00:04:24 <fizzie> SDL isn't exactly known for having much convenience code around the low-level API in the "core".
00:04:46 <pikhq_> This is a case of SDL not merely lacking convenience code.
00:04:56 <pikhq_> This is a case of SDL being *harder than the APIs it's wrapping*.
00:05:14 -!- oerjan has quit (Ping timeout: 245 seconds).
00:13:13 -!- oerjan has joined.
00:14:49 * oerjan has a hunch he lost connection, but isn't sure
00:15:51 <oerjan> As opposed to logging out voluntarily, that is.
00:17:55 -!- hagb4rd has quit (Quit: hagb4rd).
00:19:22 -!- MatressDude has joined.
00:23:00 -!- MDude has quit (Ping timeout: 246 seconds).
00:24:39 <fizzie> I would think the reasonable way to implement a synchronous sound API in SDL would be to build it on top of the asynchronous one, instead of trying to map those calls to something backend-specific that might not even exist, in which case it's pretty close to convenience code from the implementer's point of view.
00:30:45 -!- pikhq has joined.
00:30:45 -!- pikhq_ has quit (Ping timeout: 252 seconds).
00:32:46 -!- calamari has quit (Quit: Leaving).
00:35:36 -!- calamari has joined.
00:37:39 -!- calamari has quit (Client Quit).
00:37:51 -!- calamari has joined.
00:47:02 -!- hagb4rd has joined.
01:14:23 <Sgeo> tswett, monqy MUFFIN
01:27:01 <monqy> what type of muffin
01:30:56 <Sgeo> More specifically, a FFMUFFUFFIN
01:47:28 -!- calamari has left ("Leaving").
01:55:59 <Sgeo> http://www.blue-room.org.uk/wiki/Widowmaker
01:56:04 <Sgeo> http://www.youtube.com/watch?v=A25FRpkbDxU
02:39:41 -!- hagb4rd has quit (Ping timeout: 276 seconds).
03:31:23 -!- hagb4rd has joined.
03:33:03 -!- Madoka-Kaname has joined.
03:49:09 -!- calamari has joined.
03:50:25 -!- calamari has left.
04:48:32 -!- oerjan has quit (Quit: Good night).
04:57:25 -!- MatressDude has changed nick to MSleep.
05:13:07 -!- asiekierka has joined.
05:17:32 -!- monqy_ has joined.
05:17:47 -!- monqy has quit (Disconnected by services).
05:17:52 -!- monqy_ has changed nick to monqy.
05:25:16 <Sgeo> monqy, tswett muffin
05:25:40 <monqy> is it a better flavor this time
05:25:49 <Sgeo> It's cake flavored
06:06:37 -!- augur has quit (Ping timeout: 245 seconds).
06:32:56 -!- itidus20 has joined.
06:33:45 -!- augur has joined.
06:35:47 -!- itidus21 has quit (Ping timeout: 245 seconds).
06:46:48 -!- MoALTz has quit (Ping timeout: 244 seconds).
06:47:13 -!- MoALTz has joined.
07:16:01 -!- pikhq_ has joined.
07:16:15 -!- pikhq has quit (Ping timeout: 244 seconds).
07:17:29 -!- MSleep has quit (Read error: Connection reset by peer).
07:17:47 -!- MDude has joined.
07:45:49 -!- Deewiant has quit (Read error: Connection reset by peer).
07:50:47 -!- Deewiant has joined.
09:06:17 -!- elliott has joined.
09:29:28 <lambdabot> elliott: You have 3 new messages. '/msg lambdabot @messages' to read them.
09:32:20 -!- cheater_ has joined.
09:32:29 <elliott> `addquote <itidus21> it sounds awful though i know.. the thought of a whole life without any hope of lawn chairs in hawaii sipping alcohol and watching bikini clad women dance with maracas
09:32:37 <HackEgo> 826) <itidus21> it sounds awful though i know.. the thought of a whole life without any hope of lawn chairs in hawaii sipping alcohol and watching bikini clad women dance with maracas
09:33:05 <elliott> <lambdabot> fizzie said 12h 31m 19s ago: FWIW, the esocolors in L*a*b* are (138.4, -29.2, 30.6), (149.6, -17.9, 59.0), (143.8, -2.7, 38.9) and (135.5, 4.6, -12.0). http://zem.fi/~fis/esocolor.png shows the
09:33:05 <elliott> <lambdabot> slice of L*a*b* for L* = 141.8 (mean of those) that's representable in sRGB. The red dots are the colors' a* and b* coords.
09:33:25 <elliott> fizzie: That seems a reasonable enough range. How viable do you think your circle-in-L*a*b* idea would be?
09:33:39 <elliott> I hate having to choose more than one variable.
09:35:02 <elliott> 20:49:45: <oerjan> <MDude> I'd say most economic theories work if you presume human nature magically conform to thier expectations. <-- as i and probably others have said before, every form of government works if all people are perfect, including nazism.
09:35:16 <elliott> oerjan: Uhh, unless you're Jewish, or gay, or...
09:35:38 <elliott> oerjan: If by "perfect" you mean "Nazis", then yes.
09:35:40 -!- cheater has quit (Ping timeout: 265 seconds).
09:35:45 <elliott> That's quite a nonstandard definition.
09:38:22 <shachaf> elliott: Wait, how do <--- arrows work with multiple people being quoted?
09:39:05 <shachaf> elliott: Anyway, tell me about coinductive functions.
09:41:50 <elliott> shachaf: http://stackoverflow.com/questions/9819578/whats-the-meaning-of-io-actions-within-pure-functions
09:42:00 <elliott> I felt like hurting you, so I linked that.
09:42:19 -!- hagb4rd has quit (Ping timeout: 246 seconds).
09:44:49 <shachaf> The title looks promising.
09:45:09 <shachaf> "whats-the-meaning-of-io-actions" -- conal "conal" conal?
09:45:49 <shachaf> This person is distinctly not conal. :-(
09:49:24 <elliott> How can I give a monad the ability to execute actions from another one then, by giving it the possibility to pattern match against the values it contains? – Riccardo 2 mins ago
09:49:30 <elliott> Do you think I can just ignore this question?
09:49:49 <shachaf> I think you already failed.
09:50:00 <shachaf> See, I wasn't even seeing that question until you put it in here.
