00:00:05 <elliott> pikhq: I propose that we now talk about the services manager to stop us both going mad and to let me blab about Kitten more.
00:00:44 <elliott> pikhq: Okay so... now I have to serialise my damn thoughts
00:01:28 <elliott> pikhq: You know where you pass the runlevel to init?
00:01:30 <elliott> In the kernel command line.
00:01:41 * Sgeo goes to install Portal
00:02:47 <elliott> Clearly pikhq has no idea.
00:03:25 <elliott> Let us kill pikhq, olsner!
00:05:22 <elliott> I think pikhq's IRC client is so lagged that his reply isn't getting through because it's running under 10 QEMUs as part of his multilib torture test.
00:08:52 <elliott> <elliott> pikhq: You know where you pass the runlevel to init?
00:08:52 <elliott> <elliott> In the kernel command line.
00:08:57 <elliott> Let's see if he's remembered!
00:09:54 <Gregor> I appear to be writing a wind quintet for some reason.
00:09:57 <elliott> pikhq: So, yes, clearly you do.
00:10:16 <elliott> pikhq: Well, guess what? The service manager reappropriates the init argument for something else similar to runlevels but NOT. Isn't that AMAZING
00:11:15 <elliott> pikhq: So, services, right? They are concrete things -- like a specific httpd or mail server -- but they can also be more abstract things.
00:11:33 <elliott> pikhq: You can imagine a "consoles" service which is merely a dependency on eight gettys.
00:11:46 <elliott> (Getty would be a metaservice, taking the tty as an argument. Seeing a pattern here?)
00:11:48 <Gregor> I need to put range notes on my keyboard >_>
00:11:55 <Gregor> "The bassoon does not go below this Bb"
00:12:09 <elliott> pikhq: So, for instance, imagine a service "single-user", that depends on everything required to have the system in a single-user state.
00:12:22 <elliott> pikhq: Then "cli", which would be a full, command-line system.
00:12:32 <elliott> pikhq: Then "gui", which would start your login manager or whatever else you want.
00:12:41 <elliott> pikhq: The argument to "init" -- actually the service manager -- is just *a service name*.
00:12:54 <elliott> Give the kernel "gui", you get a GUI system. "cli", CLI. "single-user", guess.
00:12:56 <pikhq> elliott: Or list thereof?
00:13:10 <elliott> (But you're probably better off making your own service depending on that list.)
00:13:27 <elliott> pikhq: And then, of course, it just starts that, which means starting all its dependencies concurrently, and then running its start script.
00:13:44 <elliott> pikhq: It orders the total list of dependencies so that everything required is started up in the order it needs to be, as parallel as possible.
00:13:48 <pikhq> I'm thinking stuff like random need to start with a certain service just once. Granted, not likely to come up, but hey, you never know.
00:13:55 <elliott> This applies even when you use ctl(8) to start and stop things after-the-fact.
00:14:48 <elliott> http://www.youtube.com/watch?v=7E8MJK4nlek what
00:14:59 <elliott> "And I know... you're Saddam Hussein's... GRAND SON."
00:15:35 <elliott> I am pretty sure this is a joke.
00:16:05 <elliott> pikhq: Anyway, services would generally have more than just start/stop/restart/reload.
00:16:14 <elliott> pikhq: Because... basically every action you might want to do should be a service action.
00:16:35 <elliott> pikhq: Anything ending in "ctl" or anything done by a signal should be an action.
00:16:53 <elliott> pikhq: I'm almost crazy enough to think that the package manager and service manager could be merged.
00:17:00 <elliott> A package representing "this service is running".
00:17:04 <elliott> Almost crazy enough. Not quite.
00:18:47 <elliott> pikhq: You may now suggest things you want in the service manager.
00:19:25 <elliott> pikhq: Oh yeah, and the service files would be files with actual syntax, not shell scripts made to do awful, awful things.
00:19:48 <elliott> pikhq: Sure, you'll probably have shell scripts in the actions, but that doesn't mean the service has to be one.
00:20:35 <elliott> pikhq: Probably the structure will be /sv/foo/{some-name-signifying-that-it's-the-main-config-file,start,stop,some-action}
00:20:44 <elliott> (Why have a top-level directory? Why not?)
00:22:54 <Sgeo> http://www.thelocal.de/national/20101013-30455.html
00:23:03 <Sgeo> Proof that it isn't just the US that's insane
00:23:46 <elliott> "It's alright, Germany does it do!"
00:23:54 <elliott> I imagine far more than 1/4 of Americans are xenophobic.
00:24:14 -!- wareya has quit (Read error: Operation timed out).
00:24:29 -!- FireFly has quit (Quit: swatted to death).
00:25:03 <pikhq> elliott: Republican party.
00:25:37 <Sgeo> There may be some Republicans who are Republicans due to fiscal beliefs
00:25:42 <Sgeo> My dad comes to mind
00:25:59 <elliott> Right. Your dad is always a paragon of reasonability, you've certainly shown that...
00:26:51 -!- wareya has joined.
00:26:57 <pikhq> Sgeo: Problem being, the Republicans have the following fiscal beliefs: SPEND MONEY. CUT TAXES. DO LINES OFF A HOOKER'S ASS.
00:27:19 <elliott> *DO LINES OFF A GAY MAN'S ASS IN A BATHROOM. DECRY HOMOSEXUALITY.
00:27:29 <oerjan> listen to pikhq, he is a man entirely without prejudices
00:27:36 <pikhq> elliott: That's not a fiscal belief, that's a social belief.
00:27:44 <pikhq> elliott: And I didn't say that the hooker was female.
00:27:50 <elliott> pikhq: Doing lines off a hooker's ass is a fiscal belief?
00:28:02 <pikhq> elliott: You pay money for the lines of coke and the hooker.
00:28:07 <elliott> oerjan: I'm not sure what you're trying to say :p
00:28:33 <pikhq> elliott: s/You pay/One pays/
00:28:44 <elliott> It's so fucking pretentious.
00:29:30 <Sgeo> I love that word.
00:29:52 <elliott> That's not exactly an English word.
00:29:55 <Sgeo> Is it "Yes" and I forgot to type the accent? Is it if? Is it actually Esperanto, and "one"
00:30:24 <elliott> Gregor: That would have an accent.
00:30:25 <elliott> pikhq: BTW, with all that gcc-cross stuff...
00:30:26 -!- Ilari has quit (Read error: Operation timed out).
00:30:31 <elliott> pikhq: Does pcc use the triples?
00:30:46 <Gregor> elliott: Feh, which language has "si" with no accent, I'm pretty sure one does ...
00:31:17 <oerjan> elliott: i'm just futilely trying to use sarcasm to point out an excessive generalization of the type that i'm sure is part of what is currently poisoning the US political climate.
00:31:20 -!- Ilari_antrcomp has quit (Quit: Reconnecting).
00:31:28 <pikhq> elliott: I *think* it does.
00:31:31 <elliott> pikhq: You know that path-to-the-dynamic-linker thing?
00:31:33 -!- Ilari_antrcomp has joined.
00:31:41 <elliott> Couldn't you only hardcode the filename and let there be a search path?
00:32:07 <pikhq> Violate ELF spec, you'd need to modify the kernel.
00:32:19 <elliott> oerjan: Do you not think that *tolerating* the Republicans' utter insanity is part of that? The fact that they have a culture where the Republicans and the Democrats constitutes a LEGITIMATE TWO-PARTY SYSTEM...
00:32:35 <elliott> I'd say they need to portray both of them as goat-fucking abortions.
00:32:38 -!- Ilari has joined.
00:32:49 <elliott> Totally the plural of abortus.
00:33:04 <Sgeo> Gregor, Spanish and Esperanto, unless my memory fails
00:33:33 <pikhq> oerjan: The thing is, they actually *advocate* increasing military spending and cutting taxes.
00:33:35 <elliott> <elliott> pikhq: BTW, with all that gcc-cross stuff...
00:33:36 <elliott> <elliott> pikhq: Does pcc use the triples?
00:33:36 <elliott> <elliott> Presumably clang does.
00:34:04 <pikhq> oerjan: The lines of coke thing is just a silly stereotype about pretty much anyone with more money than anyone could possibly spend.
00:34:10 <pikhq> oerjan: Such as almost all politicians.
00:34:17 <pikhq> Erm, almost all US politicians.
00:37:12 <oerjan> elliott: pikhq: well i'm just saying you are _both_ sounding like fanatics.
00:37:23 <elliott> oerjan: i don't take any of this seriously
00:37:50 <Sgeo> Including me???
00:37:57 <pikhq> oerjan: No. Seriously. The Republicans ACTUALLY ADVOCATE THIS.
00:38:01 <pikhq> oerjan: AS THEIR PLATFORM.
00:38:16 <elliott> oerjan: i think it's pretty obvious that their political system is near-unsalvagable as they have the choice between ultra-right-wing and almost-ultra-right-wing
00:38:22 <oerjan> pikhq: i am not saying that you are lying. i am saying you _sound like a fanatic_.
00:38:24 <Sgeo> elliott, are you just trying to make sure PSOX stays dead?
00:38:31 <elliott> Sgeo: You do realise pikhq is in the US too?
00:38:51 <oerjan> it is in fact possible to tell the truth in an entirely counterproductive manner.
00:38:59 <pikhq> oerjan: Why, yes, of course I do.
00:39:03 <Sgeo> Apparently you're fundamentally opposed to Pebble too
00:39:37 <pikhq> oerjan: Are you *familiar* with what the Republicans are like?
00:39:53 <elliott> oerjan: it is also possible to vent
00:39:57 <elliott> which is clearly what pikhq is doing
00:40:13 <elliott> it's not like he's trying to convince anyone
00:40:16 <pikhq> And, yeah. I need to vent sometimes. Gaaaah.
00:40:26 -!- sshc has quit (Ping timeout: 240 seconds).
00:40:56 <elliott> http://www.acc.umu.se/~achtt315/tunguska/ Ternary system emulator.
00:41:05 <elliott> With a ternary-C-dialect compiler!
00:41:10 <elliott> I approve to an insane degree.
00:41:17 <elliott> "It is loosely based on the excellent design of the (binary) 6502-processor by MOS Technology, but entirely ternary. So instead of having two memory cell states (0, 1), it has three (-1, 0, 1)."
00:41:58 <elliott> pikhq: ^ Impossibly awesome
00:41:59 <elliott> http://www.acc.umu.se/~achtt315/tunguska/shots/tunguska_gtk.png
00:42:05 <elliott> And a command-line on this display!
00:42:48 * catseye is measuring programmer productivity in lines of coke
00:46:45 <elliott> It tries to use your proxy.
00:46:50 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
00:47:10 <elliott> Gregor: http://codu.org/projects/hackbot/fshg/index.cgi/file/faaf59ee1b1b/bin/wl
00:47:33 <elliott> `run wl 'sv sill' 2>&1 | tr -d '\n' | tail -n +10
00:47:38 <elliott> `run wl 'sv sill' 2>&1 | tr -d '\n' | tail -c +10
00:47:42 <HackEgo> (most recent call last): File "/tmp/hackenv.20289/bin/wl", line 37, in <module> q = query(continue_id) File "/tmp/hackenv.20289/bin/wl", line 29, in query response = urllib2.urlopen(url).read() File "/opt/python27/lib/python2.7/urllib2.py", line 126, in urlopen return _opener.open(url, data, timeout)AttributeError:
00:47:47 <elliott> `run wl 'sv sill' 2>&1 | tr -d '\n' | tail -c +20
00:47:51 <HackEgo> ent call last): File "/tmp/hackenv.20352/bin/wl", line 37, in <module> q = query(continue_id) File "/tmp/hackenv.20352/bin/wl", line 29, in query response = urllib2.urlopen(url).read() File "/opt/python27/lib/python2.7/urllib2.py", line 126, in urlopen return _opener.open(url, data, timeout)AttributeError: ProxyHandler
00:47:55 <elliott> `run wl 'sv sill' 2>&1 | tr -d '\n' | tail -c +50
00:47:58 <HackEgo> ckenv.20399/bin/wl", line 37, in <module> q = query(continue_id) File "/tmp/hackenv.20399/bin/wl", line 29, in query response = urllib2.urlopen(url).read() File "/opt/python27/lib/python2.7/urllib2.py", line 126, in urlopen return _opener.open(url, data, timeout)AttributeError: ProxyHandler instance has no attribute 'open'
00:48:04 <elliott> I guess that's actually my issue, huh.
00:48:10 <Gregor> `run wl 'sv sill' 2>&1 | paste
00:48:15 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.28689
00:48:37 <Gregor> That looks like your problem to me.
00:49:12 <Sgeo> Windows 7 is growing on me.
00:49:20 <Sgeo> But I'm missing a lot of stuff that Ubuntu had
00:49:23 <elliott> Every fucking thing grows on you.
00:49:31 * catseye sprays Sgeo with windicide
00:49:35 <Gregor> <Sgeo> This tumor is growing on me.
00:50:13 <elliott> `fetch http://pastie.org/pastes/1219576/download?key=1fzihgyczmg1qaxeqnpsq
00:50:17 <HackEgo> 2010-10-13 23:50:11 URL:http://pastie.org/pastes/1219576/download?key=1fzihgyczmg1qaxeqnpsq [1187/1187] -> "download?key=1fzihgyczmg1qaxeqnpsq" [1]
00:50:19 <elliott> `run cat download?key=1fzihgyczmg1qaxeqnpsq >bin/wl
00:50:33 <Gregor> ... seriously ... cat foo > bar ... not cp ...
00:50:34 <elliott> Gregor: Interwiki translate!
00:50:46 <elliott> And cp would make me have to chmod +x it again.
00:50:53 <elliott> `run rm download?key=1fzihgyczmg1qaxeqnpsq
00:50:57 <elliott> Gregor: MY LAZINESS IS UNSURPASSED
00:51:45 <elliott> Gregor: Translating English to English? :P
00:51:45 <elliott> `run wv 'en fucking la' 2>&1
00:51:47 <HackEgo> /bin/bash: line 1: wv: command not found
00:51:50 <Gregor> I don't know what this is supposed to do :P
00:52:11 <elliott> Gregor: "wl langcode foo" looks at langcode.wikipedia.org/wiki/foo
00:52:16 <elliott> Gregor: It looks for an English interwiki link
00:52:25 <elliott> if you provide a third argument, that is used instead of english
00:52:36 <elliott> Gregor: this is basically for all the times Vorpal says he doesn't know what a word is in English and I find it via interwiki
00:52:58 <elliott> `run wl 'en fucking la' 2>&1
00:53:02 <HackEgo> Traceback (most recent call last): \ File "/tmp/hackenv.21071/bin/wl", line 40, in <module> \ for link in q['query']['pages'].values()[0]['langlinks']: \ KeyError: 'langlinks'
00:53:13 <elliott> Fucking redirects to fuck, you see.
00:53:36 <elliott> There's a redirects parameter to automatically handle them.
00:53:42 -!- antivigilante has quit (Ping timeout: 240 seconds).
00:54:07 * elliott makes it "zh bicycle", "en zh bicycle"
00:54:10 <elliott> More consistent API, right?
00:55:14 <pikhq> 自転車, should be valid Written Chinese as well.
00:56:16 <HackEgo> My hovercraft is full of eels.
00:56:55 -!- petabit has joined.
00:56:57 <HackEgo> My hovercraft is full of eels.
00:57:01 <Gregor> Monty Python has defeated Wikipedia.
00:57:38 <catseye> apparently only a few european civilizations have developed the bicycle
00:57:38 <elliott> Gregor: oerjan: That is my error message, you doofuses.
00:57:56 <elliott> Gregor: i.e. "No link, here's a translation that works in any case!"
00:58:13 <oerjan> elliott: dammit i was just about to check the article for vandalism :D
00:58:14 <Gregor> elliott: Should've used "If I said you have a beautiful body, would you hold it against me? I am no longer infected."
00:58:24 <elliott> Aand why is that no output.
00:58:27 <elliott> `run wl 'en bicycle fi' 2>&1
00:58:30 <HackEgo> Traceback (most recent call last): \ File "/tmp/hackenv.21773/bin/wl", line 42, in <module> \ print link['*'] \ UnicodeEncodeError: 'ascii' codec can't encode character u'\xf6' in position 7: ordinal not in range(128)
00:58:45 <Gregor> elliott: You fail at Monty Python.
00:58:55 <catseye> Python: you fail at unicode
00:58:58 <elliott> wondering why you'd prefer that :P
00:59:11 <elliott> `fetch http://pastie.org/pastes/1219594/text?key=sggh817ogyryzpccok0bta
00:59:13 <HackEgo> 2010-10-13 23:59:07 URL:http://pastie.org/pastes/1219594/text?key=sggh817ogyryzpccok0bta [1962/1962] -> "text?key=sggh817ogyryzpccok0bta" [1]
00:59:19 <elliott> `rm text?key=sggh817ogyryzpccok0bta
00:59:23 <elliott> `fetch http://pastie.org/pastes/1219594/download?key=sggh817ogyryzpccok0bta
00:59:25 <HackEgo> 2010-10-13 23:59:19 URL:http://pastie.org/pastes/1219594/download?key=sggh817ogyryzpccok0bta [1469/1469] -> "download?key=sggh817ogyryzpccok0bta" [1]
00:59:38 <elliott> `run cat download?key=sggh817ogyryzpccok0bta >bin/wl; rm download?key=sggh817ogyryzpccok0bta
01:00:05 <HackEgo> My hovercraft is full of eels.
01:00:11 <HackEgo> My hovercraft is full of eels.
01:00:46 <HackEgo> My hovercraft is full of eels.
01:00:46 <HackEgo> My hovercraft is full of eels.
