00:02:24 -!- augur has quit (Remote host closed the connection).
00:06:29 <elliott> Doing it post-protected doesn't work either.
00:16:12 -!- cheater00 has joined.
00:19:07 -!- cheater- has quit (Ping timeout: 240 seconds).
00:22:38 <elliott> ok, lgdt _isn't_ ignoring the last byte :-/
00:27:13 <elliott> pikhq_: btw, it's actually *lidt [0]
00:27:32 <elliott> pikhq_: I bet you use AT&T syntax too.
00:28:59 <elliott> 00028083494e[CPU0 ] write_virtual_checks(): no write access to seg
00:29:00 <elliott> 00028083494e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x0d)
00:29:00 <elliott> 00028083494e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x08)
00:29:06 -!- Mathnerd314 has quit (Ping timeout: 240 seconds).
00:29:09 -!- BeholdMyGlory has joined.
00:30:16 <elliott> olsner: I'm beginning to think that this change-the-descriptor idea is really stupid.
00:32:11 -!- Behold has quit (Ping timeout: 240 seconds).
00:36:56 -!- cal153 has quit (Ping timeout: 264 seconds).
00:40:13 -!- Mathnerd314 has joined.
00:42:22 -!- oerjan has quit (Quit: Good night).
01:00:19 -!- Behold has joined.
01:04:19 -!- BeholdMyGlory has quit (Ping timeout: 272 seconds).
01:05:33 <elliott> [[When the 8042 recieves a valid scan code from the keyboard, it is converted to its set 1 equivalent. The converted scan code is then placed in the input buffer, the IBF (Input Buffer Full) flag is set, and IRQ 1 is asserted. Furthermore, when any byte is received from the keyboard, the 8042 inhibits further reception (by pulling the "Clock" line low), so no other scan codes will be received until the input buffer is emptied.
01:05:34 <elliott> If enabled, IRQ 1 will activate the keyboard driver, pointed to by interrupt vector 0x09. The driver reads the scan code from port 0x60, which causes the 8042 to de-assert IRQ 1 and reset the IBF flag. The scan code is then processed by the driver, which responds to special key combinations and updates an area of the system RAM reserved for keyboard input.
01:05:35 <elliott> If you don't want to patch into interrupt 0x09, you may poll the keyboard controller for input. This is accomplished by disabling the 8042's IBF Interrupt and polling the IBF flag. This flag is set (1) when data is available in the input buffer, and is cleared (0) when data is read from the input buffer. Reading the input buffer is accomplished by reading from port 0x60, and the IBF flag is at port 0x64, bit 1. The following assembly code illus
01:05:42 <elliott> hmm, does that imply that i have to specifically tell the keyboard to not do the interrupt method first?
01:08:47 <elliott> 00014041760i[CPU0 ] WARNING: HLT instruction with IF=0!
01:09:13 -!- cal153 has joined.
01:17:50 <elliott> hmm, maybe the keyboard thing causes a hlt in bochs
01:47:29 -!- augur has joined.
01:50:35 <Ilari> Nope, something calls HLT. And when it is called, interrupts are better to be enabled.
01:51:29 <elliott> Ilari: Nope, I did not execute hlt.
01:53:02 <elliott> So I bet the keyboard controller does it ... or something ... it's when I talk to the keyboard, anyway.
01:53:03 <Ilari> HLT somewhere in BIOS?
01:53:10 <elliott> Ilari: Nope, no BIOS calls at this point either.
01:53:17 <elliott> Lemme find the exact instruction where it happens.
01:53:26 <Ilari> Jump to hyperspace?
01:53:55 <Sgeo> Darnit, catch fire, elliott!
01:54:01 <elliott> I actually do hlt later on X-D
01:54:08 <elliott> Okay, so it's getting input from the keyboard even though there are none.
01:54:39 <Ilari> The code jumps somewhere far far away that isn't even code.
01:55:03 <elliott> Mm, seems not. Wait what ...
01:55:19 <Ilari> If there is paging enabled, usually to non-paged area (crashing instantly). But in real mode nearly every address is executable.
01:55:23 <elliott> In bytes, [0xB8000] = first char, [0xB8001] = first attribute, right?
01:55:33 <elliott> Ilari: In fact I have flat protected mode without paging.
01:56:25 <elliott> Heh, okay, so characters are actually coming through.
01:56:42 <Ilari> I think the proper way to use HLT is to first to CLI, check conditions, then STI followed _immediately_ by HLT.
01:57:44 <Ilari> Because IIRC, immediately after STI is executed, interrupts are not checked (and that would be actually important here).
01:58:40 <Ilari> Because if you check interrupts immediately before HLT, you have a race condition.
01:58:51 <elliott> Ilari: When I did that (sti hlt; I disabled interrupts way before) it actually rebooted, so I'm assuming that an interrupt came in somehow.
01:59:03 <elliott> Anyway, polling the keyboard controller itself seems like it might work.
01:59:15 <elliott> Ilari: Of course the _real_ problem will be fitting this entire Forth system into 512 bytes!
02:00:03 <elliott> (basically, boot up -> flat protected mode -> Forth compiler with very few built-in words -> type in a word, press space, it executes, then "ok" prints and you're on to the next line)
02:00:10 <elliott> In 510 bytes (2 for signature)...
02:00:47 <Ilari> There are three ways to initiate a reboot: 1) Jump to BIOS init vector in real mode (FFFF:0000). 2) Toggle the reset line (connected to keyboard controller, what else)? or 3) Triplefault the CPU.
02:01:25 <Sgeo> Is Ilari ignoring me too, or just not responding to me?
02:01:53 <Ilari> Sgeo: The only ignore I have set in this client is global CTCP ignore.
02:03:04 <elliott> Also, I believe the IBM PC design philosophy goes something like "How should we-" "Attach it to the keyboard controller."
02:03:08 <elliott> "But-" "KEYBOARD CONTROLLER!"
02:03:26 <Ilari> A20 line is also controlled by the keyboard controller.
02:03:50 <Ilari> I think CTCP is a special form of /msg.
02:03:58 <elliott> Ilari: So you cannot see /mes?
02:04:31 <Ilari> Well, whatever CTCP block in this client actually blocks (it doesn't block /me)
02:04:41 -!- Behold has quit (Remote host closed the connection).
02:04:47 <elliott> Ilari: Blocks VERSION requests, at least :P
02:05:07 <elliott> But but but HOW WILL WE EVER KNOW WHAT CLIENT YOU USE ;_;
02:06:40 <elliott> The secret code of Gregor's race reached its fifth revision today.
02:06:46 <Ilari> I configured that when some jokers did CTCP spamming to some channels (that also got chanmode +C implemented).
02:06:57 <elliott> "Gah. They totally ignored my bug report."
02:06:59 <elliott> [[It's the mysterious U+1F4A0 DIAMOND SHAPE WITH A DOT INSIDE code point, which is even more mysteriously listed as meaning "cute".
02:06:59 <elliott> It turns out it is inherited from a proprietary Japanese text messaging encoding, where it is indeed listed as meaning "cute", and represented by a tiny pixelated image, which if you look quickly at it might look a little like a "diamond shape with a dot inside".
02:06:59 <elliott> However, it is actually a crude picture of a flower.]]
02:07:44 <elliott> Gregor: FINALLY WE HAVE "PILE OF POO".
02:09:00 <Gregor> Why can't I change my skin ...
02:09:53 <elliott> Gregor: Dude, I have a skin, skins just don't load properly right now.
02:11:02 <elliott> Gregor: ... also, wrong channel :P
02:11:39 <Sgeo> To the diamond with dot inside thing?
02:26:36 -!- TLUL has quit (Quit: *disappears in a puff of orange smoke*).
02:31:23 -!- wareya has joined.
02:35:01 -!- Alyk_Meigatzroyd has quit (Ping timeout: 276 seconds).
02:50:14 <elliott> but i forgot to ask him what i was going to
02:50:15 <Gregor> DAMN YOU SOMALIAAAAAAAAAAAAAAAAAAAAAAAA
02:52:51 -!- GreaseMonkey has joined.
02:54:30 <pikhq_> elliott: Actually, IBM PC design philosophy is based around "We want this functionality. *God dammit*, the keyboard controller is the only place to add it."
02:54:40 <elliott> pikhq_: I think they secretly liked it.
02:55:40 <pikhq_> elliott: For instance, the A20 line on the keyboard controller is there because there was no way to implement the address wrapping you'd expect on an IBM PC on a 386, except to turn off the A20 line...
02:56:25 <pikhq_> And the only way to be able to *control* that without adding an expensive microcontroller was to use the only microcontroller on the motherboard which wasn't fully used: the keyboard controller.
02:58:13 <Ilari> Actually, A20 appeared on the 286.
03:01:04 <Ilari> (286 had 24 address lines)
03:02:56 <Gregor> elliott: DAMN YOU SOMALIAAAAAAAAAAAAAAAAAAAAAAAA
03:10:42 <pikhq_> Still. *That* brokenness is working around Intel's failed backwards-compatibility.
03:11:15 <elliott> Forth in 510 bytes may not be possible without significantly more juju than I possess.
03:11:59 <Ilari> So it would need to load more sectors from floppy? :-(
03:12:46 <elliott> Ilari: That may be so. I haven't started the actual Forth part yet.