09:50:19 <shachaf> elliott: Remind me what the reason that you felt like hurting me in the first place was again?
09:51:29 * shachaf is as many as four tens. :-(
09:55:16 <shachaf> elliott: "But a monad has no ability to execute the actions of another monad unless you give it that ability."
09:55:22 <shachaf> That's kind of vague and misleading.
09:55:42 <shachaf> "a monad" doesn't execute anything.
10:03:44 -!- hagb4rd has joined.
10:13:00 <elliott> shachaf: Can you suggest a less misleading reply to that nonsense?
10:14:36 <shachaf> I would suggest not saying that part at all.
10:20:26 <elliott> shachaf: I don't really expect to be able to enlighten this person if the rest of my answer didn't, anyway.
10:21:33 <elliott> How can I give a monad the ability to execute actions from another one then, by giving it the possibility to pattern match against the values it contains? – Riccardo 34 mins ago
10:21:34 <elliott> By writing methods that convert one monad into another, or do some execution. Control.Monad.ST.stToIO converts an ST computation into an IO computation, for example. – Louis Wasserman 13 mins ago
10:21:40 <elliott> I like it when other people handle questions I don't want to.
10:21:46 * elliott will avoid nitpciking about "method".
10:22:24 -!- MoALTz_ has joined.
10:24:58 -!- MoALTz has quit (Ping timeout: 252 seconds).
10:25:18 <elliott> http://www.mobiletextinput.com/Product/What_is_SlideIT/ Hmm, is this a ripoff of Swype or vice-versa?
10:26:46 -!- MoALTz_ has quit (Ping timeout: 246 seconds).
10:27:06 * shachaf rips both off, creates SwypeIT
10:28:28 -!- MoALTz has joined.
10:34:27 -!- azaq23 has joined.
10:37:43 -!- MoALTz_ has joined.
10:39:17 <hagb4rd> wertzuioknbvfderghjkoiuztrewertzuiuztrewertzuioiuztrghjkjhgfds
10:39:29 <fizzie> elliott: I'm honestly not sure how much sense a circle makes, but here are some colors picked from one: http://zem.fi/~fis/esocolorlab.html
10:39:32 <fizzie> The first table has the original colors, and then colors at angles 0, 90, 180 and 270 degrees from the about-largest circle centered at the white point that can fit in sRGB for L* = min (second line) or mean (third line) of the originals. And then there's the same thing except with a rotating 'phase'.
10:39:37 <fizzie> Technically the "perceptual distance" between any horizontal neighbours in the "phase-wheel" tables should be identical, but YMMV -- it's only a rough approximation, after all.
10:39:42 <hagb4rd> i doesn't work perfectly yet
10:39:53 -!- MoALTz has quit (Ping timeout: 260 seconds).
10:40:42 <hagb4rd> but it _is_ perfect for long words..indeed
10:44:27 <elliott> fizzie: I don't quite understand the results, but they don't look promising. :(
10:45:08 -!- MoALTz_ has quit (Ping timeout: 260 seconds).
10:45:19 <elliott> I guess science is just not equipped to handle the task "give me four different pastel-looking colours that are perceptually evenly-spaced".
10:48:45 <fizzie> Not if you actually want colours you like, perhaps. I mean, those look more or less "pastel" to me, and as for even-spacedness, well... maybe not so much, but it's at least wrong in a justified way.
10:49:38 <fizzie> Anyway, you can't have more than three unique evenly spaced points in a two-dimensional space, which is what an equal-lightness slice of L*a*b* is.
10:50:20 <elliott> fizzie: Clearly we need a... four-dimensional colour tesseract.
10:50:49 <fizzie> I guess you could stick an arbitrarily oriented regular tetrahedron in the 3D L*a*b* space, but that's probably going to give nonsense out too.
10:51:05 <elliott> fizzie: Anyway, yes, they're pastel. But they're not "primary" in the way that the existing colours are; that is, the green, yellow and blue are a lot more *strongly* different than the ones the L*a*b* circle is spitting out.
10:54:02 -!- MoALTz has joined.
10:54:04 <fizzie> That much is true. Sadly, I could only have a so large radius for a "grayness-centered" circle before going outside sRGB. The existing points are more like on an ellipse in labspace.
10:54:08 <fizzie> For the record, the one black box (at 18 degrees for L* = 88.6) would be (1.000322, 0.831912, 0.837298) or "#100d4d6" in sRGB.
10:58:30 <elliott> Would it look horrible if you picked a circle outside sRGB and then "sampled it down" to sRGB?
10:59:02 <elliott> I mean, I've seen a circle thing with "all" of L*a*b* that mapped down the stuff outside sRGB.
11:02:03 -!- MoALTz has quit (Ping timeout: 260 seconds).
11:05:58 -!- MoALTz has joined.
11:08:11 <fizzie> FWIW, I added the colors you get if you draw the ellipse that somewhat approximates the four original points (with major axis oriented along the esoyellow -> esoblue direction), and then pick points at the places where the major and minor axes cross the ellipse. But they're not especially pleasant colours either, especially the yellow after being brought down to the same level of "lightness as ...
11:08:28 <fizzie> (Of course if you *want* a vaguely urine-looking header, go ahead.)
11:10:30 -!- monqy has quit (Quit: hello).
11:10:56 <elliott> Well, the peach is kind of the odd one out as far as the colours go.
11:11:06 <elliott> But I doubt you'd get anything especially better by omitting it.
11:12:16 -!- MoALTz_ has joined.
11:13:32 -!- MoALTz has quit (Ping timeout: 264 seconds).
11:15:51 -!- MoALTz__ has joined.
11:17:43 -!- MoALTz has joined.
11:17:56 -!- MoALTz_ has quit (Ping timeout: 244 seconds).
11:20:31 -!- MoALTz__ has quit (Ping timeout: 244 seconds).
11:23:25 <elliott> Irony: http://en.wikipedia.org/wiki/Digital_object_identifier is in [[Category:Pages with DOI errors]].
11:24:08 -!- MoALTz has quit (Ping timeout: 244 seconds).
11:26:37 -!- MoALTz has joined.
11:28:39 -!- augur has quit (Remote host closed the connection).
11:35:07 -!- cheater_ has quit (Ping timeout: 272 seconds).
