00:01:33 <elliott> Gregor: More seriously, I think it'll be pretty cool. Although would you be freaked out by programs you compile ending up using a different libc to the rest of the system? 'Cause if so, well.
00:01:46 <Gregor> libc is just another library :P
00:01:48 <elliott> Gregor: (Static linking = just because a new libc comes out, doesn't mean I'm recompiling every fucking package)
00:02:01 <elliott> (but I will package the new libc)
00:02:16 <elliott> (to recompile the entire system)
00:02:39 <Gregor> I'll only like it if you use something like my Separated Packages System :P
00:02:45 <calamari> so what epic argument did I join in the midst of?
00:02:53 <elliott> Gregor: Link me to that again, I forget the specifics.
00:03:00 <Gregor> Do I have a link for it? X-D
00:03:05 <elliott> Gregor: It had a web page!
00:03:16 <elliott> Gregor: With at least ONE example!
00:03:21 <Gregor> http://codu.org/projects/trac/sps/
00:03:30 <elliott> Gregor: It was not Trac! It in fact had no CSS at all!
00:03:34 <elliott> Gregor: But this will do :P
00:03:52 <Gregor> Yeah, I have actually updated it since then. I rewrote it from scratch at least once.
00:03:55 <elliott> Gregor: Right, that horrible piece of bloat :)
00:03:55 -!- nooga has joined.
00:04:20 <elliott> Gregor: More seriously, I don't think I'll have a /usr. So, you know, you'll be messing with what / is. Which is, good luck.
00:04:40 <elliott> Gregor: Well, feel free to create a /usr. And run SPS on it. I might even package it.
00:04:51 <elliott> Gregor: You'll have to port it from apt-get though, which should be trivial :P
00:05:01 <elliott> bash-4.1 emacs-23.2 nginx-0.8.53 ruby-1.9.2-p0
00:05:01 <elliott> CLC-INTERCAL-1.-94.-2 ick-0.-2.0.29 perl-5.12.2 zsh-4.3.10
00:05:01 <elliott> egobf-0.7.1 nasm-2.09.03 Python-2.7
00:05:13 <elliott> Gregor: All installed from HTTP URLs to tarballs and nothing else :P
00:05:33 <elliott> http://www.nongnu.org/sps/ this was the page
00:05:50 <elliott> Gregor: At least you've abandoned D.
00:05:59 <elliott> Gregor: I note that your source repository is remarkably empty for SPS.
00:06:10 <Gregor> It's a good language with a community that's shit itself.
00:06:28 <Gregor> elliott: Uhh, you mean http://codu.org/projects/sps/hg/ ?
00:06:52 <Sgeo> What language?
00:06:54 <elliott> Gregor: It's a bad language with a bad toolchain and a community that has been blinded by the gigantic ejaculate of shittiness so that they can no longer see how crap their whole environment is, and patiently explain how to massage the crap into something vaguely usable to anyone who asks.
00:07:08 <elliott> Gregor: No, I meant the Trac source viewer, which inexplicably has files in it :P
00:07:11 <Gregor> It's a GOOD language. At least D1 was.
00:07:18 <elliott> Gregor: 'Snot. 'S like C++.
00:07:26 <Sgeo> Gregor, what language?
00:07:31 <elliott> Gregor: Okay, that IS the repo I was seeing.
00:09:10 <elliott> Gregor: To be honest, SPS is entirely useless for libraries because of static linking :P
00:10:36 <elliott> Gregor: As far as needing various versions of *binaries*, well... how often do you actually have that need, seriously? Enough to build an entire system around? :P
00:10:48 <elliott> Gregor: (Is putting a symlink named "gcc" somewhere in a temporary $PATH so difficult?)
00:11:11 <Gregor> elliott: It's spiritually based on a system they used at Intel.
00:11:16 <Gregor> For a normal system it's totally useless :P
00:11:36 <elliott> Gregor: Yeah, I'm sure systems as fucked up as Intel will migrate to Kitten right away :P
00:11:42 <elliott> And expect everything to work out of the box, too! :D
00:12:28 <elliott> coppro: Says the coprophiliac.
00:13:01 <calamari> http://bits.ohloh.net/attachments/5031/logo-caca.png
00:13:12 <elliott> That is indeed libcaca's logo.
00:14:20 <elliott> Oh, they changed it. Slightly.
00:15:32 <elliott> Gregor: I just realised the Kitten release day will be plagued by your silly complaints :P
00:17:39 <elliott> Gregor: (Unless you like it and promise to use it forever, and then I'll just have to deal with your bug reports)
00:17:54 <elliott> calamari: file:///home/ehird/mind-fifo
00:17:58 <elliott> (Also ~/kitten for stuff I'm hacking on)
00:18:03 <elliott> Err, not fifo. Device file.
00:18:09 <Gregor> calamari: http://www.webdesign.org/img_articles/7072/BW-kitten.jpg
00:21:33 <calamari> so if Huckabee wins and puts God into the constitution, what does that mean for us godless sinners?
00:21:48 <elliott> calamari: It means no kittens.
00:21:55 <elliott> Also, nothing (apart from the ones who live in the US).
00:22:04 <elliott> Gregor: How STRONG is your hate for ksh?
00:22:23 <Sgeo> calamari, I would hope that Congress wouldn't vote any such amendment in
00:22:43 <Sgeo> </taking-jokes-too-seriously>
00:22:58 <Gregor> elliott: Less than bash or zsh, more than csh?
00:23:11 <calamari> yeah well I'd hope that the president can't choose to kill any citizen he wants to with no oversight, but that can happen
00:23:12 <elliott> Gregor: You... hate ksh more than csh?
00:23:14 <Gregor> That was all backwards
00:23:22 <Gregor> elliott: More than bash or zsh, less than csh?
00:23:24 <elliott> Gregor: ?hsc naht... no, I still don't get it.
00:23:47 <elliott> Gregor: WELL TOUGH! cuz it totally (will) (use)[s] pdksh by default.
00:24:02 <Gregor> The Pretty Damn Kute Shell
00:24:38 <elliott> public domain ksh actually :P
00:24:54 <elliott> Gregor: Well, actually, it's more likely to be the portable version of the OpenBSD ksh.
00:26:04 <elliott> http://www.delilinux.de/oksh/
00:26:04 <elliott> https://github.com/dryfish/openbsd-pdksh
00:26:05 <elliott> http://www.wormhole.hu/~ice/ksh/
00:26:57 <elliott> "To use mksh, you only need the C runtime (and any supplemental libraries the binary was linked against) and, optionally, /bin/ed"
00:27:21 <elliott> "The first example is using BSD paxtar (MirOS BSD, OpenBSD, NetBSD®), or
00:27:21 <elliott> “bsdtar” (DragonFly); the second and third use cpio(1) or pax(1) and are
00:27:21 <elliott> for most operating systems; the fourth is for Debian." ;; there are exactly three examples listed
00:31:21 <elliott> Gregor: Aw man, I'll definitely have to include GNU make.
00:31:58 <elliott> Gregor: Unless you know of a make that's gompatible :P
00:32:13 <elliott> "pymake: A mostly GNU-compatible python implementation of `make`"
00:32:17 <elliott> GOOGLING THIS REASSURES ME
00:32:31 <elliott> Okay, slightly more inspired :P
00:32:34 <elliott> "# Parallel builds (-j > 1) are not yet supported"
00:32:36 <elliott> Inspiration utterly drained
00:34:11 <elliott> │ This option makes grep, sed etc handle rare corner cases │
00:34:11 <elliott> │ (embedded NUL bytes and such). This makes code bigger and uses │
00:34:11 <elliott> │ some GNU extensions in libc. You probably only need this option │
00:34:11 <elliott> │ if you plan to run busybox on desktop. │
00:34:19 <elliott> I like how I want that but can't if I'm not going to use glibc :P
00:35:04 <elliott> │ Store usage messages in compressed form, uncompress them on-the-fly │
00:35:05 <elliott> │ when <applet> --help is called. │
00:36:24 <Sgeo> Why would... maybe I shouldn't bother being part of this channel
00:36:45 <Sgeo> But why would things like handling embedded NUL bytes require extens... Oh. C string suckiness. Right
00:38:36 <elliott> Gregor: I certainly can't find the option to have all applets as separate binaries.
00:39:52 <elliott> │ Line editing code uses on-stack buffers for storage. │
00:39:52 <elliott> │ Symbol: FEATURE_EDITING_MAX_LEN [=1024] │
00:40:44 -!- nooga has quit (Ping timeout: 240 seconds).
00:42:04 <Vorpal> elliott, bought minecraft yet?
00:42:15 <elliott> Vorpal: Ehm, too busy Kittening. Which would you prefer? :P
00:42:30 <elliott> Vorpal: Right now I'm trying to coerce busybox to build a lot of little binaries rather than one gigantic one.
00:42:52 <Vorpal> elliott, I would prefer that you made a giant kitten model in minecraft. Or implemented kitten on that ALU in minecraft!
