00:03:25 <Phantom_Hoover> <shachaf> is there another name for functions f such that f(0) = 0
00:04:53 -!- copumpkin has joined.
00:05:21 <Phantom_Hoover> i mean they're just a constant addition away from every other function
00:07:45 <Taneb> I am looking forward to the future
00:08:14 <Taneb> Phantom_Hoover, will you come to York to my "How to make an esolang without really trying" talk, wherein I make an esolang in front of a live audience?
00:08:30 <Bike> you make it sound like you were sawing a lady in half
00:08:51 <Taneb> Bike, I'll be doing that too (no I won't)
00:09:21 <Bike> behold, the amazing ngevdello and his dancing exec bits!
00:12:38 <Bike> "Hey VCs / lunatics / 1%ers, kristallnac.ht is available for registration"
00:15:28 <Taneb> Phantom_Hoover, you are the channel regular who is most likely to attend other than me
00:21:24 -!- nooodl has quit (Quit: Ik ga weg).
01:01:26 <Phantom_Hoover> ok i can actually make that provided i don't carelessly agree to do something more important
01:03:51 <shachaf> Taneb: have you considered new york instead
01:04:05 <Taneb> shachaf, call me old-fashioned, but I prefer this one
01:07:03 -!- Tod-Autojoined has changed nick to TodPunk.
01:09:56 -!- Phantom_Hoover has quit (Remote host closed the connection).
01:27:17 -!- yorick has quit (Read error: Connection reset by peer).
01:35:52 -!- Sgeo has joined.
01:53:27 -!- tromp has joined.
01:55:32 <kmc> Lena Dunham having her own unsympathetic character on her own TV show deliver an annoying multi-minute rant in praise of Gawker and Jezebel is probably the sickest burn I've seen on TV
01:58:49 -!- zzo38 has quit (Remote host closed the connection).
02:14:10 -!- namaskar_ has quit (Ping timeout: 245 seconds).
03:17:25 <kmc> shachaf: did you know the number of km in a mile is equal to φ to within 0.5%
03:17:35 <kmc> so there are approx fib(n+1) km in fib(n) miles
03:21:10 -!- zzo38 has joined.
03:32:15 -!- Sorella has quit (Quit: It is tiem!).
04:19:33 <shachaf> kmc: yep, that's my usual heuristic
04:26:14 <kmc> also did you know that Dunning and Kruger showed a *positive* correlation between perceived and actual ability?
04:26:17 <kmc> http://www.talyarkoni.org/blog/2010/07/07/what-the-dunning-kruger-effect-is-and-isnt/
04:29:01 <kmc> oh I think I learned that here
04:29:08 <kmc> in a conversation you were part of
04:29:10 <kmc> hooray for logs
04:41:48 <kmc> i,i the meta-dunning-kruger effect: people who make stronger claims about the dunning-kruger effect know less about it
04:57:52 <oklopol_> til: the dunning-kruger effect had to do with people called dunning and kruger
04:57:59 -!- oklopol_ has changed nick to oklopol.
05:02:25 -!- v^ has quit (Quit: http://i.imgur.com/MHuW96t.gif).
05:18:59 -!- tromp has quit (Remote host closed the connection).
05:19:32 -!- tromp has joined.
05:23:29 -!- tromp has quit (Ping timeout: 240 seconds).
05:29:03 -!- Bike has quit (Ping timeout: 252 seconds).
05:29:34 -!- Bike has joined.
06:05:03 <Bike> if anyone has gross code from vidya games, tell @ibogost
06:05:36 <kmc> this sounds like a job for ais523
06:17:47 -!- guardianx has joined.
06:23:19 -!- guardianx has quit (Ping timeout: 265 seconds).
06:23:55 -!- guardianx has joined.
06:24:00 -!- guardianx has quit (Excess Flood).
06:24:30 -!- guardianx has joined.
06:24:34 -!- guardianx has quit (Excess Flood).
06:27:12 -!- guardianx has joined.
06:29:11 -!- oerjan has joined.
06:37:32 <shachaf> should i go for longer passwords using only lowercase letters
06:38:32 <oerjan> <shachaf> oerjan: someone suggested that "relevant" could mean f(0) = 0 <-- hm, plausible
06:39:24 <oerjan> shachaf: some sites won't accept that, i hear
06:39:58 <shachaf> yes but you can add decoy symbols for those
06:40:17 <shachaf> oerjan: should an infinite series like sine be considered relevant
06:41:05 <oerjan> hm then you would want to expand the concept to analytic functions.
06:42:14 <zzo38> Do you know about 6502 unofficial opcodes?
06:43:06 <oerjan> shachaf: well if relevant is to mean something different from linear, it needs to apply to some non-affine functions.
06:44:22 <shachaf> well, f(x)=x^2 or something
06:44:45 <shachaf> does this apply to just any function between vector spaces such that f(0)=0 or do you need something more
06:45:22 <oerjan> hm oh if it's vector spaces it gets even weirder.
06:45:41 <shachaf> well wasn't that the point
06:45:59 <shachaf> since you were talking about how you couldn't generalize those polynomials to n-dimensional vector spaces
06:50:24 <shachaf> the phrase "origin-preserving" does appear in some places
06:50:30 <Bike> zzo38: i've heard of em
06:50:32 <shachaf> "any origin-preserving isometry is a linear map"
06:53:06 <oerjan> well isometry is stronger than affine
06:59:58 -!- namaskar_ has joined.
