacts) that is instrumental in
08:17:39 <lambdabot> accomplishing some end [syn: {instrumentality},
08:17:46 <fizzie> Sense #3 in that tree.
08:18:20 <fizzie> All nouns derive from "entity", so you can compute the distance between any pair of nodes.
08:18:48 <fizzie> Of course that's not what it's meant for, and the height of the tree is very different for different kind of words.
08:19:37 <fizzie> And the verb trees is an entirely separate hierarchy, and is a lot flatter.
08:20:06 <fizzie> (Also doesn't have a shared root.)
08:21:25 <fizzie> (So I guess it's the verb forest instead.)
08:21:54 <fizzie> Verb forest is a popular vacation spot.
08:24:27 <fizzie> http://maraca.d.umn.edu/cgi-bin/similarity/similarity.cgi?word1=aircraft_carrier%23n%231&word2=blue%23n%231&measure=all&rootnode=yes so many measures
08:28:14 -!- Taneb has quit (Remote host closed the connection).
08:42:41 -!- heroux has quit (Ping timeout: 248 seconds).
08:44:45 -!- heroux has joined.
09:05:59 -!- epicmonkey has joined.
09:15:42 <fizzie> I just misspelled "aspell" as "aslepp". Apparently I need a spellcheck for running a spellcheck.
09:17:33 <elliott> fizzie: maybe your brain is telling you you should be aslepp
09:21:18 <fizzie> "invididual". I must've been aslepp also typing this.
09:27:26 -!- Jafet has quit (Quit: Leaving.).
09:45:44 -!- olsner has quit (Ping timeout: 260 seconds).
09:56:13 <fizzie> http://sprunge.us/FgUM this guy sounds frustrated.
09:59:08 <fizzie> It was copy-pasted from a typesetteded file.
09:59:21 <fizzie> http://www.ctan.org/tex-archive/macros/latex/contrib/IEEEtran/testflow/ that is.
10:06:43 -!- hagb4rd has quit (Quit: hagb4rd).
10:13:56 <FreeFull> fizzie: Think individual since you don't want to divide a person
10:30:01 -!- MindlessDrone has joined.
10:30:28 -!- mnoqy has quit (Quit: hello).
10:51:12 -!- Phantom_Hoover has joined.
11:59:05 <fizzie> It's the year 2013, and the ä in my supervisor's name got corrupted in IEEE's electronic copytheft form.
12:13:26 -!- hagb4rd has joined.
12:13:37 -!- asdfasd_ has joined.
12:23:10 -!- augur_ has changed nick to augur.
12:23:35 -!- Frooxius has quit (Read error: Connection reset by peer).
12:24:01 -!- Frooxius has joined.
12:39:24 -!- carado has joined.
12:53:06 -!- boily has joined.
12:55:59 -!- metasepia has joined.
13:03:34 <boily> bon matin tous! first day of the year without any coat, frock, vest, hoodie, nothing!
13:06:09 <fizzie> Shirt, pants, underwear... anything?
13:07:21 <elliott> don't question it too much. you don't want to truly know about quebec.
13:09:32 <fizzie> The guy who usually sits at the desk next to mine is currently at Vancouver, Canada; ICASSP 2013 is being arranged there.
13:11:05 <boily> fizzie: I had a tee, pants, shoes an my baseball cap. couldn't wear less; too much police presence lately.
13:17:53 <fizzie> They have +28 °C at friggin' Rovaniemi at the moment. That's, like, right at the arctic circle.
13:18:56 <metasepia> CYUL 301300Z 26011KT 15SM SCT025 SCT240 20/16 A3004 RMK CU3CI1 SLP171 DENSITY ALT 600FT
13:19:14 <metasepia> EFRO 301250Z AUTO 16005KT 070V220 CAVOK 26/M01 Q1018
13:19:28 <fizzie> Okay, that says 26; but Foreca said 28.
13:19:51 <boily> we're at 45N, and rovaniemi's at 66N, and it's warmer than here???
13:20:20 <fizzie> Weather: it makes no sense.
13:20:53 <fizzie> +30.7 is their highest-ever recorded temperature, they're kind of close.
13:21:12 <fizzie> (And -47.5 °C their lowest.)
13:24:37 <boily> rovaniemi's current relative humidity is at 16.9%. that's impossible here.
13:31:36 -!- `^_^v has joined.
13:36:45 -!- Koen_ has joined.
14:17:57 -!- Taneb has joined.
14:22:43 -!- Koen_ has quit (Quit: Koen_).
14:24:14 -!- TeruFSX has joined.
14:33:53 -!- oerjan has joined.
