00:01:35 <HackEgo> ¡Bienvenido al centro internacional para el diseño y despliegue de lenguajes de programación esotéricos! Por desgracia, la mayoría de nosotros no hablamos español. Para obtener más información, echa un vistazo a nuestro wiki: http://esolangs.org/wiki/Main_Page. (Para el otro tipo de esoterismo, prueba #esoteric en irc.dal.net.)
00:02:10 -!- anthony2 has left.
00:03:59 <oerjan> i wonder if we're included in some kind of list of "spanish" irc channels
00:04:04 -!- snuffeluffegus has quit (Quit: Leaving).
00:04:42 <kmc> I think we decided it's because the name starts with 'es'
00:05:23 <oerjan> yes, but maybe there's still some autogenerated list somewhere
00:06:13 -!- muskrat has quit (Quit: Leaving).
00:12:56 -!- Uguubee111117 has quit (Quit: Uguubee111117).
00:15:17 -!- Uguubee111117 has joined.
00:15:45 -!- Sgeo has joined.
00:18:14 -!- oerjan has quit (Quit: Buenas noches).
00:21:59 <kmc> http://upload.wikimedia.org/wikipedia/commons/b/bb/Speed1c.png
00:24:34 -!- Uguubee111117 has quit (Read error: Connection reset by peer).
00:27:48 -!- Uguubee111117 has joined.
00:28:48 -!- mnoqy has joined.
00:29:05 <kmc> nope, just a 14-speed internal hub gear for bicycles
00:29:11 <kmc> http://en.wikipedia.org/wiki/Rohloff_Speedhub
00:30:07 <kmc> it's about $1200 apparently
00:33:43 <kmc> "Rohloff still claim never to have had a hub fail and the greatest distance claimed for a Rohloff hub in 2009 was 145,000km"
00:46:53 <Sgeo> syntax-case is probably confusing to talk about. syntax-case isn't all there is to syntax-case, and arguably the syntax-case in the syntax-case mechanism is optional
00:58:21 -!- Sprocklem has quit (Ping timeout: 272 seconds).
01:07:19 -!- Phantom_Hoover has quit (Quit: Leaving).
01:15:52 -!- yorick has quit (Remote host closed the connection).
01:45:10 <Bike> never fails? for a mechanical gear? that would be impressive
01:48:09 <Bike> i guess it's impressive on its own though, that's a pretty shiny lookin diagram
01:49:05 <Bike> oh, no external components, i guess that would help lasting
01:54:17 -!- shikhin_ has quit (Read error: Operation timed out).
02:07:10 <Sgeo> elliott: http://blog.cryptographyengineering.com/2013/10/lets-audit-truecrypt.html?showComment=1381802478312
02:07:18 <Sgeo> err, why was showComment in that URL
02:08:01 -!- shikhin_ has joined.
02:11:59 -!- mnoqy has quit (Quit: hello).
02:13:12 -!- shikhin_ has quit (Ping timeout: 252 seconds).
02:14:29 -!- shikhin_ has joined.
02:15:40 -!- Uguubee111117 has quit (Quit: Uguubee111117).
02:21:27 <shachaf> i bought a dvd and now i remember my dvd drive doesn't work
02:24:56 <pikhq> They cost like $15.
02:25:44 -!- shikhin__ has joined.
02:26:45 -!- shikhin_ has quit (Read error: Operation timed out).
02:27:34 -!- shikhin_ has joined.
02:29:20 -!- shikhin__ has quit (Read error: Operation timed out).
02:35:16 -!- Uguubee111117 has joined.
02:41:06 <HackEgo> 4000000000000000/45359237 (approx. 8.818490487395103e7)
02:43:40 <quintopia> pikhq: can you transliterate the name taku yushihara for me? not being nipponese myself, i'm not even sure which is the family name...
02:44:07 <Bike> can you even reliably make up kanji for a given name
02:44:50 <pikhq> And I can only *guess* that that name is supposed to be Taku YUSHIHARA, rather than TAKU Yushihara.
02:45:08 <pikhq> (Yushihara seems more last-name-ish)
02:45:47 <quintopia> well, i want to look him up on facebook, but i don't think he used the english phonetic...
02:46:21 <pikhq> WWWJDIC doesn't have an entry for YUSHIHARA, so Imma guess it's a stupidly rare surname.
02:46:34 <pikhq> In which case the god damned kanji could be anything.
02:46:53 <pikhq> And there's like a dozen possible kanji for Taku.
02:47:16 <pikhq> Which is either a given name or a family name.
02:47:32 <pikhq> So, without asking the person him/herself, I cannot possibly know.
02:47:57 <pikhq> Japanese names suck.
02:48:42 <pikhq> Basically, it is nigh-impossible to map between readings and characters for names.
02:48:54 <Bike> it's just lossy innit
02:49:30 <Bike> i never really understand when it comes up in the animes. i just assume all japanese people wear nametags
02:49:58 <pikhq> No, I mean that any given reading could have multiple character choices, and any given characters could have multiple possible readings.
02:50:46 <quintopia> what an ambiguous typographical system
02:51:04 <quintopia> not that english is too much better
02:51:58 <Fiora> it's really only quite that bad for names
02:52:16 <Fiora> it's not impolite to ask someone how to write their name, if I remember right it's pretty common
02:52:56 <shachaf> pikhq: how do you write your name
02:53:00 <pikhq> The normal writing system has funny weird cases where it's ambiguous, but for a competent speaker it's not a problem.
02:54:16 <shachaf> My speakers are not competent. :-( Sometimes they make clicking or buzzing noises for no obvious reason.
02:54:25 <shachaf> Well, I think the buzzing is software.
02:55:26 <Bike> how do you wriiiiite it though
02:55:40 <Bike> but the kanji!
02:55:46 <Bike> katakana are for like, foreiegners, maaaaan
02:56:21 <pikhq> What am I, Japanese?
02:56:32 <Bike> just like my animes
02:57:17 * pikhq suspects "Josiah Worcester" is way too English to be Japanese.
02:57:57 <Bike> do you pronounce "worcester" in the pleb way or the massachusetts way
02:58:20 <pikhq> Vaguely like "Wuhster"?
02:58:50 <Bike> therrrrre you go
02:58:57 <pikhq> Always amused me that people have trouble with Worcestershire.
02:58:59 <Bike> rhoticism is pleb though.
02:59:31 <pikhq> Or that people like inserting an h in the spelling.
02:59:37 <pikhq> "Worchester" it is not.
03:00:07 -!- NihilistDandy has quit.
03:01:35 <pikhq> I can freaking spell it for people and they insert an h.
03:01:48 <pikhq> And then there's people who somehow screw up my first name.
03:01:59 <pikhq> C'mon, "Josiah" is not a rare name.
03:02:13 <shachaf> I have no idea how to pronounce it.
03:03:52 <pikhq> Weird anglicisation of יֹאשִׁיָּהוּ
03:12:37 -!- shikhin_ has quit (Ping timeout: 272 seconds).
03:15:04 -!- Sprocklem has joined.
03:16:19 <Sgeo> In... high school or middle school, I forget which, some librarian needed to write down my name, so I told her my name, and she asked "man?", and I thought she was asking if I was a man, so I said yes, and she wrote "Goldman"
03:17:06 <Bike> that's why i always introduce myself as Darkness O'Corvidkill.
03:23:54 <Fiora> Ebikeony Darkness Raven Way
03:31:08 -!- Rugxulo has joined.
03:33:39 -!- Rugxulo has quit (Client Quit).
03:37:36 <quintopia> it seems like there should be a simpler way to make polyvariadic functions
03:42:36 -!- muskrat has joined.
03:44:43 -!- muskrat has quit (Client Quit).
03:49:11 -!- nisstyre has joined.
04:17:10 -!- asie has joined.
04:26:35 -!- shikhin has joined.
04:26:38 -!- conehead has quit (Quit: Computer has gone to sleep.).
04:30:09 -!- zzo38 has joined.
04:31:40 <zzo38> When I write my name in Japanese I write using the katakana for "Aaron" and the kanji for "Black".
04:32:06 <zzo38> Do you ever write your name in other languages than English and (if applicable) your language?
04:32:23 <pikhq> Yes, I write my name in Japanese.
04:32:37 <pikhq> English is my native language. :P
04:34:16 <shachaf> zzo38: how do i write my name in your language
04:34:37 <pikhq> Last name as typically transcribed by the handful of people needing to talk about Worcester, first name as it appears in Japanese Bible translations.
04:34:49 <zzo38> I saw that in the log but I forgot that is what it means
04:35:03 <zzo38> shachaf: I don't know how *you* write your own name.
04:35:15 <kmc> zzo38: http://mjg59.dreamwidth.org/27881.html Implementing UEFI Boot to Zork
04:35:43 <zzo38> kmc: Yes I have seen UEFI to Zork
04:36:29 <pikhq> Also has the nice property that my name actually fits in 6 characters and isn't hard to pronounce for Japanese people.
04:36:59 <pikhq> Dear god, I cannot imagine having a transcription of a name like "Aldwinkle" in Japanese.
04:37:23 <zzo38> Although I think a more accurate interpreter than Frotz should be used; most (even Infocom's own interpreters) have errors, partially due to errors in a Z-Machine Standards Document, Infocom's own documents, and other things; even my interpreters (Fweep and Aimfiz) have what I found to be a few errors (I will fix them), despite being more accurate in general
04:38:54 <Bike> hm wi is obsolete or something isn't it
04:38:56 <zzo38> pikhq: Yes, that is why my brother's name could fit in Super Smash Brothers Melee but not Brawl, despite Brawl allowing five letters and Melee only having four.
04:39:01 <pikhq> Bike: "Arudouxinkuru"
04:39:11 <Bike> pikhq: my next band
04:41:12 <zzo38> I use a kanji in my name although I think usually non-Japanese people don't do that (my brother also just transliterates the sound into katakana, although I prefer to use the kanji meaning "black")
04:41:32 <shachaf> is your brother ever in this channel
04:44:28 <zzo38> How often are quit messages in the quote file?