01:00:54 <HackEgo> My hovercraft is full of eels.
01:01:04 <elliott> There are surprisingly few interwikis on [[en:fuck]].
01:01:35 <elliott> Because it's an article about the word, not the meaning. Blarh.
01:01:41 <elliott> `wl en simple Category_theory
01:01:44 <HackEgo> My hovercraft is full of eels.
01:01:58 <oerjan> that pretty much sums up the theory, yeah
01:02:01 <Gregor> `wl en simple Jesus_of_Nazareth
01:02:13 <pikhq> The Japanese article has a full description of how to use "fuck" idiomatically in English. It's kinda funny.
01:02:38 <oerjan> Gregor: Jesus_of_Nazareth redirects to Jesus, i'm pretty sure
01:03:01 <HackEgo> Traceback (most recent call last): \ File "/tmp/hackenv.22666/bin/wl", line 55, in <module> \ for link in q['query']['pages'].values()[0]['langlinks']: \ KeyError: 'langlinks'
01:04:00 <elliott> `fetch http://pastie.org/pastes/1219610/download?key=vfm9cgewz89yc6tolaprwq
01:04:15 <HackEgo> 2010-10-14 00:04:09 URL:http://pastie.org/pastes/1219610/download?key=vfm9cgewz89yc6tolaprwq [1559/1559] -> "download?key=vfm9cgewz89yc6tolaprwq" [1]
01:04:17 <elliott> `run cat download?key=vfm9cgewz89yc6tolaprwq >bin/wl; rm download*
01:04:18 <Gregor> `wl en simple General_relativity
01:04:33 <oerjan> nothing simple about it
01:04:49 <HackEgo> /tmp/hackenv.22947/bin/wl: error while loading shared libraries: /tmp/hackenv.22947/bin/wl: file too short
01:05:10 <elliott> `run cat download?key=vfm9cgewz89yc6tolaprwq >bin/wl; rm download*
01:05:12 <catseye> "file too short". love it.
01:05:17 <elliott> `fetch http://pastie.org/pastes/1219610/download?key=vfm9cgewz89yc6tolaprwq
01:05:22 <Gregor> elliott: Good job breaking it, hero.
01:05:24 <HackEgo> 2010-10-14 00:05:18 URL:http://pastie.org/pastes/1219610/download?key=vfm9cgewz89yc6tolaprwq [1559/1559] -> "download?key=vfm9cgewz89yc6tolaprwq.1" [1]
01:05:32 <elliott> `run cat download?key=vfm9cgewz89yc6tolaprwq.1 >bin/wl; rm download*
01:05:33 <catseye> "file surpasses komologorov lower bound"
01:05:57 <HackEgo> My hovercraft is full of eels.
01:06:41 -!- petabit has quit (Read error: Connection reset by peer).
01:07:12 <HackEgo> My hovercraft is full of eels.
01:07:34 <elliott> `wl en it Dirty_Hungarian_Phrasebook
01:08:32 <elliott> Now let's replace Google Translate's backend with it.
01:09:32 <HackEgo> My hovercraft is full of eels.
01:10:01 <elliott> I MUST CREATE EXPLICIT LATIN WITH MY PROGRAM
01:11:05 <HackEgo> My hovercraft is full of eels.
01:11:14 <elliott> Shouldn't it be hu en, really?
01:11:40 <elliott> `wl en nl The_Treachery_of_Images
01:13:20 -!- antivigilante has joined.
01:13:26 <catseye> `run 'wl en pt Bicycle' | bzip2
01:14:57 <HackEgo> My hovercraft is full of eels.
01:14:57 <HackEgo> My hovercraft is full of eels.
01:15:21 -!- sshc has joined.
01:17:11 <Gregor> `wl en ja Cheeseburger
01:19:09 -!- elliott has quit (Quit: Connection reset by peer review).
01:19:17 <HackEgo> My hovercraft is full of eels.
01:21:05 <HackEgo> My hovercraft is full of eels.
01:21:44 <pikhq> I "love" how wrong that renders here.
01:21:48 <HackEgo> My hovercraft is full of eels.
01:22:08 <HackEgo> My hovercraft is full of eels.
01:22:28 <HackEgo> My hovercraft is full of eels.
01:23:22 <HackEgo> My hovercraft is full of eels.
01:23:44 <HackEgo> My hovercraft is full of eels.
01:23:47 <HackEgo> My hovercraft is full of eels.
01:24:44 -!- oerjan has quit (Quit: Good night).
01:24:45 -!- Sgeo has quit (Read error: Connection reset by peer).
01:25:18 -!- Sgeo has joined.
01:25:37 <catseye> Sgeo: do you still have windows growing on you
01:26:13 <Sgeo> Windows is to blame for all random disconnects?
01:29:01 <HackEgo> My hovercraft is full of eels.
01:29:14 <HackEgo> My hovercraft is full of eels.
01:29:32 <Sgeo> `wl en en My hovercraft is full of eels.
01:29:35 <pikhq> http://upload.wikimedia.org/wikipedia/commons/2/27/Book-in-Xiaoerjing.png Guess what language the right page here is written in!
01:30:13 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
01:31:33 <Sgeo> Someone coerced the bot into saying that. That's all I can assume from the delay
01:32:43 <catseye> `run wget http://www.han.de/~werner/ytree-1.96.tar.gz | bzip2 | bzip2 | bzip2 | bzip2 | bzip2
01:33:17 <pikhq> Seriously, none at all?
01:33:24 <catseye> pikhq: i would guess either Xiaoerjing, based on the link, or
01:33:31 <catseye> Arabic based on what it looks like
01:33:39 -!- augur has joined.
01:33:55 <Gregor> Written in an alphabet for some reason.
01:34:53 <pikhq> Apparently, Mandarin-speaking Muslims write translations of the Koran using Arabic script, so as to allow for not botching Arabic loan words.
01:37:04 <pikhq> catseye: "小兒經" in Traditional Chinese, "小儿经" in Simplified, "Xiǎo'érjīng" in Pinyin, شِيَوْ عَر in Xiao'erjing.
01:37:08 <Sgeo> If something were written in English using Hebrew letters, I'd be able to use my limited understanding of Hebrew for something!
01:37:32 <pikhq> Sgeo: Maybe you could halfway-guess at the meaning of Yiddish.
01:38:46 -!- Mathnerd314 has quit (Ping timeout: 240 seconds).
01:40:45 <catseye> need to make roguelike with cursive alphabet now
01:41:13 <Gregor> Are there any Unicode roguelikes?
01:43:04 <pikhq> (granted, it only uses the ASCII subset, but hey.)
01:43:54 <Gregor> Are there any NOT-EXCLUSIVELY-ASCII Unicode roguelikes?
01:44:41 <Sgeo> Dungeon Crawl Stone Soup
01:45:50 <pikhq> None that I know of.
01:45:58 <pikhq> (but there should be)
01:46:04 <catseye> Hey, it's the 13th. Happy backwards Hallowe'en day.
01:46:06 <Gregor> Sgeo: /me fails to find any evidence validating that claim.
01:47:53 <Sgeo> http://crawl.develz.org/learndb/index.html#char_set
01:48:59 <Gregor> Sgeo: The fact that it's capable of using UTF-8 does little to suggest that it actually uses any Unicode.
01:51:21 <Gregor> It could just be that it normally uses "extended ASCII" so they have to be wonky if you NEED UTF-8.
01:51:39 <pikhq> Gregor: It appears to actually use some of Unicode's drawing characters when set to UTF-8.
01:51:48 -!- petabit has joined.
01:52:05 <pikhq> The ones that are analogous to the ones on code page 437.
01:52:15 <pikhq> http://upload.wikimedia.org/wikipedia/commons/f/f8/Codepage-437.png Y'know, this one.
01:53:12 -!- augur has quit (Ping timeout: 240 seconds).
01:54:02 -!- petabit has quit (Read error: Connection reset by peer).
01:58:59 -!- petabit has joined.
01:59:21 -!- petabit has quit (Read error: Connection reset by peer).
02:02:43 <catseye> the only code page that matters
02:05:02 <pikhq> Unless you don't speak a Western European language.
02:05:18 <Sgeo> WHY is Chrome telling me it will take 3 hours to download a 355mb file?
02:05:19 -!- Mathnerd314 has joined.
02:06:28 -!- petabit has joined.
02:07:10 -!- petabit has quit (Read error: Connection reset by peer).
02:09:37 -!- petabit has joined.
02:12:58 -!- petabit has quit (Read error: Connection reset by peer).
02:15:07 -!- petabit has joined.
02:20:18 <pikhq> Wow. The Euro banknote designs are brilliant. To avoid creating anything favoring a particular nation, they had an artist *create* buildings & bridges in various architectural styles.
02:20:42 <pikhq> So, rather than featuring actual landmarks, they feature things that *would be landmarks if they existed*.
02:24:19 <Gregor> It seems I have written Transitions With No Resolution: A work for woodwind quintet
02:26:48 <Gregor> It could also be called Really Obvious Variations: A work for woodwind quintet
02:31:30 -!- augur has joined.
02:45:29 <Gregor> Does anybody know enough about flutes to tell me what the reasonable /human/ upperbound for a concert flute is? (Not the Wikipedia upperbound)
02:50:10 -!- augur has quit (Ping timeout: 272 seconds).
02:57:11 <catseye> i didn't even know wikipedia could play the flute
02:58:18 <Gregor> It's not very good at it.
03:06:16 <pikhq> Gregor: 44.1 kHz. Definitely.
03:13:34 -!- augur has joined.
03:16:39 <quintopia> Gregor: Better name: A Work for Woodwind Quintopia
03:19:07 -!- Sgeo_ has joined.
03:21:14 <Gregor> http://codu.org/tmp/WoodwindQuintet.ogg <-- nonsense for woodwind quintet
03:22:25 -!- Sgeo has quit (Ping timeout: 252 seconds).
03:43:07 -!- augur has quit (Read error: Connection reset by peer).
03:46:38 -!- sftp has quit (Remote host closed the connection).
03:48:35 -!- augur has joined.
03:52:46 -!- augur_ has joined.
03:53:17 -!- augur has quit (Ping timeout: 255 seconds).
03:58:55 -!- jcp has quit (Quit: Later).
04:01:45 -!- GreaseMonkey has joined.
04:06:07 -!- petabit has quit (Read error: Connection reset by peer).
04:06:23 -!- augur_ has changed nick to augur.
04:09:58 -!- petabit has joined.
04:10:13 -!- petabit has quit (Read error: Connection reset by peer).
04:12:59 <catseye> so if I import System.Console.Readline in my hs program does that make it GPL'ed
04:13:05 -!- augur has quit (Ping timeout: 276 seconds).
04:24:01 <catseye> also why do File Browser windows not have a 'start a terminal in this directory' button
04:25:40 -!- petabit has joined.
04:25:52 <catseye> petabit: make up your mind! :)
04:30:11 <quintopia> Gregor: yep, that had some interesting moments but it pretty much went nowhere...
04:32:06 <Gregor> catseye: No, but to do so you must agree to license the /aggregate/ under the GPL. There is a distinction.
04:32:29 <Gregor> I should use a different term than "aggregate" since it sounds like "mere aggregation" which is a different thing :P
04:32:54 <Sgeo_> My CPR class has been moved to 9AM
04:33:01 <Gregor> catseye: (Also, it gets wonky in languages that aren't compiled)
04:33:02 <Sgeo_> I am barely awake at 9AMs
04:33:59 <catseye> Gregor: yeah like someone could provide an implementation of the System.Console.Readline api using libedit and WHAT THEN
04:34:08 <catseye> (well, not a weird question really)
04:35:02 <Sgeo_> Didn't the FSF claim that there was one project that went GPL solely to use ReadLine?
04:35:25 <catseye> Since this is haskell it almost makes more sense to provide some module of higher-order functions that 'readlineify' whatever primitive-input functions you give it
04:35:45 <Gregor> Sgeo_: Projects do do that, because people are stupid and don't realize the distinction between being forced to release your code under a GPL-compatible license and (impossibly) being forced to release your code under the GPL.
04:36:23 -!- petabit has quit (Remote host closed the connection).
04:37:24 -!- augur has joined.
04:41:23 <catseye> hey i want to import functions from two libraries with incompatible licenses
04:42:06 -!- augur has quit (Ping timeout: 252 seconds).
04:42:13 <HackEgo> My hovercraft is full of eels.
04:43:56 <HackEgo> My hovercraft is full of eels.
04:47:32 -!- Mathnerd314 has quit (Ping timeout: 240 seconds).
04:47:57 <HackEgo> My hovercraft is full of eels.
04:48:06 <HackEgo> My hovercraft is full of eels.
04:48:35 <HackEgo> My hovercraft is full of eels.
04:50:35 <HackEgo> My hovercraft is full of eels.
04:51:22 -!- petabit has joined.
04:51:25 * catseye was kind of expecting 'gato'
04:54:23 -!- petabit has quit (Remote host closed the connection).
04:57:35 -!- petabit has joined.
04:57:42 -!- petabit has quit (Remote host closed the connection).
04:58:09 <catseye> one of these days, petabit will make up their mind
04:59:52 -!- augur has joined.
05:00:57 <quintopia> where is an op to kicktempban when you need one
05:01:46 <quintopia> so does wl say "My hovercraft is full of eels." only if it doesn't have the word?
05:01:54 <quintopia> or does it just randomly do it from time to time?
05:05:51 -!- petabit has joined.
05:06:01 -!- petabit has quit (Remote host closed the connection).
05:08:07 -!- petabit has joined.
05:09:17 <catseye> quintopia: the hovercraft message is if the source word is found, but there is no translation found, afaik
05:09:29 <HackEgo> My hovercraft is full of eels.
05:09:37 <catseye> or if the source word is not found, it seems.
05:10:23 <catseye> My denotational arrow has ranged beyond its lower bound.
05:11:04 <catseye> No results found for "denotational arrow", sez Google
05:12:41 <catseye> denotations are functions and functions are arrows!
05:16:49 -!- augur has quit (Ping timeout: 252 seconds).
05:18:11 -!- petabit has quit (Remote host closed the connection).
05:20:32 -!- petabit has joined.
05:21:22 -!- petabit has quit (Remote host closed the connection).
05:29:02 <pikhq> Yes, (->) is an instance of arrow.
05:29:07 <pikhq> Erm, instance of Arrow.
05:30:04 <quintopia> symbolic arrows, not objectual sagitti, yes?
05:31:12 <pikhq> I think for the first time since elementary, I just had homework that I did on automatic.
05:31:25 <pikhq> Math homework, that is...
05:32:00 <pikhq> Just grab a pencil and some paper, and do a bunch of partial derivatives.
05:35:01 -!- antivigilante has quit (Ping timeout: 265 seconds).
05:59:07 -!- Sgeo_ has quit (Read error: Connection reset by peer).
06:00:25 -!- realazthat has quit (Read error: Connection reset by peer).
06:00:54 -!- realazthat has joined.
06:03:36 -!- antivigilante has joined.
06:18:37 -!- augur has joined.
06:20:51 -!- Nostro has joined.
06:21:48 -!- Nostro has left (?).
06:35:30 -!- augur has quit (Remote host closed the connection).
06:35:49 -!- augur has joined.
06:37:32 -!- augur has quit (Remote host closed the connection).
06:37:40 -!- augur has joined.
06:43:40 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
06:43:57 <HackEgo> My hovercraft is full of eels.
06:44:09 <Vorpal> `wl sv en Ålartade fiskar
06:44:11 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
06:44:23 <Vorpal> elliott: it seems broken
06:44:28 <Vorpal> `wl sv en "Ålartade fiskar"
06:44:30 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
06:44:32 <Vorpal> `run wl sv en "Ålartade fiskar"
06:44:33 <HackEgo> My hovercraft is full of eels.
06:46:02 <quintopia> you must be drinking the wrong beer
06:46:09 <Vorpal> quintopia, or rather, yes. It works very well to destroy your liver and increase your weight
06:46:56 <Vorpal> quintopia, why would I be?
06:47:14 <Vorpal> (though you are correct, I am)
06:47:49 <quintopia> any kind of excessive alcohol intake would destroy your liver, so I assume you can't be anti-beer in specific.
06:48:08 <Vorpal> indeed, I'm anti-alcohol in general
06:48:34 <Vorpal> but I don't try to actively convince other people about it unless they bring the whole thing up.
06:49:03 <quintopia> you'll live 20 years longer! you'll get to know the joy of being 95 creaky old years!
06:50:27 <quintopia> unfortunately, you won't be able to enjoy it with tasty alcoholic beverages, so will it be worth it?
06:51:39 <Vorpal> who knows, maybe. bbl, university →
07:00:51 -!- fungot has quit (Read error: Operation timed out).
07:02:05 -!- fizzie has quit (Ping timeout: 276 seconds).
07:04:37 -!- fizzie has joined.
07:05:00 -!- tombom has joined.
07:08:55 -!- wareya has quit (Quit: leaving).
07:21:28 -!- FireFly has joined.
07:22:00 -!- wareya has joined.
07:47:53 -!- wareya has quit (Quit: leaving).
07:48:24 -!- wareya has joined.
07:50:11 -!- tombom has quit (Quit: Leaving).
07:59:40 <augur> i need appropriate hacker music
07:59:43 <augur> someone give me hacker music
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:09:04 <Vorpal> fizzie, http://sporksirc.net/~anmaster/images/misc/uni1.jpg
08:09:44 <Vorpal> will read any comments when I get back, on tethered phone atm.... so kind of slow and not a good idea in the long run and so on.
08:10:01 <fizzie> It's a bit on the blurry side, isn't it?