14:39:22 <oerjan> <boily> Taneb: let us not immediately jump to presumptuous conclusions. itidus may has become a very powerful lich for all we know here. <-- the idea of a lich with all of itidus's powers is rather horrible to contemplate hth
14:40:40 <oerjan> "sorry for that, i am not good with necromancy spells"
14:42:59 <oerjan> actually, maybe we should have a "which current or former #esoteric regular would be most scary as a lich" competition.
14:45:11 <oerjan> a swattable lich isn't scary, duh
14:46:32 <boily> what about a Canadian lich, eh?
14:50:03 <oerjan> obviously canadian liches don't exist.
14:50:38 -!- nooodl has joined.
14:54:12 -!- sampilix has joined.
14:58:23 -!- oerjan has quit (Quit: Laterë).
15:08:37 -!- sampilix has quit (Quit: Page closed).
15:25:38 -!- conehead has joined.
15:33:03 -!- Frooxius has quit (Read error: Connection reset by peer).
15:34:23 -!- epicmonkey has quit (Read error: Operation timed out).
15:52:37 -!- asdfasd_ has quit (Ping timeout: 250 seconds).
16:05:14 <kmc> perhaps Postel's Law and... whatever the opposite of Postel's Law is... are different equilibria of the same game
16:05:48 <kmc> if senders are mostly conformant, it's better for receivers (more secure, less effort) if they simply drop any malformed traffic
16:06:18 <kmc> but if senders are inconsistent and sloppy, receivers have to deal with that, and the first one to break compatibility is punished severely
16:06:33 <kmc> and naturally that means there's no pressure for senders to be less sloppy
16:06:46 <boily> kmc: I wouldn't say more secure. less effort of course, but what about corner cases, overflows, malformed packets, apocalyptic stuff, aliens?
16:07:27 <kmc> isn't it better to refuse to touch malformed packets, rather than making a wild-ass guess?
16:11:24 -!- jsvine has quit (Quit: Leaving.).
16:13:12 <kmc> for example there was a worm of sorts that crapflooded Freenode a while back
16:13:49 <kmc> that worked by having a webpage that would POST a form to the IRC server, which would ignore all the HTTP headers and execute IRC commands embedded in the form text
16:14:20 <kmc> it would be better if the IRCd just dropped any connection that starts with an unrecognized IRC command
16:14:27 <kmc> but if you do things like that, you lose users
16:14:55 <kmc> (I think the solution is to drop any connection that starts with "GET" or "POST" or possibly other HTTP verbs)
16:15:02 <kmc> (the one they actually implemented, I mean)
16:15:47 <kmc> you get similar problems if you have a network service on any port in your domain, that will echo text sent to it (e.g. in an error message)
16:16:32 <kmc> a browser can interpret the response as a HTTP/0.9 headerless response, and execute script tags embedded within
16:18:47 <Taneb> There will always be dodgy connections and solar radiation flipping bits
16:19:07 <Taneb> Perhaps a "I'm not sure I heard that right, could you send it again?" message
16:21:18 <kmc> but there will also be events that cut the connection entirely
16:21:22 <kmc> so robust clients will try again
16:21:36 <kmc> obviously this is a little different in situations where the message getting through is critical
16:21:50 <kmc> i'd argue you still want a rigid 'inner' packet and then error correcting codes around that
16:21:58 <kmc> haven't thought much about it though
16:22:00 <kmc> back later
16:30:45 -!- cpressey has joined.
16:31:27 <HackEgo> olist: shachaf oerjan Sgeo FireFly
16:42:56 <cpressey> <tswett> However: in ZFC, every infinite set has non-trivial ultrafilters. <- axiom of choice hth
16:43:41 <cpressey> actually that a complete guess on my part, i'm sure an actual mathematician will pummel me into paste with the Real Reason
16:47:01 <cpressey> <Sgeo> head insert tail block select list name <Sgeo> :/ that's not really code to show people imo <- right, we can't have that sort of thing going on CAN WE
16:51:25 <cpressey> @tell oerjan I just found a cupcake, is it yours? I assume not but I can't tell for sure
16:51:26 -!- cpressey has quit (Quit: leaving).
16:54:52 -!- jsvine has joined.
16:57:10 -!- epicmonkey has joined.
17:12:08 -!- zzo38 has joined.
17:16:12 -!- Taneb has quit (Remote host closed the connection).
17:25:06 <tswett> Yeah, the axiom of choice is what gives you non-trivial ultrafilters. In ZF, you can't prove that there are non-trivial ultrafilters.
17:32:39 <tswett> You know what, lemme stick this thing in the topic.
