00:00:08 <oerjan> http://oldweb.cecm.sfu.ca/cgi-bin/isc/lookup?number=1.4331274267223117&lookup_type=simple
00:00:52 -!- alise_ has quit (*.net *.split).
00:00:52 -!- MizardX has quit (*.net *.split).
00:00:52 -!- Deewiant has quit (*.net *.split).
00:00:53 -!- alise_ has joined.
00:00:59 <Sgeo> !haskell foldl1 (\x y -> x + 1/y) [1..50]
00:01:00 -!- MizardX has joined.
00:01:00 -!- Deewiant has joined.
00:01:08 -!- MizardX has quit (*.net *.split).
00:01:09 -!- Deewiant has quit (*.net *.split).
00:01:18 <alise_> !haskell foldr1 (\x y -> x + 1/y) [1..50000]
00:01:21 <Sgeo> !haskell foldr1 (\x y -> x + 1/y) $ reverse [1..50]
00:01:35 <facsimile> !haskell foldr1 (flip (\x y -> x + 1/y)) $ reverse [1..50]
00:01:37 -!- MizardX has joined.
00:01:37 -!- Deewiant has joined.
00:01:59 <alise_> oerjan: what is that site anyway? it's like Plouffe's inverter but less comprehensive.
00:02:15 <facsimile> I USED plouffes inverter the other day!
00:02:32 <oerjan> oh, i think i was looking for plouffe's inverter, but google failed me then
00:02:41 <facsimile> I was trying to find a number expressed in terms of sqrt(13) and mysearch program didnt get it... but plouffes inverter did
00:02:49 <alise_> oerjan: it is a better result though
00:02:55 <alise_> because plouffe's had a ton of irrelevant crap
00:03:04 -!- alise_ has left (?).
00:03:07 -!- alise_ has joined.
00:04:08 <oerjan> alise_: ah well all the other hits are just more continued fraction stuff...
00:04:16 <alise_> I have this urge to write a program to approximate curves to write a little silly mathematical typesetter.
00:04:20 <alise_> oerjan: most false positives
00:04:28 <alise_> it does mark ContFrac(naturals) though
00:05:04 <alise_> Also, I have this urge to write code for some sort of embedded CPU to make an almighty CASulator.
00:05:15 <oerjan> ouch the links to OEIS are borken
00:05:41 <oerjan> alise_: i presume most of those sequences are things _starting_ with 1,2,3... for a good number of terms
00:08:52 <oerjan> oh the continued fraction for e isn't that simple, it has 1,1 between 2n terms
00:10:15 <oerjan> tanh(1) is close, except with only odd numbers
00:10:32 <alise_> now find one with only even numbers, and define the continued fraction interleaving operation
00:12:41 * oerjan realizes he wrote borken above without noticing
00:13:26 <oerjan> (tanh(1/n) and others are listed at http://en.wikipedia.org/wiki/Continued_fraction#Regular_patterns_in_continued_fractions)
00:14:24 <alise_> Hey, the Bessel function is what it was looked up as.
00:14:28 <oerjan> ah what the inverters gave were those bessel functions, wasn't i... right
00:16:09 <alise_> It's a nice number. I like it.
00:16:30 <alise_> I think, it being unnamed, I will name it Hird's constant. Unless I think of something better to give that name to in my time, in which case it won't be called that any more.
00:18:59 <oerjan> http://en.wikipedia.org/wiki/Khinchin%27s_constant
00:19:07 <oerjan> (no it's not the same one)
00:19:17 <oerjan> but i found it interesting
00:19:23 <alise_> Its symbol is \mathcal{H}.
00:20:37 <alise_> http://1.618034.com/blog_data/math/formula.5771.png
00:21:15 <nooga> nooga :: Time -> Object
00:21:27 <nooga> nooga saturday_night = beer
00:21:37 <alise_> doesn't work; saturday_night is a variable
00:22:03 <oerjan> you want SaturdayNight, obviously
00:22:25 <nooga> nooga SaturdayNight = Beer
00:23:25 <alise_> Generalised Hird constant:
00:23:41 <alise_> H_n = n + (1 / (n + 1 + (1 / n + 2 + (1 / (n + 3 + ...
00:24:22 <alise_> Sometimes I think I may be just that little bit /too/ clever.
00:26:11 <alise_> Why does Mathematica fail so hard that it won't even evaluate H numerically?
00:26:23 <alise_> In[6]:= N[ContinuedFractionK[k, {k, 0, Infinity}]]
00:26:23 <alise_> Out[6]= ContinuedFractionK[k, {k, 0., \[Infinity]}]
00:28:43 <alise_> RSolve[hird[n + 1] == n + 1 + (1/hird[n]), hird, n][[1]]
00:28:44 <alise_> Out[15]= Function[{n}, (BesselI[-1 + n, -2] + n BesselI[n, -2] +
00:28:44 <alise_> BesselK[-1 + n, 2] C[1] + n BesselK[n, 2] C[1])/(BesselI[n, -2] +
00:29:03 <oerjan> alise_: erm precedence fail in that H_n i think
00:29:30 <oerjan> presumably you want 1 / (, not (1 /
00:29:31 <alise_> yeah, missed a paren before (n + 2
00:30:23 <alise_> Hird[n_] := (BesselI[-1 + n, -2] + n BesselI[n, -2] +
00:30:23 <alise_> BesselK[-1 + n, 2] + n BesselK[n, 2])/(BesselI[n, -2] +
00:30:36 <oerjan> alise_: hm you realise H_n = n + 1/H_{n+1} ?
00:30:46 <facsimile> I'm reading a book about finite group theory
00:30:56 <alise_> RSolve[hird[n + 1] == n + 1 + (1/hird[n + 1]), hird, n][[1]]
00:30:56 <alise_> Out[25]= Function[{n}, 1/2 (-Sqrt[5 + 2 (-1 + n) + (-1 + n)^2] + n)]
00:31:00 <alise_> that is even nicer!!!!
00:31:15 <alise_> no silly Bessel nosiree just good ol' sqrts
00:31:37 <alise_> THAT IS NOT WHAT I WANTED
00:31:58 <alise_> I made another mistake lol
00:32:14 <alise_> ReplaceAll::reps: {hird[n]==n+1/hird[1+n]} is neither a list of replacement rules nor a valid dispatch table, and so cannot be used for replacing. >>
00:32:30 <alise_> OERJAN COME UP WITH A NICE EXPRESSION FOR IT THX
00:32:42 <alise_> when in doubt, consider a mathematician to be your computer.
00:33:05 <alise_> HEY it's a law, I named it so I can't do the work about it
00:33:20 <alise_> that responsibility falls unto /you/, who will be shafted by history
00:33:38 <oerjan> !haskell scanr1 (\n x -> n + 1/x) [1..50]
00:33:40 <EgoBot> [1.4331274267223117,2.30878937306624,3.2384534159000498,4.193691234094797,5.162856257658906,6.14038425280805,7.123306068860268,8.109900909526315,9.09910577000782,10.090229861703257,11.082805416334763,12.07650470540765,13.071091440344299,14.066391047318163,15.062271803120291,16.05863247717872,17.05539400888454,18.052493764882254,19.049881490555215,20.047516400759932,21.045365053054066,22.043399768719727,23.041597443938652,24.039938643221006,25.038406899
00:34:26 <oerjan> erm the last few terms are not exact
00:34:30 <oerjan> !haskell scanr1 (\n x -> n + 1/x) [1..100]
00:34:31 <EgoBot> [1.4331274267223117,2.30878937306624,3.2384534159000498,4.193691234094797,5.162856257658906,6.14038425280805,7.123306068860268,8.109900909526315,9.09910577000782,10.090229861703257,11.082805416334763,12.07650470540765,13.071091440344299,14.066391047318163,15.062271803120291,16.05863247717872,17.05539400888454,18.052493764882254,19.049881490555215,20.047516400759932,21.045365053054066,22.043399768719727,23.041597443938652,24.039938643221006,25.038406899
00:34:45 <alise_> is that the successive hird numbers?
00:34:51 <nooga> !sadol !=$3000$3000
00:34:54 <alise_> if so, may i note that you forgot H_0
00:34:56 <oerjan> the last one just looked truncated
00:35:07 <oerjan> !haskell scanr1 (\n x -> n + 1/x) [0..100] -- sheesh
00:35:08 <alise_> !haskell scanr1 (\n x -> n + 1/x) [0..100]
00:35:10 <EgoBot> [0.697774657964008,1.4331274267223117,2.30878937306624,3.2384534159000498,4.193691234094797,5.162856257658906,6.14038425280805,7.123306068860268,8.109900909526315,9.09910577000782,10.090229861703257,11.082805416334763,12.07650470540765,13.071091440344299,14.066391047318163,15.062271803120291,16.05863247717872,17.05539400888454,18.052493764882254,19.049881490555215,20.047516400759932,21.045365053054066,22.043399768719727,23.041597443938652,24.0399386432
00:35:10 <EgoBot> [0.697774657964008,1.4331274267223117,2.30878937306624,3.2384534159000498,4.193691234094797,5.162856257658906,6.14038425280805,7.123306068860268,8.109900909526315,9.09910577000782,10.090229861703257,11.082805416334763,12.07650470540765,13.071091440344299,14.066391047318163,15.062271803120291,16.05863247717872,17.05539400888454,18.052493764882254,19.049881490555215,20.047516400759932,21.045365053054066,22.043399768719727,23.041597443938652,24.0399386432
00:35:16 <alise_> !haskell scanr1 (\n x -> n + 1/x) [0..100] :: CReal
00:35:41 <alise_> egobot doesn't have it
00:35:41 <oerjan> i doubt that EgoBot has CReal
00:35:50 <alise_> <alise_> > scanr1 (\n x -> n + 1/x) [0..10] :: [CReal]
00:35:50 <alise_> <lambdabot> [0.6977746579640063874581706013218497243509,1.4331274267223150332875975712205324867433,2.3087893730662226112174662595471111580419,3.2384534159002329348068701459991873036222,4.19369123409157731148406
00:35:50 <alise_> <lambdabot> 60758323982791002,5.1628562577447335811648079306071871127633,6.1403842495719992391097584173482975080845,7.1233062330623306233062330623306233062331,8.1098901098901098901098901098901098901099,9.1,10.0]
00:35:59 <alise_> it seems i have found quite some nice numbers!
00:36:08 <oerjan> also, isn't CReal somewhat lazy, in which case you should be able to use [0..]
00:36:24 <alise_> H_1 = I_0(2)/I_1(2); H_2 = I_1(2)/I_2(2)
00:36:49 * alise_ chops off the first term again
00:37:11 <alise_> so H_(n+1) = I_n(2) / I_(n+1)(2)
00:37:20 <oerjan> or wait CReal cannot know that the list doesn't contain something enormous negative far beyond
00:37:31 <alise_> which, err, begs the question,
00:37:55 <alise_> I do not think there is I_-1
00:37:59 <nooga> !sadol :o:g:a0~n5?=#_4"+94SaturdayNight!"4Beer0 nooga "+94SaturdayNight
00:38:21 <alise_> I_-1(2)/I_0(2) = I_1(2)/I_0(2)
00:38:32 <alise_> so basically apply abs
00:38:57 <alise_> or, you know, you don't have to
00:39:19 <alise_> Out[37]= 500.001996000004
00:39:25 <alise_> Hird[n] is almost n as n gets big enough
00:40:04 <alise_> and if n is sufficiently big, 1/(n+1) is tiny
00:41:15 <nooga> could you write a function that finds any equilibrium of given list in haskell?
00:41:37 <alise_> Out[46]= BesselI[1, 2]/BesselI[2, 2]
00:41:37 <alise_> Out[47]= BesselI[2, 2]/BesselI[1, 2]
00:42:17 <alise_> H only clashes with harmonic numbers outside of typesetting, \mathcal ofc
00:43:55 <alise_> wait that -n conjecture is actually false.
00:45:05 -!- coppro has quit (Remote host closed the connection).
00:45:32 <nooga> i hate Bessel and Tchebyshev
00:45:34 <oerjan> <uorygl> Wow, it appears that over half of all Finns know English.
00:45:41 <alise_> Erm, heh, there are even real-indexed Hird numbers!
00:45:53 <oerjan> it's a mandatory school subject in norway, i presume it's the same in finland and sweden
00:46:04 <nooga> it's even the same in Poland
00:46:07 <oerjan> i think nowadays from first grade, even
00:46:21 <oerjan> (it was fourth grade back when i did it)
00:46:25 <alise_> there are even imaginary-indexed Hird numbers
00:46:26 -!- coppro has joined.
00:46:44 <nooga> yeah, but here they "teach" English instead of teaching it
00:46:46 <oerjan> nooga: what's an equilibrium?
00:47:52 * oerjan somehow avoided most of bessel and tchebyshev in his education, despite getting a math doctorate
00:48:00 <oerjan> unless i've just forgotten it
00:48:10 <uorygl> alise_: what is this Hird function?
00:48:13 <nooga> oerjan: electronics, filters
00:48:16 <alise_> uorygl: Not a function; Hird numbers.
00:48:30 <uorygl> What is this Hird numbers?
00:48:51 <alise_> It started with H = [1; 2, 3, 4, 5, ...] (continued fraction). Now I have successfully generalised it to H_n for natural n = [n; n+1, n+2, n+3, ...]; Mathematica decided to take that further. Definition:
00:49:19 <oerjan> nooga: i mean the answer is obviously yes if it's computable at all
00:49:27 <alise_> H_n = I(n-1, 2)/I(n, 2); I is the modified Bessel function of the first kind.
00:49:34 <alise_> Where I(a,b) = I_a(b).
00:49:38 <nooga> oerjan: equilibrium is an index i in array a that satisfies a[0..i] = a[i..last]
00:49:54 <alise_> Known properties: H_n gets close to n as n increases; H_-n = 1/H_(n+1).
00:50:07 <nooga> i'm just curious how it would look in haskell
00:50:12 <oerjan> nooga: what the heck is a[0..i] and a[i..last]
00:50:23 <alise_> uorygl: If you want to play along at home, and have a copy of Mathematica:
00:50:25 <alise_> Hird[n_] := BesselI[n - 1, 2]/BesselI[n, 2]
00:50:52 <alise_> uorygl: In LaTeX, write \mathcal{H}_n.
00:51:11 <oerjan> nooga: i'm sure i could write it if you could explain it, unless it involves fourier transforms or something like that (and even then, but i would probably be too lazy for that)
00:51:18 <uorygl> I don't even know if my university has Mathematics.
00:51:19 <nooga> oerjan: a[0]+a[1]+a[2]+...+a[i] = a[i]+a[i+1]+a[i+2]+...+a[n]
00:51:35 <alise_> uorygl: Well, you can also use Wolfram Alpha for most of it.
00:51:41 <alise_> http://1.618034.com/blog_data/math/formula.5801.png
00:51:47 <alise_> I did, of course, shamelessly name these after myself.
00:51:56 <oerjan> nooga: aha, so the sum to the left == sum to the right? but ther can only be one such index afaict
00:52:00 <alise_> Hird's constant/number is H_1.
00:52:12 <alise_> The whole H_n shenanigans is the Hird numbers.
00:52:18 <alise_> H_72 is the seventy-second Hird number.
00:52:35 <coppro> just out of curiosity; isn't the convention to let someone else name stuff after you?
00:53:14 <Sgeo> I'd imagine such "convention" has been broken before
00:53:36 <nooga> oerjan: usually i don't talk with doctors of mathematics at night ;D
00:53:37 <alise_> but fuck it, I'm never going to get another chance at getting something named after me
00:53:40 <alise_> so who can blame a man?
00:53:55 <Sgeo> alise_, so you DO like fame
00:53:58 <coppro> you're (arguably) in high school...
00:53:58 <alise_> not as if anyone /cares/ about these numbers, being as they are basically poor approximations to n
00:54:06 <alise_> Sgeo: of the mathematical sort, sure...
00:54:06 <coppro> you've got many years ahead of you
00:54:21 <alise_> I can call whatever else I discover the Hird2 number
00:54:21 <uorygl> n! is just a poor approximation to n^n.
00:54:24 <coppro> plus, doesn't this seem like a waste of a good name?
00:54:28 <alise_> uorygl: No it isn't :P
00:54:39 <alise_> coppro: it's not as if anyone will use this name
00:54:45 <alise_> this is more of a joke than anything
00:54:59 <uorygl> a+b is also a poor approximation to max(a,b).
00:55:11 <nooga> oerjan: btw. a[0..i].sum = a[i..-1].sum is Ruby-like notation
00:55:28 <alise_> uorygl: Everything is a poor approximation to either infinity or 0.
00:55:31 <nooga> i'm not sure if Ruby has Array#sum method by default
00:55:33 <oerjan> !haskell let equil l = elemIndex 0 (scanl (-) (sum l) l) in equil [1,5,3,-2,2,7]
00:55:42 <uorygl> Everything is a poor approximation to both. :P
00:55:59 <alise_> uorygl: Hey, hey, 0 is not a poor approximation to infinity, and infinity is not a poor approximation to 0.
00:56:11 <alise_> They are two ends of a circle, infinitesimally distanced; they are GOOD approximations of each other!
00:56:37 <uorygl> No, they're really pretty bad approximations of each other.
00:56:44 <uorygl> In what metric are they close?
00:56:52 <oerjan> !haskell let {equil l = elemIndex 0 (scanl (-) (sum l) l)} in equil [1,5,3,-2,2,7]
00:56:52 <nooga> alise_: oh crat, I could swear that i used it several times, maybe it was defined in my code somewhere
00:57:27 <alise_> uorygl: whatever extension of the reals has infinity but not -infinity, is defined as being a circle where 0 is one end and infinity is what you get at the other side or something
00:57:27 <oerjan> hm there must be an actual error there but damn if i can see it
00:57:30 <alise_> joined at both ends or something?
00:57:36 <alise_> I don't remember exactly
00:57:40 <alise_> but there's your metric
00:57:49 <alise_> take the naturals, tie both ends together
00:57:51 <alise_> there's your number-circle
00:57:57 <alise_> 0 and infinity are right next to each other, Q.E.D.
00:58:33 <oerjan> !haskell equil l = elemIndex 0 (scanl (-) (sum l) l); main = print $ equil [1,5,3,-2,2,7]
00:58:54 <oerjan> !haskell import Data.List; equil l = elemIndex 0 (scanl (-) (sum l) l); main = print $ equil [1,5,3,-2,2,7]
00:59:30 <alise_> http://en.wikipedia.org/wiki/Real_projective_line ;; so, drop the negatives, and 0 becomes right next to infinity
00:59:59 <nooga> i see problems with computational complexity of that thing
01:00:02 * Sgeo remembers reading about that in Flatterland
01:00:49 <Sgeo> Oh, and I sort of fantasized about something like a "number circle" when I was in my pseudomathematician days
01:01:22 <uorygl> Define infinity equal to e^(-pi*sqrt(163)). Boom, 0 and infinity are really close to each other.