03:12:58 <elliott> I'm just thinking that all my protected mode stuff is already 51 bytes or so.
03:13:06 <elliott> Keyboard handling will probably take about 30 at least.
03:13:21 <elliott> Then I have to fit the compiler in, and enough basic words to be able to actually do anything...
03:13:30 <elliott> Not that I'm going to give up.
03:13:31 <pikhq_> How small could you make a compressor? >:D
03:13:37 <elliott> pikhq_: Definitely not small enough.
03:13:39 <pikhq_> Well, decompressor, really.
03:13:47 <elliott> It's not like x86 machine code would RLE well.
03:14:05 <pikhq_> Yeah, so it'd have to be a vaguely intelligent compression scheme.
03:15:03 <pikhq_> You have to admit that it'd be pretty awesome to have a self-extracting boot sector.
03:15:12 <Ilari> Heh. Boot code that would read the entiere floppy into extended memory, switch to pmode and then jump to the image. :-/
03:16:05 <pikhq_> elliott: Hmm. Pre-Forth threaded code?
03:16:19 <elliott> pikhq_: That's... assembly that calls a lot of functions.
03:16:49 <elliott> pikhq_: I'm reading the free sample chapter of RTK 1 right now, btw.
03:16:55 <elliott> And wondering whether to learn the kana first or after.
03:17:21 <pikhq_> Only really a gain if you use each sequence of threaded code more than once, though.
03:18:03 <elliott> pikhq_: "1 is 1 on its side. 2 is two 1s. 3 is three 1s. 4 is "mouth" plus "human legs"."
03:18:25 <pikhq_> elliott: Mnemonics ≠ etymology.
03:18:35 <pikhq_> Also, all this is from China.
03:18:47 <elliott> pikhq_: I don't wanna learn Japanese and then see 4 and think "MOUTHS AND LEGS".
03:18:58 <pikhq_> elliott: You won't. Not for long, anyways.
03:19:02 <elliott> What's "hello", "mosquito" + "hell" + "burn" + "unicorn" + "concrete" + "smiles"?
03:19:21 <elliott> Haha it even looks like a mouth devouring a pair of innocent legs. This is going to be bad for my sanity.
03:19:30 <elliott> pikhq_: It's assuming I know any kanji at all; so not fair.
03:20:39 <pikhq_> Well, if you knew any, they'd be, at a minimum, 一二三四五六七八九十.
03:21:01 <elliott> "6: The primitives here are top hat and animal legs."
03:21:06 <elliott> HOW DO YOU INVENT TOP HATS BEFORE THE NUMBER "SIX"
03:21:18 <pikhq_> elliott: THAT'S NOT THE BLOODY ETYMOLOGY.
03:21:35 <elliott> "And, yes, this part should be six inches high" "Sorry, what?" "Six in-" "'S...ix'?"
03:21:52 <pikhq_> elliott: Also, the names for the kanjoids are largely Heisig inventions.
03:21:54 <elliott> "We can make top hats five inches high, and seven inches high, what is this 'six'?!?!?!?!?!"
03:22:43 <pikhq_> ("mouth" and "legs" aren't, though considering 四 as composed of them is.)
03:22:53 <elliott> I like how 7 and spoon would be confusable with bad handwriting.
03:24:05 <elliott> For the Japanese, "8" is "infinity".
03:24:11 <elliott> ME NO SO GOOD WITH NUMBERS
03:24:23 <pikhq_> Not really; they tend to use Arabic numerals more than Chinese.
03:24:45 <pikhq_> Also, 見 and 貝 are a much better example of "confusable with bad handwriting".
03:25:23 <elliott> "As a primitive, we shall use this kanji to mean baseball team or simply baseball. The meaning, of course, is derived from the nine players who make up a team."
03:25:52 <pikhq_> elliott: Actually, baseball is more popular in Japan than in America.
03:28:14 <elliott> So wait, is that what it actually means, or just the mnemonic?
03:28:18 <elliott> If the latter, I prefer "waffle" :P
03:28:23 <pikhq_> It actually means rice field.
03:29:12 <elliott> pikhq_: But I want it to mean waffle instead.
03:29:19 <elliott> Also, how often does THAT get used alone nowadays :P
03:29:38 <pikhq_> elliott: Extensively in names.
03:30:08 <elliott> Agh, I'm really worried that learning a graveyard tombstone image for "old" will end up turning my mind into a slurry of ridiculous associated visions when processing words :P
03:30:28 <elliott> "five mouths: 2 nostrils, 2 ears, and 1 mouth."
03:30:30 <elliott> Great job at counting, bro.
03:30:36 <elliott> (Don't correct me, I'm just mocking.)
03:30:55 <elliott> Zwaarddijk: Because it sounds like fun. (Whatcha got against Japanese?)
03:31:12 <Zwaarddijk> why learn japanese when the world has fantastic things like nahuatl to offer
03:31:21 <elliott> And because learning another completely different writing system and a language sounds like more fun than just learning another language with a similar alphabet.
03:31:37 <elliott> Zwaarddijk: Because I'm a mainstream loser.
03:31:44 <Zwaarddijk> (I've tried learning nahuatl. I can't parse it. the grammar entirely evades me. it's like you take the model of the world that Finnish has and invert it)
03:32:05 <pikhq_> Oh, it's that strongly synthetic?
03:32:20 <elliott> pikhq_: Wait... how is "I" made up of four mouths?
03:32:28 <elliott> I see one mouth, and then a weird shape on top that maybe has most of one mouth in it.
03:32:45 <pikhq_> elliott: Which number?
03:32:48 <Zwaarddijk> the inflections have roles that are like the complete opposite of what Finnish does
03:32:55 <elliott> pikhq_: It's not an actual numeral if that is what you mean
03:33:18 <pikhq_> elliott: Uh, "five mouths". It has 五 and 口.
03:33:32 <elliott> pikhq_: OH, literally "five" + mouths.
03:33:39 <elliott> Not "five compositions of the 'mouth' character".
03:34:08 <pikhq_> Though 品 is three compositions thereof. :P
03:34:11 <elliott> At least I'm decomposing things that an hour ago would have looked atomic to me, even if they are simple...
03:34:21 <elliott> pikhq_: What does that mean, "gargantuan Cthulhuesque horror from hell"?
03:34:32 <pikhq_> It's on the next page.
03:35:16 <elliott> Great... "month" can mean "flesh". Apparently Heisig won't tell me why until a later chapter.
03:35:32 <elliott> pikhq_: SO KANA FIRST OR KANA LATER
03:35:48 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAA
03:36:12 <pikhq_> It'll be absurdly easy later, and somewhat annoying now.
03:36:41 <pikhq_> The kana, in a very technical sense, are kanji, you see.
03:37:02 <elliott> pikhq_: How many Bible references do I have to plan for, I realise this guy is a philosopher of religion but that's two in a row :P
03:37:14 <pikhq_> elliott: Uh, I think there's maybe 3 or 4 more?
03:37:31 <elliott> "Bright = sun + moon" is more clearly explained as "The sun and the moon are the two bright things in the sky." than that reference to me :-P
03:37:48 <pikhq_> Feel free to ignore his suggestions if you have something better.
03:38:22 <elliott> "And if you've ever held a diamond up to the light," NOT ALL OF US ARE AS RICH AS YOU, HEISIGGY
03:38:41 <elliott> tl;dr three suns would sparkle a lot because your eyes would burn.
03:41:53 -!- augur has quit (Remote host closed the connection).
03:41:55 <elliott> pikhq_: Does "frame N" refer to the kanji-box marked N?
03:42:25 <pikhq_> elliott: Uh, it should, yeah.
03:42:51 <pikhq_> I'm... Not sure on the reasoning for that either.
03:43:14 <pikhq_> It has nothing to do with the phrase "the land of the rising sun", BTW.
03:43:32 <elliott> Ehh, Heisig says it's that kanji which is the nickname for the flag.
03:43:38 <elliott> Which would imply to me that it _does_ have something to do with that phrase.
03:44:25 <elliott> "This character is a sort of nickname for the Japanese flag with its well-known emblem of the rising run."
03:45:12 <elliott> "Nightbreak" is an elegant character.
03:45:29 <elliott> I also kind of worry that I'll end up decomposing kanji into the _English_ mnemonics, which would impede my fluency... but I suppose I will grow out of that, so to speak.
03:46:04 <elliott> Gall bladder = part of body + nightbreak. Not even gonna try and understand that one.
03:46:11 <pikhq_> Oh, sure enough, 旭 is the first bit in 旭旗, which refers to the 日の丸.
03:46:33 <elliott> pikhq_: "Sun rising land", by any chance?
03:46:39 <elliott> At least, I recognise sun.
03:46:49 <elliott> And that last one looks like... that one... 9? Plus an extra line.
03:46:53 <pikhq_> elliott: No, that's "the circle of the sun".
03:47:01 <pikhq_> Which is the usual name for the Japanese flag.
03:47:09 <elliott> What is the last character, and was my guess of 9 + an extra line (on the curvy swoop) correct?
03:47:17 <elliott> (I forget 9 already. But hey, this is "progress".0
03:47:24 <pikhq_> 丸 is "round", and it is 9 + a stroke.
03:47:41 <pikhq_> "The land of the rising sun" is a poetic translation of 日本国, the official name for the country of Japan.