07:00:21 <HackEgo> 25146/15625 (exactly 1.609344)
07:01:37 <Bike> it's cool how on one level that's "less than 1% different" and on another it's "pointless coincidence"
07:02:19 <lambdabot> [0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,...
07:02:46 <lambdabot> [1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,17...
07:03:27 <oerjan> > zipWith(flip(/))`ap`tail$fix$(1:).scanl(+)1
07:03:28 <lambdabot> [1.0,2.0,1.5,1.6666666666666667,1.6,1.625,1.6153846153846154,1.6190476190476...
07:04:01 <oerjan> > drop 6$zipWith(flip(/))`ap`tail$fix$(1:).scanl(+)1
07:04:02 <lambdabot> [1.6153846153846154,1.619047619047619,1.6176470588235294,1.6181818181818182,...
07:04:49 <oerjan> ok there isn't even a much better fibonacci quotient.
07:05:42 <oerjan> a F(n+1)/F(n) that's a better approximation than phi itself.
07:06:06 <Bike> i thought it was exactly phi, in the limit
07:06:18 <Bike> oh, wait, i see, ok
07:06:22 <oerjan> Bike: i'm talking about the mile -> km conversion
07:06:33 <Bike> yeah i geti tnow
07:06:34 <oerjan> i suppose 1.6153846153846154 is the closest one
07:10:08 <oerjan> > let cf x | isInfinite x = [] | otherwise = f : cf (1/(x-f)) where f = fromIntegral$floor x in cf (25146/15625) :: Rational
07:10:09 <lambdabot> with `GHC.Real.Ratio GHC.Integer.Type.Integer'
07:10:26 <oerjan> > let cf x | isInfinite x = [] | otherwise = f : cf (1/(x-f)) where f = fromIntegral$floor x in cf (25146/15625 :: Rational)
07:10:27 <lambdabot> No instance for (GHC.Float.RealFloat GHC.Real.Rational)
07:10:27 <lambdabot> (GHC.Float.RealFloat GHC.Real.Rational)
07:11:01 <oerjan> > let cf x | f == x = [f] | otherwise = f : cf (1/(x-f)) where f = fromIntegral$floor x in cf (25146/15625 :: Rational)
07:11:02 <lambdabot> [1 % 1,1 % 1,1 % 1,1 % 1,1 % 1,3 % 1,1 % 1,2 % 1,7 % 1,1 % 1,1 % 1,15 % 1]
07:12:28 <oerjan> > let cf x | d == 0 = [f] | otherwise = f : cf (1/d) where f = floor x; d = x - fromIntegral f in cf (25146/15625 :: Rational)
07:12:29 <lambdabot> [1,1,1,1,1,3,1,2,7,1,1,15]
07:13:16 <oerjan> > 1+(1/(1+1/(1+1/(1+1/(1))))) :: Rational
07:13:39 <oerjan> > 1+(1/(1+1/(1+1/(1+1/(3))))) :: Rational
07:14:37 <oerjan> > 1+(1/(1+1/(1+1/(1+1/(1+1/3))))) :: Rational
07:14:58 <oerjan> > 1+(1/(1+1/(1+1/(1+1/(2))))) :: Rational
07:15:29 <shachaf> > iterate (\x->1+1/x) (1::Rational)
07:15:30 <lambdabot> [1 % 1,2 % 1,3 % 2,5 % 3,8 % 5,13 % 8,21 % 13,34 % 21,55 % 34,89 % 55,144 % ...
07:16:10 <shachaf> > map numerator $ iterate (\x->1/(1+x)) (1::Rational)
07:16:11 <lambdabot> [1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,17...
07:18:53 -!- Tritonio has quit (Ping timeout: 245 seconds).
07:18:53 <oerjan> > minimumBy (comparing $ \(m,n) -> abs $ m/n - 25146/15625)) [(m,n) | n <- [1..20], m <- [0..40]]
07:18:54 <lambdabot> <hint>:1:58: parse error on input `)'
07:18:57 <oklopol> anyone wanna whip up a haskell solution to counting the number of brainfuck programs with matching brackets without . and ,
07:19:10 <oerjan> > minimumBy (comparing $ \(m,n) -> abs $ m/n - 25146/15625) [(m,n) | n <- [1..20], m <- [0..40]]
07:19:32 <oerjan> > sortBy (comparing $ \(m,n) -> abs $ m/n - 25146/15625) [(m,n) | n <- [1..20], m <- [0..40]]
07:19:33 <lambdabot> [(29.0,18.0),(21.0,13.0),(8.0,5.0),(16.0,10.0),(24.0,15.0),(32.0,20.0),(13.0...
07:20:01 <oklopol> like p(n) = 4p(n-1) + sum_{i = 0}^{n-2} p(i) p(n-i-2)
07:20:13 <oklopol> can you do this with list magic
07:20:29 <oerjan> > sortBy (comparing $ \x -> abs $ x - 25146/15625) $ (sqrt 5 + 1)/2 : [m/n | n <- [1..20], m <- [0..40]]
07:20:31 <lambdabot> [1.6111111111111112,1.6153846153846154,1.618033988749895,1.6,1.6,1.6,1.6,1.6...
07:21:39 <oerjan> > sortBy (comparing $ \x -> abs $ x - 25146/15625) $ (sqrt 5 + 1)/2 : [m/n | n <- [1..50], m <- [0..100]]
07:21:40 <lambdabot> [1.6097560975609757,1.608695652173913,1.608695652173913,1.6111111111111112,1...
07:22:07 <oerjan> phi seems to do pretty well in the comparison.
07:22:53 -!- FreeFull has quit.