08:10:30 <Vorpal> fizzie, stitched as rectilinear btw
08:10:38 <Vorpal> fizzie, to straighten up perspective
08:10:58 <fizzie> Yes, it looks that way. Very architectural feel to it.
08:11:08 <Vorpal> fizzie, well yes, that was intentional
08:11:49 <Vorpal> fizzie, also I would show you the first attempt at stitch, but the connection is too slow for me to want to upload more than one image.
08:11:54 <Vorpal> fizzie, it was... curvy
08:11:57 -!- realazthat has quit (Read error: Connection reset by peer).
08:12:03 <Vorpal> like, too much distortion correction
08:12:22 -!- realazthat has joined.
08:12:37 <Vorpal> fizzie, straight lines near the corners looked like they were distorted by relativistic effects!
08:15:00 <Vorpal> fizzie, also I have some material to stitch an image of a long articulated bus here. The colour scheme they use is a kind of pink-lilac around here.
08:15:14 <Vorpal> don't have time to do that now though
08:16:35 <Vorpal> bbl, disconnecting from ssh tunnel to bouncer
08:29:11 -!- atrapado has joined.
08:33:13 -!- FireFly has quit (Quit: swatted to death).
08:40:17 -!- ais523 has joined.
08:41:28 -!- zeotrope has joined.
08:45:44 -!- oerjan has joined.
08:56:58 -!- dbc has quit (*.net *.split).
08:56:58 -!- yiyus has quit (*.net *.split).
08:56:58 -!- atrapado has quit (*.net *.split).
09:07:33 -!- GreaseMonkey has quit (*.net *.split).
09:07:34 -!- sshc has quit (*.net *.split).
09:07:34 -!- oklopol has quit (*.net *.split).
09:07:35 -!- mtve has quit (*.net *.split).
09:07:35 -!- SimonRC has quit (*.net *.split).
09:07:36 -!- quintopia has quit (*.net *.split).
09:07:36 -!- cal153 has quit (*.net *.split).
09:07:36 -!- EgoBot has quit (*.net *.split).
09:07:36 -!- Deewiant has quit (*.net *.split).
09:07:36 -!- olsner has quit (*.net *.split).
09:07:36 -!- Vorpal has quit (*.net *.split).
09:07:37 -!- augur has quit (*.net *.split).
09:07:38 -!- coppro has quit (*.net *.split).
09:07:38 -!- Quadlex has quit (*.net *.split).
09:07:40 -!- Quadrescence has quit (*.net *.split).
09:07:42 -!- wareya has quit (*.net *.split).
09:07:44 -!- Leonidas has quit (*.net *.split).
09:07:44 -!- aloril has quit (*.net *.split).
09:07:45 -!- rodgort has quit (*.net *.split).
09:07:45 -!- ineiros has quit (*.net *.split).
09:07:46 -!- yiyus_ has quit (*.net *.split).
09:21:21 -!- oerjan has quit (Quit: leaving).
09:55:41 -!- atrapado has joined.
09:55:41 -!- dbc has joined.
09:55:41 -!- yiyus has joined.
10:00:15 -!- wareya has joined.
10:00:15 -!- augur has joined.
10:00:15 -!- sshc has joined.
10:00:15 -!- oklopol has joined.
10:00:15 -!- quintopia has joined.
10:00:15 -!- cal153 has joined.
10:00:15 -!- EgoBot has joined.
10:00:15 -!- aloril has joined.
10:00:15 -!- Wamanuz3 has joined.
10:00:15 -!- Deewiant has joined.
10:00:15 -!- rodgort has joined.
10:00:15 -!- Quadrescence has joined.
10:00:15 -!- olsner has joined.
10:00:15 -!- coppro has joined.
10:00:15 -!- mtve has joined.
10:00:15 -!- ineiros has joined.
10:00:15 -!- Vorpal has joined.
10:00:15 -!- SimonRC has joined.
10:00:15 -!- Leonidas has joined.
10:00:15 -!- yiyus_ has joined.
10:00:15 -!- Quadlex has joined.
10:24:22 -!- zeotrope has quit (Read error: Operation timed out).
10:37:35 -!- augur has quit (Remote host closed the connection).
10:37:43 -!- augur has joined.
11:02:02 -!- blaize has joined.
11:09:55 -!- blaize has left (?).
11:27:50 <cheater> so i am working at this energy company
11:27:53 <cheater> and the lights are flickering
11:28:00 <cheater> wonder if they paid their bill. LULZ!
11:57:18 -!- FireFly has joined.
13:45:56 -!- ais523 has quit (Read error: Connection reset by peer).
13:46:54 -!- ais523 has joined.
14:09:31 -!- Slereah has quit (Ping timeout: 240 seconds).
14:16:00 -!- Slereah has joined.
14:22:12 -!- pikhq has quit (Ping timeout: 240 seconds).
14:33:50 -!- Slereah has quit (Ping timeout: 276 seconds).
14:34:23 -!- Slereah has joined.
15:00:42 -!- antivigilante has quit (Ping timeout: 240 seconds).
15:13:57 -!- antivigilante has joined.
15:27:39 <quintopia> i amresentation on a neaural net experiment next thursday
15:42:10 -!- cpressey_ has joined.
15:52:53 -!- MigoMipo has joined.
15:57:54 -!- MigoMipo has quit (Remote host closed the connection).
15:58:27 -!- MigoMipo has joined.
16:03:14 -!- Phantom_Hoover has joined.
16:17:00 <quintopia> http://www.skepticmoney.com/fort-worth-city-councilman-joel-burns-tells-kids-it-gets-better-at-meeting/
16:29:27 <Slereah> That's because this whole channel is made of nerds
16:29:43 <Slereah> obv. they're gonna side with the nerds
16:30:52 <quintopia> i like to side with the underdogs whenever possible
16:31:12 <quintopia> i'm not really all that nerdy, as far as steretypical nerd behavior goes
16:31:15 <Slereah> Are you siding with the neonazis? :o
16:31:30 <quintopia> but i pretty much was in middle school, so that's fair
16:31:52 <quintopia> i said "whenever possible" for a reason
16:32:00 <Slereah> http://abstrusegoose.com/260
16:32:08 <Slereah> "My apologies to Jeff Kinney. I actually think his book, is awesome, but it’s clearly biased against dumb jocks.
16:32:12 -!- sftp has joined.
16:32:14 <quintopia> it's not possible for me to side with the immoral, rude, or otherwise douchey on principle
16:32:51 <quintopia> these kids he's talking about were not even nerds
16:32:57 <quintopia> he even said he himself played basketball
16:33:07 <quintopia> they got picked on solely because they were gay or seemed gay
16:33:58 <quintopia> i'll takes gays over jerks and bigots any day
16:34:12 -!- Slereah has quit.
16:35:01 -!- Slereah has joined.
16:35:31 <Gregor> <Phantom_Hoover> And kind, pleasant bigots?
16:35:43 <quintopia> hmm, i'd probably take the kind, pleasant bigot over the douchey gay, but I would never be close friends with either
16:35:53 <Slereah> Yeah, I find his broad generalization of bigots to be a bit biggoted
16:36:01 <quintopia> but i think the idea of a kind, pleasant bigot is kind of an oxymoron
16:36:19 <Gregor> Y'know, like the old southern rocking-chair bigot. "Back in my day, the negros tended to the land. But today they run 'round killin'. Oh well, would ya like some pie?"
16:36:53 <quintopia> hehe, sure i'll take some pie ... to go.
16:37:08 <Slereah> Or the grandmothers who are physically incapable of stopping using the word negro
16:37:24 <Slereah> "Oh dear, such a nice young negro!"
16:39:13 <quintopia> you can be totally not racist while still not being able to use the currently PC terms
16:40:58 <Slereah> But what if she thinks they are a bit simple minded :o
16:41:07 <Slereah> Abraham Lincoln was a racist.
16:41:25 <Slereah> He did not believe in the equality of races, but that they should have equal rights :o
16:41:30 <quintopia> in 1866, almost all white americans were racist. what of it?
16:41:41 <Slereah> Well, that is not the point.
16:42:00 <Slereah> Are they all people you would not befriend? :o
16:42:27 <Gregor> Well, Lincoln was a republican too. Damned republicans.
16:43:19 <quintopia> well, that was back in the day when republicans were more liberal than democrats iirc
16:43:42 <quintopia> but yes, i would have a hard time befriending any of those people if they were alive in the modern world
16:43:53 <Gregor> Yeah, since then both terms have turned into complete mockeries of themselves :P
16:55:40 -!- tombom has joined.
16:57:13 <quintopia> if you get your jollies from being up the mentally-challenged, you are definitely a bad person
16:59:23 <Phantom_Hoover> The kind of person who thinks that float is a good type for currency variables.
16:59:55 <quintopia> if by that you mean "people who are of normally intelligent but are too close-minded to ever consider doing anything but what requires the least mental effort" then probably not. those people deserve ridicule.
17:00:19 <quintopia> but you have to have a lot of contact with someone before you can justifiably label them that way
17:01:07 <ais523> heh, I was teaching people about floating point inaccuracy today
17:01:15 <ais523> and even managed to slip in 850*77.1 as an example
17:02:05 <quintopia> yes. i have never heard of this example myself.
17:02:15 <ais523> the sum's famous for Excel 2007 getting the answer wrong
17:02:43 <ais523> the just-below-65535 number that's the result of the sum was handled incorrectly, somehow, by the binary->decimal conversion
17:02:53 <ais523> and it displayed the number, as a string, as "100000"
17:03:05 -!- elliott has joined.
17:03:55 <ais523> it is with float arithmetic, assuming I've remembered the sum correctly
17:04:04 <ais523> and I have, just checked
17:04:15 <ais523> the answer's 65535 exactly only in normal maths
17:04:28 <ais523> floats can't represent 77.1 exactly, so it gets rounded down very slightly
17:05:21 <quintopia> are there any production systems that use fractional representations instead of floats at a low level?
17:05:39 -!- augur has quit (Ping timeout: 260 seconds).
17:06:18 <elliott> 17:31:33 <Sgeo> Someone coerced the bot into saying that. That's all I can assume from the delay
17:06:29 <quintopia> Does anyone here like Janssons Frestelse?
17:07:35 <elliott> 17:41:13 <Gregor> Are there any Unicode roguelikes?
17:07:44 <elliott> NetHack's rogue level, if you convert it!
17:08:22 <elliott> 18:20:18 <pikhq> Wow. The Euro banknote designs are brilliant. To avoid creating anything favoring a particular nation, they had an artist *create* buildings & bridges in various architectural styles.
17:08:22 <elliott> Just like Toy Story 4: Debian Versioning.
17:09:10 <elliott> 20:32:06 <Gregor> catseye: No, but to do so you must agree to license the /aggregate/ under the GPL. There is a distinction.
17:09:22 <elliott> rms, and his lawyer, got CLISP relicensed as gpl
17:09:26 <elliott> just because it *could* be linked with readline
17:09:32 <elliott> GPL: Scarier fucking shit than you think.
17:09:44 <Gregor> Wait ... that doesn't even make sense.
17:09:52 <elliott> But the FSF lawyer agreed.
17:09:57 <elliott> And, y'know, that's an actual lawyer.
17:10:03 <elliott> Gregor: Oh yeah, and clisp even CAME with a libnoreadline.a.
17:10:16 <elliott> That, when linked with clisp, provided it same-named functions as readline has, but that just used dumb stuff.
17:10:26 <elliott> Since it's "intended" to be linked with readline...
17:10:37 <Phantom_Hoover> So wait, any software that allows itself to be used with RMS' spawn has to be GPLed?
17:10:40 <elliott> I suggest not trying to make sense of the law because it makes none.
17:10:46 <elliott> Phantom_Hoover: Pretty much, yes.
17:10:58 <elliott> Of course rms might have a harder time convincing a court...
17:11:02 <elliott> but the FSF lawyers certainly think so
17:11:05 <elliott> and it's certainly his intention.
17:11:11 <elliott> Phantom_Hoover: Nobody else was doing it, really.
17:11:14 <elliott> This was in the early 90s.
17:11:31 <Gregor> elliott: Was it under something non-GPL-compatible before? That's the only explanation I can remotely think of ...
17:11:54 <elliott> Gregor: The code was readable-and-patchable-but-proprietary, I think.
17:12:09 <elliott> But readline wasn't distributed with it, and it provided a full method to link a perfectly-working CLISP.
17:12:21 <elliott> It's just that you could s/libnoreadline.a/libreadline.a/ and have it use readline.
17:12:26 <elliott> And since this was "intended" functionality...
17:12:30 <Gregor> Well, you've walked into much murkier waters now.
17:12:40 <elliott> The fucking thing didn't distribute readline.
17:13:04 <ais523> my own personal theory about this, is that if a user linked it with readline, it would be them who was breaking the law
17:13:07 <elliott> rms is arguing that since *IT CALLED FUNCTION PROTOTYPES THAT HAPPEN TO BE IN READLINE* and because Bruno Haible *thought in his mind* that people would link it with readline, it was copyright violation.
17:13:16 <elliott> This is *fucking insane* and not murky at all.
17:13:18 <ais523> but the software itself wasn't
17:13:26 <elliott> tl;dr stop using the GPL, it's utterly insane
17:13:33 <ais523> so perhaps they could be found guilty of facilitating copyright infringement, but nothing else
17:13:51 <ais523> anyway, the GPL is relatively sane; some of its advocates aren't
17:14:02 <elliott> 20:35:45 <Gregor> Sgeo_: Projects do do that, because people are stupid and don't realize the distinction between being forced to release your code under a GPL-compatible license and (impossibly) being forced to release your code under the GPL.
17:14:10 <elliott> You can't use readline in a BSD-licensed project.
17:14:38 <Gregor> "I don't agree. My lisp.a is not a "work based on libreadline.a". What I distribute is a "mere aggregation" of lisp.a and libreadline.a - the latter with source." // this clisp email discussion is suggesting something very different from what you're saying
17:14:53 <elliott> Gregor: Yes -- he then offered to stop doing that.
17:14:57 <elliott> And rms said that that would not suffice.
17:15:32 <Gregor> In spite of your statements to the contrary, this IS a murky area. What he's distributing is COMPILED BINARIES, not source.
17:15:42 <elliott> Gregor: *HE THEN OFFERED TO STOP DOING THAT*
17:16:22 <Gregor> As far as I can tell, he offered to stop distributing libreadline.a, distribute a libclisp.a that was still compiled with readline headers, and a libnoreadline.a that provides the same interface.
17:16:29 <elliott> 22:44:32 <Vorpal> `run wl sv en "Ã…lartade fiskar"
17:16:41 <ais523> elliott: here's a fun license question: NetHack is under the NHPL, a trivial variant of the GPL0. Suppose I disassemble one of the executables that nethack.org provides, to extract a library function from it that wasn't in the source they provided
17:16:48 <elliott> Gregor: You can't exactly prove it was compiled with those headers.
17:16:49 <ais523> can I post that disassembly?
17:16:55 <elliott> That doesn't affect the binary.
17:17:10 <elliott> i thought you were asking like
17:17:30 <Gregor> elliott: It can affect the binary. And you DON'T know, that's the whole issue. Your source is still being intermixed with GPL'd source before creating a .o.
17:17:30 <elliott> ais523: wait, they include extra things in the binary builds? or is this a hypothetical?
17:17:40 <ais523> elliott: it's the libc, it's linked statically
17:17:51 <ais523> they probably didn't even notice they were theoretically supplying it
17:17:53 <elliott> Gregor: Yeaaah I don't believe a handful of function prototypes can be copyrighted.
17:18:10 <elliott> ais523: depends on the libc license, surely
17:18:37 <ais523> it's a license that lets them ship binaries, at least
17:18:41 <ais523> I am also completely muddled
17:18:52 <ais523> especially as I have the source for the libc in question, but not the exact version
17:18:56 <ais523> and the asm doesn't match the source I have
17:19:38 <Vorpal> <elliott> 22:44:32 <Vorpal> `run wl sv en "Ã…lartade fiskar" <-- unicode fail
17:20:00 <elliott> Vorpal: what's it actually?
17:20:06 <elliott> i'll see if it works used properly
17:20:23 <Vorpal> `run wl sv en Ålartade_fiskar
17:21:12 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
17:21:16 <Gregor> Yeah, elliott, you're majorly oversimplifying the issue. You are allowed to distribute a piece of software under the MIT license that uses a GPL'd library. If you distribute a binary /linked to use the GPL library, whether it includes its implementation or not/, then you must release that binary under the GPL. But you can release the source under MIT, somebody can rip out the GPL dependency, and then they can produce an MIT-redistributable binary. RMS wanted C
17:21:16 <Gregor> LISP released under the GPL and not some compatible license because RMS is RMS, not because that was the legal obligation.
17:21:29 <elliott> en is totally useless there
17:21:44 <elliott> Gregor: It's not linked to that.
17:21:55 <elliott> Vorpal: because it's automatic
17:22:05 <elliott> Gregor: Also, rms explicitly states that his lawyer said that Bruno had to relicense as *GPL*.
17:22:06 <Gregor> I didn't say linked to
17:22:11 <elliott> Gregor: Linked to use. Define this.
17:22:33 <Gregor> elliott: Compiled using copyrighted headers and linked against the interface of.
17:22:46 <elliott> "Linked against the interface of"
17:22:53 <elliott> so if someone releases a set of GPL'd functions
17:23:02 <elliott> you can't use another library that has functions with the same prototype
17:23:06 <elliott> because that counts as using gpl'd software?
17:23:07 <ais523> I think the current wisdom is to do with the definedness of the API
17:23:26 <Gregor> elliott: You can't link against THAT HEADER.