00:43:02 <elliott> Vorpal: There's a CPU now.
00:43:07 -!- augur has quit (Remote host closed the connection).
00:43:20 <elliott> Vorpal: It had a whopping 8 bytes of RAM, which is apparently going to be doubled.
00:43:32 <Vorpal> elliott, could take a lot of work
00:43:45 <elliott> Vorpal: Not really, it's just tedious (copying the blocks).
00:43:48 <elliott> Of course if you used an editor...
00:44:02 <pikhq> elliott: You should exclusively support x86_64.
00:44:16 <elliott> pikhq: And then what of all the other machines?
00:44:32 <elliott> pikhq: You do realise that all good ThinkPads were 32-bit?
00:44:54 <elliott> Vorpal: screw ARM. no really, screw ARM :P
00:44:57 <pikhq> Vorpal: Everyone knows there can only be one architecture.
00:45:06 <elliott> Vorpal: You can't make "an ARM OS", you have to make 50 ARM OSes pretending to be one.
00:45:39 <Vorpal> elliott, um you could just put the custom code in one place?
00:45:46 <elliott> "a minimal static blog generator written using old-school unix tools (make, ksh, m4, awk, procmail and a pinch of elisp)"
00:45:54 <elliott> make okay, ksh okay, m4 okay, awk okay, ... procmail?!
00:45:57 -!- zzo38 has joined.
00:46:20 <Vorpal> elliott, I mean, linux pretends to be not only x86 and x86_64, but also ARM and what not
00:46:33 <elliott> Vorpal: I do not understand your words. :P
00:47:00 <elliott> Gregor: Gaah, why did you make my brain believe BusyBox would be nice and simple; the damn thing has an implementation of dpkg!
00:47:04 <Vorpal> <elliott> Vorpal: You can't make "an ARM OS", you have to make 50 ARM OSes pretending to be one. <-- so linux is more than 50 OSes (remember the other arches) pretending to be one?
00:47:14 <Gregor> elliott: I never intended to imply it was simple at all :P
00:47:27 <elliott> Vorpal: Except that ARM is wildly inconsistent since it's, like, the most popular CPU around and isn't even one consistent architecture :P
00:47:39 <elliott> And at least x86/64 implies something vaguely like an IBM PC in some ways.
00:47:49 <Vorpal> elliott, there were non-PCs
00:47:56 <elliott> Yes, but you can't boot i386 Linux on them :P
00:48:08 <Vorpal> elliott, it supports some of them
00:48:28 <Vorpal> elliott, some weird 32-bit NUMA system and so on
00:49:16 <elliott> Gregor: Fuck BusyBox, I already have one hellishly deep configuration set to work out (Linux) :P
00:50:09 <elliott> Gregor: Now you are legally required to point me to something that isn't BusyBox (or gnu coreutils) :P
00:50:29 <Gregor> http://www.google.com/search?q=heirloom+toolchest
00:50:35 <pikhq> elliott: You can boot i386 Linux on those odd-ball architectures.
00:50:41 <pikhq> elliott: You just have to build the kernel for them.
00:50:47 <elliott> Gregor: Yeah, I know what Heirloom is. It's what Sgeo would do, if he grew up on UNIX.
00:50:56 <elliott> Gregor: I am not aiming for nostalgia, old Unix sucked and that's the truth :P
00:51:06 <elliott> (I would like to claim the UNIX typo there was intentional.)
00:52:38 <elliott> -rwxr-xr-x 1 elliott elliott 937K Nov 10 00:50 busybox
00:52:43 <elliott> (Okay, so that's dynamic glibc.)
00:52:55 <elliott> (And it probably contains all sorts of useless crap I didn't have the patience to disable.)
00:53:04 <Sgeo> What's the point of keeping the old algorithms (which are somewhat invisible) intact, but removing some nostalgia-inducing limitations?
00:53:17 <elliott> Gregor: I tried compiling the FreeBSD core utilities on OS X once -- the same damn OS, you will recall.
00:53:21 <elliott> Gregor: FIRE AND MOTHERFUCKIN' BRIMSTONE.
00:53:27 <elliott> That code is soooooooo not portable.
00:53:35 <Sgeo> Oh wait, this is actually for "real work"?
00:53:41 <Gregor> elliott: How 'bout NetBSD!
00:53:46 <Gregor> "Linux: Of course it runs NetBSD!"
00:54:07 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
00:54:17 <elliott> fizzie: What did I call that command, again?
00:54:30 -!- EgoBot has quit (Remote host closed the connection).
00:54:40 -!- HackEgo has joined.
00:54:50 -!- EgoBot has joined.
00:56:58 <Gregor> I think the latter, like "shoe in the door"
00:57:04 <Gregor> But apparently other people think the former.
00:57:14 <elliott> Gregor: I thought it was the latter but apparently "not" http://www.wsu.edu/~brians/errors/shoe-in.html
00:57:19 <elliott> A race horse so fast that you can merely shoo it across the finish line rather than having to urge it on with stronger measures is a “shoo-in”: an easy winner. It is particularly unfortunate when this expression is misspelled “shoe-in” because to “shoehorn” something in is to squeeze it in with great difficulty.
00:57:39 <elliott> I think shoe-in has overtaken in the public consc... consciousness of the small portion of the public that says that.
00:58:14 <Gregor> elliott: That description of unfortunateitude makes no sense, since it ignores the expression "shoe in the door" ...
00:58:30 <elliott> Gregor: THARS PRESCRIPTIVISM FOUR YAR
00:59:05 <Gregor> Prescriptivists suck. And not in the good way. I'm spelling it "shoe-in" just to be anti-prescriptivist ... in a prescriptivist kind of way.
00:59:20 <elliott> I'll prescribe YOUR ivist.
01:02:48 <elliott> Gregor: Wikipedia trying to counter our argument for little-endian systems because of some misguided notion of neutrality: "On the other hand, in some situations it may be useful to obtain an approximation of a multi-byte or multi-word value by reading only its most-significant portion instead of the complete representation; a big-endian processor may read such an approximation using the same base-addr
01:02:48 <elliott> ess that would be used for the full value."
01:02:52 <elliott> I wonder if anyone has ever used that ever :P
01:03:53 <Gregor> Yeaaaaaaaaaaaaaaaaah, nice nonsense hypothetical there :P
01:12:01 <elliott> Sgeo: your wiki reign is finally over
01:12:05 <elliott> Sgeo: nethackwiki just moved off wikia
01:12:39 <elliott> now let's watch wikia do their usual thing and try and split the community
01:12:57 <Gregor> Nethackwiki should totes be a Hackiki :P
01:13:18 <elliott> [[Wikia staff members (ie Sannse) have been deleting ANY post which mentions eyestrain or headaches on the community blogs. There have been several ophthalmologists from Britain who've posted that the skin has caused eyestrain to those who have dominant right eyes, and that the eyestrain could easily cause a headache. All posts were deleted with the reason being "spam".]]
01:13:23 <elliott> It is absolutely insane, Wikia.
01:13:32 <Sgeo> elliott, I'm an admin there too
01:13:34 <elliott> Then it was "we made a new skin. FUCK YOU, YOU WILL LIKE IT."
01:14:05 <elliott> Then it was "don't like it? Moving off? We refuse to give you anything, and we WILL continue using your URL and name and promoting it over yours. Also, we registered [yourwiki].com because fuck you. You can't have it. [This actually happened]"
01:14:19 <elliott> Now it's "Expert? Think we're damaging eyes? FUCK. YOU."
01:14:29 <Sgeo> Also, I did get someone more experienced with such moves to help contribute advice
01:14:37 <Sgeo> And elliott still has me blocked
01:14:46 * Gregor plays websplat on the Nethack wiki.
01:15:08 <Sgeo> Gregor, get elliott to either unignore me or read the logs of just now
01:15:09 <elliott> What, http://nethackwiki.com/wiki/Main_Page?
01:15:16 <elliott> Gregor: any new havenworks scores?
01:15:29 <Gregor> elliott: None better than what'd been done before :P
01:15:43 -!- Sgeo has changed nick to CheckLogs.
01:15:46 -!- CheckLogs has changed nick to Sgeo.
01:15:56 <elliott> Gregor: Did Mr. Inhuman play again?
01:15:59 <elliott> Actually, I saw that nick change.
01:17:27 <elliott> (15:56:29 * Sgeo is just glad humanity hasn't been sterilized <-- wut)
01:17:46 <Sgeo> SG-1 "2010" reference
01:17:48 <elliott> 15:36:59 <Sgeo> <Sgeo> A ... geeky person I know seems to have found an issue with common definitions of turing-complete
01:17:56 <zzo38> I added two configured channels now
01:17:58 <elliott> Hardly; ais523 has been over all these ambiguities with his proof.
01:18:11 <elliott> That's as far as I'm reading :P
01:18:27 <Sgeo> I was hoping you'd be reading my response to the wiki stuff >.>
01:18:48 <elliott> 17:18:27 <Sgeo> I was hoping you'd be reading my response to the wiki stuff >.>
01:19:35 <Sgeo> Is elliott having trouble counting without Captain Obvious?