07:24:01 <oklopol> 220284348986588570109202836338028914428611773535981066182485818891641176575745435677614245375605705664256074439978888016219589144792715586214584933803945623110548077752330775651922506462688159269237015622883861450993208900404333264
07:28:18 <oerjan> i assume you are already memoizing already you couldn't possibly get to 5000
07:29:33 <oklopol> i also tried out a c bignum library, and got up to 50 or something
07:29:52 <oklopol> python give 1000 in a few seconds
07:30:23 <oklopol> (probably i just did something stupid because i sort of suck at c)
07:31:15 <oerjan> i'm pretty sure i've done catalan numbers as a list thing in haskell before but i cannot quite remember how i did that awkward sum.
07:31:55 <Bike> wheres the SymbolicCombinatoric module
07:32:09 <zzo38> Now I figured out on Astrolog when is Chinese New Year.
07:32:25 <oklopol> yeah this is the number of dyck words with extra symbols
07:32:31 <oerjan> zzo38: i think it has already been this year?
07:32:46 <oklopol> so sort of catalan numbers
07:33:14 <zzo38> oerjan: No. It is later (if the Chinese timezone is used, I think Jan.31)
07:37:22 <zzo38> I made up a diagram to show you how to calculate when is Chinese New Year (the diagram is not displaying the data for this year though): http://zzo38computer.org/img_14/chinese-new-year.png
07:42:07 <lambdabot> [[],[1],[1,2],[1,2,3],[1,2,3,4],[1,2,3,4,5],[1,2,3,4,5,6],[1,2,3,4,5,6,7],[1...
07:43:29 <oerjan> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p
07:43:30 <lambdabot> [1,4,17,76,354,1704,8421,42508,218318,1137400,5996938,31940792,171605956,928...
07:43:54 <oerjan> oklopol: are those correct
07:44:18 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p; p !! 20
07:44:19 <lambdabot> <hint>:1:90: parse error on input `;'
07:44:37 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 20
07:44:53 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 20
07:44:59 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 1000
07:45:00 <lambdabot> 1851699348848864557490460908252351435791528503792571199732642872987429667855...
07:45:15 <oklopol> 1851699348848864557490460908252351435791528503792571199732642872987429667855741056388118709506982301095273520239166445812465897225719714673518034688729346409566576494246766257017500088160821295706471401620743174768643242408712933258876221056914708216651234496955612303387531937013570229817305356203053671545992668489792589165268920056950337807083735642100968509541109361559746272395310718713159517363957691130968711391511344404455840470000377647503645773
07:45:16 <oklopol> 009181430981878971527595122036960829733406983371159554875983183416111068844077682114144155308181843356898547292029738655566107038367415868264323343437633171416678659424310520616791395366627206645082341823517027845344762094860602519075347723961239585700915967561030319441407129604596884739279594357333381658012980964032832
07:47:29 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 1000
07:47:31 <lambdabot> 1851699348848864557490460908252351435791528503792571199732642872987429667855...
07:47:43 <oerjan> it does look pretty fast
07:47:57 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in (p !! 1000 + p !! 1000 + p !! 1000 +p !! 1000 +p !! 1000)
07:48:00 <lambdabot> 9258496744244322787452304541261757178957642518962855998663214364937148339278...
07:48:08 <oerjan> although for real timing you'd want to compile it properly i assume
07:48:35 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 3000
07:48:42 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 2000
07:48:49 <oklopol> > let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in p !! 1500
07:48:51 <lambdabot> 1200860333331673794565127667310636780014056128877285462588690504707901944941...
07:49:01 <oerjan> `run echo 'main = let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in ' >Test.hs
07:49:25 <oerjan> `run echo 'main = let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in print $ p !! 1000' >Test.hs
07:49:44 <HackEgo> [1 of 1] Compiling Main ( Test.hs, Test.o ) \ \ Test.hs:1:83: \ Not in scope: `inits' \ Perhaps you meant `init' (imported from Prelude)
07:50:05 <oerjan> `run echo 'import Data.List(inits);main = let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in print $ p !! 1000' >Test.hs
07:50:34 <HackEgo> [1 of 1] Compiling Main ( Test.hs, Test.o ) \ Linking Test ...
07:50:58 <oerjan> i don't remember how to do timing
07:51:00 <Bike> `run time echo is it a shell command
07:51:02 <HackEgo> is it a shell command \ \ real0m0.002s \ user0m0.000s \ sys0m0.000s
07:51:31 <HackEgo> 18516993488488645574904609082523514357915285037925711997326428729874296678557410563881187095069823010952735202391664458124658972257197146735180346887293464095665764942467662570175000881608212957064714016207431747686432424087129332588762210569147082166512344969556123033875319370135702298173053562030536715459926684897925891652689200569503378070837356
07:51:37 <oerjan> `run time ./Test | tail -1
07:51:40 <HackEgo> 18516993488488645574904609082523514357915285037925711997326428729874296678557410563881187095069823010952735202391664458124658972257197146735180346887293464095665764942467662570175000881608212957064714016207431747686432424087129332588762210569147082166512344969556123033875319370135702298173053562030536715459926684897925891652689200569503378070837356
07:52:01 <Bike> output wants to be free
07:52:20 <oerjan> `run time ./Test >/dev/null | tail -1
07:52:24 <HackEgo> \ real0m2.075s \ user0m2.000s \ sys0m0.400s
07:53:30 <fizzie> Huh, what sort of character encoding is this. It looks like latin-1 except all non-ascii letters have the wrong case.
07:54:21 -!- guardianx has quit (Remote host closed the connection).