17:32:53 -!- tswett has set topic: ♪ LLVM, Z-machine, Nintendo Famicom, etc ♪ | http://underhanded.xcott.com/?page_id=5 | http://codu.org/logs/_esoteric | There exist Turing machines that halt in some models of ZFC, but not others..
17:37:40 <tswett> Turing proved that the problem of determining whether or not a Turing machine halts is undecidable. Thus, there must exist some Turing machine (call it T) such that there is neither a proof that it halts nor a proof that it does not halt.
17:38:01 <tswett> Gödel's completeness theorem states that in certain theories (including ZFC), if a statement is true in all models of the theory, then there is a proof of it.
17:38:27 <tswett> However, there is no proof that T halts, nor a proof that T does not halt. Therefore, it must not be true that T halts in all models of ZFC; and it must not be true that T fails to halt in all models of ZFC.
17:39:45 <coppro> you have two different definitions of undecideable there
17:39:50 <coppro> and you're conflating them
17:40:10 <tswett> I think the two definitions are equivalent in this case.
17:40:17 <tromp> > map (chr. (+96)) [23,1,12,20,5,18,19]
17:40:34 <tswett> The first is "there is no algorithm determining it", and the second is "there is no proof of it or its negation", right?
17:41:09 <tswett> If, for all Turing machines, there were either a proof that the machine halts, or a proof that the machine does not halt, then we would, by searching through all these proofs, be able to determine whether the machine halts or not.
17:41:26 <shachaf> FireFly: Hey, we should change `olist to print the argument.
17:41:39 <tswett> So the first undecidability implies the second undecidability.
17:41:46 <tswett> Except I feel like I'm doing something fallacious here.
17:42:41 <tswett> I'm assuming that if there's a proof that the machine halts, then the machine halts, and that if there's a proof that it doesn't halt, then it doesn't halt.
17:42:41 <boily> > [prec] <*> "uspnq;!xiz!xbmufst@"
17:42:48 <boily> > [pred] <*> "uspnq;!xiz!xbmufst@"
17:43:25 <tromp> i was looking at this tattoo http://blogs.discovermagazine.com/loom/files/2008/07/god-the-void.jpg
17:43:29 <tswett> It's certainly true that if a machine halts, then there is a proof that it halts. So if there's a proof that a machine doesn't halt, then it must not halt, because otherwise ZFC would be inconsistent.
17:43:52 <shachaf> FireFly: I assign you to do it. Just echo "$@" somewhere.
17:43:53 <tswett> What I'm not sure about is the proposition that if ZFC has a proof that a machine halts, then the machine halts.
17:44:01 <shachaf> FireFly: Also change `emptylist and `smlist etc., of course.
17:44:25 <HackEgo> echo -n "$(basename "$0"): "; tail -n+2 "$0" | xargs; exit
17:44:42 <boily> tromp: makes sense.
17:46:21 <FireFly> `run sed -i '1s/: /\$@: /' bin/olist
17:46:35 <HackEgo> echo -n "$(basename "$0")$@: "; tail -n+2 "$0" | xargs; exit
17:46:46 <zzo38> If ZFC has a proof that the machine halts, but it doesn't even though it runs forever, then either ZFC doesn't apply to Turing machines, or the machine is broken, isn't it?
17:46:55 <coppro> tswett: it is in fact possible to prove the incompleteness theorem through reduction to the halting problem
17:47:02 <coppro> through iteration of every proof
17:47:11 <FireFly> `run sed -i '1s/\$@/ &/' bin/olist
17:47:18 <HackEgo> echo -n "$(basename "$0") $@: "; tail -n+2 "$0" | xargs; exit
17:47:58 <tswett> zzo38: yeah, in this case, I would say that ZFC "doesn't apply" to Turing machines in the sense that ZFC makes a false statement about them.
17:48:20 <tswett> coppro: can you do that without assuming that if ZFC proves that a machine halts, then the machine halts?
17:48:42 <tswett> Anyway, I think Gödel's incompleteness theorem does actually prove my thing here.
17:49:03 <coppro> tswett: you're assuming consistency here
17:49:10 <coppro> which is entirely reasonable
17:49:27 <tswett> Gödel's incompleteness theorem implies that there must exist some Turing machine (call it T) such that there is neither a proof that it halts nor a proof that it does not halt[citation needed].
17:49:37 <tswett> Gödel's completeness theorem, blah blah blah.
17:49:54 <tswett> coppro: are you saying that consistency implies that if ZFC proves that a machine halts, then the machine halts?
17:49:58 <zzo38> If it halts, that is a proof that it halts. Isn't it?