01:20:00 <coppro> elliott should take some C&O lessons
01:20:36 <coppro> elliott: combinatorics & optimization
01:20:45 <elliott> Gregor: [[In Related languages section, after the phrase "Many people at various times have tried to extend brainfuck to make it easier to program in", I would like to delete "but such efforts have been compared to trying to make a luxury car by gluing parts onto a skateboard" and put in this place "BrainSub is the first one to achieve this goal in 2007"; unless someone have reasons to not to do this c
01:20:46 <elliott> hange. What do you think? Aacini 03:50, 18 July 2007 (UTC)]]
01:20:49 <elliott> coppro: Yes, I am not sure why you said that.
01:20:51 <Sgeo> He needs to learn to count first
01:21:15 <Gregor> elliott: laaaaaaaaaaaaaaaawl
01:21:31 <coppro> Sgeo: no, that's where he goes to learn to count
01:21:42 <coppro> oh right, elliott has Sgeo on /ignore
01:21:47 <coppro> that's why he doesn't get it
01:22:36 <Gregor> This BrainSub person is an idiot :P
01:23:11 -!- elliott has quit (Quit: Leaving).
01:24:56 <Sgeo> Let's put COBOL in a modern runtime! Oh, wait, that's been done for real. For serious reasons.
01:25:21 <Sgeo> COBOL.NET and BrainSub were made by the same person!
01:25:29 <Sgeo> Can't you see?
01:28:14 -!- zzo38 has quit (Remote host closed the connection).
01:31:50 <Sgeo> I think Gregor has his ignore list hardlinked to elliott's
01:32:11 <Sgeo> Acting like it sometimes though
01:53:24 * Sgeo catches the end of Time's Arrow
02:02:20 -!- Sasha has quit (Read error: Connection reset by peer).
02:05:23 -!- Sasha has joined.
02:10:14 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.12/20101026210630]).
02:16:04 * Sgeo immediately notes an averted trope in this SGU episode
02:19:01 <Sgeo> The ship did not stop moving when engines went off
02:19:14 <coppro> I need to watch this season
02:23:25 <coppro> hmm.. 3 weeks from now
02:23:30 * coppro debates torrenting now vs. then
02:29:59 -!- Mathnerd314 has joined.
02:32:01 <Sgeo> Is it possible to torrent now, and then later only the relevant episodes?
02:33:27 <Gregor> That sentence makes no sense, but I THINK you're talking about downloading only particular files via *torrent, which you can do with certain clients.
02:33:36 <Gregor> Such as most clients, for example.
02:41:41 <Sgeo> This episode is somewhat painful
02:42:02 <Sgeo> Just.. plot hole wise. Or maybe I don't remember some previous episodes well enough
02:45:52 <coppro> but that requires getting two torrents
02:47:22 <pikhq> Wow. Car manufacturers design 3 different versions of their cars: US, rest of the world right-hand-drive, rest of the world left-hand-drive.
02:47:59 <pikhq> Because, just to be contrary (honest, *just to be contrary*), our safety standards are completely different from every other countries'.
02:48:14 <pikhq> (and incompatible)
02:49:18 <coppro> actually just to be incompatible?
02:49:34 <coppro> that sounds so very american
02:49:35 <pikhq> The world safety standards came *before* the US ones.
02:49:40 <coppro> examples of incompatibilities?
02:50:00 <pikhq> The types of headlights mandated in the rest of the world are banned in the US.
02:51:35 <pikhq> Also, US safety standards are based around increasing safety without increasing cost appreciably.
02:51:39 <Gregor> And yet these Zenon (or whateverTV) bulbs are allowed?
02:51:39 <Gregor> Oh wait, are they actually Xenon?
02:51:52 <Sgeo> What's bad about Xenon?
02:51:54 <Gregor> I thought it was just some brand name, never occurred to me that it may not be X-D
02:52:25 <Sgeo> Oh, the ultra bright bulbs?
02:52:31 <Gregor> Sgeo: Xenon bulbs are so fucking bright that they blind everyone unlucky enough to be in their path. So the driver may be able to see the pedestrian (unless he's blinded by the glare), but the pedestrian is blind.
02:53:36 <Sgeo> You're only supposed to use them in certain conditions, I think
02:53:45 <Sgeo> Although, drivers being drivers...
02:53:59 <pikhq> Gregor: In the rest of the world, such headlights require lens cleaners and automatic beam levelling for the purpose of reducing the glare and blinding.
02:54:06 <pikhq> Gregor: In the US, such things are optional.
03:01:47 <Sgeo> SGU SPOILER IN TRAILER FOR NEXT EPISODE FFFFFFFUUUUUUUUUUUUUUUUUU
03:03:04 <Sgeo> 's already editing it to make me wonder if I misremembered and maybe they said something else
03:03:25 <Sgeo> And of course, it's possible that what I heard happen doesn't happen for "real"
03:03:35 <pikhq> Oh, and US safety standards also have the effect of reducing fuel efficiency *while* decreasing safety.
03:03:47 <Sgeo> pikhq, how so?
03:03:54 <coppro> pikhq: but it costs no more, right?
03:04:00 <coppro> that's what's important!
03:04:39 <pikhq> Sgeo: The safety standards require less safe and less fuel-efficient designs.
03:04:57 <Sgeo> Stargate Wiki: FFFFFFFFFFUUUUUUUUUU even more
03:05:13 <coppro> and you went to a wiki
03:05:14 <pikhq> In particular, our mandate for 1940s headlight designs up until the early 90s made creating aerodynamic cars almost impossible.
03:05:17 <Sgeo> I am caught up
03:05:31 <coppro> then wth did the wiki do?
03:05:37 <pikhq> We motherfucking hate logic.
03:05:53 <Sgeo> "**** will have a *-episode story arc"
03:06:25 <Sgeo> Although actually, the given number doesn't quite match ... um, actuality
03:07:24 <Sgeo> Wait, n-episode might mean she's not involved in every episode
03:07:35 <Sgeo> So the n only includes episode's she's in?
03:08:22 <Sgeo> Still, I heard what I heard. And that n is a small number
03:08:43 <Sgeo> I gave away more than I intended to
03:10:44 <pikhq> US classification of weapons makes no fucking sense.
03:11:24 <pikhq> All GPS receivers capable of functioning above 18km altitude and 515 m/s are considered weapons under US law.
03:11:59 * Sgeo has no use for such a devic... _capable_?
03:11:59 <coppro> what about Galileo receivers?
03:12:14 <pikhq> As such, all consumer GPS receivers have code for disabling that.
03:12:29 <pikhq> Sgeo: Yes, *capable*.
03:13:04 <Sgeo> I had this missile lying around, and I wanted to target a specific location, so I got a GPS. Sadly, they disabled the ability to work under the conditions I need.
03:13:08 <pikhq> coppro: No such stupidity.
03:13:18 <pikhq> Sgeo: Doesn't stop cruise missiles.
03:13:27 <pikhq> Sgeo: Just ballistic missiles.
03:13:42 <pikhq> I'm pretty sure someone with a ballistic missile could make a GPS receiver.
03:14:14 <Sgeo> Doesn't the military have the ability to disable GPS or something?
03:14:51 <pikhq> coppro: Actually, they just disabled the feature that gives the military higher accuracy than consumer devices.
03:15:15 <pikhq> coppro: They can still selectively *disable* GPS in a region except for the US military.
03:20:15 <pikhq> Oh, awesome. Soviet Russia did a global navigation satellite system as well.
03:20:18 <pikhq> It still functions.
03:20:53 <pikhq> It fell into disrepair in the 90s, but Russia committed to restoring it by 2010 back in 2000.
03:22:12 <Sgeo> Did they succeed?
03:22:38 <pikhq> Sgeo: It covered the world once again in September.
03:22:50 <pikhq> Had full coverage of Russia earlier.
03:23:54 <Sgeo> How do sattelites manage to have full coverage of exactly one area of the globe for a significant period of time?
03:23:55 <pikhq> They're also about to launch more satellites that will broadcast on the same frequencies as GPS, GALILEO, Compass, et al.
03:24:14 <pikhq> (so that a multistandard receiver would be easier)
03:24:53 <Sgeo> Hypothetically, could someone launch a satellite that sends fake GPS or GALILEO signals?
03:25:44 <pikhq> Sgeo: Actually, it's more that each satellite has coverage of a *path*. Full coverage in an area involves being able to access more than one satellite at the same time 24/7, by having the paths intersect just right.
03:26:39 <pikhq> Sgeo: Not very well. One of the things that those satellites do is broadcast information about *all* the satellites (rough orbital path, general system health, etc.).
03:29:46 <pikhq> And how they actually get the positioning down is by each satellite having an atomic clock and broadcasting a time signal.
03:31:24 <pikhq> Account for the speed of light, you get the distance from each satellite, triangulate, voila.