03:48:07 <elliott> What is the literal translation?
03:48:19 <pikhq_> "The root of the sun".
03:48:31 <pikhq_> Well, "The country of the root of the sun."
03:49:04 <pikhq_> Quite obviously coming from the Chinese seeing the sun rise in the east, out towards Japan.
03:50:18 <pikhq_> "Japan" itself is 日本 after traversing through a variety of languages.
03:51:06 <elliott> pikhq_: What does /that/ translate to?
03:53:39 <elliott> Welp, I suppose I should start writing these things down in a day or two when I buy it and start properly.
03:53:46 <elliott> My handwriting is atrocious though.
03:54:12 -!- elliott has quit (Quit: Leaving).
03:54:45 <pikhq_> Don't anyone tell elliott this, but there's a lot of atrocious handwriting in Japan.
03:55:17 <Sgeo> Feel free to tell clog to tell him
04:09:45 -!- augur has joined.
04:11:13 -!- aloril has quit (Ping timeout: 264 seconds).
04:21:47 -!- azaq23 has quit (Ping timeout: 240 seconds).
04:23:02 -!- azaq23 has joined.
04:23:27 -!- aloril has joined.
04:27:59 -!- aloril has quit (Ping timeout: 250 seconds).
04:41:21 -!- aloril has joined.
04:41:28 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAAAAAAA WHYYYYYYYYYYYYYYYYYYY
04:48:50 <pikhq_> Gregor: What's this about Somalia, anyways?
04:49:02 <Gregor> pikhq_: You'll see SOON ENOUGH
04:49:12 <Gregor> Any time within the last 24 hours lololol DAMN YOU SOMALIA
04:49:50 * pikhq_ concludes that Somalia has gotten a fucking government, making it less valuable in thwarting anarchists
04:50:13 <Gregor> Fucking government: The best kind of government?
04:51:13 <pikhq_> I approve of government-by-orgy.
04:59:47 <pikhq_> Nay, I heartily applaud it.
05:15:42 -!- asiekierka has joined.
05:42:09 -!- asiekierka has quit (Ping timeout: 240 seconds).
05:43:01 -!- wareya has quit (Remote host closed the connection).
05:49:47 -!- wareya has joined.
06:15:15 -!- augur has quit (Ping timeout: 276 seconds).
06:15:47 -!- augur has joined.
06:46:26 -!- GreaseMonkey has quit (Quit: The Other Game).
06:47:05 -!- augur has quit (Ping timeout: 276 seconds).
06:48:49 -!- augur has joined.
07:42:14 -!- oerjan has joined.
07:44:29 -!- cheater00 has quit (Ping timeout: 240 seconds).
07:45:52 -!- cheater00 has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:14:44 -!- azaq23 has quit (Quit: Leaving.).
08:15:47 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.16/20101130074636]).
08:22:24 -!- wetneb has joined.
08:27:41 -!- copumpkin has quit (Ping timeout: 264 seconds).
08:28:21 -!- copumpkin has joined.
08:39:56 -!- wareya_ has joined.
08:39:57 -!- wareya_ has quit (Client Quit).
08:58:51 -!- wetneb has quit (Quit: Leaving.).
09:35:52 -!- MigoMipo has joined.
09:45:19 <quintopia> if Kalastajat is fishermen, what is Kalastajatorppa?
09:45:43 <oerjan> `translatefromto fi en torppa
09:46:36 <fizzie> 1. croft -- (a small farm worked by a crofter)
09:46:42 <fizzie> 1. A fenced piece of land, usually small and arable and with a crofter's dwelling thereon.
09:47:10 <fizzie> a. A piece of enclosed ground, used for tillage or pasture: in most localities a small piece of arable land adjacent to a house.
09:47:21 <fizzie> 2. A small agricultural holding worked by a peasant tenant; esp. that of a crofter n.1 in the Highlands and Islands of Scotland (see quot. 1851).
09:47:27 <quintopia> so a Kalastajatorppa is a fish farm?
09:48:11 <fizzie> Well, to me it evokes the image of a sort of rustic cottage-style building owned by a fisherman. (I mean, if we're going literally.)
09:48:24 <fizzie> (It's also a relatively expensive restaurant.)
09:49:17 <fizzie> The name is, I think, a bit whimsical.
09:49:45 <fizzie> "Kalastajatorppa (Fisherman’s cottage) park in 1915" translates a Wikipedia article about the region.
09:50:13 <fizzie> "Hotel Kalastajatorppa (Fisherman’s cottage), Kalastajartorpantie 1 and 2-4. The new part is from 1975 and is planned by Einari Teräsvirta. The old congress part is built in 1937 and 1939, architect Jarl Eklund. The original crofter’s cottage that gave the hotel its name was demolished in 1936, but a local famous basketball team established in 1932 still has its name after that - Torpan Pojat, which means cottage boys in English."
10:06:49 -!- cheater00 has quit (Ping timeout: 240 seconds).
10:08:08 -!- cheater00 has joined.
10:14:18 -!- pikhq has joined.
10:14:59 -!- pikhq_ has quit (Ping timeout: 276 seconds).
11:43:31 -!- ais523 has joined.
11:47:21 -!- ais523 has quit (Excess Flood).
11:47:37 -!- ais523 has joined.
12:01:07 <oerjan> ais523: http://oerjan.nvg.org/esoteric/underload/minskyconv.ul
12:01:42 <ais523> wow, it is possible after all?
12:01:47 <ais523> that language deserves a name of its own
12:02:12 <oerjan> http://oerjan.nvg.org/esoteric/underload/ULDisclaim2.hs
12:04:05 <ais523> it might even be worth a research paper
12:04:09 <ais523> heh, and I just noticed the topic
12:05:13 <oerjan> ...not that i've actually checked for previous work, mind you
12:05:29 <ais523> is it exactly likely in this case?
12:05:36 <ais523> although I haven't checked either
12:07:04 <ais523> I'm not even sure what the search terms should be
12:08:11 <ais523> there isn't anything relevant on Google Scholar for "underload programming language", but I'd have been massively surprised if there were
12:12:23 <oerjan> "Much of the original work on concatenative language theory was carried out by Manfred von Thun.
12:23:44 -!- pikhq_ has joined.
12:23:58 -!- pikhq has quit (Ping timeout: 264 seconds).
12:29:57 -!- aloril has quit (Ping timeout: 252 seconds).
12:33:28 -!- BeholdMyGlory has joined.
12:39:28 -!- augur has quit (Remote host closed the connection).
12:40:43 <ais523> oerjan: can you explain how your !-replacement works?
12:43:13 -!- aloril has joined.
12:45:09 <oerjan> like with the ~:!()^ case, it is a mechanism for turning a program into an equivalent one which can be run in either such a way as to emulate the original program or to delete itself.
12:45:44 <oerjan> however without ~ it gets tricky to actually give the program an argument to test for what to do
12:45:56 <ais523> yep, I can see how that would be problematic
12:46:21 <ais523> because the only way to delete code is to get it to run in such a way as to cancel itself out, as ^ is the only data-deleting command
12:49:06 <oerjan> however inspired by the :!()^ minsky automaton, we do a change of perspective: : can be seen as a command that gives the element on the top of the stack an extra argument, namely _itself_.
12:50:25 <oerjan> the next clue is to consider when the top of the stack is of the form (^P). then you get to run P either once or twice by using ^ or :^, respectively.
12:51:12 <ais523> by adding two copies to the program if it's :^
12:52:40 <oerjan> now consider a program R containing only : or ^, and not ending with : (R could be empty)
12:54:05 <oerjan> now construct L as follows: for every block of :'s followed by ^ in R, L contains ()(^), for every single ^ L contains (), in the reverse order of the corresponding R parts
12:54:28 <oerjan> then we can see that LR is equivalent to the null program.
12:54:53 <ais523> and LRR is thus equivalent to R
12:55:40 <oerjan> incidentally whenever LR is equivalent to the null program, R is equivalent to something of that :^ form. (L might not be just ()'s and (^)'s though)
12:56:07 <oerjan> (this can be seen by reducing all quotes in R away.)
12:57:08 -!- Sgeo has quit (Ping timeout: 240 seconds).
12:57:24 <oerjan> anyway, {R} = L()(^R) is now a program such that {R}^ is null and {R}:^ is R
12:58:11 <oerjan> we should notice that ^ and :^ are _themselves_ of this R form. this means it will be convenient to use such conditionals to control each other.
12:59:22 <ais523> and the rest is just trying to get your existing Minksy machine into that form?
13:00:05 <oerjan> not quite. we also need to handle quotes, which certainly cannot be R's.
13:01:14 <oerjan> so consider the more complicated case of a program P that cannot be canceled only by something on the left, but which can be canceled by something on the left and right _combined_. i.e. LPR is null for some L and R.
13:01:57 -!- Mathnerd314 has quit (Ping timeout: 264 seconds).
13:02:18 <oerjan> also let L'R be null (we could make L' = LP but P may be much larger than necessary for this.)
13:02:55 <oerjan> then we can define {P} = LR'()(^RP)
13:03:28 <oerjan> and once again, {P}:^ is P
13:03:31 -!- cheater00 has quit (Ping timeout: 240 seconds).