01:01:45 <uorygl> But yeah, defining 0 and infinity to be right next to each other is a great way to get them to be right next to each other.
01:01:51 <alise_> In[90]:= Hird[Hird[Hird[Hird[n]]]]
01:01:51 <alise_> BesselI[-1 + BesselI[-1 + BesselI[-1 + n, 2]/BesselI[n, 2], 2]/
01:01:51 <alise_> BesselI[BesselI[-1 + n, 2]/BesselI[n, 2], 2], 2]/
01:01:51 <alise_> BesselI[BesselI[-1 + BesselI[-1 + n, 2]/BesselI[n, 2], 2]/
01:01:52 <alise_> BesselI[BesselI[-1 + n, 2]/BesselI[n, 2], 2], 2], 2]/BesselI[
01:01:54 <alise_> BesselI[-1 + BesselI[-1 + BesselI[-1 + n, 2]/BesselI[n, 2], 2]/
01:01:56 <alise_> BesselI[BesselI[-1 + n, 2]/BesselI[n, 2], 2], 2]/
01:01:58 <alise_> BesselI[BesselI[-1 + BesselI[-1 + n, 2]/BesselI[n, 2], 2]/
01:01:59 <nooga> abstract algebra -> interesting, esoteric, pain in the ass when you're studying
01:02:02 <alise_> BesselI[BesselI[-1 + n, 2]/BesselI[n, 2], 2], 2], 2]
01:02:32 <Sgeo> Hm, it does make sense that inf+inf is undefined
01:02:41 <alise_> uorygl: Well, if we consider the naturals as a line such that one end is 0 and the other end is infinity,
01:02:45 <Sgeo> But inf+inf != 2*inf
01:02:59 <alise_> Then a reasonable way to be able to answer questions about order on infinity would require some number to be said to be "after" infinity, I guess?
01:03:05 <alise_> I'm just looking for excuses.
01:03:14 <alise_> Sgeo: Simple answer: Add infinity, lose mathematical properties.
01:03:40 <nooga> infinity should behave like 0
01:06:22 <oerjan> !haskell import Data.List; equil l = elemIndex 0 $ zip (scanl (+) 0 l) (scanr (+) 0 l); main = print $ equil [1,5,3,-2,2,7]
01:06:44 <oerjan> !haskell import Data.List; equil l = elemIndex 0 $ zipWith (-) (scanl (+) 0 l) (scanr (+) 0 l); main = print $ equil [1,5,3,-2,2,7]
01:06:46 <uorygl> For infinity inspiration, we should look to projective geometries.
01:07:03 <uorygl> Extend y = x^2 to the real projective plane, and you'll see what infinity squared is.
01:07:25 <oerjan> !haskell import Data.List; equil l = elemIndex 0 $ zipWith (-) (scanl (+) 0 l) (tail $ scanr (+) 0 l); main = print $ equil [1,5,3,-2,2,7]
01:07:36 <uorygl> (Hint: it's infinity.)
01:08:43 <oerjan> hm i guess with negative numbers there _could_ be more than one spot
01:09:27 <oerjan> !haskell import Data.List; equil l = elemIndices 0 $ zipWith (-) (scanl (+) 0 l) (tail $ scanr (+) 0 l); main = print $ equil [1,5,3,-2,2,-2,2,7]
01:10:21 <oerjan> huh that's correct i think
01:11:01 <nooga> i think it is correct.... Also doctor of mathematics is solving my dumb, theoretical problem, hell yeah ;D
01:11:20 <nooga> i wonder how is the performance of that haskell code
01:12:42 <oerjan> well it builds sums from both the left and right
01:12:59 <oerjan> but it shares the partial sums in each direction
01:13:38 <oerjan> so shouldn't be more than linear
01:14:07 <alise_> hey oerjan didn't solve /my/ problem
01:14:08 <oerjan> the list from left is probably never physically built
01:14:49 <alise_> my Hird numbers stuff :P
01:15:08 <oerjan> i think i may have implied i don't know anything about bessel numbers
01:16:05 <Sgeo> Not particularly good computers. I doubt we can easily store GBs of data in memory for random access
01:16:22 <alise_> In[107]:= FullSimplify[Hird[n] - n]
01:16:23 <alise_> Out[107]= BesselI[1 + n, 2]/BesselI[n, 2]
01:16:24 <nooga> oerjan: so it does not build left and right sums for every index?
01:17:11 <oerjan> nooga: yes it does, but it uses recursion on the neighbors to share the work
01:18:02 <nooga> omg, haskell is so awesome
01:18:49 <alise_> In[135]:= FullSimplify[Hird[n]]
01:18:49 <alise_> Out[135]= Hypergeometric0F1Regularized[n, 1]/BesselI[n, 2]
01:18:58 <alise_> Say it with me now: WUT.
01:19:21 <nooga> i could say WUT for everything that you've pasted today
01:19:32 <nooga> because this notation means nothing
01:19:37 <oerjan> nooga: basically scanl (+) and scanr (+) are like foldl (+) and foldr (+) (which both sum lists), except they also give you a list of _all_ the partial sums
01:20:08 <oerjan> starting from left and right respectively
01:20:29 <alise_> nooga: means nothing how
01:20:36 <alise_> it's either mathematica or just plain ascii mathematical notation
01:20:55 <nooga> Hypergeometric0F1Regularized
01:21:22 <nooga> let's ask the guy with math doctorate: how do we define this thing?
01:21:29 <facsimile> I want to learn hypergeometry soon
01:21:30 -!- Patashu has joined.
01:21:41 <alise_> Hypergeometric0F1Regularized[a,z]
01:21:41 <alise_> is the regularized confluent hypergeometric function Subscript[\[Null], 0]Subscript[F, 1](a;z)/\[CapitalGamma](a).
01:21:58 <alise_> where the 0 and the 1 are subscripted
01:22:01 <alise_> I have no fucking idea either
01:23:33 <oerjan> i have a slight memory of "hypergeometric" from statistics, i think
01:24:00 <Sgeo> Dear Google: Sgeo and Geo are not synonyms
01:25:02 <alise_> Visualize confluence relation \!\(
01:25:03 <alise_> \(\*SubscriptBox[\(\[InvisiblePrefixScriptBase]\), \(0\)]\)
01:25:03 <alise_> OverscriptBox[\(F\), \(~\)], \(1\)]\)\)\[InvisibleApplication](\[Null];a;z)\[LongEqual]
01:25:03 <alise_> \!\(\*UnderscriptBox["lim",
01:25:03 <alise_> RowBox[{"p", "\[Rule]", "\[Infinity]"}]]\)\[ThinSpace]\!\(
01:25:05 <alise_> \(\*SubscriptBox[\(\[InvisiblePrefixScriptBase]\), \(1\)]\)
01:25:06 <nooga> 02:19 < facsimile> I agree with oerjan
01:25:09 <alise_> OverscriptBox[\(F\), \(~\)], \(1\)]\)\)\[InvisibleApplication](p;a;z/p):
01:25:11 <alise_> What the fuck Mathematica!
01:25:17 <alise_> Admittedly it is rendered on screen.
01:25:20 <alise_> But it doesn't make any more sense!
01:25:22 <oerjan> the hypergeometric distribution is something like the binomial distribution, except you _don't_ get to pull a number twice. or something like that.
01:26:04 <oerjan> "In probability theory and statistics, the hypergeometric distribution is a discrete probability distribution that describes the number of successes in a sequence of n draws from a finite population without replacement, just as the binomial distribution describes the number of successes for draws with replacement."
01:29:22 <nooga> alise_: email Wolfram for explanation
01:30:13 <nooga> guy is named after the element that is often found in lightbulbs, how awesome is that?!
01:31:12 <alise_> guy is named after the famous series of BesselI function numbers, how awesome is that?!
01:32:08 <nooga> if i could change my surname to Thorium or Ununhexium i would do it immediately
01:33:16 <oerjan> element 1111111 seems a little unlikely to be synthesized
01:33:36 <oerjan> hm maybe in neutron stars?
01:34:17 <nooga> btw. what's with LHC?
01:34:34 <oerjan> is there some news about it?
01:35:00 <nooga> the thing is that there is no news about it
01:35:26 <oerjan> it is presumably doing its ordinary, longtime, boring collection of data
01:35:43 <alise_> Unununununununununununununununununium
01:35:47 <oerjan> which it will take months if not years to analyze before anyone can make real news out of it
01:36:04 <alise_> not really boring is it; how many terabytes a second?!
01:36:10 -!- Gracenotes has joined.
01:36:32 <alise_> You wouldn't think colliding erections would produce so much data.
01:36:35 <oerjan> well _most_ of it is boring, since interesting collisions are probably very rare :D
01:37:00 <alise_> Indeed. That often happens with hardons.
01:37:03 <nooga> once i've known a guy who worked on analyzing CERN's data
01:37:28 <nooga> they have special grid for this task and he wrote some minor programs for it
01:37:49 <oerjan> nooga: also it is still only working at half its final intended energy, iirc, and will be doing so until end of 2011, when they will take it down to upgrade it.
01:38:12 <nooga> he said that 500TB is nothing when you're talking about CERN's toys
01:38:20 <alise_> Slereah works on the LHC iirc
01:38:30 <alise_> like in france doing some sort of theoretical work
01:38:43 <alise_> I doubt anything highly innovative
01:39:15 <nooga> other friend of mine went to some weir russian CERN equivalent
01:39:21 <oerjan> nooga: conveniently they will be skipping all of 2012 >:)
01:39:38 <oerjan> unless they start it up in december to scare everyone instead
01:39:43 <nooga> and played with some big, old accelerators and nuclear reactors
01:39:45 <uorygl> What if they forget to skip it?
01:40:09 <oerjan> uorygl: that would mean running at half energy for an extra year
01:40:11 <nooga> sadly, he remembers only vodka
01:40:19 <uorygl> (Which, like all my other stupid questions, is an intentionally stupid question.)
01:40:36 <oerjan> also, the end of the world, but we don't talk about that.
01:41:45 <alise_> Currrrrrrrrrrve approximation
01:41:51 <alise_> um... Linear interpolation!
01:42:07 <nooga> http://upload.wikimedia.org/wikipedia/commons/thumb/f/ff/Torun32m_winter.jpg/800px-Torun32m_winter.jpg :D
01:42:15 <uorygl> So, for a Computer Science minor, I can take "Database", which looks bad, "System Programming", which looks worse, or "Data Communications", which I have no idea about.
01:42:56 <nooga> the driver is 386 machine with DOS on board
01:42:58 <uorygl> The topics of the latter two classes are supposed to be abstracted out of sight. :P
01:43:46 <nooga> and some mainframe salvaged from some soviet submarine does the data analysis
01:44:06 * alise_ wonders about a glyph being ((xs,ys),(xe,ye),c).
01:44:08 <uorygl> Here's the topic for Data Communications: "An introduction to data communications techniques, particularly as applied to computer networks. Physical media and devices, data link and network protocols, and other data communications topics will be studied."
01:44:18 <uorygl> Physical media? Neat, I get to learn what a hard drive is.
01:44:18 <alise_> It's a line from (xs,ys) to (xe,ye), curved by c.
01:44:27 <alise_> So we can have negative/positive curves for inwards/outwards curves.
01:44:36 <uorygl> I guess other devices could be interesting... as could network protocols.
01:44:45 <alise_> And 1 would be a semicircle, say.
01:44:48 <uorygl> And the other two classes look icky, so I guess I'm taking this one.
01:44:51 <alise_> With -1 being the other half of a semicircle.
01:44:56 <alise_> Coords are from 0 to 1.
01:44:56 <nooga> uorygl: I've studied automatics and robotics, everything looks even worse there
01:45:41 <oerjan> alise_: (xs,ys)---c-->(xe,ye) except ---c--> is an actual curve with c above it?
01:45:46 <uorygl> The class I took this semester, "Computer Organization", was much more boring than the textbook suggests.
01:45:52 <alise_> oerjan: something like that I guess
01:46:02 <nooga> uorygl: studies are boring
01:46:04 <alise_> (0, 1/2) ->@1 (1/2, 1)
01:46:04 <alise_> (1, 1/2) ->@1 (1/2, 0)
01:46:05 <uorygl> But I've already ranted about that, so I won't rant about it again.
01:46:08 <alise_> Maybe coords from -1 to 1 is better.
01:46:13 <alise_> A lot of things seem to do that. So I will.
01:46:20 <uorygl> nooga: aren't classes pretty much all about studying?
01:46:57 <alise_> I wonder whether this is expressive enough to draw glyphs and stuff.
01:47:06 <nooga> i'm trying to write OSes, compilers and other shit that would give me credits without going to classes
01:47:08 <alise_> Might also want some sort of filling, of course.
01:47:13 <alise_> So, line thickness, basically.
01:47:26 <alise_> Filled circular stuff might be hard?
01:47:44 <nooga> MS Paint or what, alise_
01:48:05 <uorygl> Glyphs are pretty easy with Bezier curves, aren't they?
01:48:11 <alise_> and the idea is to be able to define fonts renderable quickly on, say, a device with an embedded cpu
01:48:24 <alise_> I don't think doing full bezier curves in realtime is feasible on a low powered cpu lke that
01:48:30 <alise_> this is for the CASulator
01:48:37 <uorygl> Bezier curves are pretty easy, aren't they?
01:48:41 <uorygl> They're just polynomials.
01:48:45 <nooga> it's definately possible
01:49:01 <alise_> In realtime, though, rendering a bunch of text?
01:49:15 <oerjan> alise_: if c is a curve from 0 to 1 in the complex plane, then z1 ->@c z2 could mean the curve d(t) = z1 + (z2-1)c(t)
01:49:28 <uorygl> A degree-4 polynomial can be evaluated at a single point using 4 additions and 3 multiplications.
01:49:38 <oerjan> *d(t) = z1 + (z2-z1)c(t)
01:50:07 <nooga> it is an awesome idea for using XMOS
01:50:08 <oerjan> alise_: then it's just rotation and scaling in the unique possible way
01:50:18 <alise_> uorygl: Mine sounds easier to write glyphs in by hand, though :P
01:50:28 <uorygl> So if you have a 640x480 screen, and evaluate a polynomial for every five pixels, that's 3*640*480/5 = 184,320 operations.
01:50:32 <alise_> oerjan: so it's fast, basically?
01:50:40 <oerjan> which is something postscript provides, i recall
01:51:55 <alise_> Of course, it may just be better to use bitmap fonts if we're using a regular calculator display.
01:52:27 <alise_> But nah; response time.
01:52:37 <oerjan> postscript has a translation + matrix iirc, so it can also do shears and stuff
01:52:49 <oerjan> but still linear/affine
01:52:59 <uorygl> Looks like it's impossible for me to take the class I want to take next year. No matter; I'll just take it a different year.
01:53:06 <oerjan> just multiplicatino and addition
01:53:21 <nooga> uorygl: i can't even choose my classes here
01:53:48 <nooga> and it turns out that they try to put EVERYTHING in our heads and then it gets to a point that we know NOTHING
01:53:50 <uorygl> That seems kind of strange.
01:54:00 <nooga> because it's Poland ;D
01:54:13 <uorygl> Whew, it's only Poland.
01:54:38 <uorygl> I forgot that Poland is part of Scandinavia!
01:55:18 <uorygl> No, I didn't forget anything, it's not in Scandinavia at all.
01:55:24 <oerjan> we love our poles, but they are _not_ scandinavians. unless they move here permanently, at least.
01:55:36 <uorygl> What if they move there temporarily?
01:55:55 <oerjan> then they are guest workers, usually
01:55:59 <alise_> the Poles here do all the work
01:56:03 <alise_> nooga: you should come and do the work
01:56:12 <uorygl> Where are you now, alise_?
01:56:52 <alise_> oerjan: ah, well, difference: we /don't/ love our poles
01:56:55 <uorygl> I'd expected you to be somewhere else by now.
01:57:02 <alise_> anyway the magnetic poles cannot move here permanently
01:57:08 <alise_> uorygl: A bit naive, to think moving is so easy.
01:57:21 <nooga> alise_: I don't take your jobs
01:57:36 <alise_> Never said it was my opinion :P
01:57:47 <uorygl> <false>Moving is very easy. Just put your laptop in your backpack and drive somewhere.</false>
01:57:58 <pikhq> alise_: Much like Mexicans.
01:58:04 <pikhq> DER TAKING ER JORBS
01:58:07 <oerjan> alise_: norway still has the lowest unemployment in europe. although they're probably shuffling some data around to make it lower, like everywhere else.
01:58:32 <nooga> oerjan: my brother is in norway, he's studying your language AFAIR
01:58:35 <alise_> uorygl: Here, try and find accommodation in a country you're not in. Now, go.
01:58:58 <uorygl> Sweden has a nice-looking web site, so I'll start there.
01:59:07 <nooga> and I know many Norwegians from our medical university, here in Poznan
01:59:46 <oerjan> alise_: i hear the magnetic poles may be shifting towards europe, actually
01:59:59 <uorygl> Let's see, do I want to work or study?
02:00:02 <nooga> alise_: I found accomodation in Scotland pretty fast
02:00:23 <pikhq> nooga: Where from?
02:00:26 -!- Patashu has quit (Quit: MSN = Patashu@hotmail.com , AIM = Patashu0 , YIM = Patashu2).
02:00:34 <nooga> 3 weeks, but i think it would be fully possible to stay much longer
02:00:52 <nooga> the problem was it costed something like 600 quid a month
02:00:55 <alise_> specifically because moving /twice/ sounds unfun.
02:01:25 <uorygl> I almost resisted the urge to do this:
02:01:30 <HackEgo> dollars? At todays exchange rates (2nd Dec 08) 50 = $74.75 ... US dollars? its your moms butt ... US dollars? $847.210 (as of August 4th) ...
02:01:41 <HackEgo> 600 British pounds = 888.48 U.S. dollars
02:01:52 <alise_> So, CASulator! Someone be interested? Please?
02:01:54 <pikhq> That's somewhat pricy.
02:01:58 <uorygl> Wait, where did "its your moms butt" come from?
02:02:15 <nooga> annd i'm not working in the UK so it was a little bit expensive
02:02:25 <pikhq> `calc 600 GBP in ZWL
02:02:27 <Sgeo> alise_, you're going to make a physical CASculator?
02:02:27 <HackEgo> futuresource.quote.com/reference/symlist.jsp?print=true - [37]Cached - [38]Similar
02:02:40 <uorygl> "Moving to someone in Sweden"? How does someone move to someone in Sweden?