07:54:26 <zzo38> I don't know; I can't see it.
07:54:58 <fizzie> I'd share it but it's Top Secret.
07:55:15 <fizzie> Oh, it's not as simple as the wrong case.
07:55:27 <fizzie> There's a 0xc8 (latin-1 for È) for 'é', for example.
07:56:23 <fizzie> And 0x89 for ä, and 0x88 for ö. I suppose those might be enough to find it, but I don't know of a good tool to look up character sets by example.
07:56:44 -!- aergus has joined.
07:56:47 <oerjan> didn't someone put one in HackEgo
07:58:19 <fizzie> For context, it's a CSV export of people registered on a course from the University course administration system.
07:58:40 <zzo38> I don't know what character encoding it is.
07:59:09 <oerjan> @google ancient finnish encodings
07:59:10 <lambdabot> http://scratchpad.wikia.com/wiki/Character_Encoding_Recommendation_for_Languages
07:59:10 <lambdabot> Title: Character Encoding Recommendation for Languages at Scratchpad, the home of te...
07:59:51 <fizzie> It's not the seven-bit Finnish one, I know that much.
08:00:02 <fizzie> And it's not even a legacy system. Though I guess it could be built on one.
08:01:46 <fizzie> Bah, I'll just tr it; the only special characters in it seem to be é, ä and ö. But it's still strange.
08:02:21 <zzo38> Yes you can easily convert to Latin-1 or CP437 or whatever, if that is the only non-ASCII characters it includes
08:02:36 <zzo38> Although it would still help to know what character set it really is.
08:03:40 <int-e> good \'e\"a\"o morning
08:08:34 <fizzie> I tried iconv -l | while read cs; do if [ "$(printf '\x89\x88\xc8' | iconv -f $cs -t utf8)" = "äöé" ]; then echo $cs; fi; done 2>/dev/null but it found nothing. (And the same command with the latin-1 bytes does found all kinds of compatible character sets.)
08:17:30 -!- namaskar_ has quit (Ping timeout: 245 seconds).
08:18:09 -!- conehead has quit (Quit: Computer has gone to sleep.).
08:18:56 -!- Sgeo_ has joined.
08:20:46 -!- Jafet has quit (Ping timeout: 265 seconds).
08:21:09 -!- Jafet has joined.
08:21:15 -!- Sgeo has quit (Ping timeout: 265 seconds).
08:26:13 -!- Jafet has quit (Changing host).
08:26:13 -!- Jafet has joined.
08:26:29 -!- Jafet has left.
08:31:59 <Sgeo_> http://lisptips.com/post/31567007407/putting-the-r-in-repl
08:32:10 <Sgeo_> Should I be sad that the equivalent of this works in Racket?
08:35:25 -!- aergus has quit (Ping timeout: 272 seconds).
08:40:12 -!- MindlessDrone has joined.
08:54:07 <Sgeo_> At least the GOP finally has an alternative plan. It sucks, but at least it exists
08:54:07 <Sgeo_> http://thehill.com/blogs/healthwatch/health-reform-implementation/196504-gop-senators-push-obamacare-alternative
08:54:51 <Sgeo_> "The CARE Act would require insurers to offer policies to anyone who has proof of “continuous coverage,” along with protections for those who lose their health plans for any reason. But those with pre-existing conditions who fail to maintain continuous coverage at any time could be denied coverage."
08:55:31 <Taneb> One of my friends is learning Clojure and keeps asking for my help because I know Haskell and it's all functional programming, right? and I can mostly set him on the right path but I worry that I'm gonna tell him something that is not very Clojure-y
08:55:55 <Sgeo_> Hmm, is this continuous coverage starting sometime after the implementation date? If so, that would certainly fix the obvious idiocy, and some of the criticisms on Fark
08:57:04 <Sgeo_> Although it's still problematic, one mistake and you're trapped off insurance forever?
08:57:35 <Taneb> Now I am suddenly thinking of an FRP web framework
08:57:46 <Taneb> Could that work? I dunno, I guess probably yeah
08:58:10 <Taneb> Does it provide any significant advantage?
08:58:19 <Taneb> Should I have breakfast now? Yes
09:03:03 <oerjan> Sgeo_: sorry, do us europeans it'll look idiotic regardless. even obamacare. hth.
09:18:17 -!- ais523 has joined.
09:50:54 -!- itsy has quit (Quit: itsy).
10:32:19 <ion> http://www.newyorker.com/online/blogs/borowitzreport/2014/01/stephen-hawkings-blunder-on-black-holes-shows-danger-of-listening-to-scientists-says-bachmann.html
10:34:57 <ion> Yeah, it’s satire.
10:41:34 <elliott> it's... almost funny! almost.
10:54:29 -!- zzo38 has quit (Ping timeout: 240 seconds).
11:13:08 -!- `^_^v has quit (Read error: Connection reset by peer).
11:13:33 -!- namaskar_ has joined.
11:13:55 -!- `^_^v has joined.
11:46:55 <int-e> Ok, so Hawkings writes that black holes don't have an event horizon as such; no information is lost when matter crosses it, and it's precise boundary is fluctuating rather than a sphere.
11:47:51 <int-e> I doubt that the "blunder" is an actual quote, having found no reliably looking source for it.
11:48:14 <int-e> reliable looking? grammar is hard.
11:53:03 -!- namaskar_ has quit (Ping timeout: 245 seconds).
11:58:32 <ais523> int-e: I think the story is that Hawking spent years of his life trying to work out where information went when it fell into a black hole, and then eventually concluded that the problem didn't exist in the first place
12:00:17 -!- aergus has joined.