13:03:40 <ais523> you haven't defined R'
13:03:56 <oerjan> *then we can define {P} = LL'()(^RP)
13:04:22 <oerjan> got a bit confused by L' being a function of R, more or less
13:05:07 -!- cheater00 has joined.
13:05:08 <ais523> so executing it with ^, we get LL'RP = LP, and with :^, we get LL'RPRP = LPRP = P?
13:05:38 <oerjan> note that for this complicated case we don't have a unique command to delete
13:06:01 <ais523> hmm... but we can delete LP by running R, and thus L'RR
13:06:12 <ais523> and we can run L'R in order to not delete anything
13:06:38 <ais523> ah, just thinking out loud
13:06:47 <ais523> I don't think that method would work directly
13:08:23 <oerjan> now this variability of ^R is a bit of a problem, we need a common API so to speak for things that are put on the stack
13:08:44 <oerjan> but as i said, these conditionals can control each other.
13:09:26 <ais523> ah, I see, you're using ^R to delete, and :^ to run
13:09:32 <ais523> and the issue is that ^R is not always the same thing
13:09:54 * oerjan needs a new type of bracket
13:11:13 -!- nddrylliog has joined.
13:11:26 <oerjan> since ^R is something left cancelable (by L'()), we know that {^R} is deleted by ^. this means that <P>^:^ is P
13:12:07 <ais523> surely you need L' to actually be there, in order for ^ to delete it?
13:12:46 <ais523> as in, merely being left-cancellable isn't enough for something to be deleted by ^, you need the actual left-canceller to cancel it
13:12:47 <oerjan> by ^R being left cancelable by L'() i mean that L'()^R is null
13:13:09 <ais523> oh, I missed the braces around {^R}
13:13:33 <ais523> so {^R}^ = null, {^R}:^ = ^R?
13:15:25 <oerjan> this will be the "API" for running/deleting programs of the kind that are put on the stack.
13:15:36 <oerjan> and inside translated quotes
13:15:36 -!- augur has joined.
13:15:39 <ais523> so you now have ^:^ and :^ as a second set of booleans, which unfortunately aren't left-cancellable
13:16:11 <ais523> it's ending with : that makes something uncancellable
13:17:34 <oerjan> as i mentioned above, whenever LR is null, R is automatically equivalent to these left cancelable things
13:18:13 <oerjan> (or alternatively, my haskell program constructs cancelable blocks with R's of this form, anyway)
13:19:08 <ais523> that reminds me a bit of my 2,3 proof, actually; it has cancelling-out constructions as well
13:19:28 <ais523> where it arranges sections of tape to cancel each other's effects out (although only for a finitely long length of time)
13:22:09 <oerjan> we now have the API settled, so we can translate individual commands: ! -> ^:^, ^ -> ^^:^ and : -> :
13:26:05 <oerjan> now assuming we have recursively translated nested quoted programs, the quotes in the program will be cancelable solely from the right.
13:27:29 <oerjan> ^ is cancelable from the left, as is :'s followed by ^. a : alone needs a combination.
13:27:31 -!- nddrylliog has quit (Quit: used jmIrc).
13:28:30 <oerjan> so we now split up a subprogram (with quotes already handled recursively) into blocks where each block is cancelable.
13:30:10 <oerjan> P = QP1P2..Pn. QQ', LiPiRi being null.
13:30:43 <oerjan> (the special Q is just for a bit of efficiency.)
13:31:33 <Ilari> Haha: "He also said that they understand that IPv6 is not ready because OS vendors have not implemented IPsec. Talk about a bozo filter..." (from NANOG list).
13:32:50 <oerjan> and now we define <P> = Q{P1}{:^P2}..{:^Pn}{^Rn()(^)...^R2()(^)^R1Q'}
13:32:59 -!- augur has quit (Remote host closed the connection).
13:33:27 <oerjan> (n should be >= 1 btw, we can add a dummy P1 if needed.)
13:34:47 <oerjan> and this still obeys the same API as for a cancelable P
13:35:34 <oerjan> *and now we define <P> = Q{P1}{:^P2}..{:^Pn}{^Rn...^R2^R1Q'}
13:36:01 <oerjan> erroneously mixing in an L-part there :)
13:36:41 <oerjan> ais523: and that's about it, i think
13:36:51 <ais523> yep, that seems a pretty good description
13:37:27 <ais523> it's both logical and beautifully non-obvious, and reveals something that seems moderately profound about programming
13:41:28 <oerjan> each Pi block constructed by my program tends to be of the form :^::^^:(...)...(...), with : and ^ (and sometimes S) first and quotes last
13:42:29 <oerjan> the quotes and half a final : cancel from the right, the ^'s and most of the :'s from the left
13:43:41 <oerjan> oh and Q is just quotes, thus the name :D
13:43:51 <ais523> well, I think that's provably the minimum Underload minimization
13:44:11 <ais523> to go any lower/different, you'd need to change some of the commands
13:44:27 <oerjan> yeah you should ask elliott about that, he had some suggestions
13:48:38 -!- Deewiant has quit (Quit: Viivan loppu.).
13:57:42 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAAAAA
14:22:49 -!- oerjan has quit (Quit: leaving).
14:23:58 <Ilari> Trying to figure out how to construct turing tape block in Pointer-B. What it needs to do is fairly simple. Coding it is another matter since it involves loading a number of numbers (oh, and the numbers needed depend on the amount of code loading other numbers, and sometimes also the number itself takes).
14:24:55 -!- augur has joined.
14:28:23 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
14:37:01 -!- asiekierka has joined.
14:43:19 -!- azaq23 has joined.
14:52:09 -!- copumpkin has joined.
14:57:23 -!- Phantom_Hoover has joined.
14:59:28 -!- cheater00 has quit (Ping timeout: 240 seconds).
15:01:37 -!- cheater00 has joined.
15:06:39 -!- azaq23 has quit (Quit: Leaving.).
15:09:31 <fizzie> In the "selected Unicode character names" series today: U+1F192 SQUARED COOL.
15:09:38 <fizzie> (It is, in fact, just the word "COOL" in a box.)
15:11:59 -!- pumpkin has joined.
15:13:50 -!- copumpkin has quit (Ping timeout: 276 seconds).
15:17:29 -!- cheater00 has quit (Ping timeout: 264 seconds).
15:20:19 -!- Sgeo has joined.
15:20:20 <quintopia> woah. when did the topic thing happen?
15:20:57 <fizzie> No, the next char is U+1F193 SQUARED FREE.
15:21:22 -!- ais523 has quit (Remote host closed the connection).
15:23:28 <fizzie> There's SQUARED CL, COOL, FREE, ID, NEW, NG, OK, SOS, UP WITH EXCLAMATION MARK and VS.
15:24:08 <Gregor> fizzie: Not as good as LOVE HOTEL
15:24:12 <fizzie> And SQUARE DJ, which is just DJ without the enclosing square.
15:31:13 -!- cheater00 has joined.
15:31:55 -!- copumpkin has joined.
15:33:24 -!- pumpkin has quit (Ping timeout: 252 seconds).
15:34:08 -!- pikhq has joined.
15:34:43 -!- pikhq_ has quit (Ping timeout: 272 seconds).
15:39:04 -!- augur has quit (Ping timeout: 264 seconds).
15:43:36 -!- Mathnerd314 has joined.
15:47:02 -!- cheater00 has quit (Ping timeout: 240 seconds).
15:48:18 -!- cheater00 has joined.
15:51:00 -!- copumpkin has quit (Ping timeout: 250 seconds).
15:52:31 -!- ais523 has joined.
15:56:16 -!- haze_skw has joined.
15:59:03 -!- copumpkin has joined.
15:59:03 -!- copumpkin has quit (Changing host).
15:59:03 -!- copumpkin has joined.
16:00:33 -!- haze_skw has left (?).
16:01:26 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
16:01:27 -!- Phantom__Hoover has joined.
16:08:43 -!- copumpkin has quit (Ping timeout: 260 seconds).
16:13:23 <ais523> I don't see how you plan to do S, Eodermdrome is very bad at copying things
16:14:03 <ais523> I mean, it's /possible/ due to being TC, bu doing it directly would be difficult
16:14:44 <ais523> the obvious way to copy data in Eodermdrome would be to unzip it DNA-style
16:14:52 <ais523> which might be quite interesting, actually
16:15:18 <ais523> do you know how DNA is copied?
16:15:31 <ais523> most proteins in human cells have the same localish view of reality that Eodermdrome does
16:15:46 <ais523> and no, it isn't, e.g. Minksy machines don't; but S copies things, or else shares them
16:15:49 <ais523> and sharing would be messy too
16:16:09 <ais523> trying to work out which line was which
16:16:15 <ais523> it's the basic problem of eodermdrome
16:16:50 <ais523> trying to work out which connection is which
16:17:03 <ais523> as nothing is labeled graph shapes as labels
16:17:37 <ais523> I think my client is screwing up again (due to stray touches on touchpads)
16:20:24 -!- copumpkin has joined.
16:25:20 -!- ais523 has quit (Remote host closed the connection).
16:25:25 -!- pumpkin has joined.
16:27:12 -!- copumpkin has quit (Ping timeout: 240 seconds).
16:28:00 * Phantom__Hoover wants to redirect DNA-Sharp to MONOD, but suspects that this would not go down well with ais.