03:31:44 <Sgeo> And I assume the receiver calculates where each sattelite should be? Where does that part occur?
03:32:25 <pikhq> It knows the orbital path of each satellite.
03:32:35 <pikhq> Because it got told by all the satellites.
03:33:00 <Sgeo> you said "rough" orbital path
03:33:55 <pikhq> The satellite also broadcasts its own orbital path as accurately as can be measured.
03:35:11 <Sgeo> Does the receiver have much use for the excess information (path of other satellites)?
03:35:40 <pikhq> Makes it easier to find the other satellites.
03:36:04 <pikhq> And know if some satellites are bogus.
03:44:54 <Ilari> Defintion of "weapon of mass destruction" is also quite bad. Certainly more broad than the classical definition of "ABC".
03:48:07 <Ilari> Or even the extended defintion of CBRN...
03:50:06 <pikhq> And why is encryption a form of weaponry?
03:50:15 <Ilari> Wonder if the GPS receivers are hardened against bogus data (systems receiving "trusted" data might not be)...
03:50:51 <pikhq> I'd imagine consumer receivers aren't, but military receivers are.
03:51:05 <Ilari> There was some STB that could be hung quite badly by sending random garbage as data...
03:51:36 <Ilari> Intentionally bogus data is even worse than random garbage.
03:52:20 <pikhq> The reason for that guess is simple: the military gives a shit about their systems always working, and consumer electronics manufacturers give a shit about saving pennies.
04:01:05 <Sgeo> Surely it's a matter of software to do some checking?
04:01:30 <Sgeo> Although I guess development time does cost money, so
04:01:53 <Sgeo> I don't see how it's not basically a one-time cost though
04:04:03 <pikhq> Remember: business is run by cheap-ass bastards.
04:07:08 -!- calamari has quit (Quit: Leaving).
04:07:11 <Gregor> I wonder if you could dye your hair a color that no human has, but is subtle enough the people wouldn't immediately recognize it as fake.
04:08:51 -!- augur has joined.
04:14:28 * Sgeo learns about the Nuclear Football
04:17:30 -!- bsmntbombdood has quit (Remote host closed the connection).
04:21:56 -!- zzo38 has joined.
04:52:48 <zzo38> Hello. Welcome to ;jaljsdfa;lsjfwijf.0wjt.4wtj42q3t1t3ji0gjehjNO CARRIERljq3;4tlj,,v;tc.-0935-91324`;skv a,ijgrntwerjk;
05:14:44 -!- Ilari_an1rcomp has joined.
05:16:17 -!- yiyus__ has joined.
05:16:29 -!- Ilari_ has joined.
05:16:41 -!- mycrofti1 has joined.
05:17:10 -!- fungot has quit (*.net *.split).
05:17:10 -!- yiyus_ has quit (*.net *.split).
05:17:10 -!- fizzie has quit (*.net *.split).
05:17:10 -!- Ilari_antrcomp has quit (*.net *.split).
05:17:10 -!- Ilari has quit (*.net *.split).
05:17:10 -!- mycroftiv has quit (*.net *.split).
05:17:11 -!- fizzie` has joined.
05:18:14 -!- Ilari_an1rcomp has changed nick to Ilari_antrcomp.
05:18:15 -!- Ilari_ has changed nick to Ilari.
05:25:14 -!- augur has quit (Remote host closed the connection).
05:39:58 <zzo38> Can GraphicsMagick with with MIFF format?
05:48:04 -!- Sgeo has quit (Read error: Connection reset by peer).
05:56:33 -!- augur has joined.
06:10:21 -!- augur has quit (Remote host closed the connection).
06:19:58 -!- zzo38 has quit (Remote host closed the connection).
06:30:23 -!- augur has joined.
06:40:12 -!- MigoMipo has joined.
06:44:23 -!- FireFly has joined.
06:50:31 -!- fizzie` has changed nick to fizzie.
07:30:22 -!- Slereah_ has quit (Ping timeout: 250 seconds).
07:36:39 -!- Slereah has joined.
07:40:03 -!- jcp has quit (Ping timeout: 264 seconds).
07:41:35 -!- jcp has joined.
07:43:19 -!- MigoMipo has quit (Read error: Connection reset by peer).
07:51:29 -!- wareya_ has joined.
07:54:08 -!- wareya has quit (Ping timeout: 265 seconds).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:05 <Vorpal> elliott (for log reading): yellow text: "NP is not in P!" XD
08:00:49 <Vorpal> and if he did, would it give him more money to write the paper or work on minecraft?
08:03:58 <fizzie> The Millennium prize for "P ? NP" is $1M; minecraft.net says "579053 purchases" at the moment, so with 9.95 EUR/purchase, that's gross profits of about $7.9M. Of course it's anyone's guess how large a percentage of that will be burned by the startup company; probably pretty large.
08:04:20 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.12/20101026210630]).
08:28:51 <Vorpal> fizzie, NP has a strange name considering it isn't known to not be in P
08:29:40 <fizzie> I'unno, it's not N for "non", after all.
08:31:38 <Vorpal> ah wait, I'm still half asleep
08:35:25 <Vorpal> well, bbl (university)
08:36:07 <coppro> http://www.salon.com/technology/how_the_world_works/2010/11/09/john_shimkus_god_and_noah
08:36:11 <coppro> pikhq: your government is fucked up
08:36:50 <Ilari> Heh... If there is problem in NEXPTIME that's not in EXPTIME, then NP does not equal P.
08:59:18 -!- Slereah has quit (Ping timeout: 276 seconds).
09:00:12 -!- FireFly has quit (Quit: swatted to death).
09:04:05 -!- Slereah has joined.
09:23:14 -!- Decarabia has quit (Ping timeout: 252 seconds).
09:25:52 -!- Decarabia has joined.
10:59:29 -!- Deewiant has quit (Ping timeout: 252 seconds).
11:02:22 -!- Deewiant has joined.
13:27:24 -!- sftp has joined.
13:52:46 -!- nooga has joined.
14:09:13 <Gregor> fizzie: Actually it is "non" ... non-deterministic anyway :P
14:09:37 <fizzie> Well, if you want to be a PEDANTIC PEDANTY PEDAN about it.
14:20:54 -!- elliott has joined.
14:21:07 <elliott> 21:39:58 <zzo38> Can GraphicsMagick with with MIFF format?
14:21:25 <elliott> 00:00:05 <Vorpal> elliott (for log reading): yellow text: "NP is not in P!" XD
14:21:41 <elliott> 00:03:58 <fizzie> The Millennium prize for "P ? NP" is $1M; minecraft.net says "579053 purchases" at the moment, so with 9.95 EUR/purchase, that's gross profits of about $7.9M. Of course it's anyone's guess how large a percentage of that will be burned by the startup company; probably pretty large.
14:25:52 <elliott> 19:50:06 <pikhq> And why is encryption a form of weaponry?
14:25:52 <elliott> 19:50:14 <pikhq> Fuck the US.
14:25:54 <elliott> non-US countries have that
14:27:55 <fizzie> PayPal merchant rate says it's (2.9 %+$0.30) per transaction by default, (1.9%+$0.30) for the "merchant rate" with >$100k monthly payments. Then there's a +1% cross-border fee, and +2.5% currency-conversion fee that I guess quite often would apply to someone in Sweden accepting money in EUR. And maybe other hidden fees; still, I don't think they can conceivably eat more than, say, a third.
14:28:52 <fizzie> I think quite a few indie-ish game-developers are feeling the envy. I know at least one. :p
14:33:04 <elliott> fizzie: Hell, I'm jealous of my indie-game-developer-blah friend, and he only makes on the order of thousands.
14:33:15 <elliott> (But then his games take a lot less work to make than Minecraft. :P)
14:36:41 <fizzie> "A ′′symmetric algorithm′′ employing a key length in excess of 56 bits" is part of the EU-wide export control restrictions; see the 269 pages of Regulation 428/2009 at http://trade.ec.europa.eu/doclib/docs/2009/june/tradoc_143390.pdf
14:37:13 <elliott> How to deal with crypto export laws when publishing open source software on the Internet: Ignore them. :p
14:37:20 <fizzie> There are all kinds of exceptions, of course.
14:37:27 <elliott> (Although put some kind of "If this is illegal for you then, welp" notice there.)
14:37:57 <fizzie> You are allowed to put strong encryption into a DVB set as long as it is "exclusively used for sending the billing or programme-related information back to the broadcast providers".
14:38:32 <fizzie> Also okay are things that are "specially designed and limited for banking use or ’money transactions’".
14:39:24 <fizzie> A subnote of 5A002 Note d clarifies that "settlement of fares" is a type of money transaction.
14:40:49 <fizzie> And mobile phones as long as they can only talk directly into the network operator's hardware, and can't do end-to-end encryption.
14:41:35 <fizzie> Also good: "Portable or mobile radiotelephones and similar client wireless devices for civil use, that implement only published or commercial cryptographic standards (except for anti-piracy functions, which may be non-published) and also meet the provisions of paragraphs b. to d. of the Cryptography Note ..."