11:35:34 -!- derdon has joined.
11:37:15 <elliott> http://www.reddit.com/r/haskell/comments/r8861/i_love_this_tv_show_sooooo_fucking_awesome/ I... don't understand.
11:42:02 -!- cheater_ has joined.
11:42:28 -!- hagb4rd2 has joined.
11:43:46 -!- hagb4rd has quit (Ping timeout: 246 seconds).
11:59:50 -!- MoALTz_ has joined.
12:00:21 -!- augur has joined.
12:03:00 -!- MoALTz has quit (Ping timeout: 244 seconds).
12:03:54 -!- MoALTz has joined.
12:06:15 -!- MoALTz_ has quit (Ping timeout: 250 seconds).
12:08:08 <itidus20> Bom dia, sarg. Haskell! Lindo dia para um piquenique!
12:09:05 -!- MoALTz_ has joined.
12:09:37 <itidus20> Good morning, sarg. Haskell! Nice day for a picnic!
12:10:09 -!- MoALTz has quit (Ping timeout: 250 seconds).
12:15:43 -!- MoALTz__ has joined.
12:18:23 -!- MoALTz_ has quit (Ping timeout: 250 seconds).
12:20:38 -!- hagb4rd2 has changed nick to hagb4rd.
12:20:59 -!- MoALTz__ has quit (Ping timeout: 250 seconds).
12:26:00 -!- cheater__ has joined.
12:26:42 -!- cheater_ has quit (Read error: Operation timed out).
12:30:08 -!- cheater_ has joined.
12:33:03 -!- cheater__ has quit (Ping timeout: 260 seconds).
12:41:07 -!- ais523 has joined.
13:04:53 -!- MoALTz has joined.
13:22:57 -!- MoALTz has quit (Ping timeout: 250 seconds).
13:43:20 -!- hagb4rd2 has joined.
13:44:10 -!- hagb4rd has quit (Ping timeout: 246 seconds).
13:45:59 -!- pikhq has joined.
13:46:04 -!- pikhq_ has quit (Ping timeout: 252 seconds).
14:09:54 -!- const has quit (Ping timeout: 252 seconds).
14:10:58 -!- nortti has joined.
14:11:01 -!- SimonRC has quit (Ping timeout: 244 seconds).
14:18:08 -!- SimonRC has joined.
14:23:36 -!- ais523 has quit (Ping timeout: 245 seconds).
15:03:16 <elliott> RocketJSquirrel: Did the others not work out?
15:04:20 <Sgeo> When did Magnatune start sucking?
15:08:31 <elliott> This guy on proggit is incredibly annoying.
15:21:13 -!- Slereah_ has joined.
15:22:09 -!- Slereah has quit (Ping timeout: 276 seconds).
15:24:09 * Sgeo wonders if Learn Seven Languages in Seven Weeks is any good
15:24:33 <elliott> You can't learn a language in a week.
15:24:57 <elliott> Also, everyone who has mentioned that book in #haskell has been clueless.
15:25:57 <elliott> You say that a lot. What does it mean?
15:26:41 <elliott> fizzie: What was that link to the backgroundy versions of the header colours?
15:31:06 <Sgeo> Oh, it's not called "Learn" Seven Languages in Seven Weeks
15:41:42 -!- Zuu has quit (Ping timeout: 260 seconds).
15:41:50 -!- calamari has joined.
15:44:20 -!- tzxn3 has joined.
15:45:16 -!- hagb4rd2 has quit (Ping timeout: 246 seconds).
15:47:37 <RocketJSquirrel> elliott: Their bootstrapping script only works with bash, and bash doesn't work right now due to some strange musl issue.
15:49:21 <elliott> lol, what shell are you using?
15:50:11 -!- variable has joined.
15:50:26 <nortti> real men use Minix 1 for m68k when they need to use *nix and ash is standard shell on it
15:55:20 -!- ais523 has joined.
15:56:12 <ais523> yay, teaching over for the academic year
15:56:17 <ais523> not counting potential marks entry or exam marking
15:59:05 <nortti> calamari: no. real men use Thompson shell
15:59:51 -!- Patashu has quit (Quit: MSN: Patashu@hotmail.com , Gmail: Patashu0@gmail.com , AIM: Patashu0 , YIM: patashu2 , Skype: patashu0 .).
16:00:17 <elliott> nortti: Real men use Multics.
16:00:35 -!- Phantom_Hoover has joined.
16:01:39 <nortti> elliott: No they toggle in a microkernel from the front panel
16:02:32 <ais523> real men made their computer out of transistors, and have their boot code hardcoded from memory
16:02:40 <elliott> Microkernels are for wusses.
16:04:36 <Phantom_Hoover> Real men bend silicon into shape with their bare hands and dope it with their spit.
16:05:06 <ais523> is spit n-type or p-type?
16:05:15 <ais523> it's mostly made of water and protein, right?
16:05:21 <nortti> actually they use butterflies to toggle bits in core memory of their machines built out of relays that they mined the needed minerals for
16:06:29 <elliott> Real men don't use computers.
16:06:32 <elliott> There, stupid meme killed.
16:06:41 <calamari> just use an abacus for memory and execute the instructions by hand
16:08:25 <elliott> Phantom_Hoover: Yes, yours was the only good one.
16:09:05 <ais523> the problem is that such discussions collapse into self-parody too quickly
16:09:13 <ais523> and are indistinguishable from it even if they aren't meant to be
16:09:37 <elliott> The main problem is that the meme is dumb in the first place.
16:11:05 -!- hagb4rd has joined.
16:19:40 -!- SimonRC has quit (Ping timeout: 244 seconds).
16:21:27 -!- SimonRC has joined.
16:36:07 <elliott> ais523: Hey, should I add <ref> and <math>?
16:36:24 <ais523> elliott: depends on if you think anyone would use them
16:36:35 <ais523> also, why not zzo38's custom mathy thing?
16:36:36 <nortti> elliott: isn't <math> depreciated?
16:42:10 <elliott> ais523: Because zzo38's custom mathy thing (a) needs heavy modifications to be usable on anybody else's computer by his own admission, (b) obviously doesn't support standard LaTeX syntax, and (c) has no MediaWiki extension, he wants me to write the extension for him to use it.