17:23:27 <Vorpal> elliott, strange that eel translates to "eel-like fish"
17:23:41 <elliott> Gregor: That header isn't copyrightable.
17:23:46 <elliott> Vorpal: It's just interwiki.
17:24:06 <Vorpal> elliott, strange interwiki still
17:25:02 <ais523> does `wl do "translation" via interwikis?
17:25:16 <ais523> could be a bit risky, given how interwikis work
17:25:25 <elliott> ais523: but a bit more too:
17:25:25 <Gregor> OK, I have no time for this silly argument. It is a fact that you can release source under any GPL-compatible license even if it's designed to use a GPL'd library.
17:25:28 <cpressey_> Phantom_Hoover: re how currencies are stored: oh, the stories I could tell.
17:25:32 <Gregor> And that is what I claimed.
17:25:51 <elliott> `wl fr Mon_aéroglisseur_est_plein_des_anguilles.
17:25:54 <HackEgo> My hovercraft is full of eels.
17:26:09 <ais523> elliott: there's an actual interwiki there?
17:26:17 <elliott> ais523: no, it interwikis the individual words
17:26:21 <elliott> and then looks up a word order table online
17:26:33 <elliott> ais523: (I'm joking; that's its "no translation found" error.)
17:26:49 <Gregor> elliott: From GNU's GPL FAQ: "If a library is released under the GPL (not the LGPL), does that mean that any program which uses it has to be under the GPL or a GPL-compatible license? Yes, because the program as it is actually run includes the library."
17:26:52 <elliott> `wl en fr My_hovercraft_is_full_of_eels
17:26:54 <Gregor> Note the "GPL-compatible" part
17:27:04 <elliott> Gregor: Did it say that in the 90s?
17:27:14 <ais523> `wl en it My_hovercraft_is_full_of_eels
17:27:40 <ais523> (picking Italian here, because there is actually an it: interwiki from that article on en...)
17:27:44 <Gregor> elliott: archive.org has only back to 2001, but it said it then :P
17:27:46 <HackEgo> My hovercraft is full of eels.
17:28:04 <elliott> But yeah, it works to Italian.
17:28:05 <ais523> I wonder why mine finished before yours?
17:28:13 <cpressey_> < Phantom_Hoover> I'd assume that there are lots. <-- yes, well... that's the thing see...
17:28:17 <ais523> elliott: except that it isn't a translation at all
17:28:30 <ais523> the interwiki only goes from article covering a subject, to another article covering the subject
17:28:39 <elliott> when Vorpal mentioned a fish "sill"
17:28:44 <elliott> and couldn't remember the english word for it
17:29:39 <cpressey_> < elliott> just because it *could* be linked with readline <-- but does this means that all Lisp programs that can use CLISP's readline facilities are also GPL-liable? I wouldn't think so, but...
17:30:08 <elliott> cpressey_: no, it doesn't apply to interpreters
17:30:15 <elliott> and compilers have special exceptions
17:30:34 <elliott> but true, if clisp exposes a readline api
17:30:38 <elliott> then they get the same treatment
17:30:52 <elliott> (of course if it renames the functions and jiggles the arguments it's an ~ABSTRACTION LAYER~)
17:31:15 <ais523> you know, I think RMS may be one of the only people in the world who spreads FUD about his own license
17:31:50 <quintopia> who is this root-mean-squared guy again?
17:32:38 <ais523> why would someone called "paul" have an R as their first initial?
17:32:55 <ais523> (that said, I've seen people whose initials didn't match their names before, presumably due to one using contractions and the other not...)
17:32:57 <cheater> because the additional stroke is his big penis
17:34:37 <cheater> ais523: it's easy for this to happen with titles and other things like that (e.g. Jean-Luc or d'Alembert)
17:35:03 <cpressey_> < HackEgo> Eel < HackEgo> Eel < HackEgo> Eel <-- I hear ya, dude.
17:36:07 <cpressey_> Phantom_Hoover: the stories would be too depressing
17:36:31 <elliott> Phantom_Hoover: in reply to?
17:40:55 <elliott> does anyone know how to draw a hollow square in GIMP?
17:42:53 <quintopia> um, draw a square, select the center, delete?
17:45:20 -!- cal153 has quit (Ping timeout: 255 seconds).
17:46:14 -!- oerjan has joined.
17:49:00 <cpressey_> < elliott> (of course if it renames the functions and jiggles the arguments it's an ~ABSTRACTION LAYER~) <-- sudo apt-get install rlwrap ;; making the whole thing moot for a whole array of things -- unless it's even worse than I thought
17:49:13 <elliott> cpressey_: just use libedit :P
17:49:59 <cpressey_> elliott: My haskell programs are at the mercy of the haskell implementation. Is there a pragma for "don't seat me next to rms"?
17:50:56 <elliott> cpressey_: Just use sufficiently advanced type system hackery.
17:51:04 <elliott> cpressey_: Anyway, GHC uses libedit.
17:51:35 <cpressey_> maybe i'ma gonna have to get a more recent ghc 'cuz i swear... anyway.
17:54:08 <cpressey_> Phantom_Hoover: OK, one, but it'll have to be really quick. I was once on a team with a *PhD* who defended his code change which *extracted the only-conventionally-private float member out from the currency object* and did floating point math on it, on the grounds that he was "creating an abstraction" which "the programmer might find easier to use" (presumably because he didn't understand how to use the methods of the currency object.)
17:54:26 -!- augur has joined.
17:59:47 <olsner> but but... how can breaking abstractions be construed as creating them?
18:00:40 <oerjan> http://www.mezzacotta.net/garfield/ is very topical today
18:01:14 -!- ais523 has quit (Read error: Connection reset by peer).
18:02:18 <oerjan> olsner: see: n-categories, quantum groups
18:03:14 <oerjan> and probably heaps of other examples in math
18:05:21 <cpressey_> oerjan: i want to invent "denotational arrows". they sound like they have such lovely properties.
18:13:08 -!- cal153 has joined.
18:17:59 -!- atrapado has quit (Quit: Abandonando).
18:42:29 <elliott> In which my hair is longer than reality itself: http://i.imgur.com/3tiaR.jpg
18:44:28 <fizzie> Maybe you should /nick eHAIRd, then.
18:46:27 <elliott> My hair is about that long, but I do in fact have a properly-formed mouth, non-hollow eyes, and other features.
18:50:28 <elliott> I guess my hair could be hiding the border around my torso there, actually.
18:50:34 <elliott> SO I GUESS IT IS AN ACCURATE DEPICTION AFTER ALL
18:50:45 <Gregor> Nom nom Graggo eat babies.
18:51:29 <Gregor> NOM NOM GRAGGO[R] EAT BABIES
18:53:37 <oerjan> probably just a local zombie epidemic
18:54:02 <Gregor> Local ... to your living room!
18:54:33 <oerjan> impossible. there are no babies to eat here.
18:55:25 <Gregor> Then I guess I'll just have to settle for YOUUUUU
18:59:31 -!- augur has quit (Read error: Connection reset by peer).
19:02:04 -!- augur has joined.
19:24:22 <cpressey_> one of the problems with a ring language is that you need some program (0) that both does nothing to any program (a = a + 0) and brings every program to its knees (a * 0 = 0).
19:29:29 <elliott> cpressey_: well if + is "execute in parallel, first one to finish wins" and * is "sequential execution"...
19:29:33 <elliott> iirc there are other problems with that though
19:29:39 <elliott> but by those definitions, 0 = _|_
19:29:52 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
19:30:28 <Gregor> elliott: If '+' is FIRST one to finish wins, then a + 0 is 0.
19:30:43 <elliott> Gregor: Uhh, since when does an infinite-looping program finish first.
19:30:54 <elliott> Or are you assuming 0 is halts-immediately without any reason to?
19:31:10 <Gregor> Since when is .... ohh, 0 isn't the null program, it is a program that fits its shape in this case X-P
19:32:29 <elliott> cpressey_: conal wrote a haskell package implementing an operator, let's say +, where a + _|_ = a, _|_ + b = b, and a + b = either a or b when neither a nor b is _|_
19:32:36 <elliott> cpressey_: (by just using threads)
19:32:39 <elliott> but that's basically what + is here
19:32:42 <elliott> except the a + b case would be defined
19:32:48 <elliott> it was called race or something
19:34:04 -!- Phantom_Hoover has joined.
19:39:44 <elliott> cpressey_: NetBSD's filesystem is called FFS :D
19:40:00 <Vorpal> elliott, was there any practical use for that haskell package?
19:40:14 -!- pikhq has joined.
19:40:18 <elliott> the actual semantics were that it would return whichever succeeded first
19:40:34 <elliott> so you could run two algorithms in parallel, both of which have edge-cases that blow up but the other handles
19:41:24 <cpressey_> < elliott> cpressey_: well if + is "execute in parallel, first one to finish wins" and * is "sequential execution"... <-- you just described Cabra
19:41:43 -!- antivigilante has quit (Ping timeout: 240 seconds).
19:42:17 <elliott> cpressey_: then how is it a problem for rings?
19:42:22 <elliott> also, i thought you had already done that, yeah
19:42:43 <cpressey_> elliott: yeah it's not a ring. it's a dioid. it's a problem because of distributivity iirc
19:46:49 <elliott> x SEQ (y PAR z) = (x SEQ y) PAR (x SEQ z)
19:47:00 <elliott> cpressey_: that doesn't seem so hard to me
19:47:07 <elliott> or is right-distributivity the issue?
19:47:18 <elliott> (y PAR z) SEQ x = (y SEQ x) PAR (z SEQ x)
19:47:27 <Vorpal> Gregor, btw did I mentioned that I liked that GRegor-op11-StringQuartet-VSTi-2010-10-10.ogg?
19:47:36 <elliott> cpressey_: admittedly i'm finding it a little hard to reason about this, but I don't see any problem in either of those
19:48:23 <Vorpal> Gregor, just wondering one thing: what instruments? There is obviously some strings and something that sounds like a cross between an organ and brass?
19:49:03 <Gregor> Vorpal: It's a string quartet, but the cello in this instrument set has a problem of sounding a bit brassy.
19:49:20 <elliott> Problem schmoblem, it's a FEATURE!
19:49:48 <Vorpal> Gregor, hm, also: VSTi?
19:50:15 <elliott> "a lot of people think that a fax machine sends the piece of paper rolled up really tight over the existing lines... these people are very misinformed. it's only the INK that goes over the lines." --reddit
19:50:22 <elliott> Vorpal: http://www.google.co.uk/search?&q=vsti
19:50:25 <elliott> Vorpal: http://www.google.co.uk/search?q=vsti
19:50:35 <elliott> Or even http://www.google.com/search?q=vsti
19:50:36 <Vorpal> elliott, both works :P
19:50:39 <Gregor> Vorpal: Oh, btw, http://codu.org/music/GRegor-op11-StringQuartet.ogg is a newer and somewhat improved one. And VSTi == Virtual Studio Technology instrument. It's a system for MIDI instruments that are programmatic rather than patch-based.
19:50:53 <elliott> Oops! This link appears to be broken.
19:51:05 <Gregor> * http://codu.org/music/op11/GRegor-op11-StringQuartet.ogg
19:51:22 <Vorpal> Gregor, oh I see. Do you mean synthed or something else with "programmatic" here?
19:52:14 <Gregor> Vorpal: Well, both VSTi's and patch-based systems e.g. soundfonts are synthesized, by that I mean that the instruments are actually software, and so can range from patch-based (like existing systems) to full-on physical modeling of the instrument. Usually they're somewhere in between.
19:52:36 <Vorpal> Gregor, also soundfont is sampled right?
19:53:13 <Gregor> Well, I guess that depends on your definition of "synthesized" ... they're sampled, then those samples are munged around to get the full range, looping, etc.
19:53:30 <Vorpal> oh yeah, indeed depends on definition
19:54:21 <Vorpal> Gregor, I think the piano is even better btw
19:54:39 <elliott> 3:16 is a bit jarring but I guess that's intentional :p
19:54:46 <Vorpal> Gregor, but that could just be due to the "strings sounding like brass" issue
19:56:19 <pikhq> Apparently, Firefox until 4.0 used a single heap for Javascript stuff...
19:56:35 <pikhq> I'm surprised they've even got all that still set up in a manner that it's *possible* to have a single heap for that.
19:56:44 <cpressey_> elliott: i think right distributivity was also a problem, but consider also: every program must have a unique additive inverse. for every program a, there is a unique program b that, when run in parallel, results in _|_. how?
19:56:53 <Vorpal> pikhq, do you mean the malloc() heap?
19:56:55 <pikhq> Erm. Well, not possible, but rather sane.
19:56:58 <pikhq> Vorpal: The GC heap.
19:57:07 <elliott> cpressey_: The additive inverse of a program a is a program that does the opposite of a at every step.
19:57:11 <Vorpal> pikhq, indeed, very strange
19:57:21 <pikhq> Vorpal: A single heap for *every individual bit of Javascript* running.
19:57:25 <elliott> So, for instance, if a at one step makes the flag it's looping on to false, the additive inverse will set it to true at the same time.
19:57:30 <pikhq> Vorpal: Keep in mind that the *UI* is in Javascript.
19:57:45 <olsner> it's easier to have a global heap than to keep stuff apart :P
19:58:17 <pikhq> olsner: Except that they should have been running each tab in a seperate thread for ages now.
19:58:29 <pikhq> olsner: Making a global GC heap a *royal pain* to do well.
19:58:41 <pikhq> Unless, of course, you just stop the world, defeating the whole point.
19:58:44 <olsner> really? I didn't think firefox threaded at all
19:58:57 <pikhq> That... Explains even more of its suckitude.
19:59:18 <Vorpal> pikhq, separate threads. Well, I have 38 tabs open. It's a single threaded machine. Chrome is way slower than firefox on this box
19:59:24 <Vorpal> when it comes to such things
19:59:31 <Vorpal> and it uses even more RAM
19:59:58 <pikhq> olsner: Threading should be kinda the natural structure there...
20:00:10 <Vorpal> okay, so v8d beats firefox sometimes, sure. But the whole thing is sluggish in general
20:00:12 <pikhq> Vorpal: Are you looking at top and summing the RAM reported?
20:00:27 <pikhq> Vorpal: If so, you're a moron.
20:00:32 <cheater99> you know, the funny thing about vsts is that they all sound the same
20:00:34 <Vorpal> pikhq, no, I'm checking free -m after echo 5 > /proc/sys/vm/drop_caches
20:00:38 <olsner> pikhq: opera is single-threaded :)
20:00:43 <cheater99> the only thing that changes is the interface
20:00:47 <Vorpal> with empty firefox and empty chrome open respectively
20:01:05 <pikhq> Vorpal: Ah, *empty*. See, where Firefox gets you is when you *use* it.
20:01:23 <cheater99> firefox gets so hogged up by open websites
20:01:27 <Vorpal> pikhq, test same with 10 tabs open. The difference is smaller, but still there. Also firefox is faster.
20:01:35 <olsner> I don't think threading has ever been the natural state for browsers - you're usually displaying a single tab on a single screen on (at least in the past) a single cpu
20:01:37 <Vorpal> I hesitate to call it snappy, because none of them is
20:01:40 <pikhq> Vorpal: Which Firefox, which distroy?
20:01:55 <cheater99> it's like i'm using a website and every action/click and every time i scroll makes firefox add a usleep(5) at the end of its event loop
20:01:56 <olsner> from there, you'd rather go multi-process on it than to introduce yucky deadlocky racey threads
20:01:58 <pikhq> olsner: The thing is, 1 thread *per tab* seems very natural.
20:02:01 <Vorpal> pikhq, distro is arch linux. Firefox is one called namoroka. 3.6.10
20:02:10 <pikhq> olsner: I mean, there's almost no communication between them.
20:02:16 <cheater99> olsner: all tabs are being run at the same time
20:02:19 <cpressey_> elliott: but then how does it become _|_ in the end, rather than nop?
20:02:35 <Vorpal> pikhq, and chromium version? Not sure, I uninstalled it after my tests. Was about half a month ago I tested
20:02:41 <Vorpal> and last version at that time
20:02:43 <cheater99> pikhq: sharing of caches and pre-compiled DOMs and other objects
20:02:44 <elliott> cpressey_: well if a then loops on the flag, if -a gets it false rather than a getting it true, it...
20:02:59 <pikhq> Vorpal: 7.0.517.36, built from the ebuild www-browser/chromium
20:03:02 <Vorpal> pikhq, same for firefox, so firefox might have been at 3.6.9, same first two digits
20:03:06 <elliott> cpressey_: do we need -(-a) too?
20:03:28 <Vorpal> <olsner> I don't think threading has ever been the natural state for browsers - you're usually displaying a single tab on a single screen on (at least in the past) a single cpu <-- indeed
20:03:30 <cpressey_> not saying it's not possible, obviously, just harder than it looks at first glance
20:03:55 <olsner> cheater99: what do you mean "wrong"?
20:04:03 <elliott> cpressey_: we even need foo + -foo to work even if foo is just like... increment by one
20:04:17 <elliott> cpressey_: ok, maybe 0 isn't _|_ then
20:04:18 <cheater99> olsner: "youre displaying a single tab on a single screen" is wrong train of thought.
20:04:46 <fizzie> Firefox Mobile (Fennec) has that Electrolysis thing, which puts browser UI, web-content rendering and plugins in completely separate processes; I'm not sure if it's one process per tab, chromium-like, or something else.
20:04:50 <pikhq> cheater99: Okay, but it *still* seems like you'd want to do threading for that.
20:04:57 <elliott> cpressey_: _|_ doesn't necessarily mean loop forever!