14:41:36 -!- augur has quit (Remote host closed the connection).
14:42:10 <fizzie> It's a long long document.
14:42:37 <elliott> fizzie: "Crypto is weaponry, unless you use it for things you'd use crypto for."
14:43:05 <fizzie> "Personal area network" things that involve only "published or commercial" standards are exempt as long as "the cryptographic capability is limited to a nominal operating range not exceeding 30 metres according to the manufacturer’s specifications".
14:43:24 <fizzie> "Sorry, you can't ship this: the crypto goes three metres too far."
14:46:37 <elliott> Plan: Hijack fizzie's laptop server. Use copious disk space to store Kitten packages.
14:46:38 -!- FireFly has joined.
14:46:48 <elliott> (TODO: Find out how much disk fizzie's laptop server has.)
14:46:58 <elliott> (TODO: Avoid law enforcement.)
14:47:37 <fizzie> I went to do a df for you, but apparently the vserver stuff messes that up a bit.
14:47:45 <fizzie> Filesystem Size Used Avail Use% Mounted on
14:47:45 <fizzie> /dev/hdv1 5.0G 971M 3.8G 21% /
14:48:12 <elliott> fizzie: That's quite alright! Just do a ... whatever command creates one of them servers ... for me1
14:48:51 <fizzie> I don't think I'm going to start server-hosting, sowwy.
14:49:02 <elliott> fizzie: No, you see, I'll be hijacking.
14:49:07 <elliott> So you'd have no official position at all!
14:50:10 <fizzie> In actual reality the external USB disk thing has 233G, of which 151G is used. Why is there so much stuff in there, in fact?
14:50:28 <elliott> fizzie: Yeah, that thing could store far more Kitten packages.
14:51:16 <fizzie> "Suppliers wishing to apply for authorisation should contact the competent national authorities for details of what information must be supplied." But what if everyone in my country is incompetent?
14:51:48 <elliott> (I am not quite sure what political opinion that makes.)
14:52:00 -!- jcp has quit (Read error: Operation timed out).
14:52:02 <elliott> Also there's the obvious nonsensicalness of it all.
14:52:28 <fizzie> Also the page on how export control is derived from UN resolutions is titled "How to prevent proliferation of horrific weapons, weapons of mass destruction?"
14:52:47 <fizzie> HORRIFIC 35-meter-range personal area networks.
14:53:19 <elliott> fizzie: I like the repetition.
14:53:33 <elliott> HORRIFIC WEAPONS. EWAPONS OF MASS DESTRUCTION!
14:54:40 <fizzie> Ewoks of mass destruction.
14:58:05 -!- jcp has joined.
14:59:14 <elliott> Gregor: Wait, non-GNU/Linux has a serious problem: what binutils?
14:59:20 <elliott> I don't know of any maintained ones apart from the GNU ones.
15:10:12 <elliott> <lynx> gnu/linux is about the worst host OS for it
15:10:12 <elliott> <lynx> their system headers are *so* incompatible…
15:10:12 <elliott> <lynx> gcc and glibc specifc, and all that
15:10:12 <elliott> <elliott> (non-GNU)/Linux actually :-)
15:10:17 <elliott> People as crazy as me! Whoo!
15:18:39 <elliott> Gregor: apparently in the EU interfaces aren't copyrightable
15:35:56 <elliott> │ The amount of time saved by this optimization is actually too small to │
15:35:57 <elliott> │ measure. The linker just had to search the library path to find the │
15:35:57 <elliott> │ linker script, so the dentries are cache hot if it has to search the │
15:35:57 <elliott> │ same path again. But it's what glibc does, so we do it too. │
15:38:14 -!- Sgeo has joined.
15:38:17 <elliott> 1. Compile uClibc (with gcc). 2. Compile pcc (with gcc and uClibc). 3. Compile uClibc (with pcc). 4. Compile pcc (with new uClibc and pcc).
15:49:39 <elliott> Oh well, utter fail; seems my gcc is not friends with uClibc. And I don't feel like building gcc right now.
15:52:41 <Sgeo> I lost two points on a subjective question
15:53:05 <elliott> coppro: (I'll probably use clang for C++ programs)
15:53:21 -!- augur has joined.
15:53:24 <elliott> And before you say clang has substandard C++ support, it's not nearly as substandard as pcc's C++ support, because it has none.
15:53:25 <Sgeo> "What sort of tasks is pattern matching useful for? Name three"
15:53:49 <Sgeo> http://pastie.org/1287296
15:53:57 <elliott> coppro: I could just build a gcc made to work with uClibc, but I'm trying to get away from gcc and its build system, dammit! (I could just use buildroot but I think it wanted me to build a Linux kernel and I like doing that myself thankyouverymuch).
15:54:42 <elliott> "# Supports several hundreds of packages for userspace applications and libraries: X.org stack, Gtk2, Qt, DirectFB, SDL, GStreamer and a large number of network-related and system-related utilities and libraries are supported."
15:55:12 -!- nooga has quit (Ping timeout: 250 seconds).
15:55:16 <coppro> elliott: 1. Compile clang (with gcc). 2. Compile uClibc (with clang). 3. Compile clang (with clang and uClibc). 4. Compile uClibc (with clang) etc.
15:55:17 -!- nooga has joined.
15:55:28 <elliott> coppro: step 4 is unnecessary
15:56:03 <elliott> coppro: also, i know how to build clang, i just don't want to :) Besides, I bet you that the clang you build in step 1 won't be able to build uClibc programs properly.
15:56:16 <coppro> elliott: why do you say that?
15:56:41 <elliott> coppro: because my gcc can't, and my gcc is just a regular one, like the regular clang you build in step #1
15:56:53 <coppro> elliott: why can't it?
15:56:58 <elliott> coppro: resulting programs segfault
15:57:44 <elliott> coppro: plus -nostdlib -nostdin -Lblahblah -Iblahblah blahblah/{crt1.o,libc.a} is a bit of a mouthful. it is possible i did something wrong out of laziness :)
16:02:51 -!- Sasha has quit (Ping timeout: 240 seconds).
16:04:26 <coppro> elliott: well, clang won't compile uclibc anyways :(
16:04:41 <coppro> stupid inline assembly
16:04:54 <elliott> coppro: I am not that hopeful that pcc will compile uClibc either. :)
16:05:02 <elliott> coppro: wait, clang supports inline asm right?
16:05:14 <coppro> elliott: yeah, but there are some weird gcc-only magicks
16:05:23 <coppro> I imagine pcc will choke on them too
16:05:25 <elliott> coppro: you can disable them
16:05:36 <elliott> at the cost of some speed, but what's that for anti-gcc purity?
16:05:49 <elliott> coppro: you have to make menuconfig to use uclibc anyway :) it makes some silly decisionsb y default
16:06:01 <elliott> it can't decide whether the default config is for desktopish things or embedded devices
16:07:32 <elliott> I am pretty sure that I can't avoid binutils :(
16:15:52 <Sgeo> Almost half of the people got the true/false The string "syf" matches the regular expression "sn?yf" wrong
16:17:12 <elliott> /usr/bin/gcc -c -I/home/elliott/kitten/buildroot-2010.08/output/host/include -I/home/elliott/kitten/buildroot-2010.08/output/host/usr/include -DIN_GCC -DCROSS_DIRECTORY_STRUCTURE -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wcast-qual -Wold-style-definition -Wc++-compat -Wmissing-format-attribute -pedantic -Wno-long-long -Wno-variadic-macros -Wno-overlength-strings -DHAVE_CONFIG_H -I. -I. -I/home/elliott/kitten/buildroot
16:17:12 <elliott> -2010.08/output/toolchain/gcc-4.4.4/gcc -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/. -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/../include -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/../libcpp/include -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gmp/include -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/mpfr/include -I/home/elliott/kitten
16:17:13 <elliott> /buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/../libdecnumber -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/../libdecnumber/dpd -I../libdecnumber /home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/gcc/gcov-dump.c -o gcov-dump.o
16:17:28 <elliott> coppro: Unix and C, aren't they so simple and lean and wonderful and crisp?
16:17:28 <Sgeo> I'd spam elliott in relatiation, but...
16:21:05 <elliott> "Tinycc can already rebuild itself (for x86 and arm targets), and has previously built a modified subset of an older (2.4) linux kernel. I'm upgrading it to work on more hosts (such as my x86-64 laptop), support more targets (x86-64, mips, powerpc...), and to build more software (especially a current unmodified 2.6 Linux kernel).
16:21:05 <elliott> This project is on hold. I need to replace its code generator with TCG from QEMU, and break it up into a swiss-army-knife binary that can be called as "cc", "ld", "as", "strip", and so on, as appropriate."
16:21:09 <elliott> http://www.landley.net/code/tinycc/ ... but he gave up
16:26:33 -!- Zuu has joined.
16:39:18 <Sgeo> Is there a concept of smallest regex that matches all and only strings within a set of strings?