16:42:21 <elliott> nortti: I mean, no, not that I'm aware of, what's its successor?
16:42:49 <nortti> I just thought that it wouldn't pass vadility essor
16:43:13 * elliott has no idea what essor is.
16:43:14 <nortti> s/essor/check without errors/
16:44:06 <elliott> Oh, are you referring to some HTML <math> tag? (I think it might have had such a tag in the past.)
16:44:21 <elliott> I'm referring to the MediaWiki <math> tag, which takes LaTeX and spits out a PNG on the page.
16:44:54 <nortti> html <math> tag was in html 3.0-3.2
16:45:14 <elliott> ais523: I think <ref> would be useful, since there are quite a few articles with citations done in a rather ad-hoc style (mostly the theoretical CS stuff).
16:46:09 <elliott> oh, the math extension is transitioning to MathJax, I guess I'll wait until that's done before installing it
17:11:08 <elliott> http://stackoverflow.com/suggested-edits/225891
17:13:29 -!- asiekierka_ has joined.
17:13:35 -!- asiekierka_ has quit (Remote host closed the connection).
17:16:44 <fizzie> elliott: http://zem.fi/~fis/esocolor.html has the HLS lightened versions. First table is just setting a fixed L and keeping hue/saturation, while the second table sets L to l+x*(1-l), where l is the old l and x is that value on the header row, in a sense "lightening" all the colors by as much.
17:17:33 <elliott> fizzie: It has expanded since last time, I see.
17:17:55 <elliott> I suppose lightening them with L*a*b* would go as disastrously as the other Lab experiment?
17:18:21 <fizzie> I can try it, but I doubt it differs from HLS terribly much.
17:18:45 <elliott> 0.96 from the first table and 0.60/0.70 from the second table seem good.
17:19:52 <fizzie> For completeness, I added a third table that has equal-darkening values too.
17:20:06 <fizzie> Not that you probably need those.
17:20:19 <fizzie> Except for the GOTH EDITION the 0.1 values from that might work.
17:20:34 <elliott> fizzie: Well, it could be useful for Halloween, I suppose.
17:21:30 <elliott> fizzie: If I want to make the wiki DARK and SPOOKY.
17:22:05 <elliott> fizzie: I THINK YOU'LL FIND that orig = 1.00, unsurprisingly.
17:22:11 <elliott> Maybe you should remove the orig column and flip the table. :p
17:22:25 <elliott> And then rename the 1.00 column to orig. Clearly perfection is a requirement.
17:22:50 <fizzie> 1.0 column of second table is not the most useful either.
17:22:53 <elliott> In fact, you should completely automate the table-creation process.
17:23:10 <Sgeo> tswett, is it ok that I said hi?
17:23:51 <ais523> so what are these esocolor things?
17:24:55 <elliott> Sgeo: Saying hi is forbidden.
17:24:58 <elliott> ais523: We're painting fizzie's house.
17:25:09 <ais523> not sure I believe that
17:25:23 <elliott> Some people believe the world is round, man.
17:25:27 <elliott> fizzie: Housepainting, right?
17:25:40 <ais523> elliott: does anyone /really/ believe the world is piecewise-linear?
17:33:11 <Sgeo> Q: "What about Lunar Eclipses?"
17:33:11 <Sgeo> A: A celestial body, known as the antimoon, passes between the sun and moon. This projects a shadow upon the moon.
17:34:16 <nortti> Sgeo: where is that quote
17:34:29 <Sgeo> http://www.theflatearthsociety.org/wiki/index.php?title=FAQ
17:34:54 <elliott> i still haven't forgiven sgeo for saying hi
17:35:52 <nortti> oh god. the flat earth society
17:36:14 <fizzie> elliott: Just for completeness, L*a*b*-lightened/darkened versions: http://zem.fi/~fis/esocolor3.html
17:41:13 <fizzie> The differences are kinda-sorta slight. E.g. the 0.5 columns are identical for 10 of the 12 eight-bit components, and the remaining two differ by 1.
17:41:56 <elliott> fizzie: Well, the darkening table is rather different.
17:42:08 <elliott> (And rather better, I think.)
17:44:50 <elliott> I really need to replace that peach, though.
17:45:35 -!- hagb4rd2 has joined.
17:45:40 -!- hagb4rd has quit (Ping timeout: 246 seconds).
17:46:53 <fizzie> The darkening table is different because it's done differently; it's "equally spaced points on the line between the original and the nonchromatic black", so it sorta desaturates while it's at it. I added a third table that just brings L* down to 0, but those tend to again fall outside the vaguely shaped sRGB region.
17:49:18 <elliott> fizzie: L*a*b* is scaring me a bit. I mean, I know *I* can't imagine a colour not inside sRGB.
17:49:23 <elliott> Who is to say such colours even exist?
17:49:46 -!- ais523 has quit (Remote host closed the connection).
17:50:04 <fizzie> Any laser pointer light is outside sRGB, unless it happens to exactly coincide with one of the three sRGB primaries.
17:50:48 <elliott> This is the real question.
17:51:15 <olsner> elliott: you'll love this one: https://twitter.com/#!/jakedevine/status/182865434289258496/photo/1
17:51:30 <elliott> fizzie: http://upload.wikimedia.org/wikipedia/commons/6/60/Cie_Chart_with_sRGB_gamut_by_spigget.png -- see, sRGB can even depict areas outside of sRGB.
17:51:33 <Phantom_Hoover> fizzie, are the R, G and B in sRGB defined as precise wavelengths?
17:52:10 <fizzie> elliott: They just do those fake-color images. You'll notice there's a discontinuity at the border of the sRGB triangle.
17:52:49 <elliott> fizzie: Next you'll tell me those nebulae don't *really* look that colourful.
17:53:20 <elliott> Phantom_Hoover: You forgot the s.
17:53:21 <fizzie> Phantom_Hoover: Well, they're defined in the XYZ colorspace. I suppose they're not exactly monochromatic.
17:54:04 <elliott> fizzie: Why do we even need colour spaces, man? Can't we just use wavelengths?!
17:54:11 <fizzie> I mean, they'd be at the borders of that chromaticity chart if they were.
17:54:13 <elliott> (Question more serious than I'd like.)