17:50:12 <zzo38> Even though, the proof might be too long to write down.
17:50:21 <tswett> zzo38: yes, but we're talking about the other direction. If there's a proof that it halts, does that mean it halts?
17:50:22 <coppro> I mean, the definition of "halts" is constructed from ZFC
17:50:30 <tswett> coppro: well, can you prove that?
17:51:11 <coppro> tswett: sure. Let Delta be the transition function which takes a TM + tape + state and produces the next TM + tape + state
17:51:22 <zzo38> tswett: That must be true. If you are proving it using ZFC and yet it doesn't halt, then it means that ZFC doesn't prove that Turing machines halt or not.
17:51:50 <coppro> then halting is just saying that there exists n such that Delta^n(M, t, s) has a state of ACCEPT or REJECT
17:51:53 <tswett> zzo38: I think you might be conflating definitions of "proof" there.
17:52:07 <zzo38> tswett: O, maybe...
17:52:39 <tswett> I'm using "proof" in the sense of "sequence of statements such that each statement logically entails the next" or whatever.
17:53:22 <coppro> but we assume that ZFC is consistent, so it only proves things that are true in its models
17:54:07 <tswett> Or, well, that means that ZFC has models.
17:54:13 <zzo38> tswett: And yet, I think that is a way to prove that it halts, since it too, follows the sequence of steps, "proving" each state from its predecessor, which is the only possibility. So it does logically entails the next.
17:54:19 <coppro> sure, depending exactly on the definitions
17:54:30 <coppro> the important thing is that if we have a model, then anything ZFC proves is true in the model
17:55:13 <tswett> I'm suggesting that maybe there's a Turing machine that halts in every model of ZFC, and yet does not actually halt.
17:55:16 <tswett> But lemme think about that a moment.
17:55:36 <tswett> If, in every model of ZFC, the Turing machine halts, then, in every model of ZFC, there is some number of steps after which it halts.
17:56:21 <tswett> Now, that "number" can't be a standard integer, because if it were, then that model would just be wrong. Like, it couldn't exist, or it wouldn't actually be a model, or whatever.
17:56:33 <tswett> So it has to be a non-standard integer.
17:56:47 <zzo38> Then ZFC would not be applicable to Turing machines, therefore it doesn't actually prove that a Turing machine halts, I think.
17:57:07 <tswett> So the statement "there is a Turing machine that halts in every model of ZFC, and yet does not actually halt" implies the statement "every model of ZFC has non-standard integers".
17:57:33 <tswett> Which seems rather implausible, though I can't immediately think of a disproof.
17:59:26 <zzo38> Is it possible to prove what will halt in a standard number of steps, rather than only if it is non-standard?
18:02:56 <tswett> Well, "standard" can't be defined inside the system itself.
18:03:16 <tswett> The system thinks that all of its numbers are standard.
18:03:40 <zzo38> O, is that like the axiom of choice?
18:04:23 <tswett> Well, "non-standard number" means "object that satisfies the model's 'is a number' predicate, but is not a number".
18:05:16 -!- sprocklem has joined.
18:05:25 <zzo38> O, so that is what it means.
18:06:16 <boily> I declare 無 to be a number!
18:09:56 <boily> http://www.strangehorizons.com/2000/20001120/secret_number.shtml
18:11:12 -!- hagb4rd has quit (Quit: hagb4rd).
18:13:22 -!- hagb4rd has joined.
18:36:13 -!- sprocklem has quit (Remote host closed the connection).
18:38:16 -!- cpressey has joined.
18:42:14 -!- Frooxius has joined.
18:46:59 -!- sprocklem has joined.
18:50:56 -!- Bike has joined.
18:54:52 <zzo38> Here is a Z-machine interpreter I have written: http://zzo38computer.org/zmachine/fweep.zip But it is somewhat buggy; it won't accept "y" when it asks you to quit but it does accept "yes". Are you able to find the wrong thing?
18:58:45 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.3037
19:00:35 <HackEgo> 184) * oerjan considered buying lutefisk, but apparently it cannot be prepared in microwave </bachelor frog>
19:05:38 <cpressey> isn't lutefisk that one where they bury the shark in sand for three months before eating it so that the shark urine that poisoned the shark to death leaches out of it so you can eat it?
19:05:53 <cpressey> or is that some other disgusting maritime delicacy
19:08:55 -!- TeruFSX has quit (Ping timeout: 276 seconds).
19:11:37 -!- TeruFSX has joined.
19:13:31 <boily> cpressey: you're thinking about hakárl. lutefisk is only fish soaked in lye.
19:19:19 -!- TeruFSX has quit (Ping timeout: 252 seconds).