04:44:53 <shachaf> `run grep quit quotes | wc -l
04:45:59 <zzo38> trout: I don't know Hebrew writing very well.
04:46:22 <trout> zzo38: spelt in English as Eitan
04:46:35 -!- Nisstyre-laptop has joined.
04:46:37 <trout> the englishified name is Ethan
04:46:49 <trout> pronouned as ATE a proTON
04:48:01 <zzo38> shachaf: You are just finding if it says "quit" not necessarily if it is a quit message (although maybe part messages should also be counted, if any)
04:48:05 -!- ^v has quit (Quit: Leaving).
04:48:21 -!- trout has left.
04:48:21 -!- trout has joined.
04:48:38 * trout likes his part message
04:49:22 <zzo38> There is no message.
04:49:42 -!- trout has left ("Overflow in /dev/null").
04:49:44 <zzo38> I always type in the message every time if applicable
04:49:46 -!- trout has joined.
04:49:49 <zzo38> I always type in the message every time if applicable
04:49:52 <trout> zzo38: apperently /cycle doesn't leave the message
04:49:54 <zzo38> trout: OK it is working now
04:49:57 <trout> only /part then /join
04:50:24 <zzo38> I didn't create a macro to automatically make part/quit messages
04:50:42 <trout> there is a preference for a default message
04:51:32 <zzo38> Ah, OK. The client I use has no such preference so it would have to be a macro; maybe some other clients also have preference for a default message, and some don't have.
04:52:54 <zzo38> The client I use is simple and only has a few settings (other than macros, which it also has): ANSWER, AUTOPONG, COLS, FORMAT, IDLETIMER, ROWS, SHOWTIME, and USERINFO.
04:57:25 <zzo38> I am a bit curious to know what happens in various IRC clients when you give it the command to connect without telling it what to connect to? (In some clients this might not be possible, though)
04:58:21 <zzo38> What does Xchat do in such a case, or does it disable something in a dialog box before you type it in, or what?
05:00:26 <zzo38> Do you want to change your "/cycle" macro to leave the message, then?
05:08:26 <zzo38> Haskell's Typeable is mostly useful when you subclass it. Isn't it?
05:08:40 <zzo38> I don't think it is useful by itself.
05:12:13 <zzo38> I mean to do something like class Typeable x => XYZ x where { ... } (possibly there will be other classes mentioned before => as well) rather than using Typeable somewhere directly.
05:13:09 <zzo38> Have you ever used Typeable at all?
05:13:18 <kmc> is Data.Data.Data an example of such a typeclass
05:13:23 <kmc> (best name ever for a typeclass)
05:15:09 <zzo38> It is something that uses it, but not the examples I was thinking of.
05:15:24 <kmc> what sort of example did you have in mind?
05:15:47 <zzo38> Things like Control.Exception.Exception and Graphics.DVI.Node
05:16:23 <zzo38> (I figured out this kind of technique independently from them; but these kind of things would be common in mathematics and related stuff anyways)
05:17:04 <zzo38> Although Data.Data.Data does count too
05:18:12 <zzo38> (The methods for Exception don't seem particularly meaningful)
05:25:35 <ion> fix (Data.)
05:26:18 <kmc> http://b.igdata.co/
05:27:15 <ion> http://heh.fi/gates/
05:34:20 <elliott> zzo38: the methods for Exception are used to let you do "subtyping" of exceptions
05:43:59 <zzo38> elliott: Ah, OK yes I suppose that is what it can be useful for, you seem to be correct.
05:44:07 <zzo38> I didn't think of that.
05:46:35 -!- Sprocklem has quit (Ping timeout: 248 seconds).
06:27:54 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
06:45:49 -!- Taneb has joined.
07:25:36 <fizzie> "File size: . Are you sure you want to download this Corpus now?" Well, now, I'm not sure I have . space available.
07:26:15 <zzo38> What Corpus is this and what are you downloading it from?
07:27:14 <fizzie> It's from LDC, and it's just some speech stuff.
07:28:17 <zzo38> Is it HTTP or FTP? If it is HTTP you can try a HEAD request (I did this with something else that failed to display the filesize correctly). If it is Google then a HEAD request might not work (Google has a really bad implementation of HTTP)
07:29:35 <kmc> corpus callosum
07:37:37 <Taneb> I need to write a 1000 word essay describing a "classic publication in computer science"
07:43:49 <Taneb> I'm thinking about doing it on On Computable Numbers With An Application To The Entscheidungsproblem
07:48:55 <kmc> that's a good one
07:49:12 <kmc> you could write about _Reflections on Trusting Trust_ and its relevance to the paranoid dystopian world we live in today
07:49:26 <fizzie> LDC is the Linguistic Data Consortium, and it's behind a login thing, so testing with HEAD might be a bit iffy. Though I have the approximate sizes from elsewhere, so it's no big problem that this download system doesn't show them correctly.
07:49:39 <Taneb> I... think that may be slightly to recent to be considered classical, kmc
07:51:18 * kmc looks up when it was published
07:51:29 <fizzie> Taneb: Depending on how widely "computer science" is interpreted there, there's also A Mathematical Theory of Communication, that's a classic too.
07:53:02 <Taneb> kmc, wow, that far back?
07:53:11 <Taneb> I was under the impression it was from like 2011
07:53:22 <shachaf> _Reflections on Trusting Trust_ is barely longer than 1000 words itself.
07:54:57 <kmc> sure that's why the essay is secretly a review of _Glasshouse_ and a rant about the NSA
07:58:42 <Taneb> I think one slight problem I'm having with my course is that it assumes no prior programming experience, so is teaching us all Python
08:02:57 <Taneb> That bit of the course is all very basic and going quite slowly
08:06:32 <kmc> is it time-consuming
08:07:02 <Taneb> Four hours a week, so not too much
08:09:06 <olsner> you should also include the hours spent on writing essays about stuff
08:09:17 -!- FreeFull has quit.
08:09:25 <Taneb> olsner, in the programming module there hasn't been any of that yet but it's about to start
08:10:00 <olsner> oh, that was different course from the essay you were writing?
08:11:31 <Taneb> I've got I think 6 modules
08:14:03 -!- Taneb has quit (Quit: breakfast and then life).
08:16:38 -!- zzo38 has quit (Remote host closed the connection).
08:27:44 -!- shikhin_ has joined.
08:31:11 -!- shikhin has quit (Ping timeout: 272 seconds).
09:01:34 -!- Uguubee111117 has quit (Read error: Connection reset by peer).
09:10:37 -!- Taneb has joined.
09:32:04 -!- Nisstyre-laptop has quit (Ping timeout: 260 seconds).
09:50:40 -!- Phantom_Hoover has joined.
09:53:18 <Taneb> Phantom_Hoover: did elliott want to play dwarf fortress
09:54:58 <Taneb> Well, not right now
09:55:07 <Taneb> I'm slightly in a Python practical
09:56:02 -!- MindlessDrone has joined.
10:17:30 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
10:23:59 -!- oerjan has joined.
10:32:57 -!- Taneb has quit (Ping timeout: 250 seconds).
10:37:31 -!- Phantom_Hoover has joined.
10:49:33 -!- yorick has joined.
10:58:23 -!- Taneb has joined.
10:58:37 -!- carado has joined.
10:59:00 -!- carado has quit (Client Quit).
10:59:11 -!- carado has joined.
11:03:59 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds).
11:05:25 <Taneb> I think I'll start working on this essay now
11:12:25 -!- oerjan has quit (Quit: leaving).
11:14:15 -!- Koen_ has joined.
11:28:23 -!- Phantom_Hoover has joined.
11:41:33 -!- Sgeo has quit (Read error: Connection reset by peer).
12:02:13 -!- boily has joined.
12:02:29 -!- metasepia has joined.
12:18:08 -!- boily has quit (Quit: Poulet!).
12:18:10 -!- metasepia has quit (Remote host closed the connection).
12:20:05 -!- boily has joined.
12:20:12 -!- metasepia has joined.
12:22:57 -!- Taneb has quit (Quit: Page closed).
12:26:00 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
12:37:35 -!- Taneb has joined.
12:39:22 -!- Phantom_Hoover has joined.
12:41:58 -!- FreeFull has joined.
12:42:06 -!- FreeFull has quit (Client Quit).
12:44:54 -!- nooodl has joined.
12:45:08 -!- FreeFull has joined.
13:04:12 <fizzie> How lovely, variables "hp" and "Hp" that serve different purposes.
13:06:10 -!- Phantom__Hoover has joined.
13:09:45 -!- Phantom_Hoover has quit (Ping timeout: 272 seconds).
13:14:36 <boily> fizzie: it could be even more tortuous, with subtle greek and cyrillic letters mixed in.
13:26:23 -!- mnoqy has joined.
13:37:57 -!- Phantom___Hoover has joined.
13:38:20 -!- shikhin_ has changed nick to shikhik.
13:38:23 -!- shikhik has changed nick to shikhin.
13:40:15 -!- Phantom__Hoover has quit (Ping timeout: 265 seconds).
13:42:46 -!- Taneb has quit (Quit: Taneb).
13:43:59 -!- KingOfKarlsruhe has joined.
13:53:06 -!- sebbu has quit (Ping timeout: 245 seconds).
13:58:59 -!- asie has joined.
14:15:13 -!- Koen_ has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?").
14:16:22 -!- Phantom__Hoover has joined.
14:18:28 -!- Phantom___Hoover has quit (Ping timeout: 240 seconds).
14:28:02 -!- shikhin_ has joined.
14:29:44 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
14:30:50 -!- shikhin has quit (Ping timeout: 245 seconds).
14:31:11 -!- augur has quit (Remote host closed the connection).
14:31:37 -!- conehead has joined.
14:31:43 -!- Sprocklem has joined.
14:32:04 -!- augur has joined.