02:02:43 <alise_> Sgeo: I want to! Probably I'd write code for an embedded CPU and simulate it on a PC first.
02:02:56 <Sgeo> Sounds interesting
02:03:11 <Sgeo> But unless you're planning on making a CAS from scratch, what are you going to use?
02:03:47 <Sgeo> Oh wait, it's not going to just be porting a regular CAS to an embedded CPU, is it
02:04:01 <nooga> 02:59 < pikhq> nooga: Where from?
02:04:22 <alise_> Especially as I'll need to code with the HW in mind.
02:04:54 <pikhq> `calc 888.48 * 300000000000000
02:04:56 <HackEgo> 888.48 * 300 000 000 000 000 = 2.66544 10^17
02:05:17 <pikhq> 2.66*10^17 Zimbabwe dollars a month.
02:05:23 <alise_> also is that new or old?
02:05:23 <pikhq> Now *that's* expensive.
02:05:28 <alise_> new as in just before being obsoleted
02:06:23 <pikhq> That's just before they introduced the 4th dollar.
02:07:14 <pikhq> The 4th dollar got up to an exchange rate of 255.19 to the USD before being suspended.
02:07:57 <uorygl> Zimbabwe seems to be really bad at handling their currencies.
02:08:57 <pikhq> You see, their solution to the government not being able to fund municipal projects was to print more money.
02:09:19 <uorygl> And they just keep doing that?
02:09:36 <pikhq> Mugabe is not too bright.
02:10:15 <uorygl> Why don't they kick him out?
02:10:31 <oerjan> you don't think they've _tried_?
02:10:52 <uorygl> Well, what's stopping them?
02:11:33 <pikhq> Mugabe is heavily armed and capable of rigging elections.
02:12:55 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
02:15:06 * Sgeo looks at someone else's computer through his phone
02:15:12 <pikhq> Mugabe's first action was to seize all the land from white people and reassign it to black people, in an attempt to undo the harm caused by colonialism.
02:15:20 <pikhq> For obvious reasons, this collapsed Zimbabwe's economy.
02:15:26 <alise_> Mugabe is fucking insan.
02:15:49 <alise_> Would you say awesome if you were one of the victims of his regime's anti-white violence...?
02:15:56 <uorygl> I suppose people can always vote with their feet... or not?
02:16:11 <alise_> When you're that poor?
02:16:14 <alise_> How do you propose to do that?
02:16:16 <pikhq> He's done such land reassignments in 2000 as well.
02:16:22 <alise_> Face it; real inescapable fascism does exist.
02:16:39 <alise_> The only way to end it is intervention or natural eventual collapse.
02:16:46 <pikhq> Needless to say, assigning farmaldn to people who don't farm is a completely and utterly *retarded* idea.
02:17:24 <alise_> SO GUYZ LET'S TALK ABOUT CASULATOR
02:17:27 <uorygl> I wonder if any countries pay for travel there.
02:17:57 <pikhq> There's travel bans. That close enough? :P
02:18:19 <uorygl> "Want to come to our country? We'll send an armed helicopter to you and pick you up!"
02:18:25 <uorygl> I'm confident that that's too good to be true. :P
02:18:28 <alise_> I guess the TI-Nspire CAS (http://en.wikipedia.org/wiki/TI-Nspire) is the closest thing to the CASulator.
02:18:38 <nooga> nobody even said good night
02:19:16 <pikhq> Oh, and 25% of the country has AIDS, because AIDS is apparently a western plot, and those condoms make things worse.
02:19:43 <uorygl> Yeah, Zimbabwe sounds like a really backwards place.
02:20:15 <pikhq> It's Africa. The end of colonialism there happened in such a way that *everything* got royally fucked up.
02:20:55 <uorygl> Perhaps we should bring back colonialism.
02:20:55 <pikhq> Like most other places, really.
02:21:05 <alise_> WHY DOES NOBODY LIKE CASULATOR
02:21:05 <pikhq> No, that'll make things worse.
02:22:02 <oerjan> <pikhq> Mugabe's first action was to seize... <-- well not the first, but maybe the first really stupid one. he'd had powers for decades before that
02:22:14 <pikhq> oerjan: Mmm, right.
02:22:45 <uorygl> How do independence movements get started?
02:22:45 <pikhq> I remember that vividly because it's the first action of his that is literally the *antithesis* of a good idea.
02:23:10 <pikhq> uorygl: Most of the time? People with guns that are pissed.
02:23:13 <uorygl> Armies don't simply form because a bunch of people want to be in an army.
02:23:34 <pikhq> No, armies form because they feel that they *must* fight.
02:24:41 <uorygl> How do they get their resources, and who leads them?
02:25:15 <pikhq> Any way they can, and whoever was both lucky enough and sufficiently charismatic to lead them.
02:26:31 <pikhq> And how well it ends up is, well, a crapshoot.
02:26:45 <oerjan> pikhq: i occasionally see Botswana (zimbabwe's next door neighbor) mentioned as one place in africa which is _not_ fucked up. at least as far as government is concerned.
02:26:58 <pikhq> You can end up with the US, or you can end up with North Korea.
02:27:02 <oerjan> functional democracy and all
02:27:32 <uorygl> How well would a revolution work nowadays, given that you have all this war technology?
02:27:49 <alise_> mugabe would crush any oppression
02:27:53 <pikhq> oerjan: Oh, yeah. Botswana has a stable government and everything.
02:27:56 <alise_> you simply can't get guns. and if you do, mugabe has more.
02:28:26 <alise_> also 16% of Botswana has AIDS apprently
02:28:58 <oerjan> there's always something :(
02:29:08 <pikhq> Yes, that's because AIDS is just plain *rampant* in all of Africa.
02:29:09 <uorygl> Seems you could just program your technology to only work in the hands of the government.
02:29:35 <alise_> yeah i was reading the botswana article and i was
02:29:37 <uorygl> Boom, the revolution guys have no technology, and they lose.
02:29:42 <alise_> gee whiz this place does have a good government as these things go
02:30:15 <pikhq> alise_: Having a lot of AIDS is pretty good compared to most of the rest of Africa.
02:30:33 <pikhq> After all, otherwise you have a lot of AIDS *and* everything else that could be wrong.
02:30:47 <uorygl> Africa sounds like it has a lot of stuff wrong with it.
02:31:14 <pikhq> uorygl: Colonialism ended in the 50s, and it ended roughly by "Fuck it, we're just leaving".
02:31:27 <pikhq> Bunch of war and shit, and voila. Africa.
02:31:28 <uorygl> What a great way to end colonialism.
02:31:45 <alise_> "Bunch of war and shit".
02:31:46 <uorygl> I wonder how expensive it would be to invade Africa again.
02:32:00 <alise_> You know, glossing over the details.
02:32:14 <pikhq> There's lots of them!
02:32:47 <uorygl> This thought actually ran through my head: "Let's invade Africa and govern them well until their skin--wait, skin color isn't actually determined by prosperity."
02:33:52 <alise_> uorygl: I believe you are insane.
02:36:04 <pikhq> How's about we just make South Korea an island as a show of power and call it a day?
02:37:23 -!- bsmntbombdood has quit (Remote host closed the connection).
02:38:02 <uorygl> Huh. If you're a dual citizen, and you go to prison in one country, can the other country generally pluck you out if they want to?
02:38:53 <alise_> Discussions with pikhq lead me to believe that even as a purely British citizen in another country, the only way Britain can touch me whatsoever is if I committed a crime on British soil.
02:39:01 <alise_> If I have, then they can apply for extradition.
02:39:16 <alise_> So I'd assume that someone with /dual/ citizenship would have an even stronger case to stay in the country they reside in.
02:39:53 <pikhq> alise_: They can also petition for your release *from* jail if they feel that the jailing is in some way unfair.
02:39:59 <uorygl> I commit a crime in Country A, so I go to prison there. Country B, however, doesn't consider this a crime, and would be willing to have me as a citizen.
02:40:06 <alise_> pikhq: Well that's nice :P
02:40:09 <pikhq> Such petitioning can actually get pretty serious.
02:40:16 <alise_> uorygl: No, that would not work.
02:40:23 <alise_> In fact, if you escaped to country B, country A /could/ apply for extradition.
02:40:30 <alise_> Unless the crime was serious it would be denied, though.
02:40:55 <uorygl> I believe that extradition generally requires the thing to be considered a crime in whatever country you're to be extradited from.
02:41:20 <alise_> But the point is, no, you couldn't escape prison by going to another country, I don't think.
02:41:26 <pikhq> It depends on the extradition treaty.
02:42:28 -!- Sgeo has quit (Ping timeout: 248 seconds).
02:42:42 <alise_> What, for my statement?
02:42:53 -!- facsimile has quit (Quit: facsimile).
02:43:32 <pikhq> alise_: You could escape prison by going to a country without extradition treaties.
02:43:44 <pikhq> Most such countries suck *ass* though.
02:44:01 <alise_> And it comes full-circle.
02:44:03 <uorygl> Clearly, the thing to do is install an app on my iPhone that will teleport me to the country of my desire.
02:45:41 <alise_> it was rejected by apple sorry
02:46:34 <pikhq> Also, one can avoid a death sentence from, say, the US, by going to a country that doesn't extradite unless there is *0* chance of a death sentence happening.
02:46:47 <pikhq> (Mexico, Canada, and the EU are all like this)
02:47:04 <alise_> I didn't know that. That's nice of those countries.
02:47:45 <pikhq> Human rights laws mandate it.
02:48:27 <alise_> So... apparently [i.e., reddit IAmA credibility] the guy who designed the current American flag was a child abuser.
02:48:29 <uorygl> Michigan doesn't have death sentences. Unless you commit a crime in a national park or something.
02:48:34 <alise_> http://www.reddit.com/r/IAmA/comments/c0ua8/i_am_the_sister_of_a_man_who_was_sexually_abused/
02:48:35 <pikhq> Hey, the US has no extradition treaties with Taiwan, because *you can't have treaties with a country you don't recognise the existence of*.
02:48:45 <alise_> The US doesn't recognise the existence of Taiwan????
02:49:01 <pikhq> alise_: Recognising Taiwan and China are mutually exclusive.
02:49:15 <pikhq> As each claims the other's territory as their own.
02:49:19 <uorygl> So if I find some product that says "MADE IN TAIWAN", it's a lie?
02:49:22 <alise_> Are they? Just recognise China-sans-Taiwan and Taiwan-sans-China, and if the China don't like it, sucks to be them.
02:49:27 <alise_> i.e., only the relevant territories from both
02:50:05 <pikhq> China will not have *any* relations with any organization that recognises Taiwan as an independent state.
02:50:09 <alise_> What are they listed as being in whatever database the US as?
02:50:43 <pikhq> They are listed as Chinese Taipei, Taiwan, Formosa, the Republic of China, etc.
02:51:12 <uorygl> Does that mean China will not threaten them?
02:51:43 <pikhq> Like most countries, we go to *great* lengths to have informal relations with them while making it *perfectly* clear to the PRoC that we are merely having discussions with the rebels.
02:51:55 <pikhq> Not recognising their sovereignity.
02:52:05 <alise_> <uorygl> Does that mean China will not threaten them?
02:52:17 <alise_> You can't declare war against a country you don't recognise!
02:52:17 <pikhq> This includes the UN.
02:53:33 <oerjan> alise_: but they wouldn't consider it a war against another country
02:53:49 <oerjan> just an internal chinese matter
02:54:19 <alise_> if USA recognises Taiwain, then China does-not-recognise USA
02:54:23 <alise_> therefore China cannot-declare-war-on USA
02:54:49 <pikhq> Amusingly, the ROC treads carefully to *not* declare itself an independent state.
02:54:58 <pikhq> They claim the same borders they had in the 40s.
02:55:33 <oerjan> incidentally, both chinas are members of the world trade organization
02:56:08 <oerjan> so taiwan _does_ have mandated free trade with the US and many other countries
02:56:31 <pikhq> alise_: The US is required by law to defend Taiwan if the PRoC attacks.
02:56:32 <alise_> how can china be a member of WTO if WTO recognises Taiwain?
02:56:40 <alise_> that's against the axiom of China-Taiwain calculus
02:56:45 <pikhq> Because of this, we have a permanently deployed *fleet* around Taiwan.
02:56:54 <alise_> pikhq: Just... discussing. With the rebels.
02:57:00 <oerjan> alise_: they're technically not recognizing taiwan as a _country_, just a special region, i think
02:57:01 <alise_> Our army. Is discussing with the rebels. At their location.
02:57:21 <alise_> does the chinese government ever exercise any actual power in taiwain?
02:57:26 <alise_> can't think how they'd go about doing it
02:57:34 <pikhq> alise_: The WTO deals with *trading entities*. Not countries. :P
02:57:37 -!- Sgeo has joined.
02:58:06 <pikhq> And the PROC government maintains no power over the Taiwan. Just the ROC.
02:58:36 <oerjan> alise_: on the reverse side, greenland and the færøyer are not part of the EU, despite being part of the kingdom of denmark
02:59:07 <oerjan> i wasn't even going to try to remember the english spelling.
02:59:21 <alise_> heh like that's easier
02:59:26 <pikhq> oerjan: The EU treaties generally don't apply to out-of-Europe holdings of member countries.
02:59:45 <pikhq> Also, aren't those considered seperate countries which *happen* to be also part of the Kingdom of Denmark?
02:59:57 <pikhq> Kinda like how Australia, Canada, and the UK all have the same Queen?
02:59:58 <oerjan> pikhq: færøyene are in europe, most definitely, and i think greenland was in EU, then voted to secede
03:00:10 <alise_> [[Polish pop star faces two years' prison for blasphemy; made negative statement about the Bible in TV interview.]]
03:00:31 <oerjan> they do have some free trade agreement though
03:01:19 <alise_> http://www.reddit.com/r/worldnews/comments/c1je9/polish_pop_star_faces_two_years_prison_for/ ;; oh
03:01:47 <pikhq> Taiwan's membership in the WTO is as the "Separate Customs Territory of Taiwan, Penghu, Kimmen and Matsu".
03:02:29 <oerjan> pikhq: they may be considered so _now_, but i'm not sure they were so at the time denmark joined the EU
03:02:44 <alise_> Romething is sotten in the state of Mendark.
03:02:53 <oerjan> greenland got a greater autonomy agreement just last year (or was it this winter?)
03:02:58 <alise_> ((I hear it's country matters))
03:03:58 * pikhq notes that Shakespeare must have been a great fan of puns
03:05:26 <oerjan> alise_: there have been debates about removing norway's blasphemy law recently (i'm not even sure if it's been repealed or not, but i _think_ not). it's been a sleeping law for half a century, though.
03:05:43 <alise_> pikhq: "country matters" is hardly a pun!
03:05:56 <alise_> More like pretty darn close to an outright uttering of a pretty serious expletive :-)
03:05:57 <pikhq> alise_: Cunt-ry matters.
03:06:38 <pikhq> I find it completely bizzare that the ROC *can't* change the regions that it claims.
03:06:46 <alise_> (Incidentally in the Royal Shakespeare Company production with David Tennant and Patrick Stewart that was broadcast on BBC2 on Boxing Day (-- 3 hours without interruptions, sheesh --) Tennant basically pronounces it exactly as "cuntrrrrry matters".)
03:07:06 <alise_> (Which, you know, sort of offended my innate British sensibilities. How dare they put this filth on the BBC? :P)
03:07:18 <pikhq> It they reduced their claims to just the area they control, the PROC will consider this a declaration of independence and go to war.
03:07:33 * oerjan wonders if ROC claims tibet or not
03:07:52 <pikhq> Note that the *PROC* is the government that should be declaring independence: they're the bastards who revolted against the ROC!
03:08:16 <pikhq> oerjan: Because they did in the 40s, and to not claim it would cause war.
03:08:21 <uorygl> Heh, Norway has a blasphemy law?
03:09:07 <oerjan> uorygl: well a blasphemy paragraph in a law (the law of punishment, probably)
03:09:11 <pikhq> http://en.wikipedia.org/wiki/File:ROC_Administrative_and_Claims.png
03:09:56 <uorygl> Whoa, Finland has a blasphemy law, and they actually use it.
03:10:00 <oerjan> pikhq: i wasn't sure when china started claiming tibet, they only invaded in the 50s didn't they, but i guess they only consider tibet's former independence a temporary lapse...
03:10:01 <alise_> lol the ROC just claims everything
03:10:12 <alise_> there should be a country that just claims the entire world
03:10:23 <pikhq> Sealand should do it.
03:10:26 <oerjan> and so "always" claimed it
03:10:27 <uorygl> I'm sure many nomics do that.
03:10:43 <alise_> In Finland, section 10 of chapter 17 of the penal code relate to blasphemy.[10][11] Unsuccessful attempts were made to rescind the section in 1914, 1917, 1965, 1970, and 1998.[12]
03:10:43 <alise_> In 1969, Finland prosecuted Harro Koskinen for publishing a picture of his painting called Pig Messiah, which featured a crucified pig. For violating the sensibilities of a religion, Koskinen had to pay a fine.[13][14]
03:10:44 <alise_> In 2008, the issue of religious sensibilities arose again. On 30 May 2008, Tampere District Court sentenced Seppo Lehto to two years and four months imprisonment for offences which involve hate speech and blasphemy. The court found Lehto guilty of: defamation, incitement of an ethnic group, and violating the sensibilities of a religion. The judgment said that Lehto had violated the sensibilities of Islam because he had disseminated, with insulting intentions
03:10:45 <alise_> , material which openly blasphemes and desecrates that which Muslims deem holy.[14] Outraged at the punishment of Lehto, Jussi Halla-aho, a Helsinki councilman, posted to the Internet in 2008 some controversial remarks about Islam and about Somalis. Those remarks induced the Helsinki District Court to order Halla-aho to trial.[15]
03:10:48 <pikhq> *North Korea* should do it.
03:11:13 <pikhq> "We declare war on everyone else! MWAHAHAHAH!"
03:11:28 <uorygl> I think I did that in Civilization IV once.
03:11:45 <uorygl> I don't remember what happened afterward; I think I got bored and quit.
03:12:50 <alise_> New Canada said that the only true Canadas are Old Canada and New Canada, and that all other entities using that name are misguided.
03:13:01 <alise_> (New Canada was the barely-played-at-all successor to Canada.)
03:13:04 <pikhq> http://en.wikipedia.org/wiki/File:CountriesRecognizingROC.png Such a headache.
03:13:24 <uorygl> I don't really remember Old and New Canada.
03:13:36 <alise_> Old Canada was IRCNomic, which was a right fun time apart from that UPPERCASE bullshit.
03:13:42 <alise_> And, well, everything else AnMaster did.