17:54:25 <olsner> hmm, twitter had a fair bit of wtf itself there... the text in the tweet says "pic.twitter.com/ybRMIfeR", the status bar while hovering the link says "http://t.co/ybRMIfeR", the tooltip says "https://twitter.com/jakedevine/status/182865434289258496/photo/1"
17:54:34 <olsner> and finally, following the link takes you to the link I pasted above
17:54:55 <elliott> Phantom_Hoover: Well, as we all know, purple isn't real.
17:55:00 <elliott> Phantom_Hoover: So it doesn't matter if you can't display it.
17:55:12 <elliott> Phantom_Hoover: OK, why can't we just use a combination of an arbitrary number of wavelengths?!
17:55:46 <fizzie> As soon as they start making displays that give out arbitrary spectral power distributions, I suppose we can.
17:56:01 <elliott> fizzie: The displays can just ignore the colours they can't produce!
17:56:10 <Phantom_Hoover> They had a thing in St Andrews which pulsed laser light at arbitrary frequencies.
17:56:15 <elliott> L*a*b* still exists and there aren't any displays with that! (Are there?)
17:56:43 <elliott> Phantom_Hoover: I forget, did I link you to that thing where you stare at an image and its border becomes more blue than blue itself.
17:56:46 <Phantom_Hoover> I suppose if you miniaturised the hell out of that and stuck a few of them behind each pixel, you could compound a few wavelengths (which is enough).
17:57:02 <elliott> Phantom_Hoover: No, there's a blue one.
17:57:18 <elliott> Phantom_Hoover: It's awesome.
17:57:23 <elliott> I really like the blue colour it produces.
17:57:44 <elliott> Phantom_Hoover: Grep /Eclipse of Mars/, http://www.skytopia.com/project/illusion/illusion.html
17:58:12 <fizzie> Alternatively, if you just hook the display up to the optical nerve (or I guess clamp it on the retina and add enough pixels) you can specify which color receptors to tickle, which is by definition enough to make any seeable colour for that person.
17:58:50 <Phantom_Hoover> elliott, the text accompanying it suggests that it's just not representable on a monitor.
17:58:51 <fizzie> I suppose if you aim *really well* you could do it from a distance, too.
17:59:26 <elliott> Phantom_Hoover: Uhh, the colour it produces is not representable on a monitor.
17:59:39 <elliott> It's an optical illusion; it doesn't have to have any pixels of the colour in question to make you see it.
17:59:44 <elliott> I have done it, it works perfectly.
18:00:06 <elliott> fizzie: Oh, CIE 1931 XYZ actually includes colours that don't exist. That's reassuring.
18:00:39 <Phantom_Hoover> elliott, I mean it suggests that it's not displayable by a monitor but can be realised through other non-illusory means.
18:00:51 <elliott> Phantom_Hoover: Oh, well right.
18:00:59 <elliott> Phantom_Hoover: It certainly looked more brilliant than any cyan *I've* seen.
18:01:51 <Phantom_Hoover> http://en.wikipedia.org/wiki/File:CIE_1931_XYZ_Color_Matching_Functions.svg
18:02:23 <fizzie> The color matching function approximations are IIRC really horrible.
18:02:40 <fizzie> There are "better" curves in the bits that talk about the eye.
18:02:52 <fizzie> (Not that they're any more clearly "red" or "green" or "blue" either.)
18:03:21 <fizzie> http://en.wikipedia.org/wiki/File:Cone-response.svg has some sort of normalized curves.
18:03:30 <elliott> fizzie: Are there colourspaces that contain the entire visual range?
18:03:31 <fizzie> (Don't know how good those are either.)
18:04:16 <fizzie> elliott: Well, XYZ for one, as long as it's the visual range for the CIE standard observer.
18:04:29 -!- RocketJSquirrel has changed nick to RocketJSquirrel_.
18:04:35 -!- RocketJSquirrel_ has changed nick to RocketJSquirrel.
18:05:34 <elliott> fizzie: Man, fuck that standard observer.
18:05:47 <elliott> I suppose that's why everyone specifies things in terms of XYZ?
18:06:30 <fizzie> It's a reasonable thing as long as your target market is humans.
18:06:53 <elliott> Phantom_Hoover: Did you do that eclipse of mars thing yet it's awesome.
18:06:54 <fizzie> But obviously you can't represent an arbitrary spectrum with just three components, so it's not enough for more sophisticated customers.
18:07:06 <elliott> fizzie: OK but what about aliens?
18:07:24 <elliott> fizzie: Or dogs. Is infrared in XYZ?
18:07:42 <fizzie> I suppose you'll just have to start modeling actual spectral power distributions. Some 3D modelling systems do.
18:08:08 <fizzie> That LuxRender thing, for example.
18:09:32 <fizzie> (Well, okay, they're still just sampled and quantized spectra.)
18:10:23 <elliott> fizzie: Or dogs. Is infrared in XYZ?
18:11:08 <fizzie> I think the XYZ color matching functions go to 0 pretty much at the edges of "visible light", so I'm going with no.
18:11:54 <elliott> Phantom_Hoover: http://www.skytopia.com/project/illusion/ipage-mb.html Does this one work for you? It doesn't work for me.
18:11:57 <fizzie> LuxRender has a fancy thing where if you specify a color in R, G, B, it has a whole thing -- http://www.cs.utah.edu/~bes/papers/color/ -- that it uses to make a "physically plausible" spectrum for it.
18:13:13 <Phantom_Hoover> elliott, works, I just need to change the point of focus a lot.
18:13:26 <elliott> Phantom_Hoover: Oh, wait, I think this might be a better version of the Eclipse of Mars: http://www.skytopia.com/project/illusion/ipage-et.html
18:13:28 <elliott> I think that's the one I used.
18:17:06 -!- Zuu has joined.
18:18:24 -!- cheater_ has quit (Ping timeout: 260 seconds).
18:19:19 -!- MoALTz has joined.
18:28:40 -!- asiekierka_ has joined.
18:28:40 -!- asiekierka_ has quit (Read error: Connection reset by peer).
18:30:55 -!- BullwinkleJMoose has joined.
18:33:19 -!- itidus22 has joined.
18:34:19 <nortti> BullwinkleJMoose: are you replacing all your programs with ones compiled with musl?
18:35:52 -!- itidus22 has changed nick to itidus21.