14:32:04 -!- augur has quit (Read error: Connection reset by peer).
14:32:30 -!- augur has joined.
14:33:05 -!- augur has quit (Remote host closed the connection).
14:38:16 -!- conehead_ has joined.
14:41:04 -!- conehead has quit (Ping timeout: 246 seconds).
14:41:05 -!- conehead_ has changed nick to conehead.
14:42:08 <boily> I just stumbled on a SO comment by oerjan.
14:42:28 -!- sebbu has joined.
14:43:10 -!- sebbu has quit (Changing host).
14:43:11 -!- sebbu has joined.
14:43:23 -!- shikhin_ has changed nick to shikhin.
14:51:03 -!- asie has joined.
14:57:36 -!- Phantom___Hoover has joined.
14:57:56 -!- Taneb has joined.
15:00:25 -!- Phantom__Hoover has quit (Ping timeout: 245 seconds).
15:13:15 -!- Phantom___Hoover has quit (Ping timeout: 272 seconds).
15:18:04 -!- augur has joined.
15:19:55 -!- ^v has joined.
15:25:08 -!- Phantom___Hoover has joined.
15:51:36 -!- zzo38 has joined.
16:02:48 -!- Sprocklem has quit (Ping timeout: 240 seconds).
16:06:39 -!- Sprocklem has joined.
16:11:35 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
16:15:00 -!- asie has joined.
16:15:11 -!- asie has quit (Remote host closed the connection).
16:17:33 <Taneb> Someone told me today there was one paper by Alan Turing that I was explicitly not allowed to write this essay on
16:17:44 <Taneb> I checked just now and it turns out it's the other one
16:18:30 <Taneb> Computing Machinery and Intelligence
16:18:38 <Taneb> I was going to write my essay on On Computable Numbers, with an Application to the Entscheidungs problem
16:19:03 <coppro> entscheidungsproblem is one word
16:19:19 <Taneb> coppro, yeah but I was copy-pasting it from somewhere that was wrong because I'm lazy
16:19:20 <coppro> Bike: Turing has one Paper and a bunch of papers
16:21:17 <Taneb> "You can choose any publication within the area
16:21:17 <Taneb> of computer science; the only restriction is that you cannot choose Turing’s “Computing
16:21:17 <Taneb> Machinery and Intelligence” or any publication whose main topic is the Turing test.1"
16:23:04 <zzo38> It is mentioned variables "hp" and "Hp" being different? Someone also mentioned being confusing if you mixed it with Greek, but that isn't a problem if non-ASCII characters aren't allowed in variable names (I prefer not to allow non-ASCII anywhere except comments, string literals, and character literals, and in all such cases can be any encoding following Principle of Extended ASCII; UTF-8 is one such encoding).
16:23:30 <zzo38> And I have done things where "VERSION" and "version" are two different macros.
16:25:34 <fizzie> zzo38: The code in question seemed to have a convention that if "foo" was something, "Foo" was some processed version of it; there were more pairs like that.
16:27:01 <zzo38> I suppose that can be such a convention. (In my program I had no such convention; furthermore I didn't notice I had given these things similar names until afterward)
16:31:12 <zzo38> (Also in mine they are macros, not variables, although "version" expands into something that can be used as an lvalue but never is used that way)
16:32:37 <Phantom___Hoover> Taneb, was this because they were sick of everyone writing essays on the turing test
16:32:46 <Taneb> It really sounds like it
16:33:37 <Bike> shame. i'm sure they'd love my essay on ESP
16:36:32 <Bike> computing machinery and intelligence isn't a very mathy paper, anyway, and presumably this is a math class
16:37:14 <Taneb> Bike, it's a Computer Science class
16:38:05 <Phantom___Hoover> are you doing maths with computer science or straight CS, i keep forgetting
16:39:57 <Phantom___Hoover> (it doesn't help that in warwick that's called discrete maths)
16:40:38 <mnoqy> why do you have to write an essay on a turing thing. what the heck
16:41:43 <zzo38> The descrpition given why "idso" Deadfish isn't acceptable in dc, fails to mention that it is because "s" requires to put another letter afterward.
16:41:43 <Taneb> mnoqy, I have to write an essay on a classic paper in computer science
16:42:21 <Bike> what's weird about reading old papers.
16:42:36 <mnoqy> ther'es nothing weird about reading old papers
16:42:51 <mnoqy> what's weird is having to write an essay about them for a cs class
16:43:17 <Bike> why? it gets students to read and thin about old papers.
16:43:29 <Phantom___Hoover> having students familiarise themselves with the literature on their subject of choice is indeed weird and pointless
16:44:21 -!- muskrat has joined.
16:45:10 <boily> having... students... think?
16:45:37 <Phantom___Hoover> you'd almost think not everyone did a load of amateur study outside of university
16:48:00 <Jafet> A computer science degree certifies that you have been learning the right programming languages since preschool
16:48:33 <Jafet> ...and, in this case, also that you can write an essay
16:49:12 <Jafet> Is the simplexity analysis paper considered classic
16:49:35 <Phantom___Hoover> just do church's one or something, you get bonus hipster cred for that
16:53:12 <fizzie> How about "On certain formal properties of grammars"?
16:58:41 -!- oerjan has joined.
17:00:17 -!- Phantom___Hoover has quit (Ping timeout: 272 seconds).
17:01:49 <oerjan> <boily> I just stumbled on a SO comment by oerjan. <-- i was slightly active there for a while, sadly i never properly registered, and all my points disappeared when the cookies or whatever had expired after a prolonged absense.
17:02:12 <oerjan> or well, i guess the points are still there, i just don't have access to that profile.
17:03:55 <oerjan> hm searching for "Ørjan Johansen" gives one hit which is not me :(
17:04:42 <boily> there are otherjans on the Planet? that's shocking!
17:04:56 <Taneb> oerjan, help I'm not evil enough
17:05:51 <zzo38> Not evil enough for what?
17:05:58 <Taneb> zzo38, to be a supervillain
17:06:29 -!- Nisstyre-laptop has joined.
17:07:02 <zzo38> Then don't be a supervillain!
17:08:07 <Taneb> oerjan, I don't like cats :(
17:08:15 <Taneb> I find them a bit scary
17:08:17 <oerjan> ah there's the problem.
17:08:34 <zzo38> Do you prefer rats?
17:08:36 <fizzie> You can start with a cat facsimile.
17:08:38 <zzo38> Or, perhaps, hats?
17:09:04 <oerjan> i think bats would also do.
17:09:08 <fizzie> Or with some sort of a starter cat.
17:09:19 <oerjan> although you may then need some dental adjustment.
17:09:55 <oerjan> now, bats _and_ rats, then we have a good basis for a supervillain.
17:10:37 <zzo38> You don't need bats and rats to have a basis for a supervillain.
17:10:44 <fizzie> https://dl.dropboxusercontent.com/u/113389132/Misc/20131016-starter-cat.png that's not at all what I was looking for.
17:10:54 <zzo38> Nor does it really help, but it doesn't hurt, either.
17:11:03 <oerjan> zzo38: there are of course many possible bases.
17:11:26 -!- Sprocklem has quit (Ping timeout: 245 seconds).
17:11:34 <oerjan> btw the rats in question should be plague infected.
17:11:47 <zzo38> No, plague isn't necessary.
17:13:26 <oerjan> zzo38: how can the arrival of your apparently unmanned ship in the harbor cause proper terror if it doesn't have plague-infested rats?
17:13:27 -!- adhi has joined.
17:13:43 -!- adhi has left.
17:13:58 <oerjan> we got neither a hi nor ads.
17:14:16 <zzo38> oerjan: Well, there is the other way.
17:15:11 <zzo38> If you hear ghosts in the ship then maybe it might cause "proper" terror
17:15:17 <oerjan> oh. a supervillain could also be based on vats. the larger the better.
17:15:54 <Taneb> I hear Stalin did very well with only secretariats!
17:16:09 <zzo38> Yes, secretary is good enough.
17:16:31 <oerjan> Taneb: i don't think plaits has the right pronunciation?
17:16:37 <boily> secretarian gnat-rat mutagenic hybrids. that I would be scared of.
17:16:39 <Taneb> oerjan, it does with my accent?
17:16:47 <zzo38> Rats, bats, hats, cats, secretaries, computers, etc can be even if you are not a supervillain, I suppose.
17:16:52 <oerjan> Taneb: hm i guess it's an evil accent.
17:16:52 -!- Sprocklem has joined.
17:17:01 <Taneb> oerjan, well, I am British, after all
17:17:36 <boily> stupid google translate. en:plait → fr:plait.
17:17:51 <oerjan> also a good basic supervillain policy is tits for tats
17:18:37 <boily> oh. apparently en:plaits is fr:tresses. much better.
17:19:11 <fizzie> plait /n./ 2. †c. fig. A complication, an aberration; a quirk, a snag, a wrinkle; something obscured, concealed, or hidden away, esp. in order to mislead or deceive; a kink, a dodge, a trick. Obs.
17:20:06 <oerjan> i understand aberrations may be supervillains. Mr. Kjugobe of course excepted.
17:21:04 <fizzie> 1589 G. Puttenham Arte Eng. Poesie iii. xxiv. 245 Oportet iudicem esse rudem & simplicem, without plaite or wrinkle, sower in looke and churlish in speach.
17:21:12 <Taneb> oerjan, wiktionary says "plait" can be pronounced either way
17:21:26 <zzo38> Of course it is possible, but so may someone else be supervillains. Mr. Kjugobe is of course excepted, but other thing can also be exception possibly.
17:23:25 <oerjan> i am still wondering whether sir vulpus of yafgc is an exception or an example. the claims were divided.
17:23:36 <oerjan> (and the story arc isn't over yet.)
17:23:46 <boily> fungot: kjugobe? vulpus? yafgc?
17:23:47 <fungot> boily: yes i do gregor. he wore a different hat every fnord insanely good coder, either?)