12:00:51 -!- itsy has quit (Ping timeout: 260 seconds).
12:03:12 <int-e> So he does write that there are no black holes ... in a specific sense: "The absence of event horizons mean that there are no black holes - in the sense of regimes from which light can’t escape to infinity."
12:07:56 <int-e> ais523: Right. But that doesn't sound like mainstream news to me.
12:15:16 <ais523> int-e: thus, it's satire
12:22:41 -!- oerjan has quit (Quit: leaving).
12:25:16 <ais523> quintopia: actually, complex numbers can be totally ordered via alternating digits in the decimal expansion
12:25:21 <ais523> between the real and imaginary part
12:25:31 <ais523> it's a stupid total ordering, but it obeys all the requirements to be a total ordering
12:35:01 <fizzie> It's a total ordering, but it doesn't make complex numbers an ordered field, right? (I can't figure out what that was a response to.)
12:37:51 <ais523> yeah, it doesn't add or multiply properly
12:38:00 <ais523> and it was a response to an edit summary on the wiki
12:38:19 <fizzie> Oh, that explains why grepping the logs wasn't so useful.
12:39:55 <ais523> I don't normally consider myself bound by communication medium when quoting or replying to something
12:41:44 <fizzie> I'm tempted to send an email containing only "That's fine."
12:42:44 <ais523> seems like a waste of time for all parties concerned :-)
12:47:33 <ais523> now I'm wondering how and whether you know my email address
12:47:43 <ais523> it's certainly believable that you know it, given that it's public information
12:50:27 -!- namaskar_ has joined.
12:53:03 -!- boily has joined.
13:02:48 -!- Sgeo_ has quit (Read error: Connection reset by peer).
13:04:18 -!- namaskar_ has quit (Ping timeout: 245 seconds).
13:04:18 <int-e> fizzie: in an ordered field, one has x^2 >= 0 for any x, but i^2 = -1 < 0 contradicts that.
13:08:45 -!- metasepia has joined.
15:57:44 -!- esowiki has joined.
15:57:47 -!- glogbot has joined.
15:57:47 -!- HackEgo has joined.
15:57:49 -!- esowiki has joined.
15:57:49 -!- EgoBot has joined.
15:57:49 -!- esowiki has joined.
15:58:00 -!- ^v has joined.
15:59:36 -!- Gregor has joined.
16:00:18 -!- glogbackup has joined.
16:00:37 -!- nooodl has joined.
16:03:29 <HackEgo> A racoonspirator is a collaborator wrapped in fur
16:04:01 <HackEgo> A cocoonspirator is a collaborator wrapped in caterpillar silk
16:12:09 <quintopia> ais523: my bad. i stand by my correction that Por gammer was mistaken speaking of a "maximum" complex number though.
16:12:28 <ais523> quintopia: yeah, I was complaining about the summary, not the edit
16:13:16 <ais523> was this about complex integers, or complex numbers in general, btw? if it's complex numbers in general you have the issue of precision, too
16:13:41 <ais523> (and computability, although that's unlikely to come up while programming)
16:13:57 <quintopia> ais523: Por gammer specified double precision, but my implementation is multiprecision
16:14:35 <boily> what is a gammer, and why does it gam pors?
16:14:55 <ais523> to me, "double precision" implies an actual maximum value (or in the case of complex numbers, maximum for each component)
16:15:21 -!- FreeFull has joined.
16:15:23 <ais523> because it specifies a number of bits for the exponent as well as the mantissa
16:16:41 <quintopia> this is how georgia feels today: http://theoatmeal.com/comics/log_out
16:17:08 <quintopia> ais523: which is why i ignored the that recommendation and went multiprecision
16:17:13 <metasepia> KATL 281612Z 33010KT 9SM -SN BKN028 BKN036 OVC050 M03/M15 A3019 RMK AO2 P0000
16:17:35 <quintopia> boily: i don't mean the weather. i mean the attitudes of georgians
16:18:20 <ais523> quintopia: that link looks vaguely like an XSRF attempt to me
16:19:59 <ais523> I don't think it's possible to write a client-side browser plugin like NoScript to block XSRF without strong AI, right? you can do it server-side without issues though
16:30:36 <quintopia> but NoScript does actually help against some attacks
16:31:24 <int-e> yeah, for example it helps against certain (intended!) credit card transactions ...
16:31:31 <quintopia> the wikipedia page for CSRF lists a bunch of partial client-side solutions
16:32:39 <ais523> quintopia: hmm, I thought the HTML community decided to use "X" for "cross" to reduce acronym colisions
16:32:55 <ais523> but maybe they don't for CSRF because it has no collisions anyway
16:33:13 <int-e> (There's something called "3Dsecure" that's popular in Austria at least; basically a shop makes a cross site request to the credit card companie's "secure" server, and that one does another when the transaction is done. There's of course there's Javascript involved, too.)
16:35:20 <boily> uhm. multiple points of failure, multiple parties involved, unsecured information channels... not only it sounds dumb, but dangerous too.
16:36:19 <int-e> It's ass covering by the CC company; they have logs to show that a certain transaction was acknowledged on their own servers. They don't care that I can't distinguish between entering the data on their server or on another one.
16:36:50 -!- sebbu has quit (Ping timeout: 264 seconds).
16:37:02 <int-e> But in any case I have to jump through those hoops from time to time.