16:42:06 <coppro> elliott: I imagine it exists
16:42:11 <elliott> coppro: imagine what exists?
16:43:20 -!- coppro has set topic: Number of times elliott has been confused because he /ignored Sgeo: 3 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:44:27 -!- elliott has set topic: Number of times elliott has been confused because he /ignored Sgeo: 3 | Number of times elliott has avoided banging his head against a brick wall because he /ignored Sgeo: countless | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:45:53 * Sgeo pushes elliott's head against a brick wall
16:47:29 <Sgeo> It's a more recognizable word for what you meant. You know the MasterCard commercials... then again, it's not like number of times confused is a price
16:47:58 <Sgeo> I'd say that was a Captain Obvious thing, but since elliott didn't get it
16:49:31 <elliott> coppro: oh you mean the topic?
16:49:36 <elliott> "priceless" makes no sense in that context
16:50:02 <elliott> echo "Welcome to Buildroot" > /home/elliott/kitten/buildroot-2010.08/output/target/etc/issue
16:50:37 <elliott> *ugh* and i'm left with a useless pile of disorganised software
16:56:10 <elliott> coppro: HAHAHA IT WORKED FUCK YOU
16:56:45 <coppro> elliott: Sgeo explained it... oh wait
16:57:01 <elliott> coppro: increment the counter and you die :)
16:57:08 <elliott> coppro: I assume it's a mastercard ad reference
16:57:16 <elliott> coppro: but, uhh, "priceless" makes no sense there.
16:57:29 -!- Gregor has set topic: Number of times elliott has been confused because he /ignored Sgeo: 3 | Number of times elliott has avoided banging his head against a brick wall because he /ignored Sgeo: countless | Number of days without oerjan's passionate embrace: 19 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:57:59 -!- elliott has set topic: Number of times elliott has been confused because he /ignored Sgeo: 3 | Number of times elliott has avoided banging his head against a brick wall because he /ignored Sgeo: countless | Number of days without oerjan's passionate embrace: 19 | Number of days since the topic last changed: 39 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:58:20 -!- elliott has set topic: Number of times elliott has been confused because he /ignored Sgeo: 3 | Number of times elliott has avoided banging his head against a brick wall because he /ignored Sgeo: countless | Number of days without oerjan's passionate embrace: 19 | Number of days since the topic last changed: 39 | Number of days the channel has been logged in the new clog directory: See http://tunes.org/~nef/log.
16:58:28 -!- elliott has set topic: http://tunes.org/~nef/logs/esoteric/?C=M;O=D: 34.
16:58:41 -!- elliott has set topic: 42: http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
16:58:58 -!- Sgeo has set topic: 42: http://tunes.org/~nef/logs/esoteric/?C=M;O=D | number of times elliott has changed the topic: countless.
16:59:17 -!- Sgeo has set topic: 42: http://tunes.org/~nef/logs/esoteric/?C=M;O=D | number of times elliott has changed the topic: countless (not technically).
16:59:23 -!- elliott has set topic: ASIEKIERKA FOREVER | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:00:54 -!- Sgeo has set topic: ASIEKIERKA FOREVER | http://tunes.org/~nef/logs/esoteric/?C=M;O=D | amount of topic spam: uncountably infinite.
17:01:13 <Sgeo> Obviously, that's a deliberate lie. Then again, so is infinite
17:01:27 -!- Gregor has set topic: Welcome to #esoteric, the international hub for the occult, voodoo, crystal healing, esoteric topics in computation and programming languages, magick, astrology and spiritual projection | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:01:37 <elliott> Gregor: That upsets ais523 :P
17:01:45 -!- elliott has set topic: praise be unto ais523 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:01:54 <elliott> Gregor: WE CAN'T GO ON WITHOUT OERJAN *OR* AIS523
17:02:29 -!- Gregor has set topic: Welcome to #esoteric, the international hub for the occult, voodoo, crystal healing, esoteric topics in computation and programming languages, magick, astrology and spiritual projection | Praise be unto the enlightened one, Zendu, who currently possesses the mind and being of ais523 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:02:47 -!- elliott has set topic: THERE IS NO AIS523. THERE IS ONLY ZUUL. http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:03:04 <Vorpal> <elliott> 00:00:05 <Vorpal> elliott (for log reading): yellow text: "NP is not in P!" XD <elliott> link <-- link how?
17:03:08 -!- Gregor has set topic: Welcome to #esoteric, the international hub for the occult, voodoo, crystal healing, esoteric topics in computation and programming languages, magick, astrology and spiritual projection | Praise be unto the enlightened one, Zenduul, who currently possesses the mind and being of ais523 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:03:10 <elliott> Vorpal: link to whatever page had that
17:03:11 <Vorpal> elliott, it was on the screen in minecraft
17:03:27 -!- elliott has set topic: THERE IS NO ZUUL. THERE IS ONLY GREGOR'S UNWASHED BEARD. http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:03:27 <Vorpal> ah the update was release
17:03:38 -!- elliott has set topic: THERE IS NO ZUUL. THERE IS ONLY GREGOR'S UNWASHED BEARD. http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
17:04:12 <elliott> Gregor: All purple bananas are poisonous. All Gregor's beards are unwashed.
17:04:23 <Vorpal> elliott, on the minecraft wiki version history. "Known bugs: Pressing F4 spawns a portal to the Nether near the player's location. This is probably debug code that Notch forgot to remove before the release."
17:04:37 <Vorpal> elliott, a bit like cheating though
17:04:43 <elliott> Vorpal: Probably near = nearest multiple of 8.
17:04:56 <elliott> Vorpal: Since otherwise you could get overlapping portals by making one, moving to the right a little bit, and making another.
17:05:07 <Vorpal> also texture pack support h,m
17:05:09 <elliott> Both would go to the same place; where would you end up if you went into the single Nether one?
17:05:39 <elliott> elliott@dinky:~/kitten$ ./buildroot-2010.08/output/staging/usr/bin/x86_64-unknown-linux-uclibc-gcc -static hello.c -o hello
17:05:39 <elliott> elliott@dinky:~/kitten$ ls -lh hello
17:05:39 <elliott> -rwxr-xr-x 1 elliott elliott 11K Nov 10 17:05 hello
17:05:56 <elliott> Haven't configured uClibc much or anything, and it's typical bloated gcc code. But I'm going to use this to bootstrap pcc and uClibc.
17:06:52 <elliott> "Still, two more weeks to try to finish stuff up, so I'm building packages as fast as I can. One of them is gnu libiconv. Its configure stage says:
17:06:52 <elliott> checking for iconv... (cached) no, consider installing GNU libiconv"
17:07:27 <elliott> If it was ghc, telling you to install ghc... that wouldn't be a lie.
17:07:32 <elliott> (At least SBCL can build with CMUCL and CLISP.)
17:08:16 <Vorpal> elliott, "near" seems to be "on top of you"
17:08:27 -!- Sgeo has quit (Ping timeout: 255 seconds).
17:08:33 <elliott> elliott@dinky:~/kitten/pcc/pcc$ CC=$HOME/kitten/buildroot-2010.08/output/staging/usr/bin/x86_64-unknown-linux-uclibc-gcc LDFLAGS="-static" ./configure --prefix=$HOME/kitten/stage1
17:08:36 <Vorpal> elliott, or rather: around you
17:08:47 <elliott> So you end up... inside the portal?
17:08:54 <elliott> /home/elliott/kitten/buildroot-2010.08/output/staging/usr/bin/x86_64-unknown-linux-uclibc-gcc -DLIBEXECDIR=\"/home/elliott/kitten/stage1/libexec/\" -DGCC_COMPAT -DINCLUDEDIR=\"/home/elliott/kitten/stage1/include/\" -DPCCINCDIR=\"/home/elliott/kitten/stage1/lib/pcc/x86_64-unknown-linux-gnu/0.9.9/include/\" -DPCCLIBDIR=\"/home/elliott/kitten/stage1/lib/pcc/x86_64-unknown-linux-gnu/0.9.9/lib/\" -Dos_linux -DTARGMACH=amd64 -Dmach_amd64 -I../.. -I../
17:08:54 <elliott> ../os/linux -I../../mip -I../../arch/amd64 -g -O2 -Wall -Wmissing-prototypes -Wstrict-prototypes -Wshadow -Wsign-compare -c cc.c
17:09:05 <elliott> make[2]: ./mkext: Command not found
17:10:31 <elliott> BUT WILL IT BLEND^Wwork with uClibc?
17:10:40 <Gregor> elliott: LDFLAGS is a wonky concept in modern times anyway; are ldflags the flags passed to ld, or are they flags used for linking? Since they don't accept the same input, it's a sticky situation. There's no "correct" use of LDFLAGS.
17:10:50 <Vorpal> huh, why did this one back to earth spawn deep below the ground
17:10:52 <elliott> Gregor: Did I mention fuck software?
17:10:55 <Vorpal> I thought they spawned on the surface?