17:24:15 <fungot> Taneb: hackish. maybe but it just looks... obese and confusing... i like watching that trace :)
17:24:38 <oerjan> boily: yafgc is a web comic (some times nsfw.) kjugobe is zzo38's d&d character, an illithid (and thus an aberration.)
17:25:33 <oerjan> vulpus is a character mentioned in passing in the former as part of the current arc's background story.
17:26:51 <boily> I never illithided. too much repulsion.
17:27:23 <boily> zzo38: how do you pronounce your char's name? /kjugɔbe/?
17:27:43 <zzo38> boily: Kind of like that, yes.
17:27:57 <zzo38> (Actually their full name is: Iuckqlwviv Kjugobe.)
17:28:20 <oerjan> (he was a werewolf. the not entirely answered question is whether he was a very good or a very evil werewolf. although one of the sides seems to have lost a lot of believability.)
17:28:34 <zzo38> oerjan: Is he necessarily either?
17:29:49 <oerjan> zzo38: it is of course possible that both sides exaggerated their claims.
17:29:59 * boily tries to pronounce his first name as if IPA... “juck... juckqlw *cough*...”
17:31:00 <zzo38> I made it up using a algorithm designed to make up name by rolling a d6 and d20 dice, except that I programmed it into the TI-92 calculator instead. It is OK if the name is difficult for humanity.
17:33:04 <boily> my paranoia character is named “Shlebesh Gurgudugurgudu”, with some random generation number stuck at the end.
17:33:19 <boily> (we lost track a long time ago, what with the multiple deaths and other mishaps...)
17:33:56 <oerjan> i though paranoia names were shorter than that.
17:34:22 <oerjan> and had a color character in the middle.
17:34:48 <oerjan> although i guess that's not permanent.
17:35:26 -!- Sprocklem has quit (Ping timeout: 240 seconds).
17:35:32 <boily> well, it's usually shortened to just “Shle”.
17:38:54 <zzo38> I shorten my character's name to just "Kjugobe" since there isn't anyone else of that name in context.
17:41:04 <zzo38> Although the name "Gxxyuxihuvxi" is really difficult to pronounce, so I never do.
17:42:03 <zzo38> Instead we call Gxxyuxihuvxi "the god who shall not be named because it is difficult to pronounce".
17:42:16 <oerjan> zzo38: i think it's not that hard if you do the x'es as ipa [x]
17:42:39 -!- NihilistDandy has joined.
17:43:06 <zzo38> oerjan: Maybe, but I don't think that is quite the way
17:43:11 <oerjan> assuming you can pronounce that sound in the first place
17:48:19 <zzo38> I am not sure what it is but I will look it up in Wikipedia
17:49:52 <oerjan> https://en.wikipedia.org/wiki/Voiceless_velar_fricative
17:50:25 -!- NihilistDandy has quit (Ping timeout: 272 seconds).
17:50:32 <zzo38> Yes that is the one I found
17:52:55 <zzo38> I don't really like the IPA symbols; the consonants and vowels are arranged in a grid and I would use combined symbol corresponding to what row and column of the grid they would belong in, and voice/unvoice variant, etc
17:54:36 -!- muskrat has quit (Quit: Leaving).
17:54:53 -!- conehead has quit (Quit: Computer has gone to sleep.).
17:55:28 -!- Nisstyre-laptop has quit (Ping timeout: 268 seconds).
17:55:48 <boily> zzo38: that may result in interesting compound characters, like labialized-pharyngealized-spirant labiodental syllabic plosive with no audible release.
17:56:25 <boily> (also known as the “PFSHSHSHHHTHTHT” sound.)
17:56:31 -!- NihilistDandy has joined.
17:56:41 <zzo38> boily: Yes, allowing compounds like that too
18:01:40 <oerjan> yay i managed to recover my stackoverflow user
18:05:24 -!- NihilistDandy has quit (Ping timeout: 260 seconds).
18:08:28 <Vorpal> The hell just happened to Steam? I was trying to put in some activation codes, and it was one copy and paste behind if that makes sense. Like it pasted the last code I entered before this instead. Which was bloody weeks ago. Then on the next attempt it pasted one I actually copied. And when I put in yet another code (I was adding the latest humble bundle, so there were a few to add in) it kept being one
18:09:46 <Vorpal> That was super strange
18:09:59 -!- Taneb has quit (Quit: Leaving).
18:11:28 <boily> seeing and IP in an X.0.0.0/8, for small values of X, always disturb me. I was thinking that Taneb came from the Ministry of Defence or something.
18:13:21 <fizzie> I saw one from 5.0.0.0/8 the other day.
18:13:46 <fizzie> Oh, that's also where Taneb was from.
18:13:57 <fizzie> Makes sense, they gave it to RIPE.
18:14:49 <fizzie> APNIC's 1.0.0.0/8 is even more impressive.
18:15:16 <fizzie> 1.2.3.0/24 is reserved by the "APNIC Debogon Project".
18:15:32 <fizzie> And I think 1.1.1.0/24 too.
18:15:40 -!- Sprocklem has joined.
18:22:56 -!- Nisstyre-laptop has joined.
18:25:19 <boily> Vorpal: that one ain't scary, it's a lucky IP.
18:26:12 <Vorpal> boily, oh? It is just googles public DNS server
18:26:38 <boily> I know. they solved transient problems many times here.
18:26:57 <Vorpal> I used it when the local unbound daemon fucked up once
18:27:06 <Vorpal> Since I run my own recursive resolver
18:27:17 <Vorpal> I don't think Google's DNS servers does DNSSEC does it?
18:27:20 <Vorpal> It didn't last I looked
18:27:45 <zzo38> Maybe we can make some METAFONT program for the new kind of pronounce symbols and see what compounds are made with it?
18:28:08 <boily> has anybody outside of knuth created metafont fonts?
18:28:29 <Vorpal> Why not use xelatex and plain old TTFs?
18:28:42 <zzo38> boily: Yes, I have done
18:28:54 <zzo38> Vorpal: I prefer METAFONT, I think it is better than TTF
18:29:04 <Vorpal> zzo38, on what grounds?
18:29:18 <Vorpal> TTF is significantly more popular, surely there must be some reason for that
18:29:34 <zzo38> TTF is more popular because Windows uses it.
18:30:23 <oerjan> zzo38 does have a point
18:30:31 <Vorpal> And Mac OS X. And anything based on Qt, GTK or any other modern X11 toolkit
18:30:46 <Vorpal> But why is METAFONT better than TTF?
18:31:16 <Vorpal> I'm ignorant as to their technical merits, so maybe you can summarize them?
18:32:06 <boily> Vorpal: according to the wikipédia, google's DNS has DNSSEC.
18:32:27 <zzo38> Well, METAFONT programs can include any parameters you want to, as well as kern/ligatures, and compiling for the print device (it can include decisions as to how to make it print on that device), and you can type the program to write the font in any text editor, or even interactively
18:32:41 <Vorpal> boily, ah, it says "Since May 6 2013". Pretty sure last time I looked was maybe a year ago or so
18:32:55 <boily> when I'm feeling particularly masochistic, I sometimes try to understand how fonts in LaTeX work. never achieved that goal so far.
18:33:40 <zzo38> TTFs can be compiled into .TFM and .GF too though, like METAFONT can, and so can other formats perhaps.
18:34:04 <oerjan> <zzo38> The descrpition given why "idso" Deadfish isn't acceptable in dc, fails to mention that it is because "s" requires to put another letter afterward. <-- well i suppose they only need one reason why it breaks, although you can add that if you want to.
18:34:35 <Vorpal> zzo38, And TTF doesn't support arbitrary parameters?
18:34:37 <zzo38> I have created a font in METAFONT for chess variants, and a TeX macro package to use them (including parsing algebraic chess notation)
18:35:17 <zzo38> Vorpal: I don't think so; I think TTF has a fixed set of parameters and you cannot use them for whatever purpose you want to either.
18:35:47 <Vorpal> zzo38, Also, how does LaTeX know how to make use of all those extra parameters in METAFONT?
18:35:56 <Vorpal> Without manually telling it to that is
18:36:41 <zzo38> Vorpal: The parameters are compiled externally and then the resulting font is given a name.
18:37:54 <Vorpal> Ah, that seems like a downside compared to TTF. With TTF I don't have to manually generate the font in 22pt bold italic for 600dpi. I seem to remember having to do such stuff for pre-XeLaTeX
18:38:25 <zzo38> Vorpal: Well, actually it usually automatically generates the fonts if needed, if the parameter file exists, which they usually do.
18:38:58 <zzo38> Even if it doesn't exist, you can set the font magnification in TeX, so you don't need a parameter file for individual magnifications of a font.
18:39:15 <Vorpal> zzo38, If I generate a PDF I want the font to be resizable to whatever the viewer has available. It could be for printing, or displaying on a low res PC monitor. Or a high res tablet screen (my Nexus 10 tablet has 300 dpi for example)
18:39:48 <zzo38> DVI does that doo; the rendered fonts aren't stored in the DVI, they are separate so that they can be rendered for the printer in use.
18:40:05 <zzo38> (This also results in smaller file sizes since you don't need to store a copy of the font with each document)
18:40:11 <Vorpal> zzo38, So the document isn't self-contained? It needs the fonts to be distributed alongside it?
18:41:27 <zzo38> Vorpal: Yes, although the Computer Modern fonts are generally available anyways in such case. I don't find it so impractical; I think not being self-contained like this is in fact better (although you can distribute the fonts in a .ZIP if you want to; there are also .DVI extensions to support embedded fonts but I prefer non-embedded fonts)
18:43:14 <Vorpal> zzo38, I generally prefer PDF because I know it will look exactly the same everywhere, and everyone can open it.
18:43:22 <zzo38> METAFONT also makes the fonts much easier to modify.
18:43:34 <zzo38> Vorpal: Yes, and I find that to be exactly the problem; not everyone has the same model of printer!