20:04:59 <elliott> cpressey_: it can just mean fail
20:05:09 <cheater99> pikhq: i'd just have separate threads for separate domains
20:05:16 <elliott> a + 0 = a; -- 0 raises the error but + ignores it and picks the other
20:05:16 <Vorpal> fizzie, sounds like a rather large and invasive change
20:05:23 <elliott> a * 0 = 0; -- a runs, then 0 runs and flags an error
20:05:35 <olsner> cheater99: IMO the common case is that you have only one tab visible at a time
20:05:46 <fizzie> Vorpal: Maybe, but they've implemented it in the recently-released beta.
20:05:47 <elliott> cpressey_: then we just have, for every instruction a, -a which is a version of a, say, with the Flag An Error For No Reason flag on
20:05:48 <cheater99> olsner: and that doesn't mean jack
20:05:52 <elliott> cpressey_: so -(-a) = a because it flips it twice
20:05:54 <cheater99> olsner: because of what i told you earlier
20:05:58 <pikhq> olsner: Aaaand all the tabs *still have things to do*.
20:05:59 <Vorpal> fizzie, as in "difficult to keep code synced to standard browser"
20:06:00 <elliott> and -a always throws an error for a "positive" (non-erroring a)
20:06:08 <pikhq> olsner: Javascript doesn't just stop because the page isn't being rendered.
20:06:17 <pikhq> olsner: Nor does Flash (the abomination).
20:06:47 <fizzie> Vorpal: I think at least the "out-of-process plugins" bit is something they're trying to put into the desktop browser too.
20:06:50 <cheater99> which is the crucial matter to showing how failed all the internet standards are
20:07:05 <pikhq> Basically, to do a tabbed interface, you are having to either implement something vaguely similar to multiple processes or *actually do that*.
20:07:15 <cheater99> this single situation shows they were all invented by idiots and cobbled together by human-monkey crosses
20:07:19 <pikhq> Because it's all running stuff.
20:08:00 <cheater99> they all use the exact same framework, exact same objects just copies of them, exact same layout
20:08:03 <pikhq> Presumably Firefox, Opera, etc. are just implementing a scheduler on a single thread...
20:08:11 <cheater99> yet, they all have to be calculated 20 times!
20:08:20 <Vorpal> fizzie, has there been any updates on the n900-replacement front? I guess you are likely to know this better than me.
20:08:29 <olsner> pikhq: so they're cooperatively threaded then :P
20:09:01 <fizzie> Vorpal: The plugin thing seems to be the first bullet point in the "what to release in Firefox in 2010 after 3.6" roadmap: "Multi-process plugins (Flash crashes/instability don't bring down the browser)"
20:09:08 <pikhq> olsner: Which is, ah, moronic.
20:09:21 <fizzie> Vorpal: Do you mean hardware-wise, or software-wise?
20:09:22 <Vorpal> pikhq, they could be timeshared?
20:09:41 -!- realazthat has quit (Read error: Connection reset by peer).
20:09:45 <cheater99> pikhq: i love the fact that in firefox 3.5 the freaking websites and http/rendering engine lock up the fucking gui
20:10:02 <olsner> pikhq: among other things, such a structure allows you to run on operating systems that don't have threads :)
20:10:05 <Vorpal> fizzie, hm isn't plugins separate processes already? I seem to remember that killall java works well when java in firefox hangs
20:10:07 -!- realazthat has joined.
20:10:27 <cheater99> olsner: which is something that is only interesting to you personally.
20:10:35 <fizzie> Vorpal: Possibly some plugins (like java) do their own "another process + IPC" thing, but it's not comprehensively so.
20:11:02 <pikhq> olsner: Yes, yes, the *highly important* DOS port.
20:11:27 <pikhq> Vorpal: That's just how Java's implemented.
20:11:28 <fizzie> At least nspluginwrapper needs to do a separate process; possibly Linux Flash in general. I think on Windows it might not necessarily be.
20:11:35 <pikhq> cheater99: Only if you use nspluginwrapper.
20:11:46 <cheater99> pikhq: which happens without my knowledge
20:11:50 <Vorpal> fizzie, so what about the hw replacement thing?
20:11:53 <pikhq> fizzie: No, it is very much just run in the same process.
20:11:58 <Vorpal> fizzie, I haven't heard anything more about n9
20:12:37 <olsner> cheater99: it is interesting for everyone who wants a browser on such an OS :)
20:12:39 <elliott> cpressey_: TOTALLY WILL get speccing
20:12:58 <elliott> cpressey_: (even if giving every command an error flag is weird...)
20:13:14 <fizzie> Vorpal: Well, they officially announced the Symbian^3 "flagship" models back then, I think it was somewhat discussed. All the rest is just rumours; there's been some more N9 rumours too, including specs, but they're all out-of-someone's-hat stuff.
20:13:53 <Vorpal> fizzie, presumably they are not just going to drop meego?
20:14:18 <fizzie> Vorpal: I guess it's pretty safe to say that N9 will do capacitive multitouch; rumour has it it'll be built on the Qualcomm SnapDragon 1GHz platform, and do MeeGo.
20:14:41 <Vorpal> fizzie, does n900 use resistive?
20:14:59 <Vorpal> fizzie, so the n9 won't work with gloves on in the winter?
20:15:23 <fizzie> Maybe if you get one of those capacitive-stylus tips for each of your fingertips in the glove.
20:15:29 -!- wareya_ has joined.
20:15:36 <cheater99> capacitive stylus works with capacitive
20:15:47 <fizzie> cheater: Yes, but we were talking about the N9.
20:16:51 <fizzie> Vorpal: Well, the N900 PR1.3 firmware, which is supposed to be out "soon", will make it "easy" (blog-post quotations) to dual-boot between Maemo and latest-weekly-MeeGo-image (via kexec, I think), so there's something going on on the MeeGo front.
20:17:23 <fizzie> Vorpal: Oh, and latest MeeGo-on-N900 can make phone calls (with the oFono stuff) already. :p
20:17:24 <elliott> Things I didn't expect to see today: A 43-minute dramatisation of Karen Carpenter's anorexia played entirely by Barbie dolls.
20:18:37 <fizzie> Vorpal: It's "a free, open source project for mobile telephony (GSM/UMTS) applications"; an Intel/Nokia project. Something like an (dbus-based) API for apps that want to do something related to phone calls.
20:19:02 -!- wareya has quit (Ping timeout: 264 seconds).
20:19:25 <Vorpal> fizzie, UMTS, is that same as GPRS or same as 3G?
20:20:07 <fizzie> But of course it's only an application-level API, the actual low-level stacks will still probably be binary blobs.
20:20:40 <fizzie> Though I think I heard something open-sourcey modemy things in a MeeGo context too. Haven't really looked very closely at it.
20:20:47 <Vorpal> hm does it send the voice data over dbus too?
20:20:52 <olsner> sounds gnome:ish - will I have to fiddle with 7 concurrent sound API:s to make phone calls work then?
20:20:54 <elliott> Why does NetBSD suck at installing in QEMU?
20:21:07 <fizzie> Vorpal: No, that's done with pulse, I think. I'm not sure if oFono mandates that, though.
20:21:21 <elliott> (I like to think pikhq just insults me for no reason)
20:21:30 <olsner> elliott: API's and gnome'ish?
20:21:57 <fizzie> I sort of would like to see a nice telephony API; I've always wanted to stick one of those horrible navigate-through automated phone menus (maybe with some command-recognition and speech synthesis) on my phone. I don't get phone calls from anyone else than telemarketers, anyway.
20:22:17 <Vorpal> olsner, hm yeah I think that was a Swedishism
20:22:22 <olsner> I usually use "'" for that stuff in swedish, where I believe : is what you're supposed to do
20:22:38 <Vorpal> olsner, just swap the behaviours then
20:23:04 <Vorpal> elliott, "API:er" would be the way to pluarlise it in Swedish I believe
20:23:17 <pikhq> Y'know, the entire telephone infrastructure sucks.
20:23:21 <Vorpal> though I have to admit I'm not 100% certain
20:23:26 <elliott> LET'S REINVENT THE TELEPHONE
20:23:42 <Vorpal> pikhq, less so in Europe
20:23:45 <fizzie> "Please repeat the next thirty words after the beep to make me answer the phone." "Sorry, my speech recognition system was built by the lowest bidder; please repeat." "No, I still didn't quite get it: please repeat." "Sorry, reached the limit on the number of attempts you are allowed to do. Better luck next time."
20:23:45 <elliott> LET'S REINVENT ELECTRICITY
20:23:57 <olsner> elliott: at least for acronyms with suffixes it's ':', gnomeish would just be gnomskt or something like that
20:24:06 <pikhq> For some reason it makes sense to run copper wires from a gigantic Internet-connected computer to a microphone & speaker, alongside the Internet signal.
20:24:29 <pikhq> Rather than just... Use the Internet for all but the last few feet.
20:24:45 <Vorpal> pikhq, historic reasons
20:24:54 <pikhq> (yes, pretty much everything but the last mile is already on the Internet)
20:25:08 <Vorpal> pikhq, also, for me the internet signal goes over that copper wire (ADSL)
20:25:15 -!- cpressey_ has quit (Ping timeout: 240 seconds).
20:25:29 <pikhq> Vorpal: Yes, and it should damned well be replaced.
20:25:57 <Vorpal> pikhq, it works well. *shrug*
20:25:59 <pikhq> Especially in the US, where fiber-to-the-home was *paid for 15 years ago*.
20:26:17 <fizzie> Also there's the reliability argument; no-one wants packet loss on their emergency calls.
20:26:20 <coppro> the taxpayer, probably
20:26:22 <pikhq> Vorpal: The United States of America.
20:26:37 <Vorpal> pikhq, and it never happened?
20:26:43 <pikhq> Vorpal: We gave a few *billion* in tax credits to phone companies for the purpose of funding fiber-to-the-home.
20:26:47 <Vorpal> pikhq, then why the fuck don't they demand the money back
20:26:54 <pikhq> The phone companies responding by doing fuck all.
20:27:03 <elliott> Vorpal: because the US is a corporatocracy
20:27:07 <Vorpal> pikhq, didn't they give it with a clause attached?
20:27:17 <pikhq> They did. But the US is a corporatocracy.
20:27:21 <Vorpal> elliott, oh good point
20:29:00 <pikhq> And now we wonder why the Internet sucks.
20:30:40 <pikhq> Another thing: there is *absolutely no point in all* in how cable service is usually done here. We use a few channels on the cable for Internet, and all the rest for TV. Instead of multicast IP, and just offering a few gigabits/sec Internet...
20:31:05 -!- cpressey_ has joined.
20:31:11 <cpressey_> 1791035613 [main] irssi 3988 exception::handle: Exception: STATUS_ACCESS_VIOLATION
20:31:11 <Gregor> pikhq: Hey! I need my TV.
20:31:15 <pikhq> Of course, to make that sane we'd need to have multicast on the public Internet.
20:31:26 <pikhq> Gregor: Imagine plugging an Ethernet jack into it.
20:31:35 <Vorpal> pikhq, Aren't there a limit to number of multicast groups iirc?
20:31:44 <Gregor> pikhq: No, you'd only need multicast on the cable networks' local networks.
20:31:57 <Gregor> Hell, you'd really only need broadcast flood.
20:32:03 <pikhq> Gregor: No, you see, the idea is you get the TV signal directly from the broadcaster.
20:32:03 <Vorpal> pikhq, also there is a sense to that. You would easily overrun the backbone
20:32:12 <Vorpal> if every american had a few gbps to his computer
20:32:30 <Vorpal> pikhq, you would need to scale up the backbone a LOT
20:32:32 <pikhq> Gregor: Imagine being able to watch any TV signal *in the world*.
20:32:35 <pikhq> Vorpal: We should!
20:32:56 <elliott> Fuck building things around TV.
20:33:17 <elliott> We dictate your audio-visual schedule for the day! Put us in your living room.
20:33:20 <pikhq> Vorpal: If the US cut off, say, 1/4 of the "defense" budget, it could pay for the necessary upgrades this year.
20:33:25 <elliott> Vorpal: I have no problem at all with watching tons of TV material.
20:33:31 <elliott> I just dislike the scheduled format.
20:33:39 <Vorpal> pikhq, fat chance though
20:33:45 <elliott> pikhq: You fool! That goes to the Stargate program.
20:34:04 <Vorpal> elliott, did I say something else?
20:34:25 <elliott> cpressey_: It is actually a very good show, despite Sgeo liking it :)
20:34:26 <pikhq> Vorpal: If the phone companies actually used the gigantic funds they get to increase infrastructure, *they* could have it done by now.
20:34:39 <pikhq> elliott: The Stargate program is a broom closet. Honest.
20:34:52 <elliott> pikhq: Biggest. Broom closet. EVER.
20:35:06 <elliott> Cheyenne Mountain Broom Closet
20:35:08 <Vorpal> pikhq, yeah, Well they did in Sweden to some degree. Except we had less gigantic funds. Oh and I found out my ISP is IPv6 ready except for last mile basically.
20:35:23 <Vorpal> they are working on that now, last I heard
20:35:29 <elliott> I want to start an ISP sometime.
20:35:51 <elliott> Actually I wish I was around in the days where an ISP could consist of a rack in your basement and a line from the phone company.
20:35:55 <pikhq> Vorpal: God dammit if we could just support multicast on the public Internet.
20:36:07 <elliott> You know, when you actually had tiny little -- even "indie" -- ISPs.
20:36:12 <Vorpal> pikhq, hm, that would only work well for scheduled tv
20:36:14 <pikhq> Vorpal: That would actually make there be much less bandwidth usage.
20:36:29 <elliott> Vorpal: xs4uk is cool but MEH :P
20:37:04 <Vorpal> pikhq, it doesn't work for the case of 10000 people watching a youtube video, because most of them would out of sync with each other
20:37:06 <elliott> Vorpal: I don't plan to live in the UK :P
20:37:29 <pikhq> Vorpal: You can actually do multicast file sharing. Repeat the broadcast continually to host it. The problem is that it can't stream too well.
20:37:31 <Vorpal> elliott, well, about starting an ISP I mean. You don't have a sane one over there in UK do you?
20:37:35 <elliott> I'll probably go to university in another country, and I don't feel a particular inclination to stay in the UK when I leave.
20:37:46 <elliott> Vorpal: http://www.bogons.net/
20:37:52 <pikhq> Vorpal: Would work wonderfully for things like a podcast, though.
20:38:04 <elliott> Vorpal: also https://www.bethere.co.uk/web/beportal/homepage
20:38:08 <Vorpal> pikhq, depends on if you watch podcast in real time or not
20:38:22 <elliott> Vorpal: less indie, but faster and cheaper
20:38:27 <elliott> Vorpal: and still pretty savvy
20:38:28 <pikhq> Vorpal: No, no, you just *use the multicast to download it*...
20:38:34 <pikhq> Vorpal: As opposed to http...
20:38:41 <elliott> pikhq: HTTP over Multicast :D
20:38:58 <pikhq> Or any other such prerecorded thing where you can expect there to be a lot of people downloading it within a certain time frame.
20:39:33 <elliott> pikhq: Implement a BitTorrent extension for multicast seeders.
20:39:39 <Vorpal> <elliott> Vorpal: http://www.bogons.net/ <-- better web page design than the other one! ;P
20:39:39 <pikhq> elliott: Beautiful.
20:39:53 <elliott> Combine this with web seeds -- which already exist -- and the regular tracker/DHT thing, and tada.
20:40:01 <Vorpal> pikhq, yeah like the iwc podcasts! ;P
20:40:02 <elliott> Just take down the multicast seed after a short while.
20:40:12 <elliott> Vorpal: be redesigned recently
20:40:13 <pikhq> On IPv4, you've got a /4 worth of possible multicast groups.
20:40:17 <elliott> it's ... not the best redesign
20:41:00 <elliott> Vorpal: http://images.trustedreviews.com/images/article/inline/1595-website.jpg their old design, silly ad but more reasonable design
20:41:05 <pikhq> On IPv6, you've a /8 worth of possible multicast groups.
20:41:19 <elliott> Vorpal: Oh yeah, and they have an IRC server for users: http://irc.beusergroup.co.uk:8080/?channels=Be
20:41:33 <elliott> Well, okay, run by the usergroup, but Be links to it from their site.
20:41:34 <pikhq> Oh, wait. Only 112 of the bits are for actual groups.
20:41:39 <pikhq> 2^112-1 possible groups.
20:41:47 <pikhq> Which is still quite absurd.
20:42:23 <pikhq> ... 2^122-1 possible groups *per scope*.
20:42:56 <pikhq> Interface-local, link-local, admin-local, site-local, organization-local, and global scopes are currently defined. Each with their own group-space.
20:43:21 <Vorpal> pikhq, what scope is ::1 ?
20:43:35 <pikhq> Vorpal: Not a multicast address.
20:44:02 <Vorpal> <pikhq> 2^112-1 possible groups. <pikhq> ... 2^122-1 possible groups *per scope*.
20:44:14 <Vorpal> pikhq, so less in total than per scope?
20:44:32 <Vorpal> pikhq, which is the actual number then?
20:44:44 <pikhq> 2^112-1 per scope.
20:44:58 <Vorpal> pikhq, also wtf is "admin-local"?
20:45:11 <Vorpal> or wait, I never seen interface-local either
20:45:43 <Vorpal> [1732343.022354] atkbd serio0: Unknown key pressed (translated set 2, code 0x7c on isa0060/serio0).
20:45:44 <Vorpal> [1732343.022364] atkbd serio0: Use 'setkeycodes 7c <keycode>' to make it known.
20:45:46 <pikhq> Local to the group of administrative configuration.