16:28:20 -!- Slo912 has joined.
16:34:29 <Gregor> Why do we immediately jump all over anybody who joins :P
16:37:13 -!- Vorpal has quit (Quit: ZNC - http://znc.sourceforge.net).
16:42:20 -!- Slo912 has quit (Quit: Slo912).
16:45:11 -!- Vorpal has joined.
16:51:54 <Vorpal> Phantom__Hoover, err what? (note I have no scrollback. I just rebooted after a kernel upgrade)
16:52:36 -!- Sgeo has quit (Ping timeout: 240 seconds).
16:53:08 <Vorpal> Phantom__Hoover, where is elliott's logging bot btw?
16:55:29 <Vorpal> I'm now on fancy 2.6.37.2 btw
16:55:47 <Vorpal> still seems like the same old world
16:55:48 -!- Deewiant has joined.
16:56:13 <Vorpal> most importantly... Still no flying cars.
17:02:54 -!- augur has joined.
17:06:46 <olsner> hmm, elliott disappeared
17:12:11 <Vorpal> how strange. radvd randomly unconfigured itself. As in the config file was wrong.
17:12:17 <Vorpal> well time to track that down...
17:13:22 <Vorpal> (good thing I keep /etc in version control!)
17:14:04 <Vorpal> the package moved from aur to main distro repos. And the config file location changed. Right.
17:19:20 -!- ais523 has joined.
17:25:03 -!- oerjan has joined.
17:27:27 <oerjan> <quintopia> woah. when did the topic thing happen?
17:27:51 <oerjan> i've been working on it for about the past week
17:28:39 <oerjan> and yesterday my conversion program was finished, so i could test the theory
17:29:04 -!- wetneb has joined.
17:29:23 <oerjan> wet nebs, the best kind of nebs
17:29:58 <Ilari> APNIC down 0.04: 8k to Bangladesh, 512k+2k+1k to Japan, 2x64k+2x1k+2x256 to Australia, 64k+2k to Indonesia, 4k to China. No IPv6 allocations.
17:30:19 <ais523> Ilari: do you think APNIC's going to run out first? RIPE and ARIN have relatively scary burn rates too
17:31:32 <oerjan> <ais523> and sharing would be messy too
17:31:44 <oerjan> hm it might still be more efficient than copying...
17:32:32 <oerjan> might leak memory, i imagine
17:32:59 <Gregor> Non-leakiness is not a requirement for TC :P
17:33:02 <ais523> yep, you'd get detached bits of graphs
17:33:08 <ais523> Gregor: we know Eodermdrome's TC anyway
17:33:41 <Ilari> ais523: I hope those aren't even remotely as scary as APNIC. And I don't think those are.
17:34:25 -!- Gregor has left (?).
17:34:29 <oerjan> ais523: although you might manage reference counting too... well if you can fit it within the 26 letters.
17:34:30 -!- Gregor has joined.
17:34:48 <ais523> you could count references in unary, but it'd still be nasty
17:34:56 <oerjan> although hm, deleting something when reference count reaches 0 is probably as hard as that DNA zipping
17:36:50 <Vorpal> oerjan, is the topic about the subset of underload that consists of :()^ ?
17:36:55 <ais523> it'd be more efficient though
17:37:01 <ais523> Vorpal: I'm not oerjan, but yes
17:37:08 <ais523> well, you're allowed to put things inside the parens
17:37:21 <ais523> but just :, ^, and more parens
17:37:22 <Vorpal> ais523, "<oerjan> <quintopia> woah. when did the topic thing happen?" "<oerjan> i've been working on it for about the past week"
17:37:39 <Vorpal> ais523, so that is what I responded to :P
17:38:41 <Vorpal> so why do you expect me to ask you specifically? Or did you apologize for answering in his place? (Why would someone apologize for that?!)
17:39:01 <oerjan> Vorpal: http://oerjan.nvg.org/esoteric/underload/minskyconv.ul and http://oerjan.nvg.org/esoteric/underload/ULDisclaim2.hs
17:39:16 -!- pumpkin has quit (Ping timeout: 252 seconds).
17:39:27 <oerjan> the first one is a minsky machine example
17:39:34 <Vorpal> will look at those a bit later
17:40:08 <ais523> Vorpal: I was apologising for answering in his place
17:40:09 -!- copumpkin has joined.
17:40:18 <ais523> and I generally do that when answering a question specifically aimed at someone else
17:40:23 <oerjan> Vorpal: also ais523 is the only one who has got a full explanation (unless you read the logs) so he is somewhat qualified
17:40:28 <ais523> there was a "oerjan," at the start of the line
17:40:57 <oerjan> (i _suppose_ others might have been reading too, but they didn't say so)
17:42:11 -!- cheater00 has quit (Ping timeout: 240 seconds).
17:44:43 -!- cheater99 has joined.
17:45:13 <Ilari> December-February (90 days) current unreserved pool and linear estimate: RIPE NCC: 1.612/3.30 (184 days). ARIN: 0.947/4.19 (398 days). APNIC: 3.693/3.37 (82 days!)
17:45:45 <Ilari> Of course, that doesn't take panic and demand shifting into account. But APNIC is depleting far far faster than other RIRs.
17:46:48 -!- augur_ has joined.
17:46:57 -!- augur has quit (Read error: Connection reset by peer).
17:46:58 <Ilari> Darn that I can't get good figures about free pool size for any other RIR than APNIC.
17:48:51 <Ilari> Anyway, the errors in pool sizes are by far insufficient for it to really be even near.
17:50:00 <pikhq> Oh jeeze. Peak oil is insanely close.
17:50:42 <pikhq> Estimates start at the end of this year.
17:50:48 <Ilari> Are you sure it even is in the future?
17:51:11 <pikhq> That is, if it hasn't happened right about now.
17:51:13 -!- pumpkin has joined.
17:51:16 <pikhq> Which is genuinely possible.
17:51:42 <pikhq> Shit's going to get *baaaad*.
17:51:48 <ais523> what definition of peak oil are you using, btw?
17:52:19 <pikhq> ais523: No further increase in oil production, and *likely* a begin in decrease of oil production.
17:52:38 <ais523> are people really insane enough to rely on oil production /increases/?
17:52:38 <pikhq> Erm, start in decrease.
17:52:53 <ais523> that's like the definition of a deficit, the amount of extra debt a country gets year on year
17:53:09 <ais523> the fact that the definition even exists is scary
17:53:12 <pikhq> ais523: Keep in mind that much of the world-wide economic development has relied on increases in oil production.
17:54:15 -!- copumpkin has quit (Ping timeout: 260 seconds).
17:54:18 <pikhq> If oil production begins to *decrease*, the US is going to have some *really* bad shit happening.
17:55:07 <ais523> it could start by increasing the price of oil to something sane
17:55:40 <ais523> it's held artificially high in the UK via taxes, which among other things means that a decrease in production, causing a price rise, could be at least partially absorbed by lowering taxes
17:55:44 <pikhq> Political suicide.
17:56:20 <pikhq> Recall that Americans, as a whole, *depend* on oil being cheap.
17:56:27 <Ilari> And whenever peak oil is/was one has to remember oil production (by almost any defintion) has been aproximately flat since 2004/2005.
17:56:53 <pikhq> Ilari: Ah, fuck, you're right.
17:57:21 <ais523> hmm, this probably explains the rise of biodiesel in the US
17:57:32 <ais523> there aren't many other explanations for it, as it seems to make not much economic sense
17:57:47 <ais523> but if it's a case of relaxing capacity issues on crude supply, there might be a reason for it
17:58:01 <Ilari> And biofuels stuff is completely insane. The EROEI is just plain horrible.
17:58:11 <ais523> and you can just conclude lobbyists or something for why it's done in such an insane way
17:58:31 <pikhq> ais523: Corn subsidies weight the whole thing against sanity.
17:58:51 <pikhq> Remember, the cost to produce corn is negative.
17:59:03 <Ilari> And looks like the economy is now in oscillatory mode caused by hitting oil supply constraints...
17:59:52 <pikhq> Ilari: Actually, the severe oscillation can be more accurately blamed on unregulated finance.
18:00:06 <pikhq> Though oil isn't helping things any.
18:00:21 <Ilari> And of course, the unrest in muslim countries doesn't help matters any.
18:00:47 <oerjan> i recall there was an oil price inflation just before the financial crisis hit
18:01:02 <oerjan> but the crisis sort of nullified it, iirc
18:01:03 <pikhq> Yeah. The economic situation is the result of a whole *ton* of stupid decisions and just straight-up bad fortune.
18:01:22 <Ilari> Of course, push things to extreme and disasters happen.
18:01:27 <pikhq> oerjan: Yeah, but it certainly didn't *cause* the financial crisis, even in the US.
18:02:05 <oerjan> pikhq: No, i meant it the other way around, the financial crisis stopped it
18:02:19 <pikhq> Yeah, it seems to have put a temporary hold on it.
18:02:35 <pikhq> Though we're seeing the price of oil skyrocket again.
18:02:51 <ais523> hmm, did the financial crisis actually hurt anyone, as in deaths, etc.?
18:02:56 <ais523> or did it just reduce quality of life?
18:03:12 <pikhq> ais523: In the US, it certainly caused deaths.