16:38:32 <ais523> int-e: in the UK, it's usual (but not universal) for the merchant's site to redirect to the credit card company's site
16:38:51 <ais523> who can either approve it immediately and redirect back, or ask for some extra information for security purposes
16:38:57 <ais523> I'm not sure how it works behind the scenes
16:39:06 <FireFly> Some swedish banks allow one to configure a personal secret, and have that secret shown on all "login-to-bank" pages to authenticate when paying for something or so
16:41:37 <Bike> my credit union has that, but only for logins into itself, since card transactions are through visa. :V
16:42:33 <ais523> FireFly: to prove it's a real bank screen?
16:42:42 -!- Slereahphone has joined.
16:42:47 -!- sebbu has joined.
16:42:49 <ais523> Bike: that doesn't actually work, though, because a malicious website could just try to log in as you
16:42:56 <ais523> it would fail, but it'd be able to get the secret before it failed
16:44:05 <int-e> ais523: well the idea is that before one enters their own data one verifies that the secret being displayed is correct.
16:44:21 <int-e> which just makes the attacker's job a bit more complicated but never mind that.
16:44:56 <int-e> the real question isn't whether it's secure, but whether a court can be convinced that it's secure.
16:44:58 <ais523> int-e: yeah, the problem is that you need the secret for the user to log in, but you need the login details to protect the secret
16:45:32 <ais523> doing it on payments works a bit better, though, because an attacker would have to submit a fake payment to get at the secret
16:46:00 <ais523> and if you do that in any sort of volume, the bank will get suspicious, because it's observable from their point of view and something that needs monitoring anyway
16:46:47 <int-e> It's not completely stupid. But it's very annoying.
16:59:16 <quintopia> ais523: one could have a rotating secret that is never the same twice. like "the 4th digit of your social and third letter of your mother's maiden name" today and "the last digit of your phone number and the third digit of your account number" tomorrow. if you ever see the same secret twice, you're being scammed
16:59:44 <ais523> quintopia: doesn't help against a MitM attack, which is what I was thinking of
16:59:54 <ais523> although this would be MitM-via-not-reading-the-URL
17:00:10 <ais523> as opposed to MitM-via-redirecting-the-connection
17:00:29 <quintopia> ais523: well, unless its MitM-by-rewriting-your-local-DNS-cache
17:01:07 <ais523> how does that attack work? I'm aware of the existence of DNS poisoning but don't understand it
17:01:11 <ais523> actually I'll just read Wikipedia
17:02:12 <ais523> oh, apparently it's just "redirect traffic aimed at a DNS server, then it gets cached"
17:02:17 <ais523> I was assuming it was something cleverer than that
17:02:26 <quintopia> i'm not sure how a MitM attack works when the information is securely transmitted
17:02:59 <ais523> it's when a DNS server asks you where a site is, you send back info on other sites too
17:03:34 <ais523> when DNS server A asks DNS server B where a site is, the attacker sends back a reply to A, forged to come from B
17:03:57 <ais523> you'd need to approximately guess the timings, but that's not a real obstacle to an exploit
17:04:37 <ais523> I guess this falls into the general class of TCP security issues that is "you trust the return address on a packet without asking the person who claims to have sent it whether it's genuine"
17:08:58 <quintopia> but then how did that attack that posioned the cache on an ADSL router via an emailed img link work?
17:10:27 <ais523> quintopia: I guess the img link was to a domain that the client hadn't seen before, thus forcing a DNS request to a server owned by the attacker
17:10:38 <ais523> from whatever the router's upstream DNS was
17:11:16 <ais523> and simultaneously giving the attacker the identity of the user (via something encoded in the subdomain, which their malicious DNS sever could see), the IP of the upstream DNS (it's making the request), and the timing of the request
17:11:55 <ais523> the email would also have to contain a link to an image on a legitimate site, but that's easy
17:12:19 <ais523> then if the malicious server knows the IP of the legit's site DNS server, it has all the information it needs to spoof it quite convincingly
17:15:20 <ais523> that actually is a really interesting attack
17:15:47 <ais523> the obvious fix is for DNS requests to contain an arbitrary string that the DNS server echos back in the response
17:20:13 <Taneb> This evening I had a rather boring lecture and ended up thinking about 0xn matrices
17:20:19 <Taneb> Conclusion: they work but are useless
17:31:03 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
17:32:22 <FreeFull> Taneb: A matrix that can't store data?
17:33:10 <Taneb> kmc, the fox was in York (well, just outside York)
17:33:22 <Taneb> It still knows its dimensions
17:34:23 <ais523> Taneb: foxes are quite common in Birmingham, although they frequently stay hidden, especially during the daytime
17:34:35 <ais523> I once had to take a long detour once because there was a fox in an alleyway in the middle of the day
17:34:46 <ais523> normally they just run away if you approach them, but they can be dangerous if cornered
17:35:00 <FreeFull> Taneb: What happens when you multiply a 4x0 matrix with a 0x4 matrix?
17:35:15 <Taneb> You get the 4x4 zero matrix, I think
17:35:49 <FreeFull> I wonder if these sort of flat matrices would be useful for matrix truncation/extension
17:36:00 -!- Slereahphone has joined.
17:36:35 -!- yorick has joined.
17:37:36 <ais523> sounds like it might be useful in linear-logic-like applications
17:37:56 <ais523> to have a safe and simple way to destroy data
17:38:24 <ais523> although simply multplying it by zero and adding it to something you were using anyway is probably easier
17:41:41 -!- conehead has joined.
17:43:42 -!- ais523 has quit.
17:45:36 -!- FreeFull has quit (Read error: Connection reset by peer).