20:45:47 <Vorpal> that happens from time to time
20:46:00 <Vorpal> I have no clue what key
20:46:24 <Vorpal> I seem unable to trigger it intentionalluy
20:46:31 <fizzie> pikhq: The current IPv6 addressing architecture spech limits the group ID to 32 bits; "While this limits the number of permanent IPv6 multicast groups to 2^32 this is unlikely to be a limitation in the future. If it becomes necessary to exceed this limit in the future multicast will still work but the processing will be sightly slower."
20:46:47 <fizzie> (It's because the mapping from IPv6 multicast groups to Ethernet MAC addresses takes the 32 lowest bits.)
20:46:51 <pikhq> fizzie: That's silly. They reserve 112 bits for the group ID.
20:47:06 <fizzie> pikhq: Yes, and top 80 bits of those 112 are "reserved must be zero".
20:47:13 <fizzie> http://www.ietf.org/rfc/rfc2373.txt 2.7.2.
20:47:34 <pikhq> Still, that's IPv4-space of possible multicast groups...
20:48:03 <Vorpal> how does that work with the MAC thingy then...?
20:48:07 <pikhq> Which is probably quite reasonable for the amount of people that will want to actually broadcast stuff.
20:48:29 <Vorpal> pikhq, unless something changes. Which is quite possible
20:48:37 <Vorpal> like a new thing using multicast is invented
20:48:53 <pikhq> Vorpal: That 4 billion people want to broadcast from.
20:48:53 <fizzie> Vorpal: Upper 80 are zero, lower 32 are taken as the MAC address.
20:48:54 <Vorpal> and we need to go ipv7 or whatever
20:49:05 <elliott> why on earth does linux use letters for drive names?
20:49:09 <Vorpal> pikhq, "we will never need 4 billion IPs"
20:49:19 <Vorpal> pikhq, "640 kB is enough for everyone"
20:49:27 <Vorpal> yeah, we heard this before a number of times!
20:49:33 <elliott> /dev/sda0 is silly, e.g. /dev/sd0p0 would be better
20:49:39 <fizzie> Vorpal: Anyway, like it says it's "easy" (for some values of) to extend the addressing to full 112 bits without having to move from IPv6 to anything.
20:49:44 <Vorpal> elliott, you could change it, udev
20:49:55 <elliott> Vorpal: yeah i'll just break everything for a minor issue :)
20:50:21 <pikhq> Vorpal: There's also the thing about how IPv4 does not have a 1-to-1 mapping from IPs to devices, unlike group IDs would.
20:50:22 <fizzie> Wasn't it so that devfs already made the paths more "logical"? (Then those went away when devfs died.)
20:50:32 <pikhq> Vorpal: Every time you create a subnet, you've made 2 addresses unusable.
20:50:33 <elliott> fizzie: you have the LOVELY LOVELY GUID PATHS
20:50:38 <elliott> but only people with OCD use those
20:50:45 <Vorpal> elliott, I'm not sure that stuff will break, it already handles stuff like /dev/md0, /dev/hda, /dev/sda, /dev/sr0 and several more
20:50:53 <Vorpal> I believe some hardware raid have their own too
20:50:55 <elliott> Vorpal: yeah but e.g. default arguments to tools
20:51:01 <fizzie> elliott: Yes, but there was in devfs something like .../bus0/lun0/disk0/part0.
20:51:02 <elliott> it's still a bit pointless
20:51:13 <elliott> fizzie: Looks like OpenFirmware disk identifiers.
20:51:21 <fizzie> "So, whereas /dev/hda4 was used previously, we now have /dev/ide/host0/bus0/target0/lun0/part4. This is far more easy... no, don't argue with me... it is easier... ah whatever! :) "
20:51:31 <Vorpal> elliott, sure you need to update your boot loader config and fstab. Possibly a few more things, but nothing more than if stuff changed from /dev/sda to /dev/sdb
20:51:38 <elliott> Also the hda/sda thing is a bit silly.
20:51:47 <elliott> "We totally had to change all the names, don't you get it? We CHANGED INTERNAL IMPLEMENTATIONS."
20:52:14 <Vorpal> elliott, um. Well that name change proves that changing it again would be trivial
20:52:20 <pikhq> elliott: The internal implementation no longer makes a distinction between IDE and SCSI and SATA.
20:52:32 <Vorpal> just a case of changing bootloader config and fstab
20:52:33 <fizzie> Doesn't the Ubuntu installer put UUID paths to /etc/fstab by default?
20:52:41 <elliott> debian-installer probably does too
20:52:47 <pikhq> (it's all ATA over the wire)
20:52:58 <Vorpal> fizzie, except it breaks in a lot of cases. Well "a lot" is maybe stretching it
20:53:03 <elliott> pikhq: Better question - why did it ever distinguish them?
20:53:08 <Vorpal> dm-crypt, lvm, software raid
20:53:09 <elliott> Why didn't SATA just get put into the hd* namespace?
20:53:10 <pikhq> elliott: Hysterical raisins.
20:53:19 <fizzie> Well, for LVM it's all just /dev/mapper/volgroup-volname.
20:53:19 <Vorpal> you can't do the UUID stuff reliably then
20:53:23 <elliott> And why is it acceptable in /dev/ to indicate hierarchy by name prefixes?
20:53:43 <Vorpal> elliott, good question, it is moving away from that though, /dev/usb and so on
20:53:45 <pikhq> Vorpal: The UUID stuff is part of the filesystem, not the device.
20:53:49 <fizzie> cheater99: "Logical Unit Number" or some-such; it's for something like the different drives in a CD jukebox that uses a single SCSI device.
20:53:59 <elliott> Probably Unix 0.00000001 didn't have nested directories.
20:54:40 <cheater99> elliott: no. it's just that resolving an inode was sloooooooooooooooooow on winchesters.
20:55:32 <elliott> Vorpal: Like that system emulator.
20:55:40 <Vorpal> elliott, hm? which one?
20:56:00 <elliott> 16:40:56 <elliott> http://www.acc.umu.se/~achtt315/tunguska/ Ternary system emulator.
20:56:00 <elliott> 16:41:05 <elliott> With a ternary-C-dialect compiler!
20:56:00 <elliott> 16:41:10 <elliott> I approve to an insane degree.
20:56:00 <elliott> 16:41:17 <elliott> "It is loosely based on the excellent design of the (binary) 6502-processor by MOS Technology, but entirely ternary. So instead of having two memory cell states (0, 1), it has three (-1, 0, 1)."
20:56:00 <elliott> 16:41:58 <elliott> pikhq: ^ Impossibly awesome
20:56:01 <elliott> 16:41:59 <elliott> http://www.acc.umu.se/~achtt315/tunguska/shots/tunguska_gtk.png
20:56:04 <elliott> 16:42:02 <elliott> It even has a display!
20:56:06 <elliott> 16:42:05 <elliott> And a command-line on this display!
20:56:17 <Vorpal> elliott, anyway, balanced ternary is IMO one of the most confusing numeral systems ever invented.
20:56:26 <Vorpal> but balanced ternary? blergh
20:56:32 <oerjan> "Why must californian grape farms be regularly sprayed with sedatives?"
20:56:33 <elliott> I don't know if it actually uses it as -1.
20:56:38 <elliott> It's used as a third logical state, at least.
20:56:59 <Vorpal> elliott, perfect for SQL then ;)
20:57:33 <elliott> "They're all crazy there."
20:57:44 <cpressey_> a number in balanced ternary: 1,0,1,-1,-1,0,1
20:57:52 <Vorpal> elliott, Umeå Universitet. WAY up north
20:58:05 <elliott> Vorpal: And ... what is that supposed to mean?
20:58:05 <fizzie> udev changelog in gentoo: "This release removes the devfs names for the tty and consolde devices, and the symlinks that were implementing the LSB standard names. We only implement the LSB names now, which saves over 3Mb of RAM on /dev." -- that's quite a lot of names.
20:58:21 <Vorpal> elliott, they have to be crazy, or they wouldn't live where you can get -50°C in the winter!
21:00:03 <fizzie> I seem to recall that Debian had the devfs names provided by udev rules at some point as a compatibility kludge too. They don't seem to be there any more now, though.
21:00:49 <oerjan> elliott: no it's because of hysterical raisins, silly
21:00:50 <Vorpal> elliott, I heard a story from someone who was up there one winter (due to work, he had no choice). It was -49°C outside. He went from the hotel to the opposite side of the square it was located at. By the time he reached that side his moustache had frozen solid from the exhaled wet air.
21:01:30 <Vorpal> elliott, yeah, anyone insane enough to live in such conditions has to be crazy.
21:01:40 <elliott> Anyone insane enough ... has to be crazy.
21:02:20 <Vorpal> elliott, only lunatics would be insane enough though!
21:02:22 <elliott> Heh, Wikipedia using zorkmid as a metasyntactic currency.
21:02:33 <Vorpal> elliott, on which page?
21:02:36 <fizzie> There's still the /dev/disk/by-path/ symlinks; but then it's something like /dev/disk/by-path/pci-0000:00:11.0-scsi-0:0:0:0-part1 instead of /dev/sda1.
21:02:37 <elliott> http://en.wikipedia.org/wiki/Balanced_ternary
21:02:44 <elliott> Paragraph starting "Similarly, a currency system using ternary values would save visits to the bank".
21:03:00 <elliott> "Donald Knuth has pointed out that truncation and rounding are the same operation in balanced ternary — they produce exactly the same result."
21:03:42 <elliott> "Multiplication by two can be done by adding a number to itself." Really now.
21:04:33 <cpressey_> "The sky would remain blue and essentially unchanged."
21:05:39 <oerjan> there _might_ be more double rainbows, though
21:07:25 <elliott> cpressey_: I'm installing NetBSD in a VM to test.
21:07:36 <fizzie> elliott: Of *course* it runs NetBSD.
21:08:04 <elliott> fizzie: McDonald's -- I'm Lovin' It
21:08:08 <elliott> (Let us all quote slogans!)
21:08:16 <elliott> http://upload.wikimedia.org/wikipedia/commons/b/bd/RonaldMcDonald-trademarkia-originaltrademark1967.jpg dear god
21:08:20 <fizzie> "Housut pois ja hoitoon."
21:08:30 <pikhq> Your mother, all night long.
21:08:59 <Vorpal> So, the purpose is to provide a simple and accessible, yet powerful playground for ternary computing for the man in the street (with a decent understanding of assembly programming and general computer infrastructure). " XD
21:09:24 <elliott> fizzie: Interesting slogan...
21:09:31 <elliott> `wl fi Housut_pois_ja_hoitoon
21:10:02 <oerjan> no, finnish has no articles.
21:10:17 <Vorpal> <elliott> http://upload.wikimedia.org/wikipedia/commons/b/bd/RonaldMcDonald-trademarkia-originaltrademark1967.jpg dear god <-- sure it isn't vandalism?
21:10:38 <Vorpal> elliott, hm, what about a 4-state logic system?
21:10:41 <elliott> Vorpal: two revisions, only extra one is a whitespace crop
21:11:26 <elliott> "Their symbol is $, although this should not be taken to mean the zorkmid exchange rate is pegged to the US dollar."
21:11:54 <elliott> Command: /bin/sh MAKEDEV all
21:12:12 <oerjan> they're using a currency basket, of course
21:12:13 <Vorpal> elliott, anyway, what about a 4-state logic system, instead of the classical binary or ternary ones
21:12:28 <Vorpal> elliott, or why not a decimal computer. That would be interesting
21:12:37 <Vorpal> 10 logical states, I wonder what you could use that for
21:12:41 <elliott> Vorpal: http://en.wikipedia.org/wiki/Decimal_computer
21:12:45 <elliott> Many early computers, for example the ENIAC, IBM 702, IBM 705, IBM 650, IBM 1401, IBM 1620, IBM NORC, IBM 7070, IBM 7080, UNIVAC I, UNIVAC II and UNIVAC III used decimal arithmetic (IBM 1401 addresses were a combination of decimal and binary arithmetic).
21:12:54 <HackEgo> My hovercraft is full of eels.
21:12:59 <Vorpal> elliott, not decimal logo though?
21:13:19 <elliott> `run wl fi Housut_pois_ja_hoitoon 2>&1
21:13:35 <HackEgo> My hovercraft is full of eels.
21:14:06 -!- GreaseMonkey has joined.
21:14:06 -!- GreaseMonkey has quit (Changing host).
21:14:06 -!- GreaseMonkey has joined.
21:14:08 <elliott> `run wl en fi article 2>&1
21:14:20 <Vorpal> elliott, what is the point of BCD for general computing.... BCD for controlling 7-segment displays: sure. But for general computing?
21:14:45 <elliott> cpressey_: Quote from the NetBSD installer:
21:14:54 <Vorpal> elliott, but I mean, x86 supports it, and lots of mainframes does
21:14:58 <Vorpal> there has to be a reason
21:15:00 -!- cal153 has quit.
21:15:05 <elliott> YES LET'S ENCRYPT MY PASSWORD WITH DES
21:15:19 <elliott> Vorpal: hysterical ones relating to raisins.
21:15:28 <Vorpal> elliott, go for whirlpool
21:15:33 <elliott> Oh yeah, it even says that DES will only encode the first eight characters.
21:15:35 <Vorpal> elliott, why? the name is cool
21:15:36 <elliott> Vorpal: I quoted all the options.
21:15:40 <elliott> I'm going to go with Blowfish.
21:15:52 <elliott> There's probably some way to do it.
21:15:54 <elliott> Just not in the installer.
21:16:05 <Vorpal> elliott, on linux you would set that in PAM
21:16:07 <elliott> I don't think there are any known flaws in Blowfish.
21:16:19 <elliott> And, well, Bruce Schneier.
21:16:28 <oerjan> google translate doesn't seem to know "hoisut"
21:16:40 <elliott> oerjan: it said "remove pants for a treat" for me
21:16:46 <elliott> Vorpal: "Blowfish provides a good encryption rate in software and no effective cryptanalysis of it has been found to date. However, the Advanced Encryption Standard now receives more attention."
21:16:50 <elliott> so blowfish should be good
21:16:58 <elliott> especially with 128 rounds
21:17:06 <Vorpal> elliott, "There remains no known way to break the full 16 rounds, apart from a brute-force search."
21:17:18 <elliott> Vorpal: 128 rounds in NetBSD
21:17:20 <Vorpal> elliott, more rounds can sometimes be bad iirc
21:17:30 <elliott> Vorpal: I sorta trust the BSD guys to not pick a retarded number of rounds.
21:17:46 <Vorpal> elliott, for some cryptos there is some best number somewhere, less is worse, but also more is worse
21:17:57 <elliott> Vorpal: Oh, and bcrypt is Blowfish.
21:18:05 <fizzie> elliott: To be completely honest, it's not actually a slogan.
21:18:11 <elliott> Vorpal: So there's not really anything to worry about.
21:18:15 <elliott> fizzie: I *did* Google it.
21:18:36 <elliott> Vorpal: "Housut pois ja hoitoon."
21:19:07 <elliott> I wonder what /bin/sh is on NetBSD?
21:20:11 <elliott> Phantom_Hoover: is not part of BSD.
21:20:12 -!- French has joined.
21:20:12 <French> This spam brought to you by: #freenode! Remember, node freely or don't node at all!
21:20:13 -!- French has left (?).
21:20:13 -!- Aeryn has joined.
21:20:44 -!- Jaxton has joined.
21:20:44 <Jaxton> This spam brought to you by: #freenode! Remember, node freely or don't node at all!
21:20:44 -!- Jaxton has left (?).
21:20:58 <Vorpal> elliott, ksh is probably pdksh?
21:21:09 <olsner> ok, that was funny exactly once
21:21:23 <olsner> now I'm slightly worried there will be hundreds of spams
21:21:26 <Aeryn> This spam brought to you by: #freenode! Remember, node freely or don't node at all!
21:21:26 -!- Aeryn has left (?).
21:21:26 <elliott> pdksh is derived from the bsd ones
21:22:52 <elliott> "Although the Blowfish-based system has the option of adding rounds and thus remain a challenging password algorithm, it does not use a NIST-approved algorithm. In light of these facts, Ulrich Drepper of Red Hat led an effort to create a scheme based on the SHA-2 (SHA-256 and SHA-512) hash functions."
21:23:04 <elliott> Ulrich Drepper, always making sure we use only NIST-approved hashing algorithms.
21:23:12 <fizzie> Debian's dash is a successor of NetBSD's ash (which is what /bin/sh is), I think.
21:23:32 <elliott> Because if there's one organise we want to trust about cryptography, it's the NSA!
21:23:52 <Vorpal> fizzie, is that a different dash than what ubuntu uses for /bin/sh ?
21:23:59 <fizzie> Vorpal: It's the same.
21:24:35 <pikhq> Vorpal: Definitely the same Debian Almquist Shell.
21:24:39 <elliott> Yup, NetBSD's X11 is from June 2008.
21:25:12 <pikhq> elliott: So far, the NSA's suggestions towards cryptography have *just* been improvements (as far as we know).
21:25:23 <Vorpal> pikhq, hm, who was Almquist? I mean, I heard the name before. Both in computing context and outside it
21:25:28 <fizzie> "Ubuntu: Stealing from Debian since 2004."
21:25:33 <elliott> pikhq: Yes, but... they also had the ability to break -- uh, what was it? DES?
21:25:42 <Vorpal> the latter obviously due to it being a Swedish familyname
21:25:58 <elliott> Vorpal: Kenneth Almquist, author of ash.
21:26:10 <Vorpal> elliott, from US or Sweden? (I guess it is one of those)
21:26:13 <elliott> Anyway, Debian doesn't use ash as /bin/sh I think.
21:26:24 <olsner> so he changed his first name to "Debian" and rewrote it as dash? :)
21:26:44 <pikhq> elliott: As far as is known, they did not have the ability to break DES until the EFF showed it could be done.