03:13:57 <alise_> New Canada I believe you played for perhaps three posts, maybe 1.5 serious.
03:14:07 <oerjan> pikhq: um, what's the legend on that?
03:14:57 <alise_> strong green = embassy
03:14:59 <alise_> http://en.wikipedia.org/wiki/Foreign_relations_of_the_Republic_of_China
03:16:44 <pikhq> Strong green = embassy, light green = "just talking with the rebels, friend PROC."
03:16:58 <oerjan> http://en.wikipedia.org/wiki/Foreign_relations_of_the_Republic_of_China#List_of_countries_with_diplomatic_relations_with_the_ROC
03:17:17 <oerjan> that legend claims light green = diplomatic relations but not embassy
03:17:24 <oerjan> presumably consulates?
03:17:30 <pikhq> Unofficial diplomatic relations.
03:17:42 <pikhq> Things like the "American Institute in Taiwan".
03:18:08 <pikhq> Which is itself a *massive* clusterfuck.
03:18:36 <pikhq> The American Institute is a US corporation, wholy owned by the US government, and staffed by federal employees.
03:19:23 <pikhq> Like I said, *great* lengths to claim we're just "talking with the rebels".
03:19:54 <alise_> Let's just buy out China, Zimbabwe and North Korea.
03:20:14 <uorygl> China sounds really expensive.
03:20:18 <uorygl> Really, really expensive.
03:20:25 <uorygl> Like, the most expensive single object in the world.
03:20:30 <uorygl> Except it's *multiple objects*.
03:20:57 <alise_> uorygl: Underneath China there is an inscription in the earth: "MADE IN CHINA"
03:21:15 <alise_> So it's probably cheap tat.
03:21:26 <oerjan> the wall of china is basically the edge of the M
03:21:32 <uorygl> Just like the inscription on the International Space Station.
03:21:33 <oerjan> the only part that shows
03:21:51 <uorygl> And the inscription on me that says "Made in the U.S.A.".
03:22:30 <alise_> Say, what laws apply on the ISS?
03:22:49 <uorygl> I'd say each person is bound by the laws of whatever country they're from.
03:23:06 <alise_> http://upload.wikimedia.org/wikipedia/commons/f/fa/ISS_26.07.07.jpg which part is the ISS?
03:23:12 <pikhq> alise_: I think we actually *could* buy out North Korea quite feasibly.
03:23:34 -!- augur has joined.
03:23:40 <oerjan> pikhq: except north korea would bail out of the agreement, _after_ taking the money
03:24:18 <uorygl> Gee, I don't remember the ISS looking quite so acular.
03:24:22 <alise_> (with a really big gun)
03:24:35 <alise_> I don't think the line is meant to be the ISS
03:24:35 <augur> WHATS THIS ABOUT LINGUISTICS?
03:24:41 <alise_> that would be ridiculous, surely
03:24:41 -!- Oranjer has left (?).
03:24:41 <HackEgo> * ketorolac tromethamine: nonsteroidal anti-inflammatory drug (trade names Acular and Toradol) that is administered only intramuscularly \ [13]wordnetweb.princeton.edu/perl/webwn \
03:24:48 <alise_> I think uorygl just made up "acular"
03:24:53 <pikhq> alise_: The ISS is under *very confusing* jurisdiction.
03:24:53 <uorygl> Yes, I'm afraid I did.
03:25:01 <uorygl> However, someone who knows enough Latin will surely know what it means.
03:25:06 <uorygl> You might go ask in ##latin.
03:25:08 <alise_> pikhq: Orally? (O'Rly)
03:25:29 <uorygl> Anyway, as for the ISS, we should ask an astronaut.
03:26:05 <uorygl> Who should we ask? Mike? Jeff?
03:26:08 <alise_> it would be fun if countries were on this stonkin' great big global country-stock market
03:26:09 <oerjan> maybe they're an anarchosyndicalist commune.
03:26:16 <alise_> and you could buy out countries and stuff
03:26:27 <alise_> /and/ they live in zero gravity, too
03:26:36 <pikhq> "each partner shall retain jurisdiction and control over the elements it registers and over personnel in or on the Space Station who are its nationals."
03:27:17 <pikhq> So... The *modules* are under the jurisdiction of whoever made them, but the *astronauts* are under the jurisdiction of their country of origin.
03:27:55 <pikhq> Which means food can change jurisdiction by being eaten.
03:28:09 <uorygl> It would be a lot easier to just have a bunch of National Space Stations.
03:28:17 <alise_> What jurisdiction is the resulting feces?
03:28:31 <alise_> uorygl: Or just make it an anarchosyndicalist commune.
03:28:40 <pikhq> uorygl: Or just make it an *independent nation*.
03:28:49 <uorygl> But each National Space Station would be staffed by approximately half a person.
03:29:10 * uorygl goes to howmanypeopleareinspacerightnow.com to see how many people are in space right now.
03:29:15 <alise_> Call the nation Space; define it to include everything except Earth.
03:29:30 <pikhq> We are at war with Space.
03:29:34 <alise_> This has the interesting consequence that Earth is not in Space.
03:29:46 <uorygl> Six. And there are five space agencies contributing to it?
03:29:52 <uorygl> Each National Space Station would get 1.2 staff members.
03:30:06 <uorygl> Earth would be an enclave of Space, though.
03:30:09 <pikhq> uorygl: One such space agency is the ESA.
03:30:20 <alise_> pikhq: I wonder what diplomatic relationships with Space would look like.
03:30:23 <HackEgo> * metal or plastic sheath over the end of a shoelace or ribbon \ * ornamental tagged cord or braid on the shoulder of a uniform \ [14]wordnetweb.princeton.edu/perl/webwn
03:30:33 <alise_> Would it recognise Taiwan or China???
03:30:43 <pikhq> Yes, this means some modules are under 11 nation's jurisdiction simultaneously.
03:30:57 <oerjan> apparently aculus means aglet, metal tag, so acular would mean looking like a metal tag, presumably
03:31:01 <uorygl> Space would not allow people to recognize China as an independent nation.
03:31:23 <uorygl> By 'acular', I meant 'needle-shaped'.
03:31:23 <alise_> What's Space's GDP????
03:32:22 <uorygl> Well, I know what its GNP is. Zero; nobody is a citizen of outer space.
03:32:47 <uorygl> You know, a lot easier than making Space an independent nation would be making the ocean an independent nation.
03:32:51 <uorygl> The United States of The Ocean.
03:33:20 <alise_> That would be awesome.
03:33:42 <alise_> In fact, define the Ocean to include most of Space, too, under that whatsitsname doctrine down to hell and up to heaven property law thing.
03:34:07 <alise_> So basically, start at any point in the ocean, and travel outwards. Every point travelled to by this procedure is part of the Ocean.
03:34:19 <uorygl> The thing is, though, Earth rotates.
03:34:45 <alise_> Then what's part of the Ocean changes as that happens.
03:35:38 <alise_> Eh. The plates move too, and stuff.
03:35:42 <alise_> Although a little bit slower.
03:35:47 <pikhq> Man, whole galaxies shift jurisdiction daily.
03:36:30 <alise_> Now define everything not in the Ocean to be part of the nation Outer Space.
03:36:34 <alise_> Outer because it's outside of the Ocean.
03:36:38 <uorygl> That must be confusing for people living in those galaxies.
03:37:00 <oerjan> uorygl: especially since they won't be told for millions of years
03:37:14 <alise_> One of the two recognises Taiwan; the other recognises China. Since which parts of space recognise who changes quite a lot, due to changing jurisdiction, just schedule your diplomatic meetings correctly and a landmass can have full relations with both.
03:37:17 <uorygl> "Let's see. It's 4:20 PM; that means we're currently in a nation where it's legal to smoke pot."
03:37:59 <alise_> I always thought it was 4:20 am.
03:38:17 <uorygl> That's the *other* nation where it's legal to smoke pot.
03:38:18 <alise_> I guess I couldn't imagine stoners being up at such a normal hour as 4:20 pm.
03:39:14 <alise_> Anyway, smoking pot is legal in both the Ocean and Outer Space.
03:42:22 <coppro> alise_: are there any^H^H^H^H^H^H^H^H^H^H^H^H^Hwhat's the best weighted graph module in Haskell?
03:43:04 <alise_> Something in http://hackage.haskell.org/packages/archive/pkg-list.html
03:43:18 <alise_> http://hackage.haskell.org/package/fgl, http://hackage.haskell.org/packages/archive/pkg-list.html#Graphs
03:43:28 <alise_> apart from that, dunno
03:43:59 <alise_> http://haskell.org/ghc/docs/6.12.1/html/libraries/containers-0.3.0.0/Data-Graph.html
03:45:26 <coppro> Data.Graph is unweighted
03:45:27 <alise_> If you don't all talk about CASulator I'm going to be very sad.
03:45:34 <alise_> coppro: I gave you two other links.
03:45:43 <coppro> I'm going through them one by one
03:45:46 <alise_> Ctrl+F "graph" gave tons of links on the pkg-list; not sure how many are relevant.
03:46:32 <coppro> irritating; so is the fgl
03:47:36 <Sgeo> This is what a classmate I'm trying to help copied off the blackboard. I can;t make much sense of it: http://pastie.org/private/afxzkbjoimumy3q4tk4kw
03:48:09 <alise_> coppro: Just write your own?
03:48:12 <alise_> It's easy in Haskell...
03:48:24 <alise_> Sgeo: they have a broken mind. Ignore them
03:48:27 <alise_> coppro: easier to write than find
03:48:35 <coppro> what sort of programmer would I be if I wasn't?
03:48:49 <alise_> just keep changing the data structure to be more general and abstract until the functions on them are so trivial they're barely even worth being called functions
03:48:53 <alise_> then you will have reached the epitome of laziness
03:48:59 <Sgeo> alise_, he swears that this is what was on the board
03:49:10 <coppro> and they should all be point-free, right ;)
03:49:19 <alise_> because all you do is write down the most trivial naive definition you can think of, write down the obvious operations, then repeatedly apply "strip data constructors and add type parameters"
03:49:33 <alise_> just like "oh it really did say Error 43.4" no it said 455.3
03:49:43 <Sgeo> alise_, or maybe the teacher's insane?
03:49:54 <alise_> more likely people are idiots
03:51:24 <alise_> I should really make a nice little editor sometime.
03:56:36 <alise_> Hey, I don't think anyone has named an editor iv or x yet. :-)
04:06:21 -!- adu has joined.
04:08:25 -!- soupdragon has joined.
04:09:05 <alise_> soupdragon: whoa the name is back
04:11:08 <soupdragon> http://www.ted.com/talks/ray_kurzweil_on_how_technology_will_transform_us.html
04:11:17 <adu> why is the topic backwards?
04:11:44 <soupdragon> When he mentions the human genome project -- just listen it's brilliant
04:11:59 <oerjan> adu: unicode right-to-left marker, i think
04:13:24 <adu> oh, so are you saying my client doesn't recognize it?
04:13:32 <Sgeo> adu, actually, it does
04:14:02 <oerjan> no _my_ client doesn't recognize it. i found out from the logs.
04:14:17 <adu> oh, so without the R2L marker, the topic would look right?
04:14:38 <adu> then why not remove the R2L marker?
04:15:15 <oerjan> it's a free channel. i cannot _type_ unicode either.
04:15:53 <Sgeo> adu, try selecting the topic
04:16:01 <oerjan> just as long as you keep the log link
04:17:27 <adu> this is what I get for the hexdump: http://pastebin.com/7mm1p5Q1
04:17:57 <adu> what is "E2 80 AE"?
04:18:41 <oerjan> i don't know, but the topic should be in UTF-8 encoding
04:19:28 <adu> E280AE = \u202E = right-to-left override
04:21:06 -!- adu has set topic: I love Unicode in my topics. | 僕が問題にユニコードが好きだ。 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
04:21:39 <adu> that's better
04:22:23 <adu> does it look the same on your client?
04:24:04 <adu> here's the new hexdump: http://pastebin.com/xLsrUVYV
04:24:39 <adu> o BTW, i love ted talks
04:26:14 <pikhq> adu: Looks *exactly* the same in my client.
04:26:32 <adu> ok, then your client doesn't recognize R2L overrides
04:26:44 <adu> it looks much better in my client
04:26:59 <pikhq> Indeed, my terminal does not.
04:27:28 <pikhq> Some "unicode terminal".
04:27:40 <soupdragon> http://www.ted.com/talks/ray_kurzweil_announces_singularity_university.html
04:27:46 <soupdragon> that is the one I should have linked earlier
04:27:51 <alise_> Kurzweil is a crackpot, be careful.
04:28:13 <alise_> Especially how he keeps reprinting his book, setting forward his failed predictions a few years, but strangely somehow always within his expected lifetime...
04:28:33 <alise_> He's clever, yes. But he's not someone you want to go to for a balanced, sane look at the singularity or related topics.
04:29:44 <soupdragon> listen to what he says about the human genome project, that is something I had been wondering about myself for a long time
04:29:49 <adu> the first thing I saw when i came in here was ".scipot ym ni edocinU evol I"
04:29:58 <alise_> http://fakescience.tumblr.com/
04:30:05 <alise_> soupdragon: no Flash or speakers, sorry.
04:30:51 <alise_> "The coffee bean has a distinctive smell that makes you forget how painful it is to be awake."
04:31:30 <alise_> http://media.tumblr.com/tumblr_kz8edy2JuW1qb25dg.jpg
04:31:54 <alise_> hahahah http://media.tumblr.com/tumblr_kyz1ogCofV1qb25dg.jpg
04:33:31 <soupdragon> alise_, try this http://math.etsu.edu/LaTeXMathML/
04:33:48 <alise_> yes i saw that in use on your blog, downside:
04:34:01 <alise_> the results are pretty ugly, formatting-wise, using the css, compared to e.g. jsmath
04:35:06 <alise_> soupdragon: jsmath is "obsoleted" by mathjax, but mathjax is uglier on this ubuntu box (prettier on os x and stuff)
04:35:37 <alise_> jsmath is like 200 kb + 3.2 mb for sprite fonts or 7.9 mb for image fonts.
04:35:41 <alise_> where do you get 80mb from
04:35:59 <alise_> also if you use unicode fonts or tex-fonts-on-user's-computer
04:36:05 <alise_> soupdragon: link to the relevant page plz
04:36:42 <soupdragon> mathjax is pretty but I am worried about the -jax bit
04:36:49 <alise_> it's not actually ajax
04:36:52 <alise_> it's just jsmath souped up a bit
04:36:53 <soupdragon> http://www.math.union.edu/~dpvc/jsMath/download/jsMath.html
04:37:01 <alise_> -- but mathjax output is really ugly on linux for some reason!
04:37:21 <alise_> "On a system with a 4K minimum file size"
04:37:26 <coppro> heh... my experiences with vim are funny
04:37:28 <alise_> the sprite fonts won't do that
04:37:32 <alise_> since they're all in one big images
04:37:37 <alise_> only disadvantage, doesn't work with older browsers
04:37:41 <coppro> half the time when I try to type something, I end up just fiddling with keys until I accidentally enter insert mode
04:37:52 <alise_> soupdragon: yeah ok the ones on their official website show the ugly, sec
04:38:05 <alise_> http://www.mathjax.org/wp-content/uploads/2010/02/Ubuntu_8.10_Firefox_3.0.6_1280x960_snapshot.png
04:38:09 <alise_> it's just blurry basically
04:38:33 <alise_> soupdragon: but: mathjax is basically just jsmath with css font embedding stuff
04:38:43 <alise_> I'd just use jsmath with sprite fonts tbh
04:39:10 <alise_> the 80mb is for image fonts because of minimum file size stuff
04:39:12 <alise_> sprite fonts are only a few megs
04:39:28 <alise_> also, the /user/ never downloads 80mb; nor do you
04:39:32 <alise_> it's just due to minimum-file-size on filesystems
04:39:38 <alise_> but seriously, just use sprite fonts
04:39:44 <alise_> nobody cares about old browsers :P
04:39:47 <alise_> i mean you certainly don't, using mathml
04:40:25 <alise_> and jsmath output looks impeccable ofc, since with sprite fonts/native tex fonts it's pixel-for-pixel like tex output
04:40:56 <soupdragon> hmm so you are saying that jsmath is best?
04:41:31 <pikhq> What's best is taking over the world.
04:42:23 <adu> what are we doing today pinky?
04:43:12 <alise_> when mathjax matures it will probably be better
04:43:30 <alise_> jsmath+sprite fonts has had a lot of time to mature, and it renders pretty much anything you throw at it well
04:44:03 <alise_> http://www.math.union.edu/~dpvc/jsMath/jsMath-lab.html
04:44:18 <alise_> lets you play around with jsmath typesetting
04:44:30 <alise_> leave it as use images for symbols mostly since that's what ~everyone (who doesn't have native tex fonts) will see
04:45:25 <oerjan> <alise_> http://fakescience.tumblr.com/ <-- i am _sure_ the sneezing one has to be true </grumble>
04:45:34 <alise_> soupdragon: although i'd change it to image fonts for everything, not just symbols
04:46:41 <alise_> soupdragon: I believe that the /ideal/ solution is probably something like jsMath with native TeX fonts enabled, and a CSS @font-face to load the computer modern font from the site
04:47:04 <alise_> that way it'd be the best possible output for everyone, and totally scalable
04:47:07 <soupdragon> I want something like this for a program editor
04:47:10 <alise_> ...in fact I think you could do it!
04:47:22 <alise_> if you just add the @font-face and force-override jsMath into native TeX fonts it should work!
04:47:29 <soupdragon> just using lower_case, titleCase and SCREAMCASE is not enough
04:48:03 <soupdragon> Things like mathbb, mathcal - are essential in program code, but for some reason missing
04:49:07 <oerjan> data UlTrAcAmElCaSe uLtRaCaMeLcAsE = UlTrAcAmElCaSe uLtRaCaMeLcAsE | AAAAAAAAAAAAAAAAAA
04:49:09 <coppro> i BeLiEvE tHe TeRm Is ChAoScAsE 9NoT rEaLlY0
04:50:40 <alise_> soupdragon: oh I'm not sure but I think you also need to get e.g. the blackboard bold font from http://www.math.union.edu/~dpvc/jsMath/download/extra-fonts/welcome.html don't quote me on this though
04:50:42 <Sgeo> Wow, a 2 hour 46min phone call
04:52:39 <oerjan> !adduserinterp chaos haskell import Data.Char; main = interact $ concat . zipWith ($) (cycle [toLower, toUpper])
04:52:52 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
04:52:58 <EgoBot> Sorry, I have no help for userinterp!
04:53:01 <EgoBot> userinterps: Users can add interpreters written in any of the languages in !help languages. See !help addinterp, delinterp, show | !userinterps. List interpreters added with !addinterp.