18:03:12 <oerjan> i'm sure there were _some_ suicides and stuff
18:03:33 <pikhq> Remember, unemployment means no non-ER health care.
18:03:47 <ais523> oh right, the US health system is insane
18:03:51 <Ilari> And it pushed financial system to really really edge. And thus the massive bailouts.
18:04:52 <pikhq> Oil is at about $100 a gallon.
18:05:26 <oerjan> http://en.wikipedia.org/wiki/Price_of_petroleum
18:05:33 <oerjan> still a bit up to the top
18:08:10 -!- augur_ has quit (Ping timeout: 240 seconds).
18:11:08 <Ilari> And of course, SAD (Standard American Diet) doesn't help things any.
18:12:27 <Ilari> It seems that out of diets that aren't blantantly massively deficient, diets that rollercoaster blood sugar, resulting in massive snacking are the absolute worst.
18:13:42 -!- ais523_ has joined.
18:14:36 -!- ais523 has quit (Disconnected by services).
18:14:37 -!- ais523_ has changed nick to ais523.
18:14:38 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
18:16:24 <Ilari> Whatever the diet, at least have a diet that one can follow and isn't blatantly deficient. Otherwise, you will end up snacking a lot, and those foods are the absolute worst.
18:16:41 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
18:18:23 <Ilari> Loads of sugar, loads of industrial trans fats (yes, those do still exist), loads of "fat" that isn't even fat anymore and other bad-for-you stuff...
18:20:58 <Gregor> SOMALIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
18:21:14 <Ilari> Oh, and apparently lots of that third group of trans fats (polyunsaturated non-conjugated trans fats), of which very little is known.
18:22:27 <Gregor> Except that they're delicious.
18:22:55 <Ilari> I don't think those things even taste good.
18:24:34 -!- cheater99 has quit (Ping timeout: 240 seconds).
18:24:38 -!- cheater00 has joined.
18:26:19 * Gregor turns back to the channel and away from his tub of Imitation Butter Flavr polyunsaturated non-conjugated trans fat meal to disagree. Then curse Somalia.
18:28:13 <oerjan> polly unsaturated, wants a cracker!
18:32:39 -!- oerjan has quit (Quit: Reboot).
18:37:26 -!- oerjan has joined.
19:01:21 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds).
19:07:42 -!- asiekierka has quit (Ping timeout: 276 seconds).
19:14:16 -!- ais523 has quit (Remote host closed the connection).
19:19:20 -!- cheater00 has quit (Ping timeout: 240 seconds).
19:21:00 -!- cheater00 has joined.
19:23:51 -!- augur has joined.
19:24:07 -!- pikhq has quit (Ping timeout: 240 seconds).
19:24:12 -!- pikhq has joined.
19:25:33 -!- pumpkin has quit (Ping timeout: 264 seconds).
19:32:23 -!- copumpkin has joined.
19:42:40 -!- pumpkin has joined.
19:43:51 -!- copumpkin has quit (Ping timeout: 260 seconds).
19:45:14 -!- Phantom__Hoover has joined.
20:06:02 -!- Sgeo has joined.
20:07:11 <Sgeo> I think my Statistics professor may be an idiot
20:07:17 <Sgeo> But I'm not sure, I want to ask in here
20:08:16 <Phantom__Hoover> a) you do, b) they are because only an idiot would voluntarily teach at a place called "Farmingdale".
20:08:55 <fizzie> To quote the 8-ball: ALL SIGNS POINT TO YES. [Disclaimer: the earlier may not be construed as the opinion of me.]
20:09:03 <Sgeo> So, problem: X represents how many times 1 appears on a fair die in 3 rolls
20:09:07 <Gregor> I'll bet Farmingdale doesn't even have a better Poultry Sciences department than Purdue.
20:09:26 <fizzie> Gregor: I bet they just Farmville all the time.
20:09:35 <Phantom__Hoover> Gregor, yeah, but Purdue has poultry science facilities beyond the dreams of avarice.
20:09:52 <Sgeo> He proceeds to show the possibility (I forget the exact term) space as such:
20:10:38 <Sgeo> Then says that the probability of P(0) = 1/8
20:10:39 <Gregor> "not 1" is a poor way to represent a fair die :P
20:10:58 <oerjan> Sgeo: i'd be worried if the etc. doesn't include 1 not1 1
20:11:07 <Sgeo> oerjan, it does
20:11:17 <Gregor> Technically he didn't say how many sides the die has.
20:11:24 <Gregor> It could be a two-sided die (also known as a coin)
20:12:17 <Phantom__Hoover> It's 1-(5/6)^3, which is trivial if you know any probability at all.
20:12:21 <Sgeo> He explains the discrepency between what the formula got and what he did there as the difference between "discrete" and "binomial" (I may have forgotten the exact term there) variables
20:12:33 <Gregor> Phantom__Hoover: Number of 1s, not 1 appearing at all.
20:12:47 <Gregor> Phantom__Hoover: (Which is even more trivial)
20:12:57 <oerjan> Phantom__Hoover: um P(0) is (5/6)^3, no 1-
20:12:59 <Gregor> Oh, or maybe I misunderstood the question *shrugs*
20:13:58 <Sgeo> He says that the 1/8 is considering it as a discrete variable, and the formula as binomial
20:14:26 <Phantom__Hoover> Sgeo, please inform him of this and that the disagreement is backed up by an unemployed mathematician, a programmer whose greatest achievement is a JS game and a 16-year-old.
20:14:29 <fizzie> I guess by his logic, P(X=3) = P(X=0) = 1/8?
20:14:47 <oerjan> Sgeo: he is wrongly assuming uniformity of the two options he has made for each coin
20:15:12 <fizzie> oerjan: Oh, so now it *is* a coin? :p
20:15:20 <Sgeo> I think I ended up questioning that in the middle of class. Once as me just being confused, and once as me trying to explain why the answers were disagreeing
20:15:52 <Phantom__Hoover> He *thinks the sum of the probabilities on a dice roll are 3.*
20:15:57 <oerjan> Sgeo: you should have made him a large bet :D
20:16:18 <Sgeo> Phantom__Hoover, where did I say he thinks that?
20:17:02 <oerjan> well he is obviously forgetting that the binomial distribution has a parameter for the probability of the even you count
20:17:32 <Phantom__Hoover> Sgeo, "fair dice" and the fact that P(X) = P(¬X) → sum is 3.
20:17:32 <fizzie> The whole "discrete"/"binomial" thing sounds rather confused too.
20:17:34 <Sgeo> oerjan, it's not like he left that out of the formula he gave us...
20:18:29 <Sgeo> According to him, (as a discrete variable), P(0) = 1/8, P(1) = 3/8, P(2) = 3/8, P(3) = 1/8
20:18:42 <Sgeo> So no, that doesn't equal 3
20:18:59 <Phantom__Hoover> Sgeo, erm, I mean in terms of the sum of the die rolling each face.
20:19:02 <oerjan> well that's of course correct for a coin.
20:19:50 <Sgeo> I kept telling him that the possibility space he should be expanding the not1, so 1 1 not1 -> 1 1 2, 1 1 3, etc. And he kept saying that then it wouldn't be binomial
20:20:16 <oerjan> Sgeo: there is absolutely no need to expand the not1 if you know what you are doing, though
20:20:44 <oerjan> you just need to take into account that the probability of a single not1 is 5/6, not 1/2
20:21:19 <Sgeo> Is there any context in which his answers make sense?
20:21:26 <Sgeo> Some weird phrasing of the question
20:21:48 <Sgeo> Besides P(success)=1/2
20:22:17 <fizzie> A coin, the fairest of dice.
20:23:17 <Sgeo> Phantom__Hoover, hmm?
20:23:32 <Sgeo> He was showing us with a coin on Monday. That wasn't what he wanted to illustrate
20:24:16 <fizzie> I think the "wouldn't be binomial if you expand the not1" does sort-of make sense, since the binomial distribution is for the number of successes of yes/no Bernoulli trials.
20:24:17 <Phantom__Hoover> Anyway, in non-idiot news, I'm thinking of doing an Eodermdrome implementation.
20:24:56 <fizzie> (But still, you can't assume p=0.5 for those trials.)
20:26:07 <Gregor> ARGH. SOMALIAAAAAAAAAAAAAAAAAA
20:27:27 -!- FireFly has joined.
20:27:37 <Gregor> Wow, somebody figured it out
20:27:55 <Gregor> I've just been screaming "SOMALIAAAAAAAAAAA" all day wondering whether anybody would "get it" :P
20:28:46 <Gregor> oerjan: Are you offended by my quest for awesome domain names?
20:28:47 -!- pikhq has quit (Read error: Operation timed out).
20:29:36 -!- pikhq has joined.
20:29:36 <oerjan> no, i was assuming you were talking about some news that somehow i never saw...
20:30:07 <oerjan> i mean there was this danish family but i didn't think that would cause such a reaction.
20:30:33 <Gregor> The news that the .so landrush is over, and they will be assigning domains within the previous 24 hours or so.
20:30:52 <Gregor> In theory I may already own libc.so
20:30:57 <Gregor> In practice I don't know.
20:31:12 <Gregor> Phantom__Hoover: Not REALLY
20:31:13 <oerjan> *mainstream news, you dolt
20:31:41 <Sgeo> Did you just accuse Gregor of not knowing of a US President who resigned?