18:37:21 -!- itidus20 has quit (Ping timeout: 245 seconds).
18:40:39 <asiekierka> i <3 you man... non-sexually of course
18:45:29 -!- nortti has quit (Quit: ChatZilla 0.9.88.1 [Firefox 11.0/20120312200651]).
18:49:09 -!- BullwinkleJMoose has quit (Quit: Perlin' up my irssi).
18:50:44 -!- BullwinkleJMoose has joined.
18:50:53 <elliott> And that night BullwinkleJMoose wept for his one true love's declaration of platonicism.
18:56:26 <itidus21> it has just occured to me that if each pixel was a mirror of some sort then it could be made to reflect some color sample
18:57:41 -!- cheater has joined.
18:58:10 <itidus21> such as an apple in a box with light shined on it magically reflected along a series of mirrors until finally reflecting off the pixels
18:59:41 <itidus21> it's newspeak for someone elses problem
19:00:26 -!- monqy has joined.
19:03:01 <itidus21> the cie 1931 color space page was too tough for me to read so i thought of that instead
19:03:33 -!- asiekierka_ has joined.
19:03:33 -!- asiekierka_ has quit (Client Quit).
19:06:02 -!- ais523 has joined.
19:12:53 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: : not found
19:13:18 <HackEgo> ais523 is ais523. This topic may retroactively become more informative if or when Feather is invented.
19:13:30 <ais523> gah, HackEgo has ruined our convention
19:13:35 <ais523> which shows how long ago we last tried to use it…
19:14:00 <ais523> come to think of it, "not found" is not really a sensible error message for trying to exec the null string
19:16:00 -!- Vorpal has joined.
19:17:33 <elliott> `run mv bin test; touch bin
19:17:36 <fizzie> Sadly, there is only ENAMETOOLONG, not an ENAMETOOSHORT.
19:17:44 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: : not found
19:17:48 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: : not found
19:20:41 <itidus21> i have thought about colors in scRGB space before.. it's hard to really express the idea to most people
19:21:05 -!- pikhq_ has joined.
19:21:53 -!- pikhq has quit (Ping timeout: 272 seconds).
19:25:39 <itidus21> but it's not as if they're of any great use... an apple would not be any more of an apple in an imaginary color
19:34:27 * itidus21 makes a post without putting my foot in my mouth.
19:37:14 -!- asiekierka_ has joined.
19:37:14 -!- asiekierka_ has quit (Remote host closed the connection).
19:37:45 -!- calamari has left ("Leaving").
19:38:01 <HackEgo> bin \ canary \ hackenv \ karma \ lib \ paste \ quotes \ share \ wisdom
19:39:12 <elliott> fizzie: Did you know: Not submitting a candidate featured language is defined as treason in 23 of the 29 states of Canada, including Finland, China, and Europe?
19:39:14 -!- audy has joined.
19:39:21 <HackEgo> audy: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page
19:39:25 <elliott> Wait, I've seen that name before. Never mind(?)
19:39:54 <audy> Yes I idle in here sometimes
19:42:10 -!- ais523_ has joined.
19:45:24 -!- ais523 has quit (Ping timeout: 276 seconds).
19:47:28 -!- hagb4rd2 has quit (Ping timeout: 246 seconds).
19:51:32 -!- MoALTz_ has joined.
19:54:28 -!- MoALTz has quit (Ping timeout: 246 seconds).
19:55:03 -!- ais523_ has changed nick to ais523.
19:56:39 -!- MoALTz__ has joined.
19:59:52 -!- MoALTz_ has quit (Ping timeout: 260 seconds).
20:19:55 -!- derdon has quit (Remote host closed the connection).
20:22:47 <elliott> > (round 48.5, round 49.5)
20:23:33 <lambdabot> Ambiguous type variable `a' in the constraints:
20:23:35 <elliott> > printf "%.0f" 48.4 :: String
20:23:43 * Sgeo raises an eyebrow. I ... get the purpose of that rule, I think, but why is it only as popular as it is, and was the choice to round to even arbitrary or is there a reason for it?
20:24:12 -!- nortti has joined.
20:24:19 <olsner> it's to make the roundings average out, or something
20:24:34 <fizzie> It's the default rounding mode in IEEE-754, that's pretty popular.
20:24:34 <Sgeo> I mean, as opposed to rounding to odd.
20:25:58 -!- oerjan has joined.
20:26:41 * oerjan smelled a gregor nick change, and was right
20:26:45 <itidus21> i was looking at a cake website.. and i had an amusing thought.. an esolang cake
20:26:49 <elliott> Your daily dose of irony: what type of dash is used in the prose of http://www.emdash.net/en/mission.html?
20:27:01 <elliott> Sgeo: I was fact-finding for my answer to http://stackoverflow.com/questions/9829769/strange-half-to-even-rounding-in-different-languages.
20:27:11 <itidus21> it could spell out happy birthday in brainfuck
20:27:53 <fizzie> Didn't I make a Befunge birthday card once?
20:27:57 <fizzie> Or was it so that I got one?
20:28:03 <fizzie> I think it was the latter.
20:28:14 <oerjan> 09:35:16: <elliott> oerjan: Uhh, unless you're Jewish, or gay, or...
20:28:14 <oerjan> 09:35:38: <elliott> oerjan: If by "perfect" you mean "Nazis", then yes.
20:28:25 <itidus21> the cakes on this website are just that inspirational..
20:28:58 <oerjan> elliott: your joke detector needs adjustment, it's a little slow.
20:29:18 <nortti> what timzone is used on those quotes?
20:30:12 <elliott> oerjan: It was clearly a ha-ha-only-serious thing.
20:34:13 * oerjan looks at elliott and fizzie's attempt to brighten colors, and wonders if there's something wrong with simply applying \ (x,y,z) -> (x + a*(255-x), y + a*(255-y), z + a*(255-z)) to an rgb triple
20:34:27 <fizzie> oerjan: It's not SEMANTIC.
20:34:54 <oerjan> fizzie: it's a convex combination of your original color and white, how is that not semantic?
20:35:01 <olsner> "semantic"? just prepend "bright" or "light", works with every color
20:36:50 -!- asiekierka has quit (Remote host closed the connection).