18:43:37 <Vorpal> Never felt the need of it
18:43:59 <zzo38> Causing it to look worse on some printers.
18:44:13 <Vorpal> Really? You mean colour matching or such?
18:44:23 <Vorpal> Isn't that up to ICC profiles?
18:44:46 <zzo38> No, I mean DPI, aspect ratio, ink thickness, and a few other related things.
18:44:50 <Vorpal> Since I don't embed bitmap fonts in the PDFs, surely it will scale properly
18:45:11 <zzo38> Well, it will scale better than bitmap fonts do, at least
18:45:25 <Vorpal> Anyway, paper aspect ratio varies yes, but isn't that fixed in the DVI?
18:45:26 <zzo38> That doesn't necessarily mean it will scale properly, though.
18:45:32 <Vorpal> It is either US Letter or A4
18:45:40 <zzo38> Vorpal: I mean pixel aspect ratio
18:48:23 <zzo38> Also, METAFONT and Plain TeX and DVI and whatever will continue to work in the same way fifty years from now, but the Lua-pdf-e-XeLaTeX probably won't work anymore because they will replace it with something else which is a bit similar but really different.
18:49:24 <Vorpal> Doubt it, PDF is a standard nowdays. There is even an archival format of PDF
18:49:35 <Vorpal> https://en.wikipedia.org/wiki/PDF/A
18:51:41 <Bike> heh, font linking mentioned right there
18:51:51 <zzo38> Maybe PDF/A might work, but not necessarily quite the same way, and source files may not work either (some old C source files already fail to work). And then you need LaTeX packages and stuff, resulting in even more confusion. But I think METAFONT, Plain TeX, and DVI will continue to work in fifty years without any difficulties.
18:52:11 <Vorpal> Bike, yes PDF supports it, but it is deprecated nowdays iirc
18:57:36 <lexande> what does the iceweasel say?
18:57:51 <Vorpal> lexande, with regards to what?
19:00:40 <Vorpal> Does anyone use JPEG 2000?
19:04:21 -!- zzo38 has quit (Remote host closed the connection).
19:10:05 -!- Sprocklem has quit (Ping timeout: 268 seconds).
19:14:40 <fizzie> Vorpal: Movie theatres, AIUI.
19:15:01 <fizzie> Vorpal: "Typical" digital movie distribution formats have each frame encoded as a JPEG2k image.
19:15:06 <fizzie> Or so I've heard from somewhere, anyway.
19:16:29 <fizzie> "Digital Cinema Initiatives (DCI), a joint venture of the six major studios, published the first version (V1.0) of a system specification for digital cinema in July 2005. -- Briefly, the specification calls for picture encoding using the ISO/IEC 15444-1 "JPEG2000" (.j2c) standard and use of the CIE XYZ color space at 12 bits per component encoded with a 2.6 gamma applied at projection. Two ...
19:16:35 <fizzie> ... levels of resolution for both content and projectors are supported: 2K (2048×1080) or 2.2 MP at 24 or 48 frames per second, and 4K (4096×2160) or 8.85 MP at 24 frames per second."
19:19:24 -!- Sprocklem has joined.
19:20:37 <Bike> wowwww, CIE is old
19:21:11 <Bike> the oldest, in fact
19:22:31 <olsner> International Commission on Illumination (CIE)
19:23:48 <Bike> the XYZ colorspace, i mean
19:24:25 <olsner> `addquote <boily> I love django.
19:24:32 <HackEgo> 1120) <boily> I love django.
19:24:36 <mnoqy> how many quotes about django is that now
19:24:57 <olsner> one that doesn't have me in it! \o/
19:25:07 <ion> Ooh, an aligned body for once.
19:25:16 <boily> are we all speaking about the same django? I was refering to the Python framework here.
19:25:32 -!- asie has joined.
19:26:36 <shachaf> i was thinking a django was a type of dog
19:26:40 <shachaf> but i guess that's a dingo
19:27:23 <Vorpal> <ion> Ooh, an aligned body for once. <-- quite
19:27:40 <oerjan> django is a type of reinhardt
19:27:52 <Vorpal> I thought it was a movie as well?
19:28:19 <fizzie> Bike: I recall someone telling me CIE 1931's frequency response curves are... not very good, physiologically speaking, according to more accurate, post-1931 measurements.
19:28:41 <Bike> fizzie: i know nothing about colorspaces :( which is probably bad since i'm interested in perception psychology
19:29:08 <boily> fizzie: of course their curves aren't good. the world was in black and white back then!
19:29:11 <Vorpal> fizzie, What about the Lab colorspace?
19:29:12 <oerjan> ion: you can increase the alignment by changing to a 6-character nick
19:29:12 <fizzie> Bike: There's also CIE 1960, CIE 1964 and CIE 1976; they keep doing it over.
19:29:26 -!- stuntaneous has joined.
19:30:15 <fizzie> Vorpal: That's CIE 1976, I think. It does perceptual uniformity better, but it's still just a nonlinear compression over CIE XYZ, which means it has the same underlying curves, AFAIK.
19:30:47 <Vorpal> fizzie, so what is a good colour space then? That covers the entire vision I mean
19:30:54 <boily> fizzie: I think the most common illuminant used in Lab is D65, which was added in '67.
19:31:09 <Vorpal> fizzie, I guess none of the RGB ones are very good?
19:31:28 <Bike> man. i don't think i even understand what a color space is.
19:31:33 <fizzie> Vorpal: The one where you store actual spectrum vectors, presumably.
19:31:54 <Vorpal> My monitors at work can do 10-bit Adobe RGB btw. Not sure if my GPU there supports that though. Wide gamut can be interesting.
19:33:10 <fizzie> Vorpal: I guess that's a matter of definition. It doesn't exist in any sort of colorspace you could get a regular computer to use, but certainly astronomers and the like use spectroscopes of all kinds.
19:33:40 <Bike> what if i hook a spectroscope to my computer to play Half Life 2 HDR
19:34:22 <fizzie> I don't know what the inverse of a spectroscope (a projector that can reproduce a desired spectrum) is called.
19:34:40 -!- KingOfKarlsruhe has quit (Ping timeout: 264 seconds).
19:35:59 <fizzie> In fact, I'm not sure if those really exist.
19:36:35 <Bike> it seems like it would be hard to make
19:37:05 <boily> google is no help on that subject :(
19:37:15 <fizzie> Bike: I think you start with a full-spectrum light source, split it up with something prismatic, put a variable-intensity filter (any old LCD) and then put the frequencies back again.
19:37:19 <fizzie> Some lenses may be involved.
19:37:22 -!- MindlessDrone has quit (Quit: MindlessDrone).
19:37:29 <olsner> https://en.wikipedia.org/wiki/Spectrometer ?
19:37:40 <fizzie> olsner: That's the wrong direction.
19:37:55 <Bike> a backwards spectrometer, olsner.
19:38:14 <olsner> > reverse "spectrometer"
19:38:32 <Bike> i think some genera of spiders have those
19:38:45 <fizzie> Invariably fatal, I'm sure.
19:38:58 <Bike> indeed, and it's hard for the spiders' families
19:39:28 <fizzie> Also I suppose the retemortceps has less obvious uses than a spectrometer.
19:40:54 -!- ais523 has joined.
19:41:12 <olsner> did "(a projector that can reproduce a desired spectrum)" refer to the thing you were looking for, or the spectroscope?
19:41:28 <fizzie> Google is indeed no real help, though "Relativistic Positron Creation Using Ultra-Intense Short Pulse Lasers" is a pretty impressive title.
19:41:28 <Bike> the thing being looked for
19:41:49 <Bike> will keep in mind when i need positrons
19:41:57 <olsner> right, I was looking at that page and thinking "yeah, this is the inverse of a projector that can reproduce a spectrum"
19:42:01 <fizzie> You can never have too many positrons.
19:42:10 <boily> http://babele.tesionline.it/babele/testo.jsp?id=58169 ← some... article, maybe? something about ypocsortceps.
19:42:43 <Bike> maybe such a thing would be useful for visual system physiology studies
19:43:04 <Bike> boily: haha what the hell
19:43:26 <boily> Bike: I don't know. I really don't know.
19:43:44 <Bike> maybe it's a test of a "grab a webpage and reverse it" thing...
19:44:03 <fizzie> Hmm. There's "Agile Spectrum Imaging: Programmable Wavelength Modulation for Cameras and Projectors".
19:44:18 <fizzie> It seems to have some wavelength masking abilities.
19:44:19 <boily> fungot: what is the positronic yield of a reverse spectroscopic arachnoid?
19:44:19 <fungot> boily: that way, exactly... i'd have said quebecois myself, but that looks less like sql :p)
19:44:44 <boily> fungot: si t'insistes sur le québécois, ça te dérangerais-tu d'explique de quossé qui se passe?
19:44:44 <fungot> boily: i'm usually writing code to get at the procedure's fnord
19:45:01 <boily> the Search for the Quintessential Fnord.
19:46:10 <fizzie> http://sprunge.us/OYfa I guess that's at least somewhat close.
19:46:17 <fizzie> (Also: rainbow plane.)
19:46:47 <Bike> one of the best mario kart levels tbh
19:47:21 <fizzie> http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=1262930 <-- oh, there we go.
19:48:21 <fizzie> To get back to the initial comment, apparently "hyperspectral image projector" is the term I was looking for.
19:48:56 -!- Sprocklem has quit (Ping timeout: 268 seconds).
19:49:04 <Bike> fizzie: that's pretty wizard lookin'.
19:49:37 <fizzie> 1024x768 pixels and 2-5nm spectral resolution for the 450-2400nm range, that sounds pretty impressive.
19:50:00 <fizzie> (Also: "supercontinuum".)
19:52:00 <fizzie> Article doesn't mention anything about playing Half Life on it.
19:53:01 <fizzie> https://dl.dropboxusercontent.com/u/113389132/Misc/20131016-hyperspectral-projector.png oh, it's that simple?