17:46:00 -!- FreeFull has joined.
17:46:52 <FireFly> Taneb: a 0xn matrix could be represented in J & friends
17:52:55 <boily> what is a 0×N matrix good for, except some horrible type-level hacking?
17:54:34 <kmc> it's the zero matrix because the sum of an empty set is 0?
17:55:32 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
17:57:19 <Taneb> boily, probably nothing
18:03:17 <nooodl> you could use 0xN matrices to represent "width component for zero matrices" and Mx0 ones to represent "height component for zero matrices" and then when you multiply them you get a MxN zero matrix
18:03:57 <nooodl> that is probably the only interesting thing they do, really? destroying data isn't very useful
18:06:23 <kmc> does a 0xn matrix still represent a linear map? i guess so but none of them are invertible
18:06:28 <kmc> except a 0x0 matrix :D
18:06:29 -!- itsy has quit (Ping timeout: 240 seconds).
18:08:41 <boily> (meanwhile, I fungotting hate bureaucracy and micromanagement.)
18:08:41 <fungot> boily: i expect it's that or the uploads, i'll do it
18:11:22 <kmc> i had a dream about a block cipher mode which is like ECB but you use a Bloom filter to detect whether you're going to repeat a block and you re-key instead
18:11:26 <kmc> I don't think this is useful in the slightest
18:13:25 <boily> people here have the weirdest dreams...
18:15:58 <kmc> fungot: what do you dream
18:15:58 <fungot> kmc: i'm guessing you're about to be one for your implementation, usually all of r5rs on the sheet of paper"
18:16:35 -!- zzo38 has joined.
18:18:35 <FireFly> fungot dreams in lisp forms? . o O ( http://xkcd.com/224/ )
18:18:35 <fungot> FireFly: mmmm......... i am just using chess because the structure of the cond
18:18:48 <FireFly> fungot thinks in lisp forms.
18:18:49 <fungot> FireFly: should it bother me. :) but no, not at work. i still don't see the importance of critical thought and logic in an argument
18:19:59 -!- zzo38 has quit (Remote host closed the connection).
18:20:26 <boily> '(fungot fungot fungot)
18:20:27 <fungot> boily: did he do that one of the things most people who speak a different dialect that allows file i/ o
18:20:35 <fungot> boily: i like keyword arguments, and so you've now got the fnord.
18:20:52 * boily has the fnord! *zelda item get melody ♪
18:21:59 <FireFly> may the fnord be with you.
18:45:10 <quintopia> boily: guess what happens when it snows in atlanta?
18:45:56 <quintopia> boily: 1) billions of traffic accidents because tire chains aren't a thing here 2) i get the day off work yay
18:46:32 <boily> quintopia: the car pileups are previsible. we had one yesterday involving about 40 of them.
18:47:06 <quintopia> previsible? what? help boily i don't speak that.
18:47:26 <boily> predictable. sorry, my French's slipping again...
18:47:45 <boily> fr:prévisible → en:foreseeable, predictable.
18:48:39 <quintopia> boily: did you foresee my not working today? because i'm not.
18:49:48 <boily> the only people who skip their regular schedules when it snows are schoolchildren.
18:50:23 <boily> otherwise, well, you've got a coat, a hat, and a shovel.
18:50:56 <boily> quintopia: where do you work again?
18:51:26 <fizzie> MATLAB/Octave support 0xN and Nx0 matrices, too.
18:52:02 <quintopia> boily: well i am a tutor. schoolchildren are ostensibly my clients
18:52:15 <fizzie> Also the 0x0 matrix, which is handy because you can concatenate the 0x0 matrix to any matrix along any dimension.
18:53:10 <fizzie> (The 0xN and Nx0 matrices, for N > 0, you can concatenate only if the dimensions match right.)
18:54:30 -!- glogbackup has quit (Remote host closed the connection).
18:56:45 <boily> quintopia: good point.
18:58:17 <int-e> I like to know that a matrix has full rank without looking at its entries.
19:00:27 <quintopia> int-e: so you only use 0xN and Nx0 matrices
19:01:42 <int-e> Unfortunately, no. It would simplify a lot of things.
19:03:02 <quintopia> int-e: or wait, you could just generate your matrix randomly. then it will almost surely be full rank!
19:04:14 <int-e> quintopia: that's a dangerous claim. I might be working over a finite field.
19:05:57 <quintopia> int-e: if you have a matrix with all nonzero rows/columns you can add a random real nonce to one row and one column and get an almost surely full rank matrix
19:06:49 <int-e> quintopia: (1,1,1;1,1,1;1,1,1) and I like to differ.
19:06:56 <quintopia> obv this won't work if you have rows that are identically the same
19:07:05 <quintopia> but that's easy enoguh to guarantee
19:07:43 <int-e> (the nonce applied at a single position doesn't help ... much. you can add it to all diagonal entries though, that will work.)
19:20:21 <quintopia> int-e: also the set of 1xN and Nx1 matrices with at least one nonzero entry is all full rank!
19:21:20 <quintopia> you could just look at their length rather than their entires
19:25:17 -!- oerjan has joined.
19:27:49 -!- MindlessDrone has quit (Quit: MindlessDrone).
19:29:27 <oerjan> boily: what large numbers
19:30:25 <oerjan> he was enumerating IO-less brainfuck programs.
19:30:46 -!- oklopol has quit (Quit: Leaving).
19:34:39 <oerjan> @tell ais523 a + bi <= c + di iff a <= c or (a == c and b <= d) is a simpler total ordering. (the one you'd get from automatically deriving Ord in haskell, although the Complex numbers don't do that out of principle.)