19:22:54 <boily> cpressey: oops. misaccented hákarl.
19:23:19 <FreeFull> Lutefisk and hakárl are nothing compared to surströmming
19:23:25 <boily> (when talking about fermented shark, you try to not make any mistake.)
19:23:53 <boily> FreeFull: all I know is that all of these have very pungent odours, from their wikipedia articles. I never experienced none of them with my nose, sadly.
19:24:30 <FreeFull> I thought lutefisk didn't have much of a smell at all
19:24:54 <FreeFull> Isn't it basically fish made into soap
19:25:21 <FreeFull> Oh, apparently it does smell strong
19:25:42 <FreeFull> Surströmming is by far the strongest though
19:25:53 <fizzie> http://sprunge.us/CUbD that was kinda funny.
19:28:37 -!- TeruFSX has joined.
19:30:51 <boily> FreeFull: I know that I do like nattō, so maybe any of them is better than what wikipédia implies.
19:33:21 -!- TeruFSX has quit (Ping timeout: 248 seconds).
19:33:49 <FreeFull> boily: surströmming's smell is worse than wikipedia implies
19:37:48 <boily> Santé Canada seems to have no notice or law concerning surströmming, so maybe it is legal for importation.
19:40:33 <boily> it seems like google is not friendly concerning that subject. the only relevant infos are from swedish tourism sites, and tourists who went there.
19:41:34 <boily> I want to buy surströmming!
19:49:10 -!- epicmonkey has quit (Ping timeout: 256 seconds).
19:53:24 -!- sprocklem has quit (Remote host closed the connection).
19:55:05 -!- zzo38 has quit (Remote host closed the connection).
19:59:47 <cpressey> Phantom_Hoover: I can't see the sea from where I am. I could walk to it, but it would take me a few hours (and I might end up at a cliff instead of a beach.) (Now you ask me: "Why not drive to the beach?")
20:02:37 <cpressey> <Phantom_Hoover> Why not drive to the beach?
20:02:52 <cpressey> HAVE YOU *SEEN* THESE ROADS???
20:03:09 <boily> are they worse than Québec roads?
20:03:45 <boily> FreeFull: I probably have found a potential hypothetic fine grocery store that may have surströmming. perhaps.
20:03:45 <cpressey> i've never been to ireland but YES THEY ARE WORSE
20:04:51 <cpressey> i may defy your command. i kind of want to see ireland and now it's nearby
20:07:01 -!- Taneb has joined.
20:11:41 <cpressey> that's not so different from here. perhaps unsurprisingly.
20:11:44 <FreeFull> boily: Never ever open it indoors
20:14:22 -!- Phantom_Hoover has quit (Remote host closed the connection).
20:14:33 <boily> FreeFull: that's what just about every source says. some even go as far as to tell you to open the can underwater.
20:16:45 <boily> I think I'll save a can for mid-July, and open it on the Mont Royal and see how long it takes for me to get beaten down by angry cops.
20:16:57 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:18:28 -!- Phantom_Hoover has joined.
20:22:54 -!- TeruFSX has joined.
20:49:41 -!- variable has quit (Ping timeout: 240 seconds).
20:52:16 -!- variable has joined.
20:53:57 -!- TeruFSX has quit (Ping timeout: 248 seconds).
21:05:23 <cpressey> http://catseye.tc/images/xyzzy/kola-kola-koski-flub-flub-flub.jpg
21:06:13 -!- boily has quit (Quit: Poulet!).
21:06:15 -!- metasepia has quit (Remote host closed the connection).
21:07:10 <shachaf> cpressey: why do you have a different 404 page for directories that exist and for files that aren't found, hth
21:12:24 <cpressey> shachaf because i'd rather spend my time drawing cartoons than debugging my htaccess sorry for your GRAND INCONVENIENCE hth
21:13:19 <shachaf> your apology doesn't cut it sorry hth
21:14:05 <cpressey> i'm not the one who decided to strip the filename off the url to see if i could get a directory listing hth
21:14:40 <shachaf> of course not, you have better ways to get directory listings
21:15:45 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.29378
21:17:02 <shachaf> cpressey: maybe i appreciated your cartoons so much that i wanted to see more hth
21:17:50 -!- sivoais has quit (Ping timeout: 256 seconds).
21:18:05 -!- sivoais has joined.
21:20:23 <cpressey> shachaf: if you really want to know, the only other file in that directory is a (scrawled) cartoon about the socks. i haven't been doing this very long, see, and i assume that at some point, i'll, like, organize them. but not yet
21:21:22 -!- Phantom__Hoover has joined.