19:54:16 <fizzie> I'm reasonably happy that my guess about how you'd go around making one was actually pretty accurate, except for substituting a micromirror thing in place of a LCD filter.
19:55:16 <olsner> simply integrating sphere the eigenspectra into abundance images
19:55:45 <fizzie> Okay, yes, they also made a 2D image out of it without using W*H copies of the entire device.
19:56:08 <fizzie> With the Spatial Engine. But the Spectral Engine looks right.
19:57:30 <Vorpal> fizzie, isn't the W*H issue solved by the phrase "within the integration time"? Basically they don't do it all at once?
19:57:34 <Vorpal> Or did I miss something
19:57:49 <Vorpal> That image didn't make a lot of sense to me though
19:58:37 <fizzie> I think it's slightly more complicated than just passing a variable-spectrum beam over the pixels.
19:59:06 <Vorpal> That is one hell of a complicated prism setup though
20:00:16 <fizzie> "When the spectral engine is coupled to the spatial engine to make the full HIP, the spectrally-programmed light globally illuminates the 1024x768 pixels of the spatial engine DMD. Gray scale at this DMD enables realistic 2-dimensional images to be displayed, as in conventional projectors. The difference here is that the time-integrated spectra projected from each spatial engine DMD pixel can ...
20:00:34 <fizzie> ... be individually controlled, using a compressive projection algorithm described previously, enabling projection of realistic spatial/spectral image cubes into sensors under test [2]."
20:00:51 -!- epicmonkey has joined.
20:01:11 <fizzie> I don't know if I want to go and see what [2] is.
20:01:20 <fizzie> There's a bit of a summary here.
20:01:26 -!- KingOfKarlsruhe has joined.
20:01:33 <fizzie> But it goes into component eigenspectra and abundance images pretty soon.
20:01:43 <Vorpal> What do those words even mean?
20:02:55 <fizzie> As far as I can tell, it's pretty much just that it generates K (optimally selected) spectra and K corresponding greyscale images, giving that way individual per-pixel weights for each K spectra.
20:03:30 <fizzie> So if you'd have just pixels with two spectra in the image, you'd get those two (at different times) out of the spectral engine, and a black-and-white picture (and its inverse) in the spatial engine.
20:03:54 <fizzie> And that's done fast enough so that it falls in the "integration time of the sensor", so that it just sees the weighted sums.
20:04:20 <fizzie> Which would probably work just fine for a human eye too.
20:04:47 <fizzie> Though if it's your eye, this part sounds potentially uncomfortable: "A sensor under test is focussed at infinity with its optical axis co-aligned with the HIP output collimator --"
20:07:16 -!- epicmonkey has quit (Ping timeout: 245 seconds).
20:07:18 -!- Sprocklem has joined.
20:07:27 <fizzie> Also apparently that's how DLP projectors do color images these days too, there's a rapidly cycling RGB filter in front of the light source and then for each color the mirror is set to have that color component of the image, so that they blend together.
20:07:31 <fizzie> So this is pretty much the same thing, except instead of fixed R,G,B primaries, it can generate an arbitrary amount of optimal spectra to reproduce the original.
20:09:21 <fizzie> (I still haven't figured out what a "supercontinuum-based spectral engine" is.)
20:09:39 <fizzie> (So far it doesn't seem to be related to the CH.)
20:12:06 <kmc> i think that's the oldest way to do color with DLP
20:12:42 <fizzie> Well, okay, you can also do three chips at the same time.
20:13:14 <fizzie> Now that you mention it, I do remember I knew this already.
20:13:22 <kmc> yeah, that's how the more expensive ones work now
20:13:41 <kmc> you can also use a rapidly changing LED or laser light source, which is like the color wheel but faster so less distracting
20:13:47 <kmc> but I kind of like the rainbow artifacts myself :)
20:14:00 <fizzie> "In optics, a supercontinuum is formed when a collection of nonlinear processes act together upon a pump beam in order to cause severe spectral broadening of the original pump beam."
20:14:56 <fizzie> Apparently it's just a way to get a flat and wide spectrum out.
20:15:31 <kmc> "severe spectral broadening" reminds me of "severe tire damage"
20:15:40 -!- carado has quit (Ping timeout: 246 seconds).
20:15:44 <kmc> "The number of mirrors corresponds to the resolution of the projected image (often half as many mirrors as the advertised resolution due to wobulation)."
20:15:49 <kmc> sure, of course, the wobulation
20:15:52 <fizzie> That's appropriate, since both end up with a flat.
20:16:07 <metasepia> Wobulation is a term which refers to the known variation (or wobble) in a characteristic.
20:16:20 <fizzie> fungot: Stop wobulating.
20:16:20 <fungot> fizzie: until well into it in pure python? i've written a few scripts for the darcs repository
20:16:33 <boily> what the fungot. how in fungot is that a real word?
20:16:33 <fungot> boily: tommo it depends on? ( re ' i have never heard of
20:16:40 <boily> fungot: me neither.
20:16:40 <fungot> boily: my point was that if you listen to _me_? going to be interesting
20:16:50 <fungot> boily: no ei kovin :dd mit vittua :dd then i might be fnord.
20:17:10 <boily> dd, finnish, and fnords. no way I'm going to follow that procedure.
20:17:12 <fizzie> fungot: I'll wash your mouth with soap if you keep using that sort of language, young bot!
20:17:12 <fungot> fizzie: shivers wrote the srfi...
20:17:50 <fizzie> The middle part is pretty much "what the fuck".
20:19:56 <boily> strangely, we don't have any standardized wtf. only «[RANDOM EXPLETIVE] c'est quoi ça?».
20:20:22 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
20:21:49 <fizzie> <[redacted]> suddenly [redacted] and fiz were there too and [redacted] drow to the garden with a car. he said to me "[translating from Finnish: what the FUCK are you doing here so early?]" (usually it's just "[translating: what the FUCK are you doing here?") and i said with a slightly ironic tone "there's suitable company for you here, nutcases escaped from a lunatic asylum." he was like "yeah ...
20:21:55 <fizzie> ... i know about that".
20:22:00 <fizzie> <[redacted]> he tried to communicate something to fiz with lip movements so that i wouldn't understand but i did.
20:22:04 <fizzie> <[redacted]> then one of the lunatics stabbed him and in a way they also stabbed me, it was like we were united. fiz and [redacted] tried to escape with a car (not [redacted]'s car but another one) and leave me there but i got into the car. don't know why [redacted] was left behind.
20:23:06 <fizzie> In retrospect, I should've given some pseudo-identifiers to all those [redacted]s.
20:23:18 <mnoqy> did redacted have a dream with you and redacted in it
20:23:32 <boily> censorship titillates me.
20:23:35 <fizzie> I think that was the case.
20:24:49 <fizzie> Well, me and redacted and redacted too.
20:25:15 <kmc> I would probably write that as "<▒▒▒▒▒▒▒▒> suddenly ▒▒▒▒▒▒▒▒ and fiz were there too and ▒▒▒▒▒▒▒▒"
20:25:17 <boily> ooooh... Nordic Redacted Ménage à Trois! sexy.
20:25:20 <kmc> because ♥ box drawing characters
20:26:26 <fizzie> "░▒▓████▓▒░ and ░▒▓████▓▒░"
20:28:42 <fizzie> ▗▘▀▖ ▌ ▗ ▖▖ ▌▖▖▄ ▖▖▝ ▄ ▗▖
20:28:43 <fizzie> ▚ ▝▖ ▛▖▌▌▞▖ ▞▌▛ ▄▌▙▌▐ ▌▌▚▌ ▘▀ ▀ ▝ ▘▘ ▝▘▘ ▀▘▀▘▝ ▘▘▄▘
20:28:52 <fizzie> It is the irssi line combining thing.
20:29:19 <fizzie> ▜▘ ▗▖▖▖▗▖▌ ▄ ▟▖ ▟▖▌ ▝ ▗▖
20:29:19 <fizzie> ▐ ▘▖▌▌▌ ▙▘ ▄▌▐ ▐ ▛▖▐ ▘▖
20:29:19 <fizzie> ▀▘ ▀ ▝▘▝▘▘▘ ▀▘ ▘ ▘▘▘▝ ▀
20:29:40 <myname> /exec -o toilet indeed
20:30:05 <fizzie> That was "rfk86ize.pl".
20:30:24 -!- shikhin has quit (Read error: Operation timed out).
20:30:27 <fizzie> I made that 4x6 pixel font for rfk86, it sort of goes well with the 2x2 blockery.
20:31:10 <ion> There’s a setting for the line combining thing.
20:31:41 <fizzie> ion: Maybe I should disable it completely, because I don't remember the last time it was actually useful. (Though the paste detection does usually work.)
20:31:51 <shachaf> why does it do the line combining thing anyway......
20:32:07 <fizzie> shachaf: It's for pasting from the backscroll, I believe.
20:32:41 <fizzie> Otherwise line-wrapped comments turn out real ugly.
20:32:46 <ion> If i, say, copy certain two lines from my buffer to the clipboard, i’ll get:
20:32:48 <ion> ke233142 fizzie ion: Maybe I should disable it completely, because I don't
20:32:50 <ion> remember the last time it was actually useful. (Though the
20:32:57 <ion> The functionality is a kluge for detecting that.
20:33:35 <fizzie> (Fun fact: Unicode blockifying script located with a find . -iname '*.pl' -print0 | xargs -0 grep '▚' since I didn't remember where I put it.)
20:33:56 <ion> find -iname '*.pl' -exec grep '▚' '{}' '+'
20:34:02 <shachaf> But then you have to figure out how to type ▚
20:34:17 <fizzie> I had gucharmap already open for ░▒▓█.
20:34:22 <ion> grep '▚' **/*.pl
20:34:44 <fizzie> ion: Too fancy by far.
20:35:04 <shachaf> ion: Will that run multiple grep processes if there are too many files to fit on one command line?