20:36:54 <fizzie> oerjan: Distances in the (gamma-compressed) RGB space are complete nonsense. (Anyway, that *is* what doing \(h,l,s) -> (h,l + a*(1-l),s) pretty much does, which is the second table.)
20:37:44 -!- MoALTz has joined.
20:37:44 <oerjan> ok well i know little about gamma
20:38:50 <RocketJSquirrel> * oerjan smelled a gregor nick change, and was right // no
20:38:59 -!- MoALTz__ has quit (Ping timeout: 248 seconds).
20:39:14 <oerjan> argh the gregors are breeding!
20:41:18 <elliott> "Thanks! ehird. You are good man."
20:41:39 <nortti> english language should in my opinion adopt term "pilkunnusija" (translates to something along the lines of commafucker)
20:41:42 <oerjan> <shachaf> elliott: Wait, how do <--- arrows work with multiple people being quoted? <-- badly, unless all quoters use them
20:42:03 <fizzie> nortti: Did you read that from cracked.com's list of words the English language is missing?
20:42:19 <oerjan> in which case <...> and <-- match like parentheses, i think
20:42:33 <elliott> fizzie: It would be amusing for a Finn to decide on that opinion only after reading an English-language piece.
20:42:47 <nortti> fizzie: did they list it there?
20:43:12 <fizzie> nortti: They did, and I saw the list being shared in some Social Media just the other day.
20:43:16 <elliott> oerjan: Is it immoral to remind an ESL speaker to mark my answer as accepted?
20:43:25 <fizzie> It's quite the synchronicity to spontaneously suggest the same thing.
20:43:42 <fizzie> (Especially since I didn't see any comma-fucking going on at the moment.)
20:43:43 <oerjan> elliott: i don't see what ESL speaker has to do with it
20:43:51 <elliott> oerjan: Neither did I, before the GUILT.
20:43:52 <nortti> fizzie: then they have made a new list
20:44:20 <elliott> oerjan: If I become Catholic, can I do whatever and have my guilt erased after-the-fact?
20:44:27 <oerjan> i suppose theoretically the person might think "accept" means "eviscerate", or something.
20:44:28 <elliott> That would be very helpful for my lifestyle.
20:44:33 <fizzie> nortti: http://www.cracked.com/article_19695_9-foreign-words-english-language-desperately-needs_p2.html #1
20:44:39 <elliott> Well, it's just they're so kind. :(
20:44:53 <fizzie> And yes, there's also an older list of 10 such words from 2009.
20:45:03 <fizzie> This one is from 28 Feb 2012.
20:45:34 <nortti> fizzie: that older one is the list that I have read
20:45:56 <oerjan> elliott: i think so, but note that if you die between the sinning and erasing for a cardinal sin, you are _really_ screwed. and that doesn't even get into latae sententiae excommunication...
20:46:08 -!- MoALTz_ has joined.
20:47:16 <elliott> oerjan: I don't anticipate dying before this answer is accepted.
20:47:45 <fizzie> oerjan: Can you just uncovert right before you intend to sin, and then reconvert just before the guilt-erasing?
20:48:18 <nortti> fizzie: It just came to mind when reading english language forums and seeing hiw english language doesn't have a good term for commafuckers
20:48:33 <elliott> nortti: "Grammar Nazi" is in widespread use.
20:48:51 <oerjan> fizzie: i think this stuff might require some element of genuine _regret_, you know.
20:48:52 -!- MoALTz has quit (Ping timeout: 244 seconds).
20:48:56 <elliott> Commafuckers is less distasteful, though. Although orthography ain'tn't grammar.
20:49:16 <oerjan> elliott: always a catch.
20:49:20 <elliott> I mean, guilt, sure; regret, no.
20:49:26 <elliott> I'd do it again, and feel bad about it.
20:50:18 -!- hagb4rd has joined.
20:51:43 <elliott> Awesome, they did it, now I can erase the evidence.
20:52:04 <elliott> I HAVE SO MANY CONFLICTING FEELINGS
20:54:36 <elliott> oerjan: http://stackoverflow.com/questions/9830006/type-signature-between-funtion-and-class/9830084#9830084 will people look back on me in the future and think "wow that ehird guy is such an asshole"
20:54:56 <elliott> the life of an overflown stack... is the most difficult life in the wold
20:55:03 <oerjan> what do you mean, "future"
20:55:26 -!- nortti has quit (Quit: nortti).
20:55:59 <elliott> what will the kangaroos think of me
20:56:48 -!- MoALTz has joined.
20:57:34 <elliott> get the feeling oerjan doesn't know much about kangaroo feelings
20:57:41 <oerjan> i don't expect the kangaroos to be most likely survivors, given that eutherians already wiped most marsupials from the _other_ continents, and humans have introduced eutherians to australia.
20:57:58 <elliott> kangaroos are pretty tough
20:58:09 <elliott> and australia is devoid of almost all civilisation
20:59:27 <elliott> TIL humans have a conservation status
20:59:33 <elliott> wonder how long it took to decide that one
20:59:39 -!- MoALTz_ has quit (Ping timeout: 250 seconds).
20:59:55 <elliott> http://upload.wikimedia.org/wikipedia/commons/7/7a/Homo_Sapien_range.png TIalsoLd there's no humans at the south pole
21:01:09 <oerjan> let me guess, "least concern, unless they destroy themselves"
21:04:09 <elliott> They should double the scale to account for species with worryingly high populations.
21:06:54 <elliott> RocketJSquirrel: Homo sapiens is, like, plural, man!
21:08:14 -!- oerjan has quit (Quit: Lost terminal).
21:10:25 -!- pikhq has joined.
21:10:34 -!- pikhq_ has quit (Ping timeout: 244 seconds).
21:15:18 -!- MoALTz_ has joined.
21:16:17 -!- Taneb has joined.
21:17:52 -!- oerjan has joined.
21:18:17 -!- MoALTz has quit (Ping timeout: 250 seconds).
21:18:23 <oerjan> almost the worst part is i _still_ don't know whether the problem is at my end or nvg's
21:18:55 <oerjan> although the webchat dropping yesterday for the first time makes me slightly more suspicious that it's at my end.
21:23:10 * oerjan wishes he lived in a universe where things didn't tend to develop irritating flaws
21:25:05 <oerjan> Taneb: no, you're not dead
21:25:22 <oerjan> well, not more dead than me, anyway
21:36:03 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
21:38:33 -!- BullwinkleJMoose has quit (Quit: leaving).