21:22:42 <HackEgo> hth is help received from a hairy toe. It is not at all hambiguitous.
21:23:35 <cpressey> `learn htdh is help received from a hairy toe belonging to Don Ho.
21:23:59 <shachaf> <Rincewind> That doesn't work.
21:24:33 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
21:24:45 <cpressey> Rincewind: I realize that. wait, where are you
21:24:59 <shachaf> cpressey: wouldn't that be htdhh or dhhth hth
21:25:49 -!- sivoais has quit (Ping timeout: 252 seconds).
21:26:04 <cpressey> shachaf: fine, but what part does Rincewind play in this, whoever they are?
21:27:25 <HackEgo> shachaf sprø som selleri and cosplays Nepeta Leijon on weekends.
21:27:52 <Taneb> `learn Rincewind is a wizzard. He likes potatoes.
21:27:53 <fizzie> HtDH is also a well-known book titled How to Design Hogprams.
21:27:54 <HackEgo> not lately; try `seen Rincewind ever
21:28:06 -!- sivoais has joined.
21:28:16 <shachaf> fizzie: I thought it was "How to Design Hotdogs".
21:28:39 <fizzie> It could be that too. I know it's all about functional condiments.
21:31:10 -!- Jafet has joined.
21:32:16 <cpressey> remind me not to search things on the internet. now I know what Nepeta Leijon is.
21:33:44 <Taneb> Nepeta Leijon is a character in Homestuck.
21:33:50 <Taneb> The joke is shachaf hates Homestuck.
21:34:26 <Jafet> Metajoke: the joke makes about as much sense as homestuck
21:34:44 <cpressey> actually, nothing about "homestuck" came up in my search
21:34:49 <cpressey> i think the metajoke is the internet
21:35:25 <Taneb> cpressey, Homestuck is part of MS Paint Adventures
21:35:43 -!- sivoais has quit (Ping timeout: 252 seconds).
21:35:54 <Taneb> I'd give it a read, it's quite fun!
21:36:00 <cpressey> Taneb: thanks for the info walking away from computer now
21:36:26 -!- Bike has quit (Read error: Operation timed out).
21:38:07 -!- sivoais has joined.
21:38:52 -!- Bike has joined.
21:45:43 -!- sivoais has quit (Ping timeout: 240 seconds).
21:47:54 <shachaf> Taneb: I don't hate Homestuck!
21:48:08 -!- sivoais has joined.
21:51:00 -!- Koen_ has joined.
21:55:15 -!- mnoqy has joined.
21:55:50 -!- sivoais has quit (Ping timeout: 248 seconds).
21:56:02 <cpressey> well THAT was odd. i just saw a UFO!
21:56:25 <cpressey> meaning, i could not identify it. probably a perfectly mundane object, but i've never seen anything like it before
21:57:21 <HackEgo> cpressey invented the esolang, the pipe cleaner and the electrical mouse.
21:57:22 <cpressey> might have been some kind of hobbyist hot air baloon. translucent orange, shaped kind of like a shopping bag without handles, with a brighter yellow-orange spherical-ish light inside near the bottom (possibly a flame, hard to tell.)
21:58:09 -!- sivoais has joined.
21:59:17 <cpressey> and i have Taneb to thank for bringing up slightly repellent internet culture (not anything against it, it's just, one odd subculture is more than enough for me) encouraging me to go for a walk
21:59:52 <cpressey> know any good search terms i could use to get images of such a thing?
22:00:04 <Bike> a ghost lantern?
22:00:10 <Koen_> well, you could try "inverted shopping bag lantern"
22:00:26 <cpressey> "chinese lantern flying" -- yeah, it looked a lot like one of those
22:00:31 <Koen_> and hopefully you'll find a thread on a forum somewhere about someone saying he's seen a ufo
22:01:39 <shachaf> since when are you in that channel
22:02:02 <nooodl> i'm in #haskell quite often now that i have it for autojoin
22:02:29 <nooodl> i don't say stuff though, because i'm bad at haskell, rip
22:02:41 <shachaf> nooodl: Oh, you're the Belgian person.
22:02:57 <shachaf> I can stop accusing everyone else of being Belgian now.
22:02:58 <nooodl> that's anothertest :-)
22:03:34 <cpressey> yeah, it was totally a flying lantern. cool :)
22:06:22 -!- sivoais has quit (Ping timeout: 276 seconds).
22:06:42 <HackEgo> Phantom Michael Hoover is a true Scotsman and hatheist.
22:08:09 -!- sivoais has joined.
22:10:32 -!- nooodl has quit (Ping timeout: 256 seconds).