20:35:45 <kmc> fizzie: haha
20:35:59 <ion> Nope. It will also just skip directories it can’t read.
20:36:22 <fizzie> It doesn't even seem to work. Is it a non-bash thing?
20:36:43 <shachaf> Probably a zsh thing in ion's case.
20:37:09 <shachaf> (ion seems like the sort of person who'd use zsh.)
20:37:10 <FireFly> that didn't work as well as I hoped it would
20:37:12 <fizzie> I'm relieved other people fail as much as I do.
20:38:03 <ion> Excluding shachaf. He never fails.
20:38:22 <ion> He fails at failing, though.
20:38:49 <lambdabot> cmccann says: < cmccann> shachaf: nobody will associate with someone who breaks the monoid laws! < shachaf> cmccann: That's why I go by a secret identity.
20:39:00 <lambdabot> cmccann says: < cmccann> shachaf: nobody will associate with someone who breaks the monoid laws! < shachaf> cmccann: That's why I go by a secret identity.
20:39:02 <lambdabot> cmccann says: < cmccann> shachaf: nobody will associate with someone who breaks the monoid laws! < shachaf> cmccann: That's why I go by a secret identity.
20:39:10 <ion> @quote cmccann shachaf
20:39:10 <lambdabot> cmccann says: some people blame themselves, some people blame the language, but the people who really know what they're doing blame shachaf.
20:39:24 <shachaf> Are those the only two cmccann shachaf quotes?
20:39:49 <boily> mwah ah ah! useless xmodmapping! I now can directly input ░▒▓█!
20:39:52 <shachaf> It seems like most of cmccann's time is spent coming up with things to say about me.
20:40:25 -!- boily has quit (Quit: chicken of the sea).
20:40:28 -!- metasepia has quit (Remote host closed the connection).
20:41:42 <fizzie> I don't know how you're supposed to connect the diagonal box drawings to anything else, though.
20:43:57 -!- shikhin has joined.
20:44:18 <fizzie> You can do something like http://sprunge.us/SYCW I guess but that's it.
20:44:32 <fizzie> (Also it renders clunky here.)
20:45:42 <ion> It renders somewhat well-ish here.
20:47:22 <ion> http://heh.fi/tmp/fizzie-x.png
20:47:42 <FireFly> What's up with your font rendering thingies?
20:47:58 <FireFly> http://i.imgur.com/NxHHzeN.png
20:48:10 <FireFly> I guess the bottom part is a bit ugly for me
20:48:30 <ion> and/or fonts
20:58:58 -!- callforjudgement has joined.
20:58:59 -!- ais523 has quit (Ping timeout: 260 seconds).
20:59:13 -!- callforjudgement has changed nick to ais523.
21:01:01 <myname> FireFly: uh? looks perfect
21:05:42 <FireFly> myname: here's a zoomed-in not-very-pretty screenshot: http://i.imgur.com/EGMcaXx.png
21:06:13 <FireFly> something's weird with the bottom, but it's not as if it's very noticeable
21:06:45 <myname> don't get your problem
21:11:07 -!- Sprocklem has quit (Ping timeout: 260 seconds).
21:16:13 <Bike> http://radar.oreilly.com/2013/10/announcing-biocoder.html i can't stop laughing
21:21:08 <shachaf> i'll have what fizzie's having
21:21:26 <fizzie> myname: The bottom is supposed to be a solid line without those per-each-character-cell discontinuities.
21:31:48 -!- ^v has quit (Quit: Leaving).
21:33:31 <kmc> @tell zzo38 gcc 4.7.2 doesn't warn me about "if ((x == 4) && (y = 0))" being always false, even with -Wall -Wextra -O3
21:34:16 <Bike> matlab's lint warns about assignment in conditionals, which is... nice considering, i guess
21:35:32 -!- ^v has joined.
21:39:44 <kmc> does it always?
21:40:43 <fizzie> "-Wtype-limits: Warn if a comparison is always true or always false due to the limited range of the data type, but do not warn for constant expressions. For example, warn if an unsigned variable is compared against zero with ‘<’ or ‘>=’. This warning is also enabled by -Wextra."
21:41:21 <fizzie> I was under the impression that there was also a "warn for constant expressions" warning with some extra heuristics to not warn for e.g. while (1), but apparently not.
21:41:53 <kmc> I like that a majority (?) of do...while loops in C are do { ... } while (0);
21:41:58 <kmc> great language feature
21:42:38 <FireFly> Speaking of which, why do C macros use do { ... } while (0) instead of just a plain block { ... } ?
21:42:55 <fizzie> FireFly: Because of semicolon problems.
21:43:08 <fizzie> FireFly: Consider if (x) MACRO(); else bar;
21:43:20 <fizzie> FireFly: Would expand to if (x) { ... }; else bar; which is a syntax error.
21:44:16 -!- mnoqy has quit (Quit: hello).
21:45:15 <fizzie> There is also the if (1) { ... } else variant which works correctly in the absence of any errors, but quietly does bad things if someone accidentally drops the semicolon after the macro. (Because the next statement will then end up in the else branch.)
21:45:59 <olsner> I like the if/else variant
21:46:19 <olsner> wouldn't ever use it though
21:46:22 <fizzie> Bike: *Can* you assign in a conditional, in MATLAB?
21:46:39 <fizzie> Bike: http://sprunge.us/QBWB
21:46:52 <fizzie> Though possibly there are more subtle ways.
21:47:06 <Bike> what about "if (x = 1)"
21:47:20 <Bike> that error doesn't look like "no assignments in a conditional", i mean
21:47:34 <fizzie> Well, you know, MATLAB and errors.
21:47:50 <Bike> "Well, you know, MATLAB and anything"
21:47:58 <fizzie> I was just under the impression that assignment was a statement and not an expression in MATLAB.
21:48:25 <fizzie> "x = 1 + (y = 2)" is equally "not a valid target for an assignment".
21:48:33 <Bike> in octave i get "warning: suggest parenthesis around assignment used as truth value" and the display happens.
21:48:49 -!- ^v has quit (Quit: Leaving).
21:49:00 <Bike> ...in that it allows assignment in conditional?
21:49:06 <fizzie> There are some subscripting things you can also do in Octave that need an assignment to a temporary in MATLAB.
21:49:16 -!- ^v has joined.
21:50:03 <fizzie> Compare e.g. http://sprunge.us/dKEQ
21:50:15 <fizzie> Arguably Octave's behaving better there.
21:50:46 <fizzie> There was some way of writing the subscripting in MATLAB "inline" too, but it wasn't nice.
21:51:53 <Bike> man, i don't even want to think about first class functions in the damn language
21:54:16 <fizzie> Oh, right, of course: subsref(f(4),struct('type','()','subs',{{1}}))
21:54:23 <fizzie> That's the equivalent of Octave f(4)(1).
21:54:30 <Bike> the lab codebase apparently won't work in octave, just stays where it is for an hour and running. so ugh
21:54:50 <Bike> and that's after finding out that octave mkdir is incompatible and fixing it.
21:54:53 <fizzie> ("subsref" being the underlying function called for x(y) syntax.)
21:59:05 -!- KingOfKarlsruhe has quit (Quit: ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]).
21:59:41 <fizzie> Another funny MATLAB/Octave difference: http://sprunge.us/KHNS
21:59:45 <fizzie> See, logspace(a, b), which is normally from 10^a to 10^b, is instead from 10^a to plain b if b equals pi. In MATLAB, it tests "b == pi || b == single(pi)", while Octave *attempts* to get by with "b == pi" because in Octave the == is automatically single-precision if either side is -- but if b is of type double but with the value equal to single(pi)...
22:00:08 <kmc> what the christ
22:00:35 <fizzie> It's also only the 'b' parameter, not the 'a' that has the 'is pi' special case, of course.
22:00:38 -!- nooodl has quit (Read error: Operation timed out).
22:00:56 <Bike> why does it even have that pi behavior
22:01:32 <kmc> it's "useful for digital signal processing where frequencies over this interval go around the unit circle"
22:01:46 <Bike> nonetheless, what the christ
22:01:58 <fizzie> Someone somewhere back in the misty days of history extensively used that "feature", and now they're stuck with it.
22:02:47 <fizzie> Personally I'd guesstimate that at least 90% of all my logspaces have been logspace(log10(x), log10(y)) because I want logarithmically spaced points from x to y, not 10^x to 10^y.
22:03:19 <Bike> fizzie: btw, "foo = ones(x,y); [bar, baz] = find(foo == 1);" is as dumb as it looks right
22:03:53 <kmc> matlab: not even once
22:04:17 <fizzie> Bike: I've never even seen that construction. It seems... very MATLAB.
22:04:24 <Bike> yes. yes indeed
22:04:33 <kmc> (hm, Rust has a feature called "once functions", I wonder if we should also have "not even once functions")
22:04:38 <Bike> i should probably start a git branch for cleaning up shit like that for microoptimization
22:04:51 <Bike> (also the other reasons to fix it)
22:05:46 <Bike> for those at home, that makes bar = a list of x indices and baz = a list of y indices, into foo, in about the least efficient way i can think of
22:06:06 <Fiora> kmc: like "gets()"?
22:06:58 <Bike> so after [foo,bar] = find(ones(2,2) == 1);, foo is [1,2,1,2] and bar is [1,1,2,2].
22:07:22 <kmc> yes, that would be a good "not even once" function :D
22:07:27 <Bike> by the way, functions are aware of how many "out arguments" they have been provided with
22:07:35 <Bike> i could go on for years and i've been doing this for two months.
22:07:44 <Fiora> kmc: sorry, bad joke without context XD
22:08:00 <Bike> "provided with" rather
22:08:19 <fizzie> Bike: I think my "intuitive" MATLAB phrasing of that would've been [bar, baz] = ndgrid(1:x, 1:y); bar = reshape(bar, [], 1); baz = reshape(baz, [], 1); if for some reason I'd've needed them as vectors.