04:53:11 <oerjan> !addinterp chaos haskell import Data.Char; main = interact $ concat . zipWith ($) (cycle [toLower, toUpper])
04:53:11 <EgoBot> Interpreter chaos installed.
04:53:27 <Sgeo> According to Reddit, non CS/Software Engineering degrees are worthless
04:53:54 <EgoBot> Interpreter chaos deleted.
04:54:22 <oerjan> !addinterp chaos haskell import Data.Char; main = interact $ zipWith ($) (cycle [toLower, toUpper])
04:54:23 <EgoBot> Interpreter chaos installed.
04:55:25 * Sgeo would seriously not think of that
04:55:27 <oerjan> !chaos test 1, 2, 3, end test.
04:55:30 <EgoBot> tEsT 1, 2, 3, eNd tEsT.
04:55:40 <Sgeo> The code you wrote
04:55:45 <Sgeo> If I wanted to get the result you got
04:56:22 <alise_> CASulator! Guys? Anyone? No?
04:56:31 <EgoBot> Interpreter chaos deleted.
04:56:33 <Sgeo> alise_, I might buy one
04:56:53 <coppro> oerjan: hold yer horses
04:56:54 <alise_> coppro: anarchy and chaos
04:56:56 <oerjan> YOU BETTER MAKE A BETTER ONE
04:57:28 <alise_> SOLVE FOR RECURSIVE FOMRULA f(x)==:f(x)
04:58:18 <Sgeo> It's midnight and I haven't had dinner
04:58:29 <alise_> It's 4:58 am and I haven't got to bed.
04:59:34 <alise_> soupdragon: btw with native tex fonts installed and jsmath i get /exceptionally/ good rendering
04:59:50 <oerjan> coppro: are you making something or not?
05:00:06 <coppro> I'm just forgetting my one-liner syntax
05:00:43 <alise_> soupdragon: two nice screenshots: http://imgur.com/1Dnsz.png; http://imgur.com/tgYHE.png
05:00:52 <alise_> all I did was press ctrl-+ and it scaled up the fonts for me
05:01:01 <alise_> ofc I had to install the tex fonts
05:01:10 <alise_> soupdragon: i think that's the fault of whoever wrote that page
05:01:18 <alise_> stuff like that certainly doesn't normally happen
05:01:22 <alise_> anyway, all of that is using native fonts, not images
05:01:25 <alise_> so it's really fast and scales
05:01:27 * Sgeo continues his unending quest of Android knowledge
05:01:30 <alise_> I think it would even print properly
05:01:39 <alise_> most people will use the sprite fonts ofc since they don't have to do anything, but still
05:02:10 <coppro> !addinterp chaos haskell import Data.Char; main = stdGen >>= (\n -> interact $ zipWith ($) (drop n . cycle [toLower, toUpper]))
05:02:10 <EgoBot> Interpreter chaos installed.
05:02:32 <EgoBot> Interpreter chaos deleted.
05:02:47 <soupdragon> "You only need to downlaod ONE of the archives. " -- yes which one?
05:03:07 <oerjan> coppro: getStdRandom, or something like that
05:03:09 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = stdGen >>= (\n -> interact $ zipWith ($) (drop n . cycle [toLower, toUpper]))
05:03:09 <EgoBot> Interpreter chaos installed.
05:03:11 <EgoBot> Interpreter chaos deleted.
05:03:15 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = stdGen >>= (\n -> interact $ zipWith ($) (drop n . cycle [toLower, toUpper]))
05:03:16 <EgoBot> Interpreter chaos installed.
05:03:28 <EgoBot> Interpreter chaos deleted.
05:03:41 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = getStdGen >>= (\n -> interact $ zipWith ($) (drop n . cycle [toLower, toUpper]))
05:03:41 <EgoBot> Interpreter chaos installed.
05:03:46 <oerjan> !haskell :t Data.Random.stdGen
05:04:29 <EgoBot> Interpreter chaos deleted.
05:05:08 <alise_> soupdragon: the split in the brackets and stuff seems to be a glitch in my browser/whatever
05:05:35 <oerjan> or randomRIO if you want a range
05:05:40 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = randomIO >>= (\n -> interact $ zipWith ($) (drop n . cycle [toLower, toUpper]))
05:05:41 <EgoBot> Interpreter chaos installed.
05:06:13 <EgoBot> Interpreter chaos deleted.
05:06:16 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = randomIO >>= (\n -> interact $ zipWith ($) (drop n $ cycle [toLower, toUpper]))
05:06:16 <EgoBot> Interpreter chaos installed.
05:06:49 <coppro> how come EgoBot chats with DCC rather than PM
05:07:06 <soupdragon> !addinterp drome haskell main = do a <- getLine ; print (a ++ a)
05:07:07 <EgoBot> Interpreter drome installed.
05:07:12 <oerjan> coppro: it may be just generating too large n
05:07:20 <EgoBot> Interpreter chaos deleted.
05:07:36 <coppro> !addinterp chaos haskell import Data.Char; import Random; main = randomRIO (0, 1) >>= (\n -> interact $ zipWith ($) (drop n $ cycle [toLower, toUpper]))
05:07:37 <EgoBot> Interpreter chaos installed.
05:07:37 <EgoBot> Interpreter drome deleted.
05:07:38 <soupdragon> !addinterp drome haskell main = do a <- getLine ; print (a ++ a)
05:07:38 <EgoBot> Interpreter drome installed.
05:07:53 <EgoBot> Interpreter drome deleted.
05:08:04 <pikhq> *Wow* it screws with Japanese.
05:08:04 <soupdragon> !addinterp drome haskell main = do a <- getLine ; putStrLn (a ++ reverse a)
05:08:05 <EgoBot> Interpreter drome installed.
05:08:06 <EgoBot> Interpreter drome deleted.
05:08:08 <soupdragon> !addinterp drome haskell main = do a <- getLine ; putStrLn (a ++ reverse a)
05:08:08 <EgoBot> Interpreter drome installed.
05:08:16 <pikhq> What is the source on toLower and toUpper, anyways?
05:08:18 <oerjan> pikhq: it doesn't understand UTF-8
05:08:35 <pikhq> oerjan: Oh, right.
05:08:41 <oerjan> pikhq: it's an IO problem, until last version or so ghc did text input in latin-1
05:08:49 <EgoBot> I roamed under it assa ti rednu demaor I
05:08:49 <pikhq> Actual proper UTF-8 handling was added in 6.12.
05:08:59 <EgoBot> iroamedunderitassatirednudemaori
05:09:06 <pikhq> Rather than "merely not breaking it most of the time".
05:09:24 <EgoBot> amanaplanaccanalpanama
05:09:42 <oerjan> coppro: DCC is better for large files, some of the commands give you program source and stuff
05:10:29 <alise_> I MUST SLEEP! 5:10 am! Soon!
05:10:34 <alise_> Argh! You fucking idiot!
05:10:56 <alise_> soupdragon: ah you are up too
05:10:59 <alise_> that makes me feel better about things
05:11:11 <alise_> [[most likely this was done automatically by our spam filtering program. the program is still learning, and may even have some bugs, so if you feel the ban was a mistake, please send a message to our site admins and be sure to include the exact name of the reddit.]]
05:11:18 <alise_> sorry for making a subreddit called jsmath to try and see if i can make it work.
05:11:22 <soupdragon> I think I would like to write up a mathematical note in jsmath
05:11:42 <coppro> so after all this, my graph module in Haskell is doing fanstastic
05:12:07 <coppro> the module declaration and a newtype
05:13:22 <alise_> replacing newtypes with types is often a good idea
05:13:25 <alise_> less cruft in using them
05:13:29 <oerjan> purely functional graphs are hard to do right, i hear
05:13:51 <alise_> because you don't have to do
05:13:56 <alise_> process (Foo x) = Foo (...)
05:14:04 <alise_> and you can write more point-free, compact definitions
05:14:34 <coppro> am happy I decided to start learning it
05:14:48 <alise_> soupdragon: irc needs tex support
05:15:01 <soupdragon> I can probably hack jsmath into this client
05:16:07 <alise_> I'd love to be able to write in proper notation rather than hackneyed ASCII crud :)
05:16:42 <EgoBot> Interpreter chaos deleted.
05:16:59 <soupdragon> actually it's not a dilemma it's just a problem I don't know how to solve
05:17:12 <soupdragon> I suppose I could use an algebra system
05:17:21 <alise_> btw you can get a latex plugin for pidgin
05:17:22 <coppro> alise_: we should design a CTCP extension for TeX mathmode
05:17:35 <alise_> coppro: easier just to replace in e.g. \(...\)
05:19:26 <oerjan> !addinterp chaos haskell import Data.Char; import System.Random; main = interact =<< zipWith ($) . map ([toLower,toUpper]!!) . randomRs (0,1) =<< getStdGen
05:19:27 <EgoBot> Interpreter chaos installed.
05:20:00 <EgoBot> Interpreter chaos deleted.
05:20:34 <oerjan> !addinterp chaos haskell import Data.Char; import System.Random; main = interact . zipWith ($) . map ([toLower,toUpper]!!) . randomRs (0,1) =<< getStdGen
05:20:34 <EgoBot> Interpreter chaos installed.
05:21:51 <EgoBot> Interpreter chaos deleted.
05:22:05 <oerjan> !addinterp chaos haskell import Data.Char; import System.Random; main = interact . zipWith ([toLower,toUpper]!!) . randomRs (0,1) =<< getStdGen
05:22:05 <EgoBot> Interpreter chaos installed.
05:23:54 <coppro> oh, right, partial application
05:25:15 <alise_> soupdragon: why aren't you in bed anyway
05:25:32 <soupdragon> too many problems though, it didn't work
05:27:24 <oerjan> it's only when you think zero is too many you have a real problem
05:28:35 <alise_> In September, 2007, the so-called "monkey tree phenomenon" caused a minor social mania in Singapore. A callus on a tree resembled a monkey, and believers flocked to the tree to pay homage to the Monkey God.[3]
05:32:21 <alise_> soupdragon: go to sleep so that i can :|
05:35:11 <alise_> soupdragon: stop torturing me, you're feeding my stupid justification of sleep-deprival by being awake
05:35:58 -!- ChanServ has set channel mode: +o oerjan.
05:36:27 <pikhq> alise_: Sleep is for mortals.
05:36:55 <alise_> besides i still have the internet
05:36:59 -!- oerjan has set channel mode: -o oerjan.
05:37:05 <pikhq> I would never have suspected you of being mortal!
05:37:20 <pikhq> I suppose next you'll tell me oerjan is Norwegian!
05:37:34 <oerjan> i suspect alise_ of having a goal to become immortal
05:37:43 <alise_> ffffffff not if it doesn't come with sleep
05:37:47 <oerjan> all that singularity stuff, you know
05:37:55 <oerjan> pikhq: there are rumors
05:38:00 <alise_> soupdragon: whyyy whyyyy
05:39:47 <soupdragon> it must have ended up in the trash somehow
05:39:59 <alise_> soupdragon: WHAT ARE YOU DOING HOW ARE YOU STILL AWAKE
05:41:30 <alise_> soupdragon: no seriouslyi want to know
05:41:51 <soupdragon> I'm going to write something up using jsmath
05:42:10 <alise_> soupdragon: you do realise i NEED to sleep
05:42:39 <pikhq> alise_: If you have biological needs, you should fulfill them.
05:42:55 <pikhq> To do otherwise is either folly or extreme determination.
05:42:58 <alise_> pikhq: CAN'T AGE OF CONSENT LOL okay cheesy joke out of the way, yes but I have a mental need to be a complete fucking idiot
05:43:04 <pikhq> Depending on the scenario.
05:43:04 <alise_> I /cannot make myself sleep/
05:43:07 <soupdragon> alise_: (x-t1)*(x-t2)=t^2-s1*t+s2 with s1 = t1 + t2, s2 = t1*t2
05:43:12 <alise_> soupdragon: NO NO NO DIE
05:43:25 <pikhq> alise_: Think you could talk to someone about that?
05:43:30 <pikhq> Like, at the unit?
05:43:35 <alise_> haha no it is not important :|
05:43:54 <pikhq> alise_: I do not demand that you make yourself sleep.
05:43:56 <soupdragon> can I continue or are you going to continue making a fuss?
05:44:02 <pikhq> However, you should turn off your computer.
05:44:11 <pikhq> ... And your iPhone.
05:44:28 <pikhq> alise_: And then turn off the light.
05:44:32 <pikhq> And stare at the ceiling all night long.
05:44:44 <pikhq> I have a sneaking suspicion you will fail in this task, and end up sleeping instead.
05:45:10 <alise_> The light shines in from outside.
05:45:18 <alise_> People are probably up, some people at least; it is morning, and I am fucking tired.
05:45:29 <alise_> Say I had a button on my desk right now
05:45:38 <alise_> I press it, and my computer is off, and I'm in bed
05:46:05 <alise_> fucking irrational mind
05:46:08 <soupdragon> <SCRIPT SRC="../plugins/noImageFonts.js"></SCRIPT>
05:46:17 <soupdragon> what the heck, why is this not in the <HEAD>?
05:46:29 <pikhq> See, what you should do, then, is develop a strong interest in the pattern of your ceiling.
05:46:35 <soupdragon> <SCRIPT SRC="../easy/load.js"></SCRIPT>
05:46:49 <alise_> soupdragon: if it's in the head
05:46:53 <alise_> the page doesn't appear until it's loaded
05:46:58 <alise_> yes html is retarded, deal with it and put it in body
05:47:11 <alise_> pikhq: What. On my computer seat?
05:47:27 <pikhq> alise_: No, lay in bed.
05:47:42 <alise_> But that involves shutting down the computer, getting into bed, and so many other things.
05:47:54 <pikhq> Get your fucking ass in bed and enjoy the comfort thereof.
05:48:19 <pikhq> I don't care if you go there with the intent to sleep, *surely* you've been sitting sufficiently long that either standing or laying down would feel pretty good by now.
05:48:38 <pikhq> (though, I suspect, as it is $fucking_early there, you will doze off quickly)
05:49:11 <alise_> i'm sorry, my brain isn't listening to you
05:49:14 <alise_> also it's light, i can't sleep in daylight
05:49:24 <alise_> [[Here are the color names most disproportionately popular among women:
05:49:27 <alise_> Okay, pretty flowery, certainly. Kind of an incense-bomb-set-off-in-a-Bed-Bath-&-Beyond vibe. Well, lets take a look at the other list.
05:49:30 <alise_> Here are the color names most disproportionately popular among men:
05:49:36 <soupdragon> man, I have no idea how to make this note
05:50:12 <coppro> f . g) b = f (g b) = (a->b) ->
05:50:18 <coppro> did NOT mean to type that
05:50:20 <alise_> pikhq: please kill soupdragon
05:50:24 <pikhq> alise_: Believe me, when you are tired enough you can sleep in the daylight.
05:50:30 <alise_> I'm not that tired yet
05:50:34 <alise_> also now kill soupdragon
05:50:40 <pikhq> But the computer is obviously not helping.
05:51:24 <pikhq> Also, seriously: have you considered talking to someone about this?
05:51:24 <alise_> http://imgs.xkcd.com/blag/misc_answers.png
05:51:32 <alise_> "i am now colour blind by choice. i can no longer see green."
05:51:35 <coppro> how do you refer to an operator literally in Haskell again?
05:51:45 <pikhq> This is probably a bit of a bad sleeping disorder, or some *really* low will-power.
05:51:56 <soupdragon> alise_, I need some sort of cheatsheet for jsmath
05:51:58 <alise_> pikhq: No, not really. I'm a teenager, we hate going to sleep and we love sleep. I am a geek with pathetically bad self-control and a severe internet addiction.
05:52:01 <alise_> soupdragon: it's just latex
05:52:12 <alise_> soupdragon: what parts
05:52:20 <alise_> pikhq: I've considered taking melatonin supplements per advice from Less Wrong
05:52:57 <pikhq> alise_: Not that I'm one to talk.
05:53:13 <pikhq> I mean, I have had weeks where I went to bed at 5...
05:53:23 <alise_> \begin{align*} and shit like that i guess you mean
05:53:42 <alise_> with \\ after each lign
05:53:56 <coppro> I like the bit about fuchsia
05:53:59 <pikhq> I really should've disambiguated that. 5AM is much easier to pull off.
05:54:04 <alise_> pikhq: both i and ais523 have been on the sleep from 8pm--4am schedule before
05:54:13 <pikhq> Though it makes classes a real bitch.
05:54:40 <alise_> I'm going to cry :| why do i have to have another sleep-deprived sunday, this never happens friday night
05:54:47 <alise_> i just have some sort of gene that refuses to sleep sat. night
05:54:48 <pikhq> alise_: I've hit an 8pm--4am schedule once. By having my sleep schedule push *forward* a bunch each day.
05:55:09 <alise_> seriously: i am going to get melatonin supplements for, like, next week.
05:55:31 <oerjan> alise_: let me guess, your brain is being geared up by using the computer a lot for the first time in almost a week?
05:55:47 <alise_> oerjan: that's a nice theory, but this sort of shit happened to me before the unit
05:56:00 <alise_> you know freerunner schedule, "be awake until you get tired and know you can sleep, then sleep?"
05:56:06 <alise_> I was on extreme freerunner
05:56:17 <pikhq> Ah, the freerunner schedule...
05:56:30 <soupdragon> alise_, http://i.imgur.com/noDzH.png this is pathetic :/
05:56:31 <alise_> "be awake until you're so incoherent that you are not capable of logical thought for more than .5s blips, then stumble to bed and sleep for 12 hours"
05:56:43 <alise_> soupdragon: hmm what did you type for that
05:56:47 <soupdragon> $$ = x^3 - (t_1 + t_2 + t_3) x^2 + (t_1 t_2 + t_1 t_3 + t_2 t_3) x - t_1 t_2 t_3 $$
05:57:19 <oerjan> soupdragon: there is no alignment between consecutive $$ or \[ blocks afaik
05:57:25 <alise_> & (x-t_1)(x-t_2)(x-t_3) \\
05:57:25 <alise_> =& x^3 - (t_1 + t_2 + t_3) x^2 + (t_1 t_2 + t_1 t_3 + t_2 t_3) x - t_1 t_2 t_3
05:57:36 <oerjan> you need to put it in some kind of environment for lining things up
05:57:43 <alise_> pikhq: I guess I may have a sleeping disorder; but I sleep fine, it's just getting to bed.
05:58:20 <pikhq> alise_: Actually. Given that you are a teenager, you most certainly do have one. It's called "being a teenager".
05:58:45 <alise_> soupdragon: I said that earlier you know :-)
05:58:53 <alise_> align* is pretty much the go-to for any alignment job in latex
05:59:22 <alise_> pikhq: yes, but having serious trouble getting to bed at ALL?