20:31:42 <Gregor> Phantom__Hoover: But the transitional government that controls about 1/4 of the capitol and a few other outposts is all about technology, so they got .so up and running :P
20:32:39 <oerjan> Phantom__Hoover: that would seem the only explanation. but didn't it do that long ago?
20:33:25 <Sgeo> http://www.fark.com/comments/5351459/Michael-Savage-I-dont-know-of-an-American-president-whos-resigned
20:33:28 <Phantom__Hoover> I am also totally disappointed at the lack of interest in my plans for an Eodermdrome implementations.
20:33:37 <Gregor> oerjan: Fine fine, but at least the money actually goes to Somalia, instead of some random company in the US or something.
20:33:58 <fizzie> Phantom__Hoover: The language is cursed: people always just talk about implementing it, but never do. (Did oklopol have an eodermdrome thing or how was it?)
20:35:16 <fizzie> Phantom__Hoover: I thought about recommending the NAUTY library, because it has a silly name, but it seems that it's only really useful for full-graph isomorphism testing, not for the subgraph isomorphism problem.
20:35:57 <Phantom__Hoover> There's an algorithm for that, but I'm failing to understand it due to procrastination.
20:36:19 <fizzie> There's an app for that.
20:37:10 <fizzie> [2008-07-17 19:06:16] < oklopol> i implemented eodermdrome
20:37:21 <fizzie> I think it was horribly inefficient, though.
20:37:21 <oerjan> Phantom__Hoover: well it is NP-complete
20:38:15 <oerjan> although with subgraph size limited to 26, maybe it's not a problem.
20:38:48 <oerjan> (well that means it's in _principle_ only a P sub-problem.)
20:45:04 -!- Zuu has quit (Ping timeout: 260 seconds).
20:45:06 <fizzie> The eodermdrome graphs-to-match-against-the-subgraphs-of-the-state-graph are also constant over the lifetime of the program, which might help in some algorithms if they happen to involve some preprocessing on them. (Can't say I've looked at subgraph isomorphism algorithms at all.)
20:46:09 <fizzie> Maybe there ought to be a "Planar Eodermdrome" variant when you want efficient code; I think it was an easier problem there.
20:48:59 <Phantom__Hoover> I was thinking that a variant where copying things was easy would be nice, but I'm not completely sure how to do that.
20:51:13 -!- Zuu has joined.
20:52:05 -!- Sgeo has quit (Ping timeout: 240 seconds).
20:56:38 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.13/20101203075014]).
20:58:01 <oerjan> it seems to me that defining what to copy gets awkward when the graph isn't acyclic
21:01:01 <Phantom__Hoover> Hm? I was thinking simply that there would be a mechanism for having a closed node appear twice in the replacement subgraph, and then deleting the original subgraph except for that node, then building the replacement for the whole graph from two such components and the rest of the replacement subgraph.
21:01:45 <Phantom__Hoover> Erm, the component is the set of vertices reachable from the duplicated node when the rest of the match subgraph is deleted.
21:02:05 -!- pumpkin has quit (Ping timeout: 240 seconds).
21:03:21 -!- copumpkin has joined.
21:03:21 -!- copumpkin has quit (Changing host).
21:03:21 -!- copumpkin has joined.
21:04:29 <fizzie> The coincidentalism is astounding: you start talking about eodermdrome, and at the same time an ad for "An excursion into algebraic tools for combinatorial problems" seminar -- which seems to be mostly about graph-related combinatorics -- drops into my INBOX.
21:05:09 <fizzie> SYN, Chron... I, city?
21:05:20 <fungot> Selected style: ct (Chrono Trigger game script)
21:05:32 <fizzie> fungot: The sword alone can't stop!
21:05:32 <fungot> fizzie: like, thanks princess. i'll take that under advisement!!
21:05:49 <fizzie> That bot needs to learn some manners.
21:06:26 <oerjan> hey fizzie is _so_ a princess name
21:08:17 <oerjan> well it would be short for fizzonica or something, but still
21:09:28 -!- Sgeo has joined.
21:09:41 <Gregor> Princess Fizzonica of Carbonium
21:10:25 <fizzie> "Disney Pretty Princess Fizzie Macic Wand Gift Set ( Fizzie Wand, 15 Tnted Bath Fizzies, Body Wash and Body Lotion )" (amazon.com)
21:11:05 <oerjan> tnt'ed bath fizzies, sounds dangerous
21:11:29 -!- copumpkin has quit (Ping timeout: 246 seconds).
21:11:58 <fizzie> For the "X-treem" bathers, for which the usual bath salts and the like are too tame.
21:12:42 <Gregor> Their bath salts are just sodium, no chlorine.
21:13:18 <fizzie> Hah, http://www.alilg.com/arcade-games/Fizzie-15731.html "Game description: Simple, fun and addictive, the goal is simply to maneuver Fizzie around the screen and capture the rising bubbles." I seem to be... quite spiky.
21:13:19 -!- BeholdMyGlory has quit (Remote host closed the connection).
21:13:42 <fizzie> "Click the mouse button to make Fizzie go Move the mouse around, and Fizzie will follow the cursor Holding the mouse button down and aiming the cursor towards a bubble is a great way to get Fizzie to catch the bubble Avoid the red enemies"
21:13:56 <fizzie> Yes, I tend to have this urge to follow the cursor around.
21:14:18 <Gregor> I assumed he means CE certification.
21:15:36 <fizzie> The CE/BCE pair is not as nice as AD/BC, since the lengths differ.
21:17:18 <Gregor> We should just use a single number starting from the big bang. DONE. Also, relativity be damned.
21:17:33 <Gregor> It's the Kelvin of time :P
21:17:47 <fizzie> "The Kelvin of time" is a great phrase.
21:19:53 <Gregor> Given current estimates, we should be able to choose a starting year within 110 million years. We could "conveniently" choose one that ends in 2011
21:22:57 <fizzie> Also forget the whole year/month/day nonsense, maybe? Just seconds (it's the SI unit of time) since the big bang.
21:25:16 -!- Sgeo has quit (Ping timeout: 240 seconds).
21:25:42 <Gregor> As a person who lives on Earth, I have certain reasons to desire the use of years :P
21:26:25 <fizzie> Goal: deprecate all legacy date formats by 2020... uh, I mean, let's say provisionally by 433620000000000000, give and take a few million "old-style years" depending on what the authorities will set the start time at.
21:27:44 <olsner> seconds are too small for most time scales, but it seems well chosen for the shortest timespan since it's easy for humans to approximate
21:30:16 <oerjan> > 433620000000000000/86400/365.2425
21:31:29 <fizzie> oerjan: It was done with the 365-day approximation, sorry about that. Still, it falls in the 13.75 ± 0.11 range, anyway.
21:31:41 <fizzie> The microfortnight, for when the seconds are almost suitable but just a bit too short.
21:31:43 -!- augur has quit (Remote host closed the connection).
21:33:26 <Zwaarddijk> who ever figured KDE4 was a good move?
21:34:24 <oerjan> white king captures black queen on E4
21:34:34 <oerjan> looks like a pretty good move to me
21:34:52 <oerjan> the notation is a _little_ off, i think
21:34:55 <Ilari> Ah, NIST got around publishing paper about SHA-3 finalists selection... Finally.
21:35:22 -!- wetneb has quit (Quit: Leaving.).
21:36:43 <Phantom__Hoover> Alternately someone who really knew what they were doing, but more probably an idiot.
21:37:48 <Zwaarddijk> yeah right, you can basically force the king to eat the queen, and then checkmate him, in some circumstances, but I'd consider that a checkmate by the point the queen threatens him already
21:39:41 <fizzie> http://www.chessgames.com/perl/chessgame?gid=1387595 -- here's a king-captures-the-queen game, courteously provided by Google, fed through an onerous Javapplet.
21:41:25 <fizzie> (I was hoping to find some sort of a number as to how often that happens in "real" chess games.)
21:57:45 -!- copumpkin has joined.
21:57:51 -!- copumpkin has quit (Changing host).
21:57:52 -!- copumpkin has joined.
21:59:30 -!- oerjan has quit (Quit: Good night).
22:03:25 -!- Wamanuz2 has joined.
22:03:55 -!- Wamanuz has quit (Ping timeout: 250 seconds).
22:09:49 -!- BeholdMyGlory has joined.
22:13:01 -!- copumpkin has quit (Ping timeout: 250 seconds).
22:13:50 -!- Patashu has joined.
22:13:51 -!- Mathnerd314 has joined.
22:16:51 -!- copumpkin has joined.
22:17:16 -!- Behold has joined.
22:17:47 -!- BeholdMyGlory has quit (Ping timeout: 250 seconds).
22:32:44 <Phantom__Hoover> http://www.reddit.com/r/IAmA/comments/fvvd4/iama_member_of_congress_rep_john_garamendi_dca_ama/
22:40:42 -!- Phantom__Hoover has quit (Quit: Leaving).
22:41:44 <Mathnerd314> q: is it true that you've been bribed into secrecy?
22:46:26 -!- copumpkin has quit (Ping timeout: 240 seconds).
22:48:04 -!- copumpkin has joined.
22:52:25 -!- elliott has joined.
22:52:27 -!- copumpkin has quit (Ping timeout: 264 seconds).