17:11:12 <elliott> Yes, I mention-fucked software.
17:11:44 <elliott> I... fucked it with... mentions.
17:14:56 <elliott> elliott@dinky:~/kitten$ ./stage1/bin/pcc -nostdinc -nostdlibs -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/uClibc_dev/usr/include -I/home/elliott/kitten/buildroot-2010.08/output/toolchain/gcc-4.4.4/include -L/home/elliott/kitten/buildroot-2010.08/output/toolchain/uClibc_dev/usr/lib hello.c -o hello
17:14:57 <elliott> /home/elliott/kitten/buildroot-2010.08/output/toolchain/uClibc_dev/usr/include/stdio.h:34: error: cannot find 'stddef.h'
17:15:13 <elliott> elliott@dinky:~/kitten/buildroot-2010.08/toolchain$ find . -name stddef.h
17:15:13 <elliott> elliott@dinky:~/kitten/buildroot-2010.08/toolchain$
17:15:32 <elliott> Oh, it's a Linux include :P
17:24:51 <elliott> elliott@dinky:~/kitten$ ls /home/elliott/kitten/buildroot-2010.08/output/toolchain/uClibc_dev/usr/lib/
17:24:51 <elliott> crt1.o crti.o crtn.o libc.so libm.so
17:25:27 <elliott> ./stage1/bin/pcc -nostdinc -nostdlib -static -I/home/elliott/kitten/stage1/lib/pcc/x86_64-unknown-linux-gnu/0.9.9/include -I$TC/uClibc_dev/usr/include -I$TC/linux/include/linux -L$TC/uClibc_dev/usr/lib hello.c $TC/uClibc-0.9.31/lib/libc.a -o hello
17:25:34 <elliott> elliott@dinky:~/kitten$ ./hello
17:40:33 <Vorpal> aaaargh I think these portals are cross connected
17:41:13 <Vorpal> as in, if I go through A in nether and then back I end up at B. If I go through B in nether and back I end up at A
17:41:58 <elliott> Vorpal: CAN I HAVE THE UPDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD
17:42:08 * elliott mauled by a wild pack of bears
17:44:04 <elliott> I was saying "can I have the update" which you have stated you won't give; it was therefore a joke :P
17:44:18 <elliott> Vorpal: Quick, copy your jars away so you can have portal-spawning action when it's fixed.
17:44:58 <elliott> Vorpal: Holy shit, the Minecraft purchase page is in Swedish X-D
17:47:19 <elliott> Vorpal: Give me a good reason not to set up an OpenGenera partition.
17:47:50 <Vorpal> elliott, you mean.. native?
17:47:57 <Vorpal> elliott, which page is in Swedish?
17:53:31 <fizzie> Hm; I also built two portals in the real world -- for faster traffic between my two buildings -- and even though they've spaced pretty far apart, something like at least a hundred steps, they both mapped to the same portal. But for me it works so that if I go into Nether from either, I get out of the single portal, but if I go in, I always end up in just one of the portals, the one I made first.
17:54:00 <fizzie> So as a fast-traffic system, it only works unidirectionally.
17:54:01 <elliott> Thank you for purchasing Minecraft! Your status should update soon, or within a week if you paid by e-check.
17:54:01 <elliott> If it doesn't, send me an email (include the transaction ID and username) and we'll sort it out.
17:54:05 <elliott> Vorpal: The PayPal page, by default.
17:58:44 <elliott> Vorpal: All I need now is a better graphics card to go with it. :p
18:02:01 <elliott> Vorpal: fancy rendering is actually quite fast for me
18:10:55 <Gregor> `echo How slow am I today?
18:11:24 <Gregor> `echo How slow am I now?
18:11:53 <elliott> Vorpal: 1. Find tiny little rock -- well, grass patch -- in a little passage of water between two islands
18:11:58 <elliott> 2. Put all your blocks down on one square
18:12:29 <elliott> 5. Watch spiders and shit on the islands below
18:13:29 <elliott> Vorpal: I'm going to F4 in the day on top of this tower :)
18:14:18 <elliott> (The shelter thing actually works well, but you have to find a square of buildableness on water close to wherever you are now)
18:14:39 <Gregor> WhatTF is so compelling about Minecraft anyway.
18:14:40 <elliott> You can't have said mhm to what I just said, you said it right as I did :P
18:14:54 <Gregor> elliott: Yes, they're delicious.
18:15:17 <elliott> Gregor: Now imagine doing Lego with huge, life-sized blocks in an endless interesting pretty sandbox.
18:15:19 <Vorpal> elliott, did you know that furnaces continue to operate while you close the dialog for them?
18:15:24 <elliott> Gregor: Also, monsters. And you can dig.
18:15:29 <Vorpal> elliott, meaning you can run several at the same time
18:15:36 <elliott> Vorpal: I haven't done furnaces yet :P
18:16:25 <elliott> I think the creepers have all figured out that I'm in the water
18:16:53 <Vorpal> elliott, also wrt running game while tabbed away: open inventory, furnace or crafting bench then just move pointer out of the window
18:17:20 <elliott> Vorpal: I've been opening inventory and then alt-tabbing away, yes. But it's hard to see what's going on if you do that (say, watching stuff at night).
18:17:50 <Vorpal> elliott, well I'm indoors and running 6 furnaces in parallel atm...
18:29:08 <fizzie> In multiplayer games, you can press 't' to open the "talk dialog", which is just a prompt at the bottom of the screen: that doesn't obscure things so much than inventories/furnaces/such.
18:29:20 <fizzie> But t doesn't do anything in a single-player session, so it's no help there.
18:29:29 -!- kar8nga has joined.
18:29:40 <elliott> fizzie: Now how do I make nights go faster?
18:29:54 -!- madbr has joined.
18:30:07 <fizzie> I don't know: I'm sure there's some sort of a hack for that.
18:30:27 <fizzie> I usually just do some mining at night: it's dark under-ground anyway.
18:30:38 <elliott> fizzie: I'm on top of a pole. :p
18:32:21 <elliott> Hey, someone finally got around to making an Emacs-esque thing that actually has a real browser in it. (And IRC too; it seems to embed gtk stuff.) In Haskell! http://www.flickr.com/photos/48809572@N02/ Bit strange though.
18:33:02 <madbr> I wonder how expensive chip pins were in the 90s
18:33:29 <elliott> Vorpal: totally gonna spawn a portal on this pole now
18:33:40 <madbr> like, if you make your mem bus 32bits instead of 16bits, that's 16 more pins of cpu and 16 more dram pins and 16 more copper tracers etc
18:33:45 <elliott> Vorpal: Suddenly, nothing happened :P
18:33:53 -!- augur has quit (Remote host closed the connection).
18:33:59 <elliott> Vorpal: I just made a portal appear on top of another portal.
18:34:07 <elliott> Vorpal: Step 1. Go up a tall pole. Step 2. Press F4 a lot.
18:34:49 <madbr> (but then again you have twice the ram bandwidth...)
18:38:43 <elliott> http://imgur.com/8C2v8.png
18:38:43 <elliott> http://imgur.com/El0iP.png
18:38:49 <elliott> http://imgur.com/CU5CA.png
18:38:53 <elliott> http://imgur.com/z03kU.png
18:39:16 <elliott> Eventually they started spawning right next to my pole (on water, even?) and then one spawned on top of me and I walked off the pole to my death by mistake.
18:40:00 * Gregor still has no comprehension of what the appeal of this game is.
18:40:36 <Gregor> I was never much for legos unless I had something very utilitarian to do :P
18:40:52 <elliott> Gregor: Well, you could drum up a fake sense of superiority to all the Lego fans or you could just accept that it appeals to everyone else* but you.
18:40:55 <elliott> *for some definition of everyone else
18:40:56 <madbr> dunno, it's creative, open ended, looks nice
18:41:10 <elliott> madbr: Has GODDAMN TERRIFYING SPIDER HISSES.
18:41:11 <Gregor> I prefer a fake sense of superiority.
18:41:21 <madbr> not if you play multiplayer
18:41:33 <nooga> let's wait for OpenMinecraft
18:42:00 <fizzie> There are enemy-spawners in multiplayer too -- in dungeons -- and the enemies they create are unkillable, except by fire.
18:43:56 <fizzie> Yes, a very general sort of bug, the server doesn't track health for anyone.
18:44:00 <fizzie> So the players are unkillable too.
18:44:30 <Vorpal> fizzie, so the enemies cause no real harm then
18:45:40 <fizzie> Creepers still blow stuff up.
18:46:18 <madbr> so can players with tnt :D
18:46:22 <Vorpal> fizzie, obsidian. I'm making the lower parts of the walls of this fort I'm working on in obsidian
18:46:40 <Vorpal> fizzie, only week point now is the doors
18:46:56 <Vorpal> hm not sure how to deal with that
18:47:02 <fizzie> Obsidian is a bit boring to mine for.