05:59:30 <alise_> soupdragon: the former
05:59:36 <alise_> t_1 t_2 is a nicer start than the more "random" t_2 t_3
05:59:46 <oerjan> soupdragon: the former is nicely lexicographic
06:00:49 <oerjan> otoh the second _does_ have its benefits if you're combining it with more things that also have a t_1 special, t_2 special, t_3 special order, i guess
06:01:08 <pikhq> alise_: I have on occasion.
06:01:23 <alise_> pikhq: I have almost universally.
06:01:42 <pikhq> alise_: Also, generally I spend at least an hour in bed before I can fall asleep.
06:01:53 <pikhq> Sometimes significantly more-so.
06:01:55 <alise_> Me too, but I would be fine if only I could get to BED.
06:02:12 <alise_> See, now I'm mentally flaking but it feels like it's "too late" to embed myself. All the damn excuses my brain comes up with.
06:02:24 <pikhq> I think a few times, I have laid in bed for 6 hours, not sleeping, and *then* slept for 8 hours.
06:02:58 <pikhq> It's like "fuck it, I'm going to fuck up my sleep schedule now."
06:03:00 <coppro> heh, (. function) or (function .) is a neat Haskell syntax trick
06:03:10 <alise_> http://www.math.union.edu/~dpvc/jsMath/download/extra-fonts/welcome.html
06:03:15 <alise_> get bbold10-sprite.zip
06:03:25 <oerjan> coppro: IT'S POINTLESS
06:03:28 <alise_> also some sans-serif, fraktur, ams greek, and what I think is \mathcal there
06:03:40 <alise_> just put them in with the sprite fonts
06:04:09 <pikhq> It's the triple-breasted whore of Eroticon Six!
06:04:31 <oerjan> !haskell :t (.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)
06:04:32 <EgoBot> (.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.) :: (b1 -> b -> c)
06:05:00 <oerjan> it gave more lines in DCC
06:05:04 <pikhq> Well, that's a nearly impossible type.
06:05:13 <pikhq> oerjan: Oh, that would help.
06:05:22 <coppro> !haskell :t (.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)
06:05:22 <oerjan> 07:04 =EgoBot> -> (a -> b1)
06:05:22 <oerjan> 07:04 =EgoBot> -> (a1 -> b)
06:05:23 <EgoBot> (.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)(.) :: (b1 -> b -> c)
06:05:57 <pikhq> (b1 -> b -> c) -> (a -> b1) -> a -> (a1 -> b) -> a1 -> c , eh?
06:06:07 <pikhq> Okay, that's *much* less crazy.
06:06:30 <soupdragon> I can't get it to display mathbb correctly
06:06:33 <pikhq> Pointless, perhaps, but much less crazy.
06:06:42 <oerjan> pikhq: it actually repeats when you add another (.)(.)(.)(.)
06:07:45 <alise_> did you install the fonts i told you to
06:08:04 <alise_> <alise_> http://www.math.union.edu/~dpvc/jsMath/download/extra-fonts/welcome.html
06:08:04 <alise_> <alise_> get bbold10-sprite.zip
06:08:04 <alise_> <oerjan> coppro: IT'S POINTLESS
06:08:04 <alise_> <alise_> also some sans-serif, fraktur, ams greek, and what I think is \mathcal there
06:08:04 <alise_> <pikhq> coppro: (.).(.)
06:08:05 <alise_> <alise_> just put them in with the sprite font
06:08:22 <alise_> because it doesn't come with jsmath, just like a few other fonts. so just put them in with the jsmath files, job done
06:09:42 -!- augur has quit (Ping timeout: 265 seconds).
06:09:54 <soupdragon> Can't load fonts/msbm10/def.js: Access to restricted URI denied
06:10:06 -!- alise_ has quit (Remote host closed the connection).
06:13:23 <soupdragon> Let us derive a formula for solutions of the general cubic with coefficients in <SPAN CLASS="math"> \mathbb{Q} </SPAN> using Galois theory.
06:13:53 <soupdragon> but it is very nice http://i.imgur.com/Dcz8N.png
06:16:31 <coppro> express the cubic formula as one equation
06:17:12 <oerjan> a x^3 + b x^2 + c x + d = 0
06:24:58 -!- augur has joined.
06:41:21 -!- adu has quit (Quit: adu).
07:01:24 -!- Sgeo has quit (Quit: Ex-Chat).
07:04:32 <soupdragon> I wish alise could be here to see this ;'(
07:30:31 <coppro> (well, for math, regular TeX will do)
07:32:59 -!- soupdragon has quit (Quit: soupdragon).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:18:53 -!- oerjan has quit (Quit: Good night).
08:47:27 -!- FireFly has joined.
09:17:39 -!- mre has joined.
09:19:40 -!- aox has quit (Ping timeout: 240 seconds).
09:19:45 -!- mre has changed nick to aox.
09:21:26 -!- mre has joined.
09:24:02 -!- aox has quit (Ping timeout: 240 seconds).
09:24:03 -!- mre has changed nick to aox.
09:35:42 -!- aox has quit (Ping timeout: 240 seconds).
09:36:28 -!- Akamaru has joined.
09:43:05 -!- aox has joined.
09:50:06 -!- BeholdMyGlory has joined.
09:56:49 <aox> Hi folks. Is anyone here into Malbolge?
10:04:16 -!- coppro has quit (Quit: I am leaving. You are about to explode.).
10:25:18 -!- cheater3 has quit (Ping timeout: 252 seconds).
10:35:39 -!- kar8nga has joined.
10:41:09 -!- Patashu has joined.
10:42:55 -!- Phantom_Hoover has joined.
10:43:11 <Phantom_Hoover> I feel that this is relevant to yesterday's discussion.
11:07:03 -!- tombom has joined.
11:08:23 -!- cheater2 has joined.
12:04:29 -!- Tritonio_GR has joined.
12:23:51 -!- mre has joined.
12:25:45 -!- aox has quit (Ping timeout: 252 seconds).
12:25:52 -!- mre has changed nick to aox.
12:30:18 -!- reitheti has joined.
12:59:23 <nooga> Phantom_Hoover: we've had like 100 discussions yesterday
13:33:34 <nooga> i've got a german guy in my flat
13:34:26 <nooga> and he talks on skype all the time and listens to techno, but then he leaves to his office because he's an accounter
13:41:18 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
13:42:05 -!- reitheti has quit (Quit: Quitte).
14:30:02 -!- aox has quit (Quit: aox).
15:12:19 -!- Patashu has quit (Quit: MSN = Patashu@hotmail.com , AIM = Patashu0 , YIM = Patashu2).
15:16:52 -!- Tritonio_GR has joined.
15:28:42 -!- alise has joined.
15:28:53 <alise> <soupdragon> Can't load fonts/msbm10/def.js: Access to restricted URI denied
15:28:59 <alise> you did something incorrectly :P
15:29:42 <alise> 01:56:49 <aox> Hi folks. Is anyone here into Malbolge?
15:29:52 <alise> nobody is in to malbolge, it is too hard to be "in" to, but yes, we have Malbolge theorists here
15:29:55 <alise> among them, oerjan
15:35:51 <alise> 22:13:23 <soupdragon> Let us derive a formula for solutions of the general cubic with coefficients in <SPAN CLASS="math"> \mathbb{Q} </SPAN> using Galois theory.
15:36:24 <alise> 23:04:32 <soupdragon> I wish alise could be here to see this ;'(
15:36:28 <alise> 23:30:21 <coppro> but it's not LaTeX
15:36:34 <alise> it renders a v. large subset of latex
15:36:37 <alise> with computer modern fonts
15:37:10 <pikhq> Good morrow, alise.
15:37:17 <alise> SOUPDRAGON DID IT WRONG
15:38:13 <alise> So, 1 + 4 + 9 + 16 + ... = 0.
15:38:29 <alise> Or, more precisely, sum(n=0, inf) n^2 = 0.
15:43:27 <alise> In[20]:= FinDeriv[If[x<0,-1,If[x==0,0,1]],x]
15:43:28 <alise> Out[20]= \[Piecewise]1x==-1||x==0
15:47:40 <alise> In[35]:= FinDeriv[3^x, x]
15:47:40 <alise> In[36]:= FinInteg[2 3^x, {x, x}]
15:47:46 <alise> it actually does proper integration on it
15:47:51 <alise> FinDeriv[x_, v_] := Simplify[(x /. {v -> v + 1}) - x]
15:47:51 <alise> FinInteg[e_, {v_, x_}] := Sum[e /. {v -> y}, {y, -\[Infinity], x - 1}]
15:55:10 -!- Phantom_Hoover has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100423140709]).
15:59:53 <alise> anyone here good with mathematica?
16:01:08 <alise> also, coppro is wrong; most of the "plain TeX" maths you use is from latex/ams-latex.
16:01:29 <alise> unless you write "a \over b" for \frac{a}{b}
16:02:23 <alise> http://www.kohala.com/start/troff/v7man/eqn/eqn2e.ps eqn is pretty fun as the non-latex formatters go
16:02:35 -!- soupdragon has joined.
16:06:31 <alise> soupdragon: you know that div math stuff?
16:06:41 <alise> maybe you have to add one more js include, don't remember
16:06:46 <alise> but you can use $foo$ and \[...\] (or $$ if you must)
16:08:07 <alise> soupdragon: http://www.math.union.edu/~dpvc/jsMath/authors/tex2math.html
16:08:17 <alise> <BODY onLoad="jsMath.ConvertTeX(); jsMath.Process()">
16:08:52 <soupdragon> so should I change <SCRIPT SRC="../jsMath-3.6e/easy/load.js"></SCRIPT> ?
16:08:59 <alise> I think it automatically loads everything you need
16:09:19 <alise> soupdragon: question
16:09:26 <alise> you have the SCRIPT at the bottom of the page right?
16:09:29 <alise> just before </body>
16:09:32 <alise> otherwise, it will lag page loading
16:09:43 <alise> so, move it there if it's not, then just change the body line to
16:09:50 <soupdragon> <SCRIPT> jsMath.ConvertTeX() </SCRIPT> works
16:10:03 <alise> you have the <SCRIPT SRC> at the bottom of the page though?
16:10:11 <alise> because browsers are stupid and stop loading the page until they completely process the script
16:10:13 <alise> so you need it at the end
16:10:19 <alise> right; but it's slower
16:10:31 <alise> soupdragon: my /second/ piece of news
16:10:43 <alise> I did finite calculus in Mathematica, and it can do indefinite finite integration!
16:12:51 <alise> soupdragon: here's the code
16:12:52 <alise> FinDeriv[x_, v_] := Simplify[(x /. {v -> v + 1}) - x]
16:12:52 <alise> FinInteg[x_, v_] := Sum[x /. {v -> a}, {a, -\[Infinity], v - 1}]
16:12:55 <alise> yeah, all two lines of it
16:13:06 <alise> caveat with FinInteg; your variable cannot be "a"
16:13:11 <alise> not sure how to avoid that problem, but who cares
16:13:38 <alise> In[73]:= FinDeriv[34^(x/3), x]
16:13:38 <alise> Out[73]= 34^(x/3) (-1 + 34^(1/3))
16:13:38 <alise> In[74]:= FinInteg[%, x]
16:13:38 <alise> Out[74]= -((34^(2/3 + x/3) (-1 + 34^(1/3)))/(-34 + 34^(2/3)))
16:13:38 <alise> In[75]:= FullSimplify[%]
16:13:44 <alise> hmm ok it should have a simplify around it
16:14:02 <alise> yikes, not even simplify does it; just fullsimplify
16:14:05 <alise> and that's slow...
16:14:09 <nooga> What's in our topic?
16:14:28 <alise> In[79]:= FinDeriv[34^(x/3), x]
16:14:28 <alise> Out[79]= 34^(x/3) (-1 + 34^(1/3))
16:14:28 <alise> In[80]:= FinInteg[%, x]
16:14:51 <alise> soupdragon: oh, and it even does ridiculously clever things
16:14:51 <alise> In[81]:= FinInteg[(2^x)/x, x]
16:14:52 <alise> Out[81]= -2^(-1 + x) HurwitzLerchPhi[1/2, 1, 1 - x]
16:16:51 <alise> soupdragon: one issue is that it can't integrate "1"
16:17:16 <alise> soupdragon: because that's "sum(a=-inf, x-1) 1"
16:17:27 <alise> nope; that simplifies to 1
16:17:53 <alise> soupdragon: I wonder what technique I should use to integrate 1?
16:18:03 <alise> soupdragon: I mean, the problem is negatives
16:18:17 <alise> ... + -3 + -2 + -1 + 0 + 1 + 2 + ... + x
16:18:27 <alise> and what mathematica doesn't realise
16:18:35 <alise> is that the negative/positive terms are irrelevant
16:18:38 <alise> depending on what input it gets, yeah?
16:18:57 <alise> soupdragon: Mathematica will usually just complain about any sum with free variables, treating them as constants and optimising anyway.
16:19:07 <alise> In[99]:= FinInteg[(1 - k)^(1 - k), x]
16:19:07 <alise> During evaluation of In[99]:= Sum::div: Sum does not converge. >>
16:19:07 <alise> \*UnderoverscriptBox[\(\[Sum]\), \(a = \(-\[Infinity]\)\), \(\(-1\) +
16:19:08 <alise> \*SuperscriptBox[\((1 - k)\), \(1 - k\)]\)
16:19:19 <alise> Anyway, there's no possible way it could evaluate it.
16:19:55 <alise> anyway as far as i can tell
16:19:57 <alise> the sum is wrong anyway
16:20:01 <alise> In[102]:= Sum[1, {k, 0, 10}]
16:20:01 <alise> In[103]:= Sum[1, {k, -1, 10}]
16:20:01 <alise> In[104]:= Sum[1, {k, -2, 10}]
16:20:04 <alise> In[105]:= Sum[1, {k, -3, 10}]
16:20:08 <alise> the result would be ... infinity
16:20:14 <alise> so in a sense this summation method CANNOT integrate 1
16:20:34 <alise> soupdragon: same result
16:20:36 <alise> you can't fool it :)
16:21:24 <alise> soupdragon: what is the definition of falling powers again?
16:24:51 <alise> so x^_0_ = 1; x^_1_ = x; x^_n_ = x * (x-1)^_(n-1)_
16:25:42 <alise> FallingPow[__, 0] := 1
16:25:42 <alise> FallingPow[x_, 1] := x
16:25:42 <alise> FallingPow[x_, n_] := x FallingPow[x - 1, n - 1]
16:27:22 <alise> App'rntly not; that yields 0 for all inputs.
16:28:04 -!- Gregor-L has joined.
16:28:45 <alise> Something is broken.
16:29:04 <Gregor-L> FEKK. How is it that I've been gone for only a few days, and already my webernets at home went kablooie
16:31:28 <alise> soupdragon: I got my falling power definition very wrong
16:31:45 <alise> prod(k=0, x-1) x-k
16:33:51 <alise> ah, so i'm doing it in reverse order
16:34:03 <alise> so let's see that's (7-0)(7-1)(7-2)(7-3)
16:34:09 -!- Gregor-L has set topic: I love Punycode in my topics. | xn--v8jad0f7b6z4eoa6v0hk534a7hlwhnnl8s. | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:34:15 <alise> so that's product of (7-k) for k in 0 to n-1.
16:34:18 <alise> but that is exactly what i have.
16:34:22 <alise> FallingPow[x_, n_] := Product[x - k, {k, 0, n - 1}]
16:34:29 <soupdragon> I wanna ask oerjan or oklopol something
16:34:49 <alise> product(k=0, 3) 7-k = (7-0)(7-1)(7-2)(7-3) = 7 6 5 4
16:34:52 <alise> what did I do wrong??
16:35:12 <alise> why is it coming out to 0?
16:35:56 <alise> wtf the product works
16:36:19 <alise> you mean FallingPow=.
16:36:21 <alise> soupdragon: ok then here is a question
16:36:37 <alise> 4 * 3 * 2 * 1 * 0 * -1 * -2?
16:36:46 <alise> i'll just restart mathematica actually
16:36:51 <alise> this is too freaky
16:37:37 <pikhq> Gregor-L: Nice work.
16:38:06 <alise> In[6]:= FinDeriv[FallingPow[x, n], x]
16:38:06 <alise> Out[6]= -(((-1)^n n Gamma[-1 + n - x])/Gamma[-x])
16:38:06 <alise> In[7]:= FinDeriv[FallingPow[n, x], x]
16:38:06 <alise> Out[7]= (-1)^x n (Pochhammer[1 - n, -1 + x] + Pochhammer[1 - n, x])
16:38:09 <alise> soupdragon: good to know right?
16:38:24 <alise> mathematica starts doing crazy shit when it doesn't assume integrality
16:38:26 <alise> soupdragon: why no :(
16:38:31 <Deewiant> Your "FallingPow" is x!/(x-n)!
16:38:45 <soupdragon> It can't realize that the best way to express it is in terms of this newly defined functin
16:39:00 <alise> soupdragon: that's because it doesn't work for non-integers, I think
16:39:12 <alise> Deewiant: Well, I know that falling powers are related to binomials.
16:39:18 <alise> So maybe it is right.
16:42:06 <alise> soupdragon: anyway of course it can't automatically do that
16:42:31 <alise> In[18]:= FinDeriv[FallingPow3[x, n], x] == n*FallingPow3[x, n - 1]
16:42:31 <alise> Out[18]= (n x Gamma[x])/Gamma[2 - n + x] == (n x!)/(1 - n + x)!
16:42:31 <alise> In[19]:= FullSimplify[%]
16:42:49 -!- Mathnerd314 has joined.
16:42:57 <alise> In[20]:= FallingPow3[x + 1, n] - FallingPow3[x, n]
16:42:57 <alise> Out[20]= -(x!/(-n + x)!) + (1 + x)!/(1 - n + x)!