22:08:23 <fizzie> Though I'm sure many would argue for [bar, baz] = ind2sub([x y], 1:x*y) instead.
22:08:56 <Bike> fizzie: i'm reasonably sure all cases of this involve really stupid iterations over a matrix of the right size anyway, so i'll have to think of a better rewrite
22:09:04 <fizzie> I have a gut feeling that all ind2sub-y solutions are sort of "officially preferred". You see them a lot in Mathworks-provided examples.
22:09:49 <Bike> like the actual code is...
22:10:52 <Bike> TnKO = ones(NAct, NTn); [iTest, jTest] = find(TnKO == 1); for iii = 1:length(jTest) ... stuff involving TnKO(:, iii) ....
22:12:05 -!- EgoBot has quit (Ping timeout: 245 seconds).
22:12:35 <fizzie> I guess TnKO is modified there somewhere? Because otherwise I guess all TnKO(:, iii)'s are the same.
22:12:50 <Bike> yeah, it's assignments to zero, it's setting up TnKO.
22:12:52 -!- EgoBot has joined.
22:13:03 <fizzie> (Really, though. MATLAB.)
22:13:32 <fizzie> The premier software package for technical computing.
22:13:59 -!- Sprocklem has joined.
22:14:02 <Bike> oh, man, let me check out this snippet to see if it does what i think it does.
22:14:05 <fizzie> I've got some lines that have triply nested bsxfuns.
22:14:55 <fizzie> SciPy has "auto-broadcasting"; if you do something with a NxM matrix and either a Nx1 or 1xM one where it'd make sense, it repeats the other automatically.
22:15:20 <Bike> yeah ok there's got to be a better way for this bit.
22:15:43 <fizzie> In MATLAB, you'd either repmat() the vector, or write it with bsxfun() instead, which generally has better performance but then all operators get replaced with cryptic names.
22:15:59 -!- oerjan has quit (Quit: Nite).
22:16:19 <fizzie> A/repmat(...) vs bsxfun(@mrdivide, A, ...).
22:16:39 <fizzie> ("Matrix right divide".)
22:16:56 <Bike> TnKO = rand(...), [iTest, jTest] = find(TnKO <= TnFraction); TnKO = zeros(...), and then it does a for loop to set the ones that were less than the fraction to one.
22:18:07 <fizzie> TnKO = rand(...) <= TnFraction; ?
22:18:08 <Bike> ah. yes. it's TnKO <= TnFraction isn't it.
22:18:15 <Bike> that's just fantastic.
22:18:52 <Bike> maybe i should compile this stuff for the daily wtf. or have they set a policy against lab code by now
22:18:53 <fizzie> You'll end up with a type-'logical' matrix out of that, instead of a type-'double' one, which theoretically could hit some corner case. :p
22:19:46 <Bike> i bet a logical matrix takes up less memory?
22:19:59 <fizzie> http://sprunge.us/IidL yes.
22:20:16 <fizzie> One eighth of a 'double', assuming no sparsity.
22:21:16 <Bike> oh i haven't even gotten into the loop that does sparse(complicated constant matrix) every loop
22:23:14 <fizzie> The for i = ...; a(x(i),y(i)) = K; end antipattern is, I think, somewhere in the official docs as an example that can pretty much always be replaced with a(...) = K; kind of a thing. (Sometimes involving some sub2ind.)
22:23:41 <fizzie> a(sub2ind(size(a), x, y)) = K; I think.
22:23:42 <Bike> what's sub2ind again
22:23:58 <fizzie> Takes subscripts and returns "linear indices".
22:24:17 <Bike> meaning, like... what.
22:24:21 <Bike> row major index?
22:24:40 <fizzie> The indexing you get if you access a 2-or-more-dimensional matrix with a single subscript.
22:24:50 <fizzie> I don't remember which way it goes.
22:25:09 <fizzie> (It's possibly a better idea to use sub2ind/ind2sub to convert, anyway.)
22:25:22 <fizzie> (Probably it goes the same way as reshaping and everything else.)
22:29:01 <fizzie> There's some way of writing that sub2ind thing that involves two subscripts, but worse in terms of performance, and I can't think of it right now. (You can't just use two vectors as subscripts directly.)
22:29:42 <fizzie> I sure hope all this MATLAB knowledge will be useful some day.
22:30:24 <Bike> do you use it for your job or somethin
22:33:56 <olsner> is matlab big in natural language stuff research?
22:34:53 <fungot> olsner: eval ( log 2)? how would i know that one.
22:36:05 <Bike> fungot just ain't got a good head for numbers
22:36:06 <fungot> Bike: i don't think i have the wrong fnord loaded, so, it might
22:36:56 <fizzie> Bike: If you can call it a job.
22:38:21 <fizzie> olsner: It's big in the field of all kinds of audio things, including speech. For more NLP-y stuff, probably not so much.
22:38:46 -!- Nisstyre-laptop has quit (Quit: Leaving).
22:39:44 <Bike> that's a job, just a shitty one, as far as i can tell
22:40:13 <fizzie> It has most of the trappings of a job, that's for sure.
22:40:25 <kmc> except the money
22:40:52 <fizzie> There's money, just less of it.
22:41:16 <Bike> Have you considered taking a part time job coaching the football team
22:41:32 <fizzie> I don't think we have one.
22:41:53 <fizzie> Sports and universities don't really mix around here.
22:42:13 <Bike> what a country.
22:42:28 <kmc> in america universities are mostly an excuse to have football, and football is mostly an excuse to drink and have riots
22:42:50 <Bike> i've seen like four different professors with that map of the highest paid official by state, it's hilarious
22:42:52 <kmc> but I think the second half holds in Europe too, different meaning of "football" notwithstanding
22:42:57 <Bike> (spoiler they're mostly college football coaches)
22:43:08 <kmc> hey some of them are college basketball coaches!
22:43:11 <kmc> one or two hockey coaches
22:43:29 <kmc> hey does anyone here know anything about moderate-to-high-end cameras?
22:43:45 <fizzie> I mean, I guess the students have all kinds of sports-related organizations (they have clubs for all kinds of hobbies) but they're really very vaguely connected to the university itself.
22:44:02 <Bike> maybe when i'm a starving grad student i'll try to strike and establish an anarcho-syndicalist university.
22:44:20 <fizzie> Also many of the clubs *are* an excuse to drink, no matter what their subject matter.
22:44:22 <Bike> http://deadspin.com/infographic-is-your-states-highest-paid-employee-a-co-489635228 ah, here's the map. "You may have heard that the highest-paid employee in each state is usually the football coach at the largest state school. This is actually a gross mischaracterization: Sometimes it is the basketball coach."
22:44:33 <Bike> kmc, public policy writer
22:45:54 <Bike> huh none of them are hockey actually
22:46:07 <Bike> "Update, June 18: We've fixed one mistake in the map, which was pointed out to us by our friends at Harper's Magazine. The highest-paid state employee in New Hampshire is now the UNH president. It's no longer the hockey coach, as we'd originally indicated"
22:48:15 <Bike> i'm kind of curious about nevada. "med school plastic surgeon"? like, some professor that teaches how to do plastic surgery?
22:52:46 <Fiora> maybe it's a school that has an actual hospital like as part of the school?
22:52:49 <Fiora> and it's a surgeon who works there
22:53:12 <Bike> i suppose that's plausible, given the hospital a block away from me
22:58:53 <Bike> oh, here's his CV.
22:59:26 <Bike> "transparentnevada.com" lists his total pay and benefits at about 1.2 mil.
23:01:00 <Bike> current positions , "Professor and Chief, Division of Plastic Surgery", "Director, Microsurgery and Hyperbaric Research Laboratory", bla bla, president of a few things, af ew professorships, and chairman of the surgery department.
23:01:30 <Bike> oh jesus this CV is thirty pages long
23:03:22 <Bike> next in the list of highest paid is an associate professor, which is kind of awesome. then the clark county fire chief, which makes sense since nevada
23:04:31 <kmc> hey Bike http://nation.time.com/2013/10/16/washington-state-approves-new-rules-for-marijuana-industry/
23:04:46 <kmc> "The rules also aim to weed out large-scale marijuana sellers by limiting the number of licenses that anyone can hold to three."
23:04:58 <kmc> lol that's like the law for alcohol in massachusetts
23:05:05 <kmc> which I think was passed by the distributors' cartel
23:05:18 <Bike> HOW MYSTERIOUS
23:05:22 <kmc> so: large-scale sellers bad, large-scale distributors fine?
23:07:20 <Bike> what's the difference exactly
23:08:08 <kmc> in MA it's about the license to operate a retail liquor outlet
23:08:27 <kmc> i dunno what the law will be for weed distributors in washington (weedshington?)
23:08:42 <Bike> so a distributor is like... selling it to the sellers, or what, though
23:09:18 <kmc> and they want to deal with a bunch of separate liquor store companies rather than the big grocery chains
23:09:25 <kmc> because the latter could negotiate better terms
23:09:29 <kmc> also did you see http://www.thebohemianblog.com/2013/09/on-smoking-weed-in-north-korea.html ?
23:09:36 <kmc> not sure how much of this to believe tbh, but it's a pretty good story
23:09:45 <Bike> bullshit, i want weed at walmart
23:10:10 <kmc> there's some rumor that tobacco companies are already registering trademarks for weed-based products
23:10:30 <Bike> good to know the war on drugs is over then :/
23:10:57 <kmc> that reminds me of that scene at the beginning of _Layer Cake_
23:10:59 <kmc> decent movie
23:11:30 <fizzie> Bike: Did you win or lose?
23:11:42 <Bike> also weed in korea is just going to make me think of https://twitter.com/Vice_Is_Hip/status/389177827524112385 then
23:11:46 <Bike> fizzie: am i a drug
23:12:16 <fizzie> I... guess? Endorphins from Biking around, that sort of thing.
23:20:49 <Bike> kmc: this article isg reat
23:46:27 -!- yorick has quit (Remote host closed the connection).