21:26:55 <Vorpal> olsner, accurate and astute as usual!
21:26:59 <elliott> pikhq: they had differential cryptanalysis, though
21:27:53 <pikhq> elliott: They *did* make an unexplained suggestion to DES that was precisely what was needed to make it resistant to differential cryptanalysis before the knowledge of that technique was public.
21:28:33 <elliott> pikhq: it must have been a coincidence :)
21:28:35 <fizzie> One of my debians has dash in /bin/sh; another doesn't. I think it's sort-of optional there, but if you do install the package, it diverts /bin/sh to itself.
21:28:57 <Vorpal> pikhq, they used ASIC for the task. In a small production run. That always makes my mind boggle
21:29:00 <pikhq> elliott: It's well-known that the NSA had knowledge of it quite a bit earlier than the general public.
21:29:16 <elliott> <Zeshawn> Is it normal for my penis to be 2 inches at the age of 12?
21:29:23 <elliott> Well that was... on-topic.
21:29:30 <pikhq> elliott: IBM was asked not to publish, you see.
21:29:36 <Vorpal> elliott, I believe a staffer told that guy to quit it.
21:30:59 <elliott> NetBSD is such a nice kernel/libc/coreutils with such a shitty distribution.
21:31:32 <elliott> (Well, okay, it isn't actually shitty. Just imperfect.)
21:31:59 <elliott> X -configure, has it ever worked in the history of things working?
21:33:14 <olsner> I've used it to configure X
21:33:26 -!- webquint has joined.
21:33:58 <webquint> do you have the pastie link for the walls ruleset in your logs?
21:34:22 <Gregor> pikhq, Vorpal: http://poll.fm/2c4e9
21:34:42 <elliott> I might have the file here
21:34:45 <elliott> Phantom_Hoover: webquint wrote a rule
21:34:52 <Vorpal> Gregor, we told you before, I forgot the numbers
21:35:07 <elliott> Phantom_Hoover: WHY SHOULD I TELL A GAY VAMPIRE ABOUT GOLLY RULES
21:35:18 <Vorpal> Gregor, let me quickly listen to them
21:35:24 <webquint> gay vampires are the best vampires
21:35:51 <Gregor> webquint: I'd rather they don't suck BLOOD is all.
21:35:59 <elliott> Gay vampire sitcom: "I vill suck yor... [KNOCK ON THE DOOR]"
21:36:03 <elliott> Gregor: DAMMIT STOP PREEMPTING ME
21:36:25 <Vorpal> Gregor, what about op1?
21:36:27 <webquint> gregor: you don't mind being sucked by vampires otherwise?
21:36:39 <Gregor> Vorpal: Opuses 1-4 DO NOT EXIST >_>
21:36:49 <Vorpal> Gregor, ah, 0 or -1 then!
21:36:52 <Gregor> webquint: Depends on how they work the fangs >_> <_<
21:37:07 <webquint> ain't you never seen true blood?
21:37:20 <Vorpal> Gregor, idea: fractional and irriational opus numbers
21:37:25 <elliott> lol the #freenode spammers are coming into #freenode
21:37:28 <elliott> and asking why #freenode is spamming
21:37:57 <elliott> * French (~French@189.188.167.108) has joined #esoteric
21:37:57 <elliott> <French> This spam brought to you by: #freenode! Remember, node freely or don't node at all!
21:37:57 <elliott> * French (~French@189.188.167.108) has left #esoteric
21:38:01 <elliott> happening all over the place
21:38:05 <elliott> now they're coming into #freenode and going
21:38:07 <elliott> WHY IS FREENODE SPAMMING ;_;
21:38:12 <Vorpal> Gregor, the op8 recording sounds good to me. At least after listening to the earlier ones
21:38:34 <Gregor> Vorpal: ... then don't vote for it?
21:38:37 <Vorpal> Gregor, I'm torn between 7 and 5. :/
21:38:40 <webquint> elliott: i can has that file now please? *insert pupy dog face*
21:38:47 <elliott> webquint: ffff i'm doing other things
21:38:58 -!- KindOne has joined.
21:39:21 <Gregor> elliott: My question is why are the French spamming? Bloody French!
21:39:42 <oerjan> Ceci n'est pas un spam
21:39:48 -!- webquint has quit (Quit: Page closed).
21:40:17 <Vorpal> Gregor, does op5 has a metronome in it? Or what is the strange clicking noise?
21:40:26 <Vorpal> hm too irregular for metronome
21:40:40 <Vorpal> Gregor, I cast my vote on 7
21:40:49 <elliott> * Schantelle has quit (Killed (idoru (Spam is off topic on freenode.)))
21:40:55 <elliott> like kindly informing them
21:41:00 <elliott> "oh, spam is a bit off-topic here"
21:41:09 <elliott> legit mistake, anyone could do it!
21:42:39 <HackEgo> My hovercraft is full of eels.
21:43:39 <Vorpal> Gregor, I love both op5 and op7. They are a lot more "action filled" than op11 certainly.
21:44:16 <HackEgo> My hovercraft is full of eels.
21:44:39 <Gregor> Vorpal: I will probably eventually get to all of them.
21:44:53 <oklopol> any math/esolanging today?
21:44:59 <Vorpal> Gregor, 7 is the one in most urgent need of it
21:45:06 <elliott> oklopol: please introduce some
21:45:20 <oklopol> also i saw two ppl having sex at a spa today
21:45:32 <oerjan> elliott: i'm surprised it's not Svenskekocken in a single word
21:45:54 <elliott> It's the Swedish cock, not the Swedishcock.
21:48:35 <oerjan> well it's usually in one word in norwegian, is all
21:49:22 <oerjan> (when referring to the muppet)
21:49:59 <elliott> catseye: Huh, NetBSD doesn't ship with pkgsrc.
21:51:31 <oerjan> same with the swedish king too
21:51:52 -!- augur_ has joined.
21:51:53 <oklopol> i can introduce a tiny bit of math: let S be a semigroup, we define an eq relation R, xRy iff xS = yS, similarly xLy iff Sx = Sy. we define H = R \cap L. greene's theorem says the maximal subgroups of S are exactly the equivalence classes X in H that \exists x, y \in X, xHyHxy or 2) there's an idempotent in X
21:52:12 -!- augur has quit (Ping timeout: 245 seconds).
21:53:32 <Vorpal> oerjan, hm? *reads up*
21:54:18 <Vorpal> oerjan, like "svenskekungen"? or what?
21:54:36 <oerjan> Svenskekongen, actually
21:54:47 <Vorpal> oerjan, would be separate words in Swedish: Sveriges kung.
21:54:51 <oerjan> possibly with a hyphen
21:54:54 <Vorpal> wouldn't even be Svenske
21:55:08 <Vorpal> svenske kocken as two words
21:55:14 <elliott> Binary packages on BSD -- cheating or cheating?!
21:55:30 <Vorpal> elliott, cheating when used for initial install!
21:55:40 <elliott> Vorpal: ...uh, all the BSDs initial-install as binary.
21:55:40 <oerjan> Vorpal: i think the swedish king may be special, we like to make fun of him ;D
21:55:44 -!- MigoMipo has quit (Read error: Connection reset by peer).
21:55:52 <oklopol> another tidbit: an isometry is a function f (between two metric spaces), d(x, y) = d(f(x), f(y)) for all x, y; the isometries from R^2 to R^2 are exactly rotations, translations and "glide reflections", where a glide reflection means you move along a line l, and then reflect around l
21:55:53 <Vorpal> elliott, you should use a stage1-like install! Or even LFS like install!
21:56:15 <pikhq> elliott: NetBSD doesn't necessarily initial-install as binary.
21:56:24 <Vorpal> oklopol, would be "Norges kung" here for your one
21:56:27 <pikhq> elliott: Download build.sh and run it. :D
21:56:30 <elliott> pikhq: it's totally weird that it doesn't come with pkgsrc though
21:56:41 <pikhq> ... Okay, so that bootstraps pkgsrc. Still.
21:56:45 <oerjan> Vorpal: http://ikkepedia.org/wiki/Knugen
21:56:49 <elliott> `addquote <oklopol> comex: what? <oklopol> *vorpal
21:56:52 <HackEgo> 240|<oklopol> comex: what? <oklopol> *vorpal
21:57:00 <pikhq> elliott: Technically, pkgsrc is a package manager for all UNIX systems.
21:57:04 <oklopol> comex: hi, tab-complete completed c to comex instead of Vorpal, dunno why
21:57:10 <elliott> There is a binary package installer but no ports by default.
21:57:11 <Vorpal> elliott, v is next to c though
21:57:17 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
21:57:27 <Vorpal> oklopol, v is next to c
21:57:27 <pikhq> Still, fetch and run ./build.sh
21:57:35 <Vorpal> oklopol, anyway "what" in response to what
21:57:47 <elliott> `addquote <oklopol> comex: what? <oklopol> *vorpal <oklopol> comex: hi, tab-complete completed c to comex instead of Vorpal, dunno why
21:57:47 <oklopol> Vorpal: you said something to me you were trying to say to oerjan
21:57:50 <HackEgo> 240|<oklopol> comex: what? <oklopol> *vorpal <oklopol> comex: hi, tab-complete completed c to comex instead of Vorpal, dunno why
21:58:01 <oklopol> "<Vorpal> oklopol, would be "Norges kung" here for your one"
21:58:22 <elliott> finland's king is called bob
21:58:27 <oklopol> i wondered whether this was about the ...autometries of R^2, or about semigroups
21:58:37 <elliott> oerjan: Pressed - President of the Swedish Association dyslexia (ed.: also known as DDR). in 1832 revealed Titten Tei that weighed uses the pseudonym King of Sweden when he is on the party.
21:58:43 <Vorpal> oerjan, you don't use "knugen" about your own one as well?
21:58:47 <HackEgo> awklisp \ babies \ bin \ cube2.base64 \ cube2.jpg \ hack_gregor \ hello.txt \ help.txt \ huh \ netcat-0.7.1 \ netcat-0.7.1.tar.gz \ out.txt \ paste \ poetry.txt \ quine \ quotes \ qw.pl \ share \ tmpdir.28221 \ wunderbar_emporium
21:58:52 <Vorpal> oerjan, heck we use it about our own!
21:58:58 <HackEgo> 20:58:53 up 16 days, 2:34, 0 users, load average: 0.85, 0.77, 0.80 \ USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
21:59:24 <elliott> comex: Yes, it really is; use `run for that; and no, it's not worth trying.
21:59:28 <Vorpal> oerjan, what would "Mil ettert mil" from that page mean?
22:00:11 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
22:00:18 <elliott> "Mil etter mil" ("Mile after Mile") was the Norwegian entry in the Eurovision Song Contest 1978, performed in Norwegian by Jahn Teigen.
22:00:22 <oerjan> Vorpal: well it would knogen for the norwegian one _if_ we used it, i guess
22:00:26 <elliott> comex: `cmd x y z = `cmd 'x y z'`
22:00:51 <oerjan> i also vaguely recall he _is_ dyslexic as well
22:00:52 <HackEgo> Linux codu.org 2.6.32-5-xen-amd64 #1 SMP Fri Sep 17 22:00:48 UTC 2010 x86_64 GNU/Linux
22:01:11 <elliott> comex: And yes, it is sandboxed more than you can imagine.
22:01:21 <oerjan> (vaguely because i'm not _entirely_ sure if it was him or his father)
22:01:36 <elliott> comex: DON'T LOOK FOR INTERNET FAME HERE, YOU WON'T BE RELEASING A HACKEGO JAILBREAK ANY TIME SOON
22:01:59 <comex> is this the same script I screwed up once? :p
22:02:07 <elliott> and the whole thing is in a plash sandbox
22:02:20 <comex> (but plash is useless without a chroot, so that's redundant)
22:02:27 <oerjan> elliott: "Pressed" is an accidental translation of "knugen" here, the real point being it's an anagram of "kungen" (sv. the king)
22:02:28 <elliott> even if you elevated to root inside plash
22:02:50 <elliott> you'd be running as a user only privileged enough to trash the bot :P
22:03:20 <pikhq> elliott: Besides which, in escaping the chroot all you'd have at your disposal is ld-linux.so.6
22:03:38 <HackEgo> bin \ dev \ etc \ home \ lib \ lib64 \ opt \ proc \ tmp \ usr
22:03:47 <Vorpal> fizzie, where is fungot?
22:04:02 <cpressey_> not that fungot ever moves as such
22:04:11 <pikhq> comex: None of those files exist in the chroot.
22:04:15 <Vorpal> cpressey_, it lost connection
22:04:26 <comex> they're through plash?
22:04:35 <pikhq> Yes, that's how plash works.
22:04:35 <oerjan> Vorpal: the norwegian king, yes
22:04:46 <Vorpal> comex, indeed. I checked some time ago using native syscall()
22:04:49 <comex> I wonder if there are any security holes in plash...
22:04:54 <Vorpal> to list what was really there
22:04:54 <comex> the server itself is probably not sandboxed
22:04:57 <pikhq> You get an almost entirely empty chroot, and all access to everything works through the patched libc.
22:05:03 <oerjan> Vorpal: also mil etter mil is "famous" for being one of norway's Eurovision zero-pointers
22:05:10 <pikhq> The server itself is a Xen VM.
22:06:10 <pikhq> The only file in the chroot is /lib/ld-linux.so.6, for the sole purpose of allowing programs to load normally.
22:06:25 <pikhq> Said ld-linux.so.6 is patched to communicate with plash.
22:06:53 <Vorpal> pikhq, it's sad plash is so debian-specific
22:07:05 <elliott> Vorpal: just run it in a debian chroot
22:07:06 <Gregor> It's wonderful that plash is so Debian-specific.
22:07:09 <elliott> even more layers of security!
22:07:17 <Vorpal> elliott, not really but meh :P
22:07:17 <oerjan> Vorpal: "how so"? it got zero points, i say :D
22:07:23 <Gregor> I have all my plashified things in a chroot *shrugs*
22:07:26 <comex> plash source is very complicated.
22:07:39 <elliott> comex: if you can break plash you have bigger targets than HackEgo.
22:07:46 <cpressey_> plash source is radioactive magic faery dust
22:07:53 <Gregor> So I assume we're talking about how to hack Codu? :P
22:07:55 <pikhq> comex: It'd actually work well if they made the patch against normal libc rather than Debian libc.
22:08:06 <fizzie> Vorpal: Oh, there was a short network break at home this morning, I guess it may have died during that.
22:08:45 <elliott> Gregor: comex is just living up to his hax0r name, you can tell by the x
22:09:02 <Gregor> Better than xXxcomexXx
22:09:06 -!- fizzie has quit (Quit: jumpin' jumpin').
22:09:11 -!- fizzie has joined.
22:09:12 -!- cpressey_ has changed nick to cpressex.
22:09:41 -!- fungot has joined.
22:09:54 <fungot> cpressex: i prefer this one: fnord/ local/ lib/ bigloo into ld.so.conf solved it. but i may be ignorant but i see np with the changes
22:10:04 -!- Gregor has changed nick to Grax0r.
22:10:07 <comex> like, where's the server part
22:10:34 <comex> oh I give up, this is too much work for IRC :p
22:10:45 <HackEgo> 218|<alise> Phantom_Hoover: Don't be nasty; he's a lunatic, not a murderer.
22:10:46 <Grax0r> That means we can't hax0r anymore.
22:10:49 <HackEgo> 54|<lacota> I guess when you're immortal, mapping your fonts isn't necessary
22:10:55 <HackEgo> 232|<Phantom_Hoover> It's only been 2 months since anyone last made a commit! <alise> WRONG 8 WEEKS
22:11:22 <HackEgo> /tmp/hackenv.29461/bin/quote
22:11:35 <comex> `strings /tmp/hackenv.29641/bin/quote
22:11:49 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/bin/quote
22:11:56 <oklopol> i don't care what YOU ppl have said
22:11:57 <comex> `run bash -c 'strings /tmp/hackenv.29641/bin/quote' 2>&1
22:11:59 <HackEgo> 48|<oklopol> i can get an erection out of a plank, you can quote me on that. 50|<oklopol> i'm not a porn star, no 53|<oklopol> anyway, torture would be fun to experience, true <oklopol> should put that on my todo list 56|<oklopol> i'm my dad's unborn sister 74|<oklopol> GregorR: are you talking about ehird's virginity or your
22:12:01 <HackEgo> 183|* Phantom_Hoover wonders where the size of the compiled Linux kernel comes from. <cpressey> To comply with the GFDL, there's a copy of Wikipedia in there.
22:12:05 <HackEgo> /usr/bin/strings: /usr/lib/plash/lib/libc.so.6: version `GLIBC_2.11' not found (required by /usr/bin/strings)
22:12:36 <Grax0r> comex: 1) It's a shell script, just read it. 2) lawl, strings doesn't work apparently, 3) that /tmp/hackenv.*** path is different for every run.
22:12:41 <HackEgo> 37|<kaelis> so, he.. uh <kaelis> basically probed me with a weasel.
22:12:52 <HackEgo> 38|<Dylan> speaking of pants <Dylan> harry potter movie
22:12:58 <HackEgo> 90|<oklopol> hmm, this is hard
22:13:06 <Grax0r> Phantom_Hoover: HackEgo responds to PM
22:13:43 <cpressex> dearnbsd, no pr0n in pkgsrc why? thx
22:14:01 -!- Grax0r has changed nick to Gregor.
22:14:07 -!- cpressex has changed nick to cpressey_.
22:14:28 <elliott> i do /etc/rc.d/network/start
22:14:34 <Gregor> Dear FreeBSD: The word Free is trademarked by the FSF. Either change your license to the GPL, thereby conforming to the trademark requirements for the word Free, or change the name of your OS.