22:54:29 <elliott> 19:54:45 <pikhq_> Don't anyone tell elliott this, but there's a lot of atrocious handwriting in Japan.
22:54:40 <elliott> pikhq: my handwriting is that of a 3 year old
22:55:23 <elliott> "US looking at new moves on Somalia piracy-Clinton"
22:55:31 <elliott> BUT HOW WILL THEY GET THEIR MUSIC AND SOFTWARE
22:57:31 <pikhq> With the other form of piracy.
23:00:03 -!- Slereah has quit (Ping timeout: 240 seconds).
23:01:59 <elliott> 09:06:46 <olsner> hmm, elliott disappeared
23:02:01 -!- copumpkin has joined.
23:02:06 <elliott> olsner: don't worry, i came back
23:02:12 <elliott> what do you wish to consult me on, pilgrim?
23:03:18 <elliott> 09:40:57 <oerjan> (i _suppose_ others might have been reading too, but they didn't say so)
23:03:27 <elliott> my future self was with you during that explanation.
23:05:33 <elliott> 10:03:47 <ais523> oh right, the US health system is insane
23:05:33 <elliott> 10:04:01 <ais523> work or die
23:05:37 <elliott> "it's not a bug, it's a feature!"
23:05:56 -!- Slereah has joined.
23:07:13 <elliott> `addquote <Phantom__Hoover> Gregor, yeah, but Purdue has poultry science facilities beyond the dreams of avarice.
23:07:23 <HackEgo> 328) <Phantom__Hoover> Gregor, yeah, but Purdue has poultry science facilities beyond the dreams of avarice.
23:08:19 <elliott> 12:14:26 <Phantom__Hoover> Sgeo, please inform him of this and that the disagreement is backed up by an unemployed mathematician, a programmer whose greatest achievement is a JS game and a 16-year-old.
23:08:24 <elliott> together, they fight crime!
23:08:39 <elliott> 12:14:47 <oerjan> Sgeo: he is wrongly assuming uniformity of the two options he has made for each coin
23:08:39 <elliott> 12:15:12 <fizzie> oerjan: Oh, so now it *is* a coin? :p
23:08:42 <elliott> yes, those pesky six-sided coins!
23:08:56 <elliott> 12:15:57 <oerjan> Sgeo: you should have made him a large bet :D
23:09:33 <elliott> 12:23:33 <Phantom__Hoover> Sgeo, die rather than coin.
23:10:44 -!- copumpkin has quit (Ping timeout: 252 seconds).
23:11:28 <elliott> 13:05:32 <fizzie> fungot: The sword alone can't stop!
23:11:29 <elliott> 13:05:32 <fungot> fizzie: like, thanks princess. i'll take that under advisement!!
23:11:29 <fungot> elliott: i'd like to see that mystical sword for myself! geez! i see you're dressing...normally again!? you're joking!?
23:11:29 <fungot> elliott: there! there it is! but by the time we're through with you, you'll be in danger. open hatch.
23:11:43 <elliott> at least if i remember my fungot correctly.
23:11:43 <fungot> elliott: that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop, crono!
23:12:26 <elliott> 13:13:18 <fizzie> Hah, http://www.alilg.com/arcade-games/Fizzie-15731.html "Game description: Simple, fun and addictive, the goal is simply to maneuver Fizzie around the screen and capture the rising bubbles." I seem to be... quite spiky.
23:12:34 <elliott> fizzie: no, that's Fizzie; you're obviously fizzie.
23:12:59 <elliott> 13:13:35 <Phantom__Hoover> Is it just me who hates "CE"?
23:12:59 <elliott> 13:14:13 <Phantom__Hoover> No, instead of "AD".
23:12:59 <elliott> THEY'RE TAKIN' OOR CHRISTIANITY
23:14:22 -!- copumpkin has joined.
23:14:58 <elliott> 14:41:44 <Mathnerd314> q: is it true that you've been bribed into secrecy?
23:15:36 <Mathnerd314> elliott: the government is clearly involved in a conspiracy
23:15:45 <elliott> Mathnerd314: No, it's not.
23:16:16 -!- MigoMipo has quit (Read error: Connection reset by peer).
23:16:16 <elliott> Mathnerd314: Burden of proof is on you; you are the one making the extraordinary claim.
23:16:49 <Mathnerd314> elliott: it is more likely for a conspiracy to exist than for one not to exist
23:16:54 <elliott> Mathnerd314: By what evidence?
23:17:50 <Mathnerd314> elliott: by the evidence that conspiracies are common
23:18:12 <elliott> Mathnerd314: What is your evidence for that? And how does it generalise to the fact that the US government is *huge* and not exactly centralised?
23:18:16 <elliott> And what exactly is this conspiracy's goals?
23:18:40 <Mathnerd314> I don't know that much about the conspiracy; just that it exists
23:20:12 <elliott> Mathnerd314: I don't know why I keep humouring you by prodding your inane assertions with requests for clarifications, you're clearly an idiot who tries to appear intellectual by spouting unsubstantiated bullshit.
23:20:55 -!- pingveno has quit (Ping timeout: 246 seconds).
23:21:28 -!- pingveno has joined.
23:21:47 <elliott> I'm sure the conspiracy is to cover up how powerful term rewriting is.
23:26:02 -!- copumpkin has quit (Ping timeout: 264 seconds).
23:30:46 -!- copumpkin has joined.
23:31:24 -!- FireFly has quit (Quit: swatted to death).
23:31:32 <elliott> pikhq: Hundred = one drop sun. WELL DUHHH
23:43:35 * Lymia sees insane people
23:43:38 <Lymia> Nothing out of the ordinary.
23:48:50 <pikhq> I really get the strong feeling that math education's ordering is based on when things were first discovered, rather than anything that makes sense at all.
23:49:04 <pikhq> On up through post-secondary.
23:50:51 <pikhq> Can anyone give me a rationale for defining many different objects and reïterating, except much more tediously, that they form a group, a monoid, a field, etc., before teaching those damned concepts?
23:51:42 <fizzie> An advocate of New Math, then?
23:51:50 <pikhq> fizzie: Not *quite*.
23:52:54 -!- shachaf has quit (Ping timeout: 272 seconds).
23:53:00 -!- shachaf has joined.
23:53:18 <elliott> pikhq: That Tae Kim fellow doesn't appear to like Remembering the Kanji
23:53:23 <pikhq> fizzie: But there is no damned reason to teach these after you're already dealing with the concept of freaking vector spaces.
23:53:40 <pikhq> elliott: Yeah, yeah, yeah, he seems to misunderstand RTK.
23:53:51 <elliott> pikhq: CLEARLY HE IS A MORON
23:54:17 <pikhq> elliott: No, his conception of RTK seems to be defined by some of the morons who have started on RTK.
23:54:21 <elliott> fizzie: the problem with new math was that it wasn't preceded by New Logic and English wasn't eliminated and ;D
23:54:36 <elliott> (ok new math was probably a legitimately bad idea, but i'm aaaall for teaching useless theory to 3 year olds)
23:55:04 <elliott> "I hope this ranks #1 in google for “Heisig douche bag” (Updated)"
23:55:08 <pikhq> elliott: Some people actually think that finishing RTK will get you full, complete literacy.
23:55:39 <elliott> "Even so, after over 50 comments, nobody has stepped forth and met my challenge by saying, “Yes, I can write whole words and sentences like a native using his methods.”"
23:55:45 <elliott> o ya, i forgot the part in the preface where he was all like
23:55:51 <elliott> and if you know all the kanji (all of them)
23:55:55 <elliott> you will be able to WRITE SENTENCES!!
23:56:10 <pikhq> When really, all it will do is make kanji not a barrier to entry.
23:56:34 <elliott> [[If that’s not enough to incite you into commenting, here some more fodder.
23:56:34 <elliott> I think it’s better to teach casual Japanese before polite Japanese. It sounds crazy I know, but first of all, it’s how all native speakers started out as kids so it can’t be that bad. Second, it’s much more useful grammatically and socially if you’re in high school or college. Finally, I worked at one of the largest, oldest, and most traditional Japanese companies in Tokyo and “business Japanese” was just putting “desu” and “
23:56:34 <elliott> masu” at the end of every sentence. The rest is knowing phrases like 「いつもお世話になっております」, honorific/humble, and vocabulary that’s too difficult for beginners anyway.]]
23:56:39 <elliott> err does heisig actually contradict that
23:56:43 <elliott> or is he just trying to get people to flame
23:56:52 <pikhq> No, Heisig doesn't ever claim that at all.
23:57:04 <pikhq> Erm, doesn't contradict it at all.
23:57:23 <pikhq> I seem to recall some interviews with Heisig saying that his further Japanese learning involved comics and talking with kids.
23:57:37 -!- pumpkin has joined.
23:57:38 <pikhq> Which suggests that he first learned casual Japanese. :P
23:58:38 <pikhq> So, yeah, Tae Kim seems to be trolling there.
23:58:51 <elliott> pikhq: hmm, do comics et al (SHUT UP I KNOW IT'S "ETC", "ET AL" IS NICER) not use the politeness crap?
23:59:16 <pikhq> elliott: Manga and anime in particular are much less likely to use the politeness crap.
23:59:47 <elliott> pikhq: what about like, novels