22:11:10 <HackEgo> Taneb is not elliott, no matter who you ask. He also isn't a rabbi although he has pretended in the past. (see also: d-modules)
22:11:23 <HackEgo> D-modules are just modules over the ring of differential operators. Taneb invented them.
22:15:11 <cpressey> wouldn't've even thought differential operators would necessarily form a ring, much less that there'd be modules over it
22:15:44 -!- sivoais has quit (Ping timeout: 240 seconds).
22:16:11 <HackEgo> elliott wrote this learn DB, and wrote or improved many of the other commands in this bot. He probably has done other things? He is also tire. And a lystrosaur.
22:18:08 -!- sivoais has joined.
22:19:19 <HackEgo> noooooodl is a stupid person
22:19:35 <Koen_> well that's not nice
22:19:57 <shachaf> Yes, but it's the same nooooooooooooooooooodl, not matter how many 'o's you write.
22:20:07 <Koen_> shachaf: but how can you do that??
22:20:31 <cpressey> it also capitalized the D in D-modules
22:20:32 <shachaf> No, `? doesn't have anything for this.
22:20:38 <shachaf> I mean it refers to the same person.
22:20:52 <shachaf> `run sed -i "s/is/ain't/" wisdom/nooodl
22:20:56 <Koen_> and added a "Michael" for Phantom_Hoover
22:21:14 <HackEgo> #!/bin/sh \ topic=$(echo "$1" | lowercase | sed "s/ *$//") \ topic1=$(echo "$topic" | sed "s/s$//") \ cd wisdom \ if [ \( "$topic1" = "ngevd" \) -a \( -e ngevd \) ]; \ then cat /dev/urandom; \ elif [ -e "$topic" ]; \ then cat "$topic" | rnooodl; \ elif [ -e "$topic1" ]; \ then cat "$topic1" | rnooodl; \ else echo "$1? ¯\
22:21:19 <cpressey> guessing you can specify what the noun it prints back is
22:21:35 <shachaf> (Well, that's most it is.)
22:21:41 <HackEgo> #!/bin/sh \ topic=$(echo "$1" | sed 's/ .*//' | lowercase) \ info=$(echo "$1" | sed 's/[^ ]* //') \ echo "$1" >"wisdom/$topic" \ echo "I knew that."
22:21:47 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/bin/%3F
22:22:11 <shachaf> cpressey: It just cats the file wisdom/blah! It's `learn that's the weird one.
22:22:17 <HackEgo> As the wisdom directory contains many files named after nicks, listing it in public annoys people. Try `pastewisdom instead.
22:22:31 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/wisdom/
22:22:41 <HackEgo> perl -pe 's/nooodl/"n@{[o x(3+rand 7)]}dl"/ge'
22:22:51 <HackEgo> noooooodl ain't a stupid person
22:22:51 <HackEgo> noooodl ain't a stupid person
22:22:52 <cpressey> shachaf: and yet, ?'s source is longer than learn's
22:22:52 <HackEgo> nooodl ain't a stupid person
22:24:18 <cpressey> hackbot filesystem is version controlled. how often does a commit happen?
22:24:30 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
22:24:46 <shachaf> That URL has a list of commits.
22:24:50 <cpressey> i guess, every time a command is executed
22:24:57 <shachaf> Every time a command changes something.
22:25:12 <HackEgo> Category theory is the theory of categories.
22:25:22 <HackEgo> Monoids are the easy version of categories.
22:25:24 <Koen_> well tha was helpful
22:25:26 <HackEgo> Monoids are just categories with a single object.
22:25:50 <shachaf> Bikes can't see red, right?
22:26:13 -!- sivoais has quit (Ping timeout: 248 seconds).
22:26:21 <cpressey> `learn Braid theory is the theory of braids.
22:26:23 <Koen_> why is shachaf sending blank lines?
22:26:53 <HackEgo> Braid theory is the theory of braids.
22:27:14 <cpressey> `run mv wisdom/braid wisdom/braid\ theory
22:27:39 <HackEgo> Braid theory is the theory of braids.
22:28:05 <HackEgo> cpressey invented the esolang, the pipe cleaner and the electrical mouse.
22:28:09 -!- sivoais has joined.