16:43:00 <alise> ^ without the fullsimplify
16:43:03 <alise> soupdragon: no, it couldn't
16:43:05 <alise> the function has to be opaque
16:43:11 <alise> otherwise it couldn't do such transformations
16:43:14 <alise> and by the end, it's been lost
16:43:25 <alise> I don't know why mathematica thinks gamma is simpler than factorial, mind
16:44:04 <alise> I mean it's not ideal I agree
16:44:11 <alise> I just don't think it would be easy to have mathematica do this
16:45:34 <alise> ok what can do it then, apart from your head
16:46:19 <soupdragon> wouldn't it be great if, when you defined a new function - it tried to figure out the algebraic properties of it
16:46:21 <alise> In[5]:= FinInteg[FallingPow[x, n], n]
16:46:21 <alise> Out[5]= E Gamma[1 + x] (1 - ExpIntegralE[n - x, 1]/Gamma[1 - n + x])
16:46:33 <alise> soupdragon: yes, but that's not really practical,
16:46:36 <alise> if every time it did a transformation
16:46:39 <alise> it had to look at EVERY function
16:46:41 <alise> to see if it would work here
16:47:05 <alise> I prefer predictability
16:48:04 <alise> In[10]:= FullSimplify[
16:48:04 <alise> FinDeriv[FallingPow[x, n], x] == n FallingPow[x, n - 1]]
16:48:25 <alise> It's a bit much to expect the computer to discover amazing new facts and identities about random functions automatically then instantly apply them everywhere
16:48:56 <soupdragon> FallingPow[x, n], x] == n FallingPow[x, n - 1] is trivial
16:50:04 <alise> it's checking the derivative
16:50:13 <alise> what you said isn't trivial it's even false :-)
16:50:28 <alise> FinInteg is a bit slow with the FullSimplify; meh
16:50:47 <alise> I don't think anyone else has done indefinite finite integration, so there
16:52:54 <alise> soupdragon: I think there probably is a way to get it to return the result in terms of FallingPower
16:52:58 <alise> AND I'M SURE DEEWIANT KNOWS IT HEM HEM
16:54:02 <Deewiant> If you find out, do tell me, though; I've wanted something similar myself on occasion
16:55:10 <alise> soupdragon: I bet CASulator could do it
16:55:48 <alise> soupdragon: no -- but I'm going to create it
16:55:58 <alise> because dammit, I want a CAS in my pocket
16:56:04 <alise> stupid numeric graphing calculators suck
16:56:27 * soupdragon wants to prove this diophantine has only one solution
16:56:53 -!- Tritonio_GR has quit (Quit: Leaving.).
16:56:59 <alise> I'M SURE A COMPUTER COULD DO THAT
16:57:38 <nooga> just put mathematica on some server and use it via gprs internet connection
16:57:43 <nooga> how hard it could be
16:57:55 <alise> soupdragon: no of course not
16:58:02 <alise> diophantine equations are not quite in the computable realm :P
16:58:08 <alise> nooga: no that's shitty, you're shitty
16:58:11 <alise> firstly, reception
16:58:23 <alise> thirdly, mathematica is not very good UI for a calculator
16:58:32 <alise> and finally, I want to write a damn CAS
16:59:41 <alise> soupdragon: on a tiny embedded calculator cpu???
16:59:47 <alise> nooga: no it's not overkill
17:00:03 <alise> using an unreliable, very very slow internet connection to talk to a $thousands package on a centralised server which will need to be expensive to handle multiple users,
17:00:11 <alise> with a bad ui not tuned to a calculator
17:00:14 <alise> /that/ is overkill
17:00:23 <alise> the basic CAS algorithms are not hard
17:00:29 <alise> it'll be a learning experience anyway...
17:00:32 <alise> CAS stuff, /and/ embedded stuff
17:00:41 -!- augur has quit (Ping timeout: 240 seconds).
17:01:08 <nooga> but cramming usable CAS into a handheld is not easy
17:01:47 <alise> which is the challenge!
17:01:59 <alise> of course I won't have every function ever, and it'll be quite slow for solving and simplification and the like
17:02:08 <alise> but it'll be nice to be able to fiddle over things like algorithm performance without being crazy
17:02:23 <alise> also, I get to draw a font (assuming I use a calculator-y display rather than a high-powered "proper" one)
17:02:29 <alise> (eink isn't an option; too slow response time for typing)
17:03:04 <nooga> you want to build the actual device?
17:03:24 <alise> well I'd write the code on a simulator on a pc first... but sure, long term sort of thing
17:03:36 <alise> I doubt it will get further than a bunch of shit tied together with wires... like, getting plastic made doesn't sound fun
17:07:03 <nooga> ARM+RAM+lcd+keyboard+card reader is not very complicated construction
17:07:23 <alise> arm is a bit much isn't it? if we want long battery life
17:07:39 <alise> I guess the TI-Nspire probably works fine with ARM
17:07:57 <alise> I was personally thinking Z80, since I've never seen or written Z80 code, seems interesting; but perhaps a bit too low powered.
17:08:47 <nooga> i'm building a 8088 based computer and IMHO even 8088 is too minimal
17:09:49 <nooga> but it sucks equally
17:10:14 <olsner> CAS? Compare-and-Swap?
17:10:25 <Mathnerd314> nooga: isn't Wolfram Alpha basically Mathematica on a server?
17:10:58 <alise> olsner: Computer Algebra system
17:11:23 <nooga> + you're wrong alise: 8088 was released in 79 and Z80 was 8080 compatible processor introdouced in 76
17:11:24 <alise> olsner: Mathematica, Maxima, Axiom, etc.; in the old says, Macsyma
17:13:26 <olsner> hmm, not familiar with that term... basically, something that does symbolic processing instead of just calculating?
17:15:11 <soupdragon> http://www.wolframalpha.com/input/?i=ContourPlot[x+%2B+3+x^2+%2B+2+x^3+%3D%3D+6+y^2%2C+{x%2C+-10%2C+10}%2C+{y%2C+-10%2C+10}]
17:17:58 <nooga> it's mathematica + really big inference engine
17:18:16 <nooga> + enormous knowledge base
17:19:15 <fizzie> It also lacks things like a session where you could define stuff in; you ony have that one query. And at least earlier it didn't evaluate arbitrary Mathematica expressions.
17:19:51 -!- Tritonio_GR has joined.
17:20:07 <Mathnerd314> actually, enormous knowledge base is in mathematica already
17:21:29 <alise> Mathnerd314: not all of it.
17:21:33 <alise> and it's hardly enormous.
17:21:37 <alise> olsner: basically, yes
17:22:18 <fizzie> I think there are some Maxima/Axiom/Sage builds/packages for Maemo, but I doubt any have very handheld-optimized UIs.
17:22:32 <alise> olsner: something that can solve equations, do precise calculations involving pi and other real numbers, let you enter "derivative of x^2" in some form and get back "2 x", and then approximate any result to any number of decimal places...
17:22:54 <Mathnerd314> alise: look at this; I think it's big: http://reference.wolfram.com/mathematica/guide/ComputableDataOverview.html
17:22:59 <fizzie> Sage's "notebook" interface is browser-based, I think. It might sort of work. Never tried it out.
17:23:10 <alise> Mathnerd314: i know of that database
17:23:16 <alise> it is sorely lacking; most of it is just toy examples
17:23:39 <alise> 99% of what wolfram says is marketing, believe the opposite until you can verify it's true yourself :)
17:23:50 <alise> fizzie: it is browser based yeah but it is kinda complex and icky
17:24:05 <alise> fizzie: there is an account-based sage notebook thing you could try that out
17:24:13 <alise> slower though of course
17:26:15 <fizzie> alise: I tried to run qbasic in the dosbox I have in this thing, but the default keyboard configuration doesn't support the "fn" key, so all you can input is a-z and .,<> -- couldn't write the "mount" command to get access to qbasic.exe, and anyway without things like digits and punctuation...
17:27:29 <fizzie> The "workaround" was "use a bluetooth keyboard", ha.
17:28:31 <alise> the problem with a casulator is all the keys you need
17:28:41 <alise> i think maybe some sort of subset of the alphabet would be nice so you could like "search" for functions
17:29:03 <alise> like press "insert", type goraph to get goldenratio (phi) or something
17:32:53 -!- oerjan has joined.
17:35:19 <alise> oerjan: ouch don't
17:35:31 <oerjan> there _is_ an Oranjer regular here you know
17:35:43 <nooga> Orajner != oerjan ?
17:35:53 <alise> couldn't you tell from the personality
17:35:58 <alise> oerjan: plz don't say lisa :P
17:36:06 <oerjan> alise: you started it ;D
17:36:16 <alise> oerjan: it is a name with bad connotations for me.
17:36:32 <nooga> DANGER DANGER DANGER
17:36:40 <alise> (it's a lisa's fault I'm in the unit)
17:36:46 * oerjan assumes it's a psychopathic psychiatrist... oh
17:36:55 <alise> nooga: if you don't know you don't want to
17:37:11 <oerjan> alise: so i was pretty close i assume
17:37:22 <alise> oerjan: pretty much.
17:47:28 -!- augur has joined.
17:52:12 -!- augur has quit (Ping timeout: 246 seconds).
18:01:49 <oerjan> Mathnerd314: you have been enlightened!
18:02:10 <oerjan> time for the second initiation rite!
18:02:24 <oerjan> bring the hippopotami!
18:02:52 <alise> soupdragon: uh oh, oerjan has got you!!
18:02:59 <alise> soupdragon: put that math note on the web, i wanna see it
18:03:23 <alise> that one you wrote with jsmath
18:04:23 <oerjan> <soupdragon> I wanna ask oerjan or oklopol something <-- that'll be 50 quibbles
18:06:06 <soupdragon> oerjan, I've figured it out by now [sort of] but if you are curious I was trying to prove that 70^2 is the only square expressible as 1^2 + 2^2 + 3^2 + ... + n^2
18:06:09 <oerjan> (conversion rates are approximate, and may not be up to date)
18:06:47 <oerjan> well step 1 would be to squish that second into a 3rd degree polynomial
18:07:29 <oerjan> also, i didn't know that, unless i've forgotten it
18:08:46 <soupdragon> Yeah, that turns it into an elliptic equation
18:09:04 <oerjan> huh? oh it requires some fancy stuff?
18:10:13 <oerjan> well for a start only one of the factors can be even
18:10:58 <oerjan> so it is divisible by either 2, 8, 32, or some odd power of 2
18:12:00 <oerjan> and also exactly one of them is divisible by 3
18:13:05 <oerjan> !haskell [(n*(n+1)*(2n+1))`div`6| n <-[0..9]]
18:13:42 <oerjan> !haskell [(n*(n+1)*(2*n+1)) `div` 6| n <-[0..9]]
18:13:44 <EgoBot> [0,1,5,14,30,55,91,140,204,285]
18:14:27 <oerjan> !haskell scanl1 (+) . map (^2) $ [0..9]
18:14:29 <EgoBot> [0,1,5,14,30,55,91,140,204,285]
18:15:42 <oerjan> well if soupdragon had it has homework, it might very well require the fancy stuff she's studying
18:16:26 <alise> soupdragon: have you got the native TeX fonts installed, or are you using the sprite ones?
18:16:46 <alise> soupdragon: yaeh they're nice
18:17:04 <alise> the sprite ones work fine for everyone else though, just pixellates on scaling, doesn't print as well and maybe looks a little bit out of place
18:17:08 <soupdragon> alise, you want the HTML? or a PDF or what?
18:17:14 <alise> but it's still better than, say, an inline image
18:17:21 <alise> soupdragon: sure the html, I just wanted to read the note :P
18:17:31 <oerjan> !haskell scanl1 (+) . map (^2) $ [0..20]
18:17:33 <soupdragon> yes but you have to read it beautifully typeset
18:17:33 <EgoBot> [0,1,5,14,30,55,91,140,204,285,385,506,650,819,1015,1240,1496,1785,2109,2470,2870]
18:17:59 <oerjan> !haskell scanl1 (+) . map (^2) $ [0..40]
18:18:01 <EgoBot> [0,1,5,14,30,55,91,140,204,285,385,506,650,819,1015,1240,1496,1785,2109,2470,2870,3311,3795,4324,4900,5525,6201,6930,7714,8555,9455,10416,11440,12529,13685,14910,16206,17575,19019,20540,22140]
18:18:09 <soupdragon> http://pastie.org/952615.txt?key=bnzqq8f8upnmvozmedqihq
18:18:37 * oerjan actually cannot be bothered
18:18:58 <oerjan> and you solved it already so
18:19:28 <alise> <soupdragon> do you have the mathbb font working
18:19:32 <soupdragon> I just found out a theorem which it lives in
18:19:47 <alise> soupdragon: btw i'm not sure but i think you can use the theorem environment stuff with jsmath
18:19:55 <alise> also darn i don't have the js files to hand :(
18:21:13 <alise> I sort of have this idea to make a nice blog with jsmath and nice typography now...
18:21:36 <oerjan> <alise> I don't think anyone else has done indefinite finite integration, so there <-- http://en.wikipedia.org/wiki/Indefinite_sum , i looked at it yesterday
18:22:33 <alise> I just use "sum(k=-inf, x-1) f'(k)", yielding f(x)
18:22:46 <alise> apart from when f'(k) = 1; then it doesn't converge...
18:24:05 <Mathnerd314> oh, I like the calculus of finite differences
18:24:25 <Mathnerd314> there was a talk on it a couple of years ago
18:25:40 <alise> soupdragon: me too what?
18:25:52 <alise> <alise> soupdragon: btw i'm not sure but i think you can use the theorem environment stuff with jsmath
18:25:53 <alise> <alise> I sort of have this idea to make a nice blog with jsmath and nice typography now...
18:25:55 <alise> <alise> I just use "sum(k=-inf, x-1) f'(k)", yielding f(x)
18:26:11 <alise> why does jsmath package depend on apache2.....
18:27:38 <alise> <SCRIPT SRC="jsmath/easy/load.js"></SCRIPT>
18:27:38 <alise> <SCRIPT> jsMath.ConvertTeX() </SCRIPT>
18:27:42 <alise> these should really be at end of body :(
18:27:43 <alise> or page load will lag
18:27:51 <alise> because browsers suck; when they see a script tag
18:27:52 <fizzie> It's just a "recommends" dependency, on "apache | apache-ssl | apache-perl | apache2 | httpd".
18:27:55 <alise> they immediately stop loading the page,
18:28:01 <alise> download the script
18:28:02 <alise> run the entire script
18:28:08 <alise> so it blocks the entire page load
18:28:34 <alise> pineapple: browsers suck; that's a general law of nature
18:28:43 <alise> soupdragon: man that's typeset well
18:28:50 <alise> I bet not even Knuth could distinguish it from TeX :-)
18:28:56 <alise> jsmath is jawesome
18:29:07 <Deewiant> That's not suckiness, there is reason to do that; if you put it up there the expectation is that you want something run at that point
18:29:25 <fizzie> Not doing it would break the broken document.write stuff, though that's a pretty sucky thing in general.
18:29:38 <alise> Deewiant: yes, but still.
18:29:43 <alise> cluttering up the body is lame.
18:29:48 <alise> i love how you can zoom in all the way and it still looks like computer modern
18:29:49 <pineapple> Deewiant: by some people, yes; i get the feeling though that others copy putting it at the top without understanding why
18:30:01 <Deewiant> I'm just saying you can't say the browser sucks because it does that
18:30:12 <Deewiant> It probably follows some spec, and if not, it does what you'd expect
18:30:32 <alise> soupdragon: btw, if you use this on anything public i strongly suggest setting the default to image fonts, not just for symbols
18:30:42 <alise> otherwise it uses the body font for variable names and shit and it looks really ugly
18:30:48 <alise> soupdragon: no no it'll still use tex fonts when available
18:30:54 <alise> just it won't fall back to image-fonts-only-for-symbols
18:31:45 <alise> oh and also it bugs everyone who reads the page to install tex fonts which 99% people really don't need to...
18:31:50 <alise> <STYLE> #jsMath_Warning {display: none} </STYLE> before including jsmath fixes that
18:32:51 <alise> oh and you also need to put
18:32:51 <alise> <SCRIPT SRC="path-to-jsMath/plugins/spriteImageFonts.js"></SCRIPT>
18:32:53 <alise> before loading jsmath
18:32:56 <alise> (yeah this is a pain but ehh)
18:33:40 <alise> soupdragon: and before /that/ script,
18:33:41 <alise> <SCRIPT> jsMath = {Font: {fallback: "image"}} </SCRIPT>
18:33:45 <alise> will make it better for people with non-tex fonts
18:35:41 <alise> The usual way to load jsMath used to be with a SCRIPT tag within the BODY of the document; loading jsMath in the document HEAD used to require additional effort, but as of version 3.0, this is no longer the case. You can now load jsMath in the HEAD and call jsMath.Process() or jsMath.ProcessBeforeShowing() in the document's onLoad handler without ever having any jsMath calls within the BODY of the document. The old jsMath.Setup.Body() call is no longer needed
18:36:00 <alise> soupdragon: does the mathbb stuff show for you?
18:36:21 <alise> just shows as bold for me
18:36:42 <alise> If you are using easy/load.js to load jsMath, you can add the new font to the loadFonts array in easy/load.js. For example,
18:36:42 <alise> loadFonts: ["cmmib10"],
18:36:42 <alise> will cause the cmmib10 font to be available on every page that uses easy/load.js. If you are loading jsMath.js by hand, then in the HTML file that will be using the new font, include the line
18:36:44 <alise> <SCRIPT> jsMath.Font.Load("cmmib10") </SCRIPT>
18:36:46 <alise> right after loading jsMath.js. That will load the data needed for jsMath to handle the cmmib10 font.
18:36:48 <alise> soupdragon: ^^ do that
18:37:52 -!- augur has joined.
18:42:44 <alise> soupdragon: Ehh who wants blackboard bold anyway right?? Stupid Bourbakian invention? ...right? Right?
18:42:58 -!- Phantom_Hoover has joined.
18:43:33 <Phantom_Hoover> http://www.conservapedia.com/Linear_algebra#Application:_Analyzing_Liberal_style_on_Wikipedia
18:44:18 <alise> soupdragon: you do not sound convince.d
18:44:58 <alise> soupdragon: just to check, you have put bbold10-sprite.zip's contents in the right place yeah?
18:45:04 <alise> not the native version
18:45:29 <oerjan> alise: sheesh you're making bourbaki roll in his grave
18:45:39 * oerjan ducks beneath a rock again
18:45:42 <alise> who likes bourbaki anyway!
18:47:10 -!- Phantom_Hoover has quit (Client Quit).
18:47:39 <oerjan> his set theory was one of the first books i found in the library when starting university
18:48:09 <alise> soupdragon: btw you can do \unicode{x2124} e.g. for blackboard Z, but tis ugly
18:48:14 -!- oerjan has quit (Quit: leaving).
18:55:06 <nooga> i don't know why but it's really awesome to ask php guys about a cool feature that php haven't got and then watch their mad reactions ;D
18:55:21 <alise> i have it almost working
18:55:25 <alise> it works with images but not tex fonts
18:55:39 <alise> it works with both
18:55:58 <alise> soupdragon: ok you need msbm10 from http://www.math.union.edu/~dpvc/jsmath/download/extra-fonts/welcome.html in both ttf and sprite
18:56:11 <alise> unpack the sprite one in jsmath/fonts; leave the directory as is don't expand the files just the directory into there
18:56:32 <alise> then edit jsmath/easy/load.js
18:56:35 <alise> and change the loadFonts line to
18:56:37 <alise> loadFonts: ["msbm10"],
18:56:48 <alise> also install the ttf one for your personal use
18:57:18 <alise> soupdragon: and this is what the head of your html page should look like: http://pastie.org/952666.txt?key=buhhmspyfe1dfqkcw54lza
18:57:26 <alise> perfect blackboard bold.