19:34:58 <FireFly> `run time ./Test >/dev/null
19:35:01 <HackEgo> \ real0m2.273s \ user0m2.100s \ sys0m0.300s
19:35:15 <HackEgo> import Data.List(inits);main = let p = 1 : zipWith (+) (map (*4) p) (map (sum . zipWith (*) p . reverse) (inits p)) in print $ p !! 1000
19:35:23 <FireFly> I think time prints to stderr..
19:35:36 <oerjan> yeah i realized that later
19:42:09 <oerjan> <boily> what is a gammer, and why does it gam pors? <-- por gammer is por gammer for por spling, hth
19:42:50 <HackEgo> time: time [-p] pipeline \ Report time consumed by pipeline's execution. \ \ Execute PIPELINE and print a summary of the real time, user CPU time, \ and system CPU time spent executing PIPELINE when it terminates. \ \ Options: \ -pprint the timing summary in the portable Posix format \ \ The value of the TI
19:42:56 <int-e> /usr/bin/time tends to be more informative than the shell builting
19:43:11 <boily> oerjan: I unnerstande.
19:43:25 <FireFly> int-e: what would provide that?
19:44:23 <HackEgo> ls: cannot access /usr/bin/time: No such file or directory
19:44:35 -!- sebbu has quit (Ping timeout: 272 seconds).
19:49:34 <int-e> ok, so there's a 'time' package in debian ...
19:52:45 -!- sebbu has joined.
19:53:22 -!- sebbu has quit (Changing host).
19:53:22 -!- sebbu has joined.
19:53:35 <int-e> > /usr/bin/time /bin/true
19:53:35 <int-e> 0.00user 0.00system 0:00.01elapsed 0%CPU (0avgtext+0avgdata 392maxresident)k
19:53:35 <int-e> 56inputs+0outputs (1major+138minor)pagefaults 0swaps
19:54:03 <metasepia> LOWI 281920Z VRB01KT 9999 FEW010 SCT015 BKN070 M00/M01 Q0999 R08/19//70 NOSIG
19:54:43 <FireFly> Hm, that does seem to be a lot of info
19:55:05 <metasepia> ESSA 281950Z 10011KT CAVOK M02/M05 Q1026 R01L/410166 R08/410150 R01R/410177 NOSIG
19:58:35 <metasepia> ENVA 281950Z 08009KT 030V130 9999 FEW050 01/M09 Q1019 WS RWY 09 TEMPO 13022G35KT RMK WIND 670FT 16008G24KT
20:03:03 <oerjan> we could need some precipitation in middle norway. another disastrous fire going on.
20:09:49 <quintopia> you didn't have any gas explosions like manitoba did you?
20:12:58 <kmc> or germany http://www.bbc.co.uk/news/world-europe-25922514
20:14:09 <oerjan> quintopia: hm no, although it was apparently close last year when there was a fire in the local paper factory
20:14:40 <oerjan> there was a gas container close to the burning building.
20:16:43 <oerjan> kmc: cow farts, or bullshit?
20:43:42 <boily> cow farts are dangerous → http://www.darwinawards.com/legends/legends2000-06.html
20:58:04 -!- typeclassy_ has joined.
20:58:30 -!- typeclassy_ has changed nick to typeclassy.
21:01:29 -!- boily1 has joined.
21:01:35 -!- metasepia has quit (Ping timeout: 272 seconds).
21:01:49 -!- metasepia has joined.
21:02:45 -!- boily has quit (Ping timeout: 272 seconds).
21:03:03 -!- boily1 has changed nick to boily.
21:06:20 <HackEgo> olist (942): shachaf oerjan Sgeo FireFly boily
21:07:10 <boily> a day with an olist is a Good Day.
21:15:35 -!- ^v has quit (Remote host closed the connection).
21:15:53 -!- ^v has joined.
21:28:03 -!- boily has quit (Quit: ROYAL CHICKEN FURNACE).
21:28:04 -!- metasepia has quit (Remote host closed the connection).
21:54:23 -!- oerjan has quit (Quit: leaving).
22:17:35 <kmc> whaaaaaaaaat AMD is going to make server-class ARM64 "Opteron A" chips now?
22:17:37 <kmc> http://community.amd.com/community/amd-blogs/amd-business/blog/2014/01/28/amd-announces-plans-to-sample-64-bit-arm-opteron-a-seattle-processors
22:33:20 <olsner> sounds potentially awesome
22:37:10 -!- nooodl has quit (Quit: Ik ga weg).
23:07:31 -!- typeclassy has quit.
23:12:41 -!- Tritonio has joined.
23:18:15 -!- aergus has quit (Ping timeout: 252 seconds).
23:29:06 -!- Tritonio has quit (Quit: Tritonio).
23:37:25 -!- zzo38 has joined.
23:43:11 -!- Sgeo has joined.
23:45:21 <Sgeo> I think I'm more likely to trust that the Racket team can come up with good tools for the manipulation of languages themselves, and less likely to trust their ability to make "ordinary" programming work well
23:56:44 <zzo38> quintopia: I don't really know. It is open source and based on Android system; the former always being an advantage and the latter advantageous when you are writing or running Android based software; I think its other design decisions aren't all that good; apparently you can't remove HDCP without opening it up though
23:57:50 <quintopia> zzo38: would you consider writing games for it
23:58:23 <zzo38> quintopia: I don't think so, but if there are interpreters/emulators for whatever I do write a game for you can try to see if you are able to run those.