22:28:32 <shachaf> `run for f in wisdom/*; do echo "$f"; cat wisdom/"$f"; echo; done | paste
22:28:38 <shachaf> `run for f in wisdom/*; do echo "$f"; cat "$f"; echo; done | paste
22:29:12 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.26955 \ cat: wisdom/lystrosaur: Not a directory \ cat: wisdom/¯\(°_o): Is a directory
22:29:32 <HackEgo> cat: wisdom/wisdom/`?: Not a directory \ cat: wisdom/wisdom/?: Not a directory \ cat: wisdom/wisdom/??: Not a directory \ cat: wisdom/wisdom/☃: Not a directory \ cat: wisdom/wisdom/⌨: Not a directory \ cat: wisdom/wisdom/🐐: Not a directory \ cat: wisdom/wisdom/$1?: Not a directory \ cat: wisdom/wisdom/aah: Not a directory \ cat: wisdom/wisdo
22:30:52 <HackEgo> Homestuck is a cult religion for disaffected teens. Gamzee drives the bus. Best summarized by http://www.mspaintadventures.com/storyfiles/hs2/05743.gif
22:31:20 -!- myndzi\ has joined.
22:32:18 -!- myndzi has quit (Disconnected by services).
22:32:30 -!- myndzi\ has quit (Client Quit).
22:32:45 -!- myndzi has joined.
22:33:36 <cpressey> `learn HtDH is a classic text on How to Design Hotdogs or possibly Hogprams. It is all about functional condiments, and was co-authored by Herrence Tao and Don Ho.
22:33:57 -!- sivoais has quit (Read error: Connection reset by peer).
22:34:20 -!- shachaf has quit (Ping timeout: 256 seconds).
22:34:23 <cpressey> damn. curse you, Terence Tao, for spelling your name with only one r
22:34:28 -!- mnoqy has quit (Quit: hello).
22:34:38 -!- shachaf has joined.
22:35:19 <cpressey> `run sed -i "s/errenc/erenc/" wisdom/htdh
22:40:14 -!- sebbu2 has joined.
22:40:14 -!- sebbu2 has quit (Changing host).
22:40:14 -!- sebbu2 has joined.
22:40:27 <cpressey> `learn ZOMGMODULES is both a small blonde veterinarian and just modules over the ring of ZOMGs (which are not presently defined.)
22:41:11 <cpressey> `learn ZOMGMODULES is both a small blonde veterinarian and just modules over the ring of ZOMGs.
22:42:10 <cpressey> Phantom__Hoover: when you find something you can't define categorically please let me know hth
22:43:08 -!- sebbu has quit (Ping timeout: 256 seconds).
22:44:32 -!- ion has joined.
22:46:03 <cpressey> `learn welcome.bork Velcume-a tu zee interneshunel hoob fur isutereec prugremmeeng lungooege-a deseegn und depluyment! Fur mure-a inffurmeshun, check oooot oooor veeki: http://isulungs.oorg/veeki/Meeen_Pege-a. (Fur zee oozeer keend ooff isutereeca, try #isutereec oon irc.del.net.)
22:46:56 -!- augur has quit (Ping timeout: 260 seconds).
22:48:22 -!- myndzl has joined.
22:49:00 -!- cpressey has quit (Quit: leaving).
22:50:04 -!- myndzi has quit.
22:51:20 -!- myndzl has changed nick to myndzi.
22:51:33 <ion> It would be very unlikely for a bot to interpret / as a command character in the same way IRC clients for humans do.
22:52:45 <fizzie> It would be very likely for HackEgo to interpret / as a directory separator, though.
22:54:19 <HackEgo> bin \ dev \ etc \ hackenv \ home \ lib \ lib64 \ opt \ proc \ sbin \ sys \ tmp \ usr \ var
22:54:44 <FreeFull> `run cat /dev/urandom > /dev/kmem
22:54:45 <HackEgo> bash: /dev/kmem: Permission denied
23:02:35 -!- jsvine has quit (Quit: Leaving.).
23:05:33 -!- myndzi has quit (Quit: .).
23:07:14 -!- myndzi has joined.
23:29:03 <Phantom__Hoover> oh my GOD i wouldn't have said those things if i knew the PRESS were watching
23:29:05 <Bike> "I'll be lurking for a few days" -- jsvine or something
23:29:24 <HackEgo> Bike: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
23:35:15 <shachaf> Did you do your interview?
23:40:57 <kmc> went pretty well I think
23:47:24 * Sgeo needs to think about quasiquotation in REBOL
23:48:21 <Sgeo> "AFAIK, only REBOL is braindead enough to dynamically parse function text like m4 does!"
23:54:17 -!- Taneb has quit (Quit: Leaving).
23:54:19 <Sgeo> One point where Tcl is better than Rebol: Tcl values are immutable.
23:54:40 <kmc> shachaf: i should ask edwardk what he thinks of rust
23:54:50 <kmc> i think he told me at one point that human beings are not smart enough to use a language with affine types
23:56:57 -!- carado has quit (Quit: Leaving).