21:42:48 -!- Taneb has quit (Read error: Connection reset by peer).
21:43:59 -!- Phantom_Hoover has joined.
21:44:09 * oerjan tries reducing putty's keepalive to 30 secs
21:44:40 <oerjan> i'm not sure if that will help, do nothing, or make things worse
21:45:24 <fizzie> It will stay alive up to 30 secs.
21:45:36 <fizzie> (Okay, that's not quite what it means.)
21:47:21 -!- hagb4rd has quit (Ping timeout: 252 seconds).
21:48:13 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
21:49:04 -!- Jafet has quit (Quit: Leaving.).
21:57:26 <ais523> btw, that gzip file that got mangled by phpBB? the person sent me the original, it was a standard gzip file after all
21:58:04 <ais523> it had a lot of NULs after the first three bytes, which presumably phpBB deleted, which is why we couldn't parse it from that
21:58:58 <fizzie> Funny, you'd think at least the modification time would have survived.
21:59:14 <fizzie> Or was that zeros too?
21:59:31 <ais523> the zeros came very early
21:59:34 <ais523> so they'd have shifted all the bytes from then on out of place
21:59:45 -!- elliott has quit (Ping timeout: 264 seconds).
22:00:09 <fizzie> But it's the "fixed" three bytes, then a flag byte that might well be a 0, and immediately after that the mtime.
22:00:44 <fizzie> I think we looked for the mtime field under the assumption that a '\0' FLG byte would've been there, but got dropped.
22:01:18 <fizzie> I suppose it's not too rare to not have a timestamp there.
22:01:51 <fizzie> The "delete nuls" is not a very invertible operation anyway, so it's a bit of a moot point.
22:05:32 -!- Phantom_Hoover has joined.
22:05:51 <ais523> oh, mtime was indeed 0, in that case
22:06:03 <fizzie> Speaking of binary data, I've been cleaning my ~, and there's a 81098752-byte file called "ut.end" in there, and I haven't managed to distinguish it from random data yet. 'file' says just "data", 'strings' or hexdump don't seem to reveal any obvious patterns in the first few hundred kb, and a "forensics" proggie called "foremost" I had (that I ran on a broken-disk image) managed to extract ...
22:06:17 <fizzie> I'm debating whether I should just delete it, or keep it secret, keep it safe.
22:06:27 <fizzie> On the other hand, 80 megs is quite a bit.
22:06:43 <fizzie> Not in the "it uses a lot of space" sense, more in the "there could be some stuff there" sense.
22:08:09 <fizzie> There's a single human-readable string I've seen in the 'strings' output, and it's the word "JUNK", the last thing 'strings' outputs.
22:08:14 <fizzie> That might be some sort of a sign.
22:09:00 <fizzie> Also there's an obviously regular-looking feature at the end, but I'm not sure what it is. It looks slightly like a sorted array of 8-byte integers.
22:09:09 <fizzie> I guess it could be offsets.
22:09:55 <fizzie> Or *something. They're a bit too large to be offsets to this file, but maybe only part of them is offsets.
22:11:21 <fizzie> The bit that actually looks like an incrementing counter goes up to 722559503.
22:11:43 <fizzie> Since the file extension is ".end", it could be just the latter piece of a CD image or something.
22:12:12 <fizzie> (Or some other 700-meg file.)
22:14:33 -!- MoALTz__ has joined.
22:15:04 <oerjan> clearly when that file is deciphered, the world ends. or is that when it is deleted.
22:17:39 -!- MoALTz_ has quit (Ping timeout: 250 seconds).
22:20:00 <itidus21> ISBN 978-0-109-8752-4 Hello Kitty Everywhere!: Haiku: Postcards in a Hinged Box
22:20:41 <itidus21> ISBN 978-0- 8109-8752-4 Hello Kitty Everywhere!: Haiku: Postcards in a Hinged Box
22:21:19 <fizzie> I was kinda-sorta wondering that it had a weird number of digits for an ISBN.
22:26:45 <fizzie> That's a funny normalization, xn = sgn(x)*log(1+abs(x)). It's like some sort of symmetric logarithm thing.
22:27:27 <fizzie> Looks nice and smooth, though.
22:30:30 -!- Vorpal has quit (Ping timeout: 276 seconds).
22:39:34 <ais523> wow, advert I just saw: "3 easy steps: 1. Click to download. 2. Run the .exe file to install. 3. Enjoy the fun content!" together with a large Download button, five stars, the company name, some version compatibility information, and a large tick for some reason
22:39:42 <ais523> no indication on what the product actually was…
22:39:54 <ais523> what are the odds it /isn't/ malware?
22:46:16 -!- MoALTz has joined.
22:48:51 -!- MoALTz__ has quit (Ping timeout: 250 seconds).
23:01:32 <Sgeo> Did I ever mention I was an idiot as a kid?
23:01:44 <ais523> Sgeo: are you an idiot now?
23:01:52 <Sgeo> ais523, I'd like to think not.
23:04:59 <Phantom_Hoover> Sgeo, you damn fool, that's at least 2 more lambdabot messages for me tomorrow morning.
23:05:13 <Sgeo> Phantom_Hoover, idgi
23:06:18 <Phantom_Hoover> elliott won't be able to resist that one, and since his sleep cycle is so insane he'll logread that.
23:08:05 <Sgeo> Surely he'll see that you saw it though?
23:08:33 <Sgeo> Or, I don't get what he's going to do. Crack jokes privately to you at my expense?
23:10:11 -!- Patashu has joined.
23:12:52 -!- tzxn3 has quit (Quit: Leaving).
23:14:04 -!- Phantom_Hoover has quit (Quit: Leaving).
23:21:46 -!- MoALTz_ has joined.
23:23:57 -!- MoALTz has quit (Ping timeout: 250 seconds).
23:28:46 -!- MoALTz__ has joined.
23:30:04 <Sgeo> tswett, monqy has been turned into a MUFFIN.
23:30:27 <tswett> The most important character in Homestuck?
23:30:53 -!- MoALTz_ has quit (Ping timeout: 250 seconds).
23:48:08 -!- ais523 has quit (Remote host closed the connection).