22:14:58 -!- cal153 has joined.
22:15:11 <elliott> cpressey_: Is... what... did you just say ipv6 for no reason? :P
22:15:45 <cpressey_> elliott: last i used it, it wanted to do IPV6 by default for everything. even if you couldn't
22:15:46 <elliott> cpressey_: I have to use ipv6 for networking to work? :p
22:15:46 <Gregor> The reason: It is the nineties. And there is time for ipv6.
22:15:55 <cpressey_> elliott: no, you have to switch it off
22:16:30 <cpressey_> elliott: you get it though, right? netBSD is THE RIGHT SOLUTION. ipv6 is THE RIGHT SOLUTION. therefore netbsd -> ipv6
22:16:37 <elliott> cpressey_: apparently it tries ipv4 after that
22:17:24 <cpressey_> whoa, google's front page is now gigantic
22:17:44 <Gregor> cheater99: alise is with all of us.
22:18:02 <olsner> alise has joined the earth-mother
22:18:13 <cpressey_> alise was split up into a hundred small pieces and we each got one
22:18:27 <Gregor> This is after joining the earth-mother, mind you.
22:18:37 <Gregor> So we all got some earth-mother too.
22:18:40 <Gregor> Awwww yeah earth-mother.
22:18:52 <Vorpal> <fizzie> Vorpal: Oh, there was a short network break at home this morning, I guess it may have died during that. <-- and the service supervisor didn't restart it?
22:19:06 <olsner> yo dawg, I heard you like earth-mother, etc
22:19:45 <Vorpal> <Gregor> Dear FreeBSD: The word Free is trademarked by the FSF. Either change your license to the GPL, thereby conforming to the trademark requirements for the word Free, or change the name of your OS. <-- err this is a joke right?
22:19:58 <Vorpal> (I can only hope so, but who knows with RMS!)
22:20:39 <elliott> cpressey_: Strange perversion I just gained: I want to write a filesystem!
22:20:56 <Gregor> elliott: You'll have to kill your wife first
22:21:01 <pikhq> Gregor: The BSD license is sufficiently Free for the FSF. :)
22:21:08 <elliott> Gregor: First I have to Russian mail-order her.
22:21:12 <pikhq> It merely goes further than they'd like.
22:21:18 <fizzie> Vorpal: It shouldn't be that hard to guess that it's still not installed as a proper service. You don't need to mention that every time.
22:21:19 <olsner> Gregor: no, that goes after writing the file system, IIRC
22:21:24 <elliott> We demand rigidly-defined areas of doubt and unFreeness!
22:21:41 <olsner> Gregor: Sweden is unanimous: After.
22:23:02 <elliott> In Sweden we are very civilised.
22:23:10 <elliott> We only kill our wives after letting them see our glorious pen-- filesystem.
22:23:18 <elliott> Ha ha, Americans! WHAT NOW
22:23:46 -!- GreaseMonkey has quit (Remote host closed the connection).
22:24:15 <Vorpal> there is an exception. You can call it ext<n+1> where n is the currently highest number in use. In that case everyone will find you too boring to bother about killing wives.
22:24:41 <elliott> Ted Ts'o has killed SO MANY of his wives, but who cares about ext4?
22:25:01 <olsner> In Sweden we are so civilised that we often don't even kill our wives!
22:25:11 <elliott> olsner: ...for the first few years!
22:25:20 <cheater99> is there some linux file system murder streak happening again?
22:25:26 <olsner> then again, we often don't write filesystems either, I may just be lacking data
22:26:16 <elliott> http://geekz.co.uk/lovesraymond/archive/so-i-married-a-kernel-programmer
22:26:24 <elliott> Everybody loves Everybody Loves Eric Raymond.#
22:26:50 <elliott> note: "not" in the caption is <blink>'d if your browser doesn't display it
22:26:52 <olsner> I wonder what would happen to a married pair of lesbian filesystem writers in this glorious wife-killing world
22:27:09 -!- Sgeo has joined.
22:27:21 <pikhq> olsner: Suicide cult.
22:28:49 <elliott> As a gay vampire, Phantom_Hoover wants none of this.
22:29:50 <olsner> most gay vampires are not eligible for this meme since they don't usually have or become wives
22:30:11 <Sgeo> Also, rages at Sine
22:30:22 <elliott> Sgeo: what have they done now?
22:30:32 <Vorpal> <elliott> As a gay vampire, Phantom_Hoover wants none of this. <-- wait, what?
22:30:45 <elliott> Vorpal: fungot called Phantom_Hoover a gay vampire once and i will never stop calling him one
22:30:45 <fungot> elliott: okay i guess i don't know...
22:30:46 <Sgeo> Failed to provide a way to connect
22:30:52 <elliott> Oh. fungot now expresses doubt that Phantom_Hoover is a gay vampire.
22:30:52 <fungot> elliott: and you subscribe to classes too? :) i don't know that about unix, just how to subclass it.
22:31:04 <elliott> fungot is now back in its native hallucinatory state.
22:31:05 <fungot> elliott: what is the simplest, and it's the kind of fnord parts. for example in haskell:
22:31:54 <Vorpal> elliott, hey, answer fungot's question, it is rude to not do so!
22:31:54 <fungot> Vorpal: there is such a thing
22:32:13 <Vorpal> fungot, such as thing as rudeness? Indeed
22:32:14 <fungot> Vorpal: by the way.) you should probably find a toc on their site about it, if its fancy you can always run it through stalin to get more people than you, too. ;p don't know about
22:32:50 <Vorpal> fungot, Through stalin hm. Need a time machine then.
22:32:50 <fungot> Vorpal: alt key of course, i think it's up to fnord
22:33:21 <Vorpal> you mean I just need an alt key and some fnord? wow
22:33:49 <olsner> fungot rhymes with ergot is a fungus, which shares a prefix with fungot
22:34:00 <Phantom_Hoover> fungot, correct, but it conceded that it wasn't sure whether I was gay. Or a vampire.
22:34:00 <fungot> Phantom_Hoover: it means experience is limited to two-argument functions)
22:34:44 <olsner> elliott: how does fungot generate this? is it some kind of markov chain?
22:34:45 <fungot> olsner: that's not particularly helpful :p those people kept beating each other on the even/ odd
22:34:56 <Vorpal> that would explain why you need to check man pages for stuff like connect() or open()
22:34:59 <olsner> fungot: sorry for not being more helpful...
22:35:00 <fungot> olsner: and i was unable to fnord revision lock ( 403 forbidden)
22:35:18 <elliott> ask fizzie for more details
22:35:23 <olsner> markov chains are funny
22:35:31 <Vorpal> olsner, read the source to find out more
22:35:33 <fungot> http://git.zem.fi/fungot/blob/HEAD:/fungot.b98
22:35:45 <Vorpal> olsner, I believe they are pre-generated by a perl script
22:35:56 <Vorpal> and it then seeks in a file (using FILE)
22:37:24 <fizzie> There's also a thing that can use any n-gram models in the standard-ish ARPA format that I wrote for using our varikn toolkit.
22:38:14 <elliott> like anyone understands that code :)
22:38:26 <fizzie> It's possible to formulate it in a markov-chainian framework, though it's not exactly the usual "dissociated-press" markov chain; there the algorithm is to copy text verbatim, and at opportunate places with some specific probability jump into another place in the text with the same context.
22:38:45 <Vorpal> elliott, um, fizzie does I think
22:38:52 <Vorpal> elliott, and I understand some parts of it
22:39:07 <elliott> "Larry Niven already invented T-Rex's Law? That's crazy! ...Why do you think he named it after me?" --T-Rex
22:39:09 <Vorpal> elliott, with a lot of work I could figure it out completely
22:39:12 <fizzie> In fungot's case it only has the limited context, not a position-in-text, and it chooses the next word based on the conditional probability from the n-gram model.
22:39:13 <fungot> fizzie: which book are you talking about? :p :) you fnord your code?
22:39:28 <olsner> I understand exactly nothing of it, but I'm guessing that ^, <, >, v change the direction of the instruction pointer's movement
22:39:50 <elliott> olsner: YAY YOU GUESSED CORRECTLY
22:39:55 <elliott> fizzie: Add a Dinosaur Comics style.
22:41:13 <elliott> Anyone got NetBSD networking working under qemu?
22:41:57 <olsner> elliott: if only alise was here, I bet alise would know exactly how to do that
22:42:21 <elliott> olsner: Too bad she's with the earth mother.
22:42:46 <elliott> Phantom_Hoover: http://xkcd.com/169/
22:42:47 <Vorpal> elliott, what is the issue?
22:42:53 <elliott> Vorpal: "It doesn't work."
22:43:02 <elliott> Vorpal: /etc/rc.d/network starts properly, no nameserver, nothing works
22:43:03 <Vorpal> elliott, shouldn't it just be to start the virtual network interface and then start dhcp on it?
22:43:13 <elliott> ifconfig -l lists only lo0
22:43:14 <Vorpal> elliott, can you ping any ips?
22:43:23 <Vorpal> elliott, ifconfig -a ?
22:43:27 <elliott> Phantom_Hoover: But you did try and mislead me based on a technicality...
22:43:38 <elliott> Vorpal: Gimme an IP to ping :P
22:43:47 <Vorpal> elliott, "your router"?
22:44:00 <Vorpal> elliott, from what I remember network config is not very automatic on netbsd
22:44:01 <elliott> Using default qemu settings
22:44:48 <Vorpal> fizzie, what is that bit about ^
22:45:01 <Vorpal> why does it parse URLs?
22:46:02 <pikhq> elliott: Y'know what'd be amazing (if somewhat silly)? A distro hosting all files using BitTorrent.
22:46:06 <Phantom_Hoover> http://www.nct.org.uk/press-office/press-releases/view/224 O.o
22:47:12 <elliott> Dammit pikhq, decentralisation is not a panacea :P
22:47:27 <cheater99> wow, i have actually found an ftp link on the internet
22:47:31 <cpressey_> i have someone else fnord my own code for me
22:47:44 <pikhq> elliott: Yeah, but BitTorrent is a good solution to hosting. Make it with HTTP seeds.
22:47:55 <fizzie> Vorpal: It's incomplete; it's supposed to have a load-code-from-URL feature at some point.
22:48:16 <elliott> I bet it's in the TWENTY-FIRST CENTURY SCIENCE course.
22:48:26 <elliott> TWENTY-FIRST CENTURY BULLSHIT ETHICAL QUESTIONS
22:48:30 <fizzie> Vorpal: Yes, though I need H from SCKE (or NSCK or whatnot) for more human-friendly URLs.
22:48:58 <Phantom_Hoover> I know people who did some sort of GCSE chemistry paper earlier this year; I should ask them...
22:49:01 <elliott> [[Charities working to support mothers who want to breastfeed are also negatively caricatured in the question, in the guise of ‘Mrs I M Right’, founder of fictional organisation ‘Responsible Mothers Are Us’.
22:49:01 <elliott> Her extreme views are framed by a reference to the fact that she has ‘made a career in ‘goodness’ and is paid from donations given to RMAU by members of the public’.]]
22:49:50 <elliott> [[Calcium carbonate occurs naturally as marble and limestone. They are important building materials and are often used for gravestones. Calcium carbonate is also an essential mineral for good health and is present in many baby foods in small amounts.My Baby Food is recommended as being the closest to a mother’s own breast milk. It is given free to mothers in the developing world – without it their babies might die of malnutrition.Responsible Mothers
22:49:51 <elliott> Are Us (RMAU) is a United Kingdom pressure group. They want to ban chemicals in baby foods. The group was founded by Mrs I. M. Right who has made a career in ‘goodness’ and is paid from donations given to RMAU by members of the public. When interviewed, she said: “Calcium carbonate is a chemical and so it is a pollutant. My Baby Food must be banned to prevent the mass medication of babies. I don’t feed my baby the stuff of gravestones.”
22:49:51 <elliott> Many people do not agree with Mrs Right’s ideas. Suggest why.]]
22:50:53 <Vorpal> fizzie, NSCK is on hold until I figure how how to support SCTP
22:51:10 <Vorpal> as in: a good API for SCTP
22:52:53 <elliott> cpressey_: what's a regular ethernet interface called on netbsd?
22:53:44 <cpressey_> network interface names were based on (get this) the driver used
22:53:58 <cpressey_> so broadcom was like 'bc0', 'bc1', etc
22:53:59 <Phantom_Hoover> elliott, so wait, 20th Century Science is /literally the only science education your school provides/?
22:54:26 <elliott> Phantom_Hoover: It may have others but -- it's the one they basically do.
22:54:37 <elliott> I am hoping to see if I can get a better course.
22:55:00 <Vorpal> elliott, yeah on freebsd it is based on driver. bc0, en0, and so on
22:55:57 <Phantom_Hoover> elliott, what about for whatever you weird people in England do after GCSE?
22:56:16 <elliott> Phantom_Hoover: I don't think there's Twenty First Century A-Level.
22:56:28 -!- tombom has quit (Read error: Connection reset by peer).
22:57:16 <elliott> But yeah, it's still the physics/biology/chemistry trio.
22:57:17 <Phantom_Hoover> True, but can you do A-Level without doing any prior stuff?
22:58:20 <elliott> Uh. Theoretically I guess.
22:58:23 <elliott> I doubt anyone would let you.
23:00:56 <fizzie> One of the BSDs used "en0" as the most-likely Ethernet interface, but I can't recall which one.
23:02:06 <Vorpal> fizzie, I seen en0 on freebsd, I seen br0 on freebsd too
23:02:39 <Vorpal> I think it was en0 for intel's gbit thingy
23:04:31 <fizzie> I do think OpenBSD had driver-specific names, with ne0 for the ne2k driver and something rather stranger for the SBUS ethernet.
23:06:25 <Vorpal> fizzie, what was SBUS?
23:06:39 <fizzie> It's a bit like PCI except in some sparc boxen.
23:06:54 <fizzie> Well, it was a router.
23:06:54 <pikhq> Hmm. I just realised that one can make trackerless torrents really easily, and that this is an awesome means of sharing files when you don't have a web host. MWAHAHAHAH.
23:07:32 <fizzie> The box had onboard le0, and then I couldn't find a plain SBUS network card, so I got one with both ethernet and an additional SCSI interface too. I don't think I ever connected anything to the SCSI side, since it had on-board SCSI too.
23:07:39 <pikhq> So, anyone want something completely random off my hard drive? I can make data URIs of the xz'd torrent file! :P
23:10:59 <pikhq> (note: probably overkill for anything I've written myself)
23:15:54 <pikhq> Aaah, wait. I could just hand the relevant magnet URI. \o/
23:16:36 <pikhq> Phantom_Hoover: Works just fine.
23:33:57 -!- sebbu2 has joined.
23:34:00 -!- sebbu2 has quit (Excess Flood).
23:34:25 -!- sebbu2 has joined.
23:34:42 -!- sebbu has quit (Ping timeout: 240 seconds).
23:41:48 <elliott> Phantom_Hoover: it works on clients with left-aligned nicks
23:42:37 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
23:46:58 <pikhq> magnet:?xt=urn:btih:12cba1cfeb9a4b96b791a5697f31075e4888814e&dn=ski&tr=http%3A%2F%2Ftracker.publicbt.com%3A80%2Fannounce&tr=http%3A%2F%2Ftracker.openbittorrent.com%3A80%2Fannounce It be a torrent for half a meg!
23:47:44 <pikhq> Sorry, for 39 kilobytes. Hooray, MORE USELESS!
23:49:36 <elliott> pikhq: But what IS it man???
23:49:39 <Sgeo> What... torrent?
23:49:43 <pikhq> elliott: My SKI interpreter.
23:49:49 <cpressey_> http://www.freebsd.org/cgi/cvsweb.cgi/src/usr.sbin/chroot/chroot.c?rev=1.13
23:49:54 <elliott> pikhq: RIP meaningful URIs
23:49:59 <pikhq> elliott: That somebody wanted a few days ago and I couldn't get uploaded to filebin for I DON'T KNOW WHY
23:50:10 <elliott> pikhq: sometime -- whenever pikhq discovered magnet links
23:50:23 <elliott> Eulogy: "I never liked knowing what I was about to click, anyway."
23:50:38 <elliott> pikhq: Also, you have http URIs in there :-O Clearly we need torrent trackers to be distributed by DHT.
23:50:56 <pikhq> elliott: The tracker exchange protocol is not commonly supported yet.
23:50:57 <elliott> cpressey_: nothin' wrong with a nicely-placed goto in C
23:51:02 <elliott> pikhq: YOU'RE not commonly supported yet.
23:51:21 <pikhq> elliott: If it weren't for that, then yeah, I'd just say "fuck the tracker URIs".
23:51:25 <cpressey_> elliott: oh, but this is argument-parsing code.
23:51:36 <elliott> pikhq: magnet:?xt=urn:btih:12cba1cfeb9a4b96b791a5697f31075e4888814e&dn=ski
23:51:57 <pikhq> elliott: Oh, fine, so you can just do without the trackers, too.
23:52:53 <elliott> pikhq: Like I said, the trackers are distributed over DHT.
23:53:31 <cpressey_> should just be syscalls + some scripting language, is what i think, every time i look at one of these sources. oh well
23:53:48 -!- cpressey_ has quit (Quit: John Freeman turned on off the computer).
23:53:53 <pikhq> elliott: I need a better program for generating torrent files. Like, one that doesn't require listing a tracker.
23:54:21 <pikhq> You realise torrent files are binary, right?
23:54:33 <elliott> pikhq: What, your terminal too much of a wimpy bitch to handle binary input?