18:58:28 -!- augur has quit (Ping timeout: 248 seconds).
19:00:07 <alise> soupdragon: change jsmath/ in that head to where your jsmath files are obviously
19:00:14 <alise> soupdragon: now will you actually do this or did I do all that work for nothing :)
19:02:54 <nooga> Mathnerd314: not on irc
19:03:24 -!- augur has joined.
19:05:17 <nooga> it's much like twitter
19:06:10 <alise> soupdragon: also if you want \binom and stuff you need http://www.math.union.edu/~dpvc/jsMath/authors/AMSmath.html
19:06:19 <Mathnerd314> nooga: except in a language I don't know :p
19:06:20 <alise> oh wait if it has align it must already been in there
19:06:28 <nooga> Mathnerd314: Polish ofc
19:06:55 <alise> loadFiles: ["extensions/AMSmath.js", "extensions/AMSsymbols.js"],
19:07:16 <alise> yay, now \therefore works
19:09:19 -!- augur has quit (Ping timeout: 260 seconds).
19:09:37 <alise> soupdragon: dammit i broke your thing by adding my thing
19:11:44 -!- hiato has joined.
19:23:15 <hiato> Don't suppose one of you is 'kirarinsnow' for GCJ?
19:23:55 <alise> soupdragon: i fixed it and now it is all pefect
19:24:30 <hiato> Whoever that was did GCJ in Piet, sed, J, Prolog, Octave and SNOBOL
19:27:11 -!- Alex3012 has quit (Ping timeout: 245 seconds).
19:29:30 <alise> soupdragon: yeah I made a lot of changes so ignore my previous advice
19:29:32 <alise> I've got everything working
19:29:35 <alise> proper blackboard bold and all
19:30:15 -!- Alex3012 has joined.
19:30:54 <soupdragon> mathbb is the sexiest thing I have ever seen
19:32:11 <alise> soupdragon: so are you prepared to do the work of downloading some files and changing two files to get it? :P
19:32:44 <alise> only works for you, y'see
19:32:54 <alise> besides, mine actually uses sprite fonts and stuff
19:33:20 <alise> soupdragon: yes, but it only shows for people without the fonts
19:33:30 <alise> the alternative is it not working at all for those people since you don't have the image fonts
19:37:02 <alise> http://pastie.org/952722.txt?key=6caif1bj8kqogegg4k6tua
19:37:20 <alise> the ultimate guide to awesome jsMath, with symbols like \therefore, proper mathbb, even \mathcal if you want, sprite font fallback, native fonts, everything
19:37:31 <alise> do that once and all you need to do in future is copy the head code to another document
19:40:58 -!- augur has joined.
19:42:49 <alise> soupdragon: slight jsMath caveat: copy and paste doesn't really work
19:43:01 <alise> but then you cannot really copy and paste typeset mathematics.
19:43:09 <alise> double clicking on an equation gets you the tex code though
19:45:48 -!- alise_ has joined.
19:49:08 -!- alise has quit (Ping timeout: 260 seconds).
19:50:48 -!- Sgeo has joined.
19:54:51 <alise_> soupdragon: wow you can even hide the mathematics while it's typesetting
19:57:35 <alise_> soupdragon: edit easy/load.js to have this instead of the :1 version,
19:57:45 <alise_> window.jsMath = {Font: {fallback: "image", Message: function () {}}};
19:57:50 <alise_> window.jsMath = {Font: {fallback: "image"}};
19:59:46 <alise_> soupdragon: you can use numbered equations!
19:59:47 <alise_> http://www.math.union.edu/~dpvc/jsMath/authors/eqn-number.html
20:00:24 <soupdragon> yeah there is no doubt about it this is the best way to do math
20:07:10 <alise_> sometime i think i will try using @font-face so that people with modern browsers get tex fonts automatically, even if they haven't installed them!
20:09:56 <alise_> soupdragon: http://91.104.225.35/maths.html i wrote some stuff about zeta summation because dammit i can't stop thinking about how to sum the reciprocals
20:10:06 <alise_> mostly just to test out how pretty i can get some stuff
20:10:17 <alise_> actually i will put in a nice mac font there
20:11:23 <alise_> soupdragon: i will see if i can fix it
20:12:12 <alise_> soupdragon: For me it displays beautifully: http://imgur.com/o21rB.png
20:13:38 <alise_> soupdragon: dunno. i'm trying to fix it now
20:13:40 <alise_> in the meantime refresh
20:13:42 <alise_> I changed the font for OS X
20:14:08 * Sgeo runs a VM he hasn't run in YEARS
20:15:03 <alise_> soupdragon: no i mean the text
20:15:07 <alise_> but the surrounding text
20:15:28 <soupdragon> oh it's changed now but it wasn't changed a moment ago
20:15:49 <alise_> is the change better? in the non-tex bits
20:15:57 <alise_> don't have an os x box to hand, a screenshot would be nice I guess :P
20:16:03 <alise_> ok right now i'll fix this alignment issue
20:17:49 <alise_> i forgot how nice hoefler text is
20:18:58 <alise_> soupdragon: and the options panel says native fonts are being used?
20:23:34 * Sgeo accidentally rtrapped himself in Win98 for a bit
20:23:37 <Sgeo> I escaped, though
20:30:04 <pikhq> soupdragon: Learn all the languages.
20:30:36 <pikhq> I suggest getting a time machine, and finding the common ancestor of all langauges, and start there.
20:30:46 <pikhq> Also, attain immortality.
20:31:21 <pikhq> You're going to be studying for at least a few centuries.
20:33:37 * Sgeo would like to learn all the programming languages
20:35:10 -!- MizardX has quit (Ping timeout: 276 seconds).
20:35:43 * Sgeo would love to learn BANCstar
20:36:22 <pikhq> Sgeo: That would actually take *significantly* less effort.
20:36:43 <pikhq> As programming languages have pretty much all been attested at one time or another.
20:36:54 <pikhq> And are generally designed to be learnable in not-too-much-time.
20:37:15 <pikhq> Though a *royal* fucking pain, I suspect someone could just about manage it in a human lifetime.
20:37:36 <pikhq> Though they'll be damned bitter about the continual invention of programming languages. ;)
20:38:15 -!- anschelsc has joined.
20:38:31 -!- anschelsc has left (?).
20:39:07 <Sgeo> To be fair to BANCstar, I wouldn't really expect an intermediate language to be decent
20:41:02 <alise_> soupdragon: i thought i worked out the sum of the reciprocals
20:41:05 <alise_> but i worked out the product
20:41:45 <fizzie> "Skeletons and maximal balls"; what a great title in this "computer vision" course lecture-slide.
20:42:15 <alise_> maximal balls are maximally manly
20:43:20 <fizzie> First-level point "Skeletons", second-level point "formation with maximal balls", third-level sub-point "the result can be non-homotopic".
20:44:14 <alise_> inf! = sqrt(2 pi); inf! = 1*2*3*4*...; 1/abcd = (1/a)(1/b)(1/c)(1/d); (1/1)(1/2)(1/3)(1/4)...=1/(1*2*3*4*...); so the product of the reciprocals is 1/sqrt(2 pi)
20:46:42 <alise_> soupdragon: I think I fixed http://localhost/maths.html
20:46:58 <alise_> make sure to cmd-refresh or whatever it is
20:47:18 <fizzie> "There's no such file in my localhost."
20:47:40 <Sgeo> Anyone have experience with ILLGOL?
20:47:58 <alise_> http://91.104.225.35/maths.html
20:48:27 <alise_> /should/ display the most beautiful rendering of mathematics ever conceived of :P
20:49:10 <alise_> View source; does it include this line?
20:49:15 <alise_> vertical-align: baseline;
20:50:27 <Sgeo> "I haven't killed anyone at ALL since I started programming in ILLGOL!"
20:50:27 <alise_> still not working? whattt
20:51:26 <alise_> soupdragon: does the view source show this:
20:52:21 <alise_> then ... what the fuck
20:52:44 <alise_> try now, tried something new
20:52:54 <alise_> it should be slightly /above/ the text now
20:53:01 <alise_> it's a bit above the baseline
20:53:04 <alise_> I'll just quickly fix that
20:54:35 <alise_> it is between 0.11 and 0.12 em too high
20:55:32 <alise_> soupdragon: refresh again :P
20:55:36 <alise_> I know this is a pain, just trying to diagnose it
20:56:10 <alise_> this would be a lot easier if I had safari to hand... try now?
20:57:58 <alise_> weird. so am i, firefox 3 here though
21:02:05 -!- alise_ has quit (Read error: Connection reset by peer).
21:02:37 <comex> A number n is irrational iff Unknown control sequence '\msbm' and Unknown control sequence '\msbm'.
21:03:49 -!- alise has joined.
21:06:39 <alise> soupdragon: i'll try to fix it later... can't seem to right now
21:06:48 <alise> http://localhost/galois.html ;; I PLAGIARISED YOU
21:09:07 -!- zzo38 has joined.
21:10:40 <zzo38> I thought about, in a D&D game, how to make a rule that if two magical effects are making anti-magic field with overlap areas, that there can be interference pattern.
21:10:54 <zzo38> In some places they constructively interfere and in others they destructively interfere.
21:11:27 <zzo38> In some places you can still cast spells but the range would be severly restricted to a few inches. If you can't cast a spell, try moving a few inches and try again.
21:12:38 -!- Oranjer has joined.
21:12:47 <alise> soupdragon: Interesting fact: John Baez founded the school of Baezian statistics.
21:14:08 <zzo38> However, in some anti-magic that doesn't count as a magical effect in that way, it will just be antimagic without the interference pattern. Another way there would be no interference pattern is if one of them is caused by an artifact and the other by a non-artifact.
21:16:28 <zzo38> Do you like this idea?
21:16:56 <alise> Absolutely! I like it and I didn't even read it.
21:20:16 <zzo38> Why didn't you read it?
21:21:16 <zzo38> Then how can you know if you like it?
21:21:29 <zzo38> If you didn't read it?
21:21:41 <soupdragon> What should I be doing for the next few hours
21:22:10 -!- hiato has quit (Quit: underflow).
21:22:31 <Oranjer> soupdragon--make a generator for something normally seen as "too vague to define"
21:24:08 <alise> zzo38: i like anything you say it's usually awesome
21:24:17 <alise> soupdragon: TYPESETTING BEAUTIFUL MATHEMATICS
21:24:26 <alise> code is fun occasionally.
21:24:28 <soupdragon> I don't have any more mathematics to write out
21:24:38 <soupdragon> I have lost that drive to write programs
21:24:54 <soupdragon> maybe I should solve a quintic in radicals
21:25:25 <alise> soupdragon: prove that there does not exist a pair of turing-complete machines/automatons A and B such that neither A nor B are Turing-complete, but feeding some code S to A (producing a potentially infinite output) and then feeding the result into B as its code yields Turing-completeness
21:25:51 <alise> (this would remove any doubt about the validity of ais523's 2,3 TM proof; since the generating automaton is not Turing-complete, and the whole construction is Turing-complete, the 2,3 TM would therefore have to be Turing-complete)
21:26:00 <zzo38> soupdragon: OK, play a game, perhaps? Such as Super ASCII MZX Town (series)? Or even make up a game.
21:26:18 <soupdragon> alise - isn't that obvious? The composition of two subturing functions is subturing
21:26:18 <Gregor-L> alise: ... you just said A and B are TC ... do you mean that A and B are non-universal TM, but their interaction is universal?
21:26:31 <zzo38> Or perhaps, play Sub-EBCDIC ZZT Village.
21:26:33 <alise> soupdragon: it isn't obvious because of infinite output
21:26:47 <alise> soupdragon: the generating automaton in ais's proof generates an infinite, non-repeating -- but still generated by a sub-TC -- computer
21:26:48 <zzo38> (Actually, "Sub-EBCDIC ZZT Village" doesn't exist)
21:26:52 <alise> this then invokes TC behaviour in the 2,3 TM
21:26:57 <alise> the argument is whether this makes the 2,3 TM TC
21:26:57 <soupdragon> zzo38, I was thinking of making a rougelike but I don't think I can really be bothered because I expect that i would give up on at before it was really good
21:26:58 <Gregor-L> One stack machine + one stack machine = one Turing machine.
21:27:06 <alise> just makes {automaton} + {2,3} TM
21:27:12 <alise> Gregor-L: No, but that isn't what I said.
21:27:14 <alise> Not just the interaction.
21:27:34 <alise> Instead of B(x), we do B(A(x)), where A's output /may be infinite and non-repeating/, but both A and B are sub-TC.
21:27:47 <alise> Prove then that B(A(x)) is not Turing-complete.
21:28:00 <alise> this has not been proved, if it has I am stating it wrong
21:28:06 <alise> because it was the fundamental disagreement about ais's proof
21:29:15 -!- zzo38 has quit (Remote host closed the connection).
21:31:24 <alise> clearly soupdragon is not nterested :P
21:33:03 <alise> if you disprove it, i.e. it does not hold for all TMs, the controversy remains :(
21:38:26 <alise> soupdragon: so... prove it :P
21:39:00 -!- Sgeo has quit (Ping timeout: 248 seconds).
21:40:32 -!- Sgeo has joined.
21:41:19 <Sgeo> FUCK MY INTERNET CONNECTION
21:41:24 <Sgeo> FUCK IT WITH A RUSTY HOSE
21:42:21 <alise> soupdragon: Why not?
21:42:31 <alise> I'd say it's very obviously intuitive. Just perhaps not true.
21:42:31 <soupdragon> I think you have to put stronge restrictions on G than just subturing
21:43:47 <soupdragon> I haven't constructed but I can imagine there could be a TC FG where F and G are subturing
21:44:04 <soupdragon> (probably exploiting the bijection between |N and |N*)
21:44:32 <alise> ok, but why do you imagine this?
21:45:12 <alise> It seems obvious to me that you cannot have complexity contained in a one way strand of communication, if the thing being communicated is not complex, and the thing it is being fed to does not have the power to create complex things,
21:45:17 <alise> then whence the complexity (Turing-completeness)?
21:49:02 <soupdragon> if you needed to calculate something, but you could only generate a bunch of stuff that included the answer - that might be subturing. If you could only find the correct answer in a stream that had it already, that would be subturing too -- but perhaps together turing
21:50:00 -!- Gracenotes has quit (Remote host closed the connection).
21:50:17 <alise> so like, that formula that gives negative -> not prime; positive -> prime
21:50:19 <alise> and a prime-checker
21:50:21 <alise> is a prime generator
21:50:27 <alise> even though neither can generate just primes?
21:50:39 <alise> I'm not sure it applies to Turing-completeness but it's a good argument
21:54:07 <Sgeo> My step-mother asked me to look up "Calorie City" as a joke grr
21:55:54 <Sgeo> She often asks me to look up places, so I thought it was just another place to look up
21:56:00 <Sgeo> So she wasted my time with a pathetic joke
21:56:52 <alise> You could not look up the places in future.
21:57:09 <alise> Anyway, how is it a joke? It lacks notable qualities, like even the seeming intent to be funny or meaningful in some way.
21:57:13 <alise> Maybe I just don't get it.
21:57:47 <soupdragon> Tell your mom it says gullible on the ceiling
21:57:56 <soupdragon> if she falls for it you have made her look up
21:58:10 <alise> no, tell her gullible is actually spelled "gulliable", and gullible is the americanisation
21:58:14 <alise> tell her to look it up in the dictionary
21:58:48 <Sgeo> alise, she always tries to get me to eat, which I guess is nice
22:02:31 <soupdragon> I am really stunned by diophantine equations
22:03:29 <uorygl> Yeah, eating is kind of a good thing.
22:07:28 -!- `q` has changed nick to Quadrescence.
22:07:44 -!- uorygl has quit (Ping timeout: 246 seconds).
22:13:20 -!- alise has quit (Read error: Connection reset by peer).
22:14:42 -!- augur_ has joined.
22:15:07 -!- augur has quit (Read error: Connection reset by peer).
22:20:35 -!- augur_ has quit (Ping timeout: 264 seconds).
22:43:07 -!- kar8nga has quit (Remote host closed the connection).
22:53:54 -!- tombom has quit (Quit: Leaving).
23:03:26 -!- oerjan has joined.
23:25:16 <oerjan> !haskell let prevSq n = (floor . sqrt $ fromIntegral n)^2 in [n - prevSq n | n <- [1..50]]
23:25:33 <EgoBot> [0,1,2,0,1,2,3,4,0,1,2,3,4,5,6,0,1,2,3,4,5,6,7,8,0,1,2,3,4,5,6,7,8,9,10,0,1,2,3,4,5,6,7,8,9,10,11,12,0,1]
23:26:03 <oerjan> !haskell let prevSq n = (floor . sqrt $ fromIntegral n)^2 in [n - prevSq n | x <- [1..50], let n = 2*x^2]
23:26:05 <EgoBot> [1,4,2,7,1,8,17,7,18,4,17,32,14,31,9,28,2,23,46,16,41,7,34,63,25,56,14,47,1,36,73,23,62,8,49,92,34,79,17,64,113,47,98,28,81,7,62,119,41,100]
23:27:19 <oerjan> !haskell let prevSq n = (floor . sqrt $ fromIntegral n)^2 in [x | x <- [1..50], let n = 2*x^2, n - prevSq n == 1]
23:28:20 <oerjan> !haskell 28*29*(2*28+1)`div` 6
23:29:38 <oerjan> !haskell let prevSq n = (ceiling . sqrt $ fromIntegral n)^2 in [x | x <- [1..50], let n = 2*x^2, n - nextSq n == -1]
23:30:08 <oerjan> !haskell let nextSq n = (ceiling . sqrt $ fromIntegral n)^2 in [x | x <- [1..50], let n = 2*x^2, n - nextSq n == -1]
23:30:47 <oerjan> !haskell 24*25*(2*24+1)`div` 6
23:35:39 <oerjan> !haskell let prevSq n = (floor . sqrt $ fromIntegral n)^2 in [x | x <- [1..1000], let n = 2*x^2, n - prevSq n == 1]
23:38:43 <oerjan> !haskell let nextSq n = (ceiling . sqrt $ fromIntegral n)^2 in [x | x <- [1..1000], let n = 2*x^2, n - nextSq n == -1]
23:44:57 -!- BeholdMyGlory has quit (Remote host closed the connection).
23:48:35 -!- aox has joined.