18:47:10 <fizzie> Well, doors don't work in multiplayer either. :p
18:47:31 <Vorpal> fizzie, true, it takes ages
18:47:37 <Vorpal> fizzie, also I play locally
18:47:54 <fizzie> Do creepers blow up iron doors?
18:48:14 <Vorpal> fizzie, so far I used peaceful mode because I have been learning how the game works.
18:48:39 -!- augur has joined.
18:48:46 <Vorpal> fizzie, seems TNT is able to blow it up
18:48:53 <fizzie> Then creepers will too, I guess.
18:48:55 <elliott> <nooga> let's wait for OpenMinecraft
18:49:02 <fizzie> I've done a bit of obsidian farming, but that's pretty slow too.
18:49:08 <elliott> Unlike you, nooga, we're really not that concerned about the price of four packs of cigarettes.
18:49:48 <elliott> Vorpal: Stacked a portal on top of a portal yet?
18:49:58 <elliott> I'm trying to get one close to me so that I can see what happens if you go into an airportal.
18:50:11 <fizzie> Probably it just maps to a single Nether portal. :/
18:50:49 <coppro> Netcraft is currently in development, btw
18:51:25 <Vorpal> elliott, no I don't want a silly looking game world :P
18:51:33 <oklopol> "<Sgeo> Is there a concept of smallest regex that matches all and only strings within a set of strings?" <<< you are given a finite set of strings, and you ask whether there's a smallest size of regex that can be used to match exactly those strings?
18:51:36 <elliott> coppro: MineBSD is dead, Netcraft confirms it.
18:51:40 <Vorpal> elliott, I actually play in a serious way
18:51:47 <Gregor> elliott: ... MineBSD???
18:51:52 <elliott> fizzie: You don't know that!
18:51:54 <Vorpal> elliott, I plan, construct, build and so on.
18:51:57 <oklopol> or are you hoping for a minimality result like the one for DFA's
18:51:58 <elliott> Gregor: Yes. It's BSD by Minecraft players.
18:52:21 <Gregor> So, BSD implemented in a buggy Java framework? :P
18:52:30 <madbr> where can I find info on designing a cpu instruction set
18:52:32 * coppro had to google it so that he could be 100% sure it doesn't actually exist
18:52:49 <elliott> Gregor: Since when is it buggy?
18:52:58 <Gregor> madbr: Look up the details on MIPS' instruction set. Expand.
18:53:10 <elliott> fizzie: Interesting choice of portal placement -- it seemingly ate up some of my huge tower and put a portal there.
18:53:18 -!- augur has quit (Ping timeout: 255 seconds).
18:53:27 <Gregor> elliott: A friend of mine told me that he checked for exception handlers and MineCraft catches an obscene number of NullPointerExceptions.
18:53:43 <madbr> gregor: doesn't the MIPS need cache to run fast
18:53:51 <elliott> I can think of a billion reasons for that and they're all to do with third-party libraries, Gregor :P
18:53:53 <coppro> Netcraft is implemented mostly in C, but with C++ to interface with Bullet
18:54:06 <Gregor> madbr: All real CPUs need a cache to run fast.
18:54:06 <oklopol> i had a dream that Gregor was in turku, and we agreed to meet, he said "you are the only pervert from the internet i'm ever going to meet irl"
18:54:31 <coppro> madbr: this is because they're lame
18:54:42 <coppro> they can't do quantum either
18:54:44 <Gregor> No, I meet lots of perverts I know from the webertubes.
18:55:13 <elliott> fizzie: Okay, gone in an airy one and arrived at a Nether portal.
18:55:22 <coppro> (btw, quantum computing is insane. And awesome. And insane. etc.)
18:55:28 * Gregor vaguely wonders if JSMIPS would run faster with some kind of caching :P
18:57:32 <elliott> Gregor: Wouldn't that just come out to an array access to cache an array access? :P
18:57:47 <elliott> Unless you have 32 thousand variables or so. And even then it'd be variable access vs. array access. :P
18:58:37 <Gregor> In fact it would be more like array access to cache /map/ access.
18:58:53 <Gregor> Since my vmem ended up having to be a map.
18:58:58 <madbr> how hard to design are caches anyways
18:59:31 <Gregor> madbr: You lurched wildly from "instruction set" to "actual chip" here :P
19:00:13 <madbr> it's hard to design an instruction set that runs fast if you don't know what sort of memory architecture you're dealing with :D
19:00:48 <coppro> netcraft = gratis minecraft
19:01:24 <madbr> gregor: my goal would be something that can do fast resampling
19:01:32 <coppro> I predict a 30% chance of actually getting something decent
19:01:42 <coppro> nooga: a bunch of people
19:01:47 <madbr> for stuff like texture mapping, music playing...
19:01:58 <nooga> who is in this bunch then?
19:02:07 <coppro> just zeeze guyz, you know?
19:02:21 <coppro> why do you care so much>
19:02:22 <madbr> like, if you can do fast enough resampling and multiplication and mixing, you don't need any real sound hardware
19:03:01 <Gregor> madbr: Can't help ya :P
19:04:16 <madbr> like, that's the problem with amigas, 286s, 386s... it has to load every instruction from DRAM and you have extra wasted instructions for managing the loop, incrementing variables etc
19:06:45 <madbr> whereas RISC instruction set makes plenty sense when the loop instructions are all cached, the DRAM access patterns come out in blocks which is ideal for stuff like EDO (ie on a Pentium or something like that)
19:08:10 <madbr> but if you don't have cache, instructions like add register, [memory] make sense
19:08:51 <nooga> coppro: cloning minecraft shouldn't be hard
19:08:59 <madbr> or even weird string instructions like STOSB
19:09:06 <elliott> minecraft is pretty complex by now
19:09:28 <elliott> Vorpal: please list the things you can do in minecraft to nooga
19:09:33 <nooga> infinite, procedural maps?
19:09:59 <coppro> minecraft does infinity really dumb-like
19:10:11 <elliott> coppro: it's been made less dumbly as of recently, no?
19:10:13 <elliott> nooga: making a good procedural map generator isn't easy. especially biomes. also, multiplayer
19:10:19 <elliott> nooga: also, good caverns and shit.
19:10:19 <coppro> elliott: by necessity, smoewhat
19:10:28 <elliott> nooga: also, the things you list aren't exactly trivial.
19:10:34 <madbr> gregor: Mostly I'm hesitating between 16bit design where you know how much cycles everything takes so you can play off that, and with lots of Amiga-like effects but there's less memory bandwidth
19:10:42 <elliott> nooga: no, it isn't; the old minecraft generators are not nearly as good as the new ones
19:11:05 * Gregor lawls at reading nooga's line as madbr's line.
19:11:40 <coppro> elliott: the server originally just retained the whole thing in memory
19:11:50 <coppro> so if you went farther and farther, it would just slow down and slow down
19:12:00 -!- Zuu has quit (Ping timeout: 260 seconds).
19:12:02 <madbr> or 32bit-ish design with something like fast-page or EDO dram and a frame buffer
19:12:08 <coppro> My understanding is that optimizations have been made, but this is still fundamentally a problem
19:13:35 <Gregor> Paging would help with that if it wasn't Java 8-D
19:14:10 <oklopol> nooga: maybe you could use sand automata for the beaches?
19:14:21 <coppro> netcraft's approach could still be exploited, but is far less dumb
19:14:43 <coppro> basically it stores the generator, the seed, and a diff
19:14:57 <coppro> so simply venturing out into infinity leaves no trace in memory
19:15:30 <coppro> (and I assume the diff is appropriately compressed so the only way to run up the memory usage is to actually make a large number of changes)
19:15:31 <oklopol> but if things are forgotten, a lot of the funness of the game is forgotten too
19:15:43 <coppro> oklopol: no, changes are retained in the diff
19:18:19 <oklopol> i got my maturity exam thingie done for my bachelor's
19:18:26 <oklopol> so now i can switch to math
19:18:56 <oklopol> that's what they call it, i don't know what it's official translation is
19:19:55 <oklopol> it's this simple test where they check you did your bachelor's yourself, and know finnish.
19:20:05 <oklopol> or the language the thing was written in
19:20:05 <nooga> coppro: that's what i would do
19:21:17 <oklopol> also i'm gonna get my first publication i think, but it's just proceedings
19:22:01 <nooga> coppro: i would be fun too make slopes as seen in TTD
19:23:03 <oklopol> did you think of matriculation exam or smth?
19:23:06 <oklopol> i don't know what a-levels is
19:23:12 <cheater99> a-levels is at the end of high school.
19:23:19 <oklopol> yeah that's called matriculation here
19:23:19 <coppro> oklopol: a-levels = high school exams, as I understand it
19:23:31 <coppro> it's called diploma exams in my home province
19:23:50 <cheater99> http://gosu-lang.org/comparison.shtml
19:24:06 <coppro> I thought that was a joke
19:24:36 <coppro> hah, I like the last one
19:24:44 <elliott> wtf happened to my ignores?
19:24:54 <coppro> also lol at 'reified generics'
19:25:00 <coppro> sounds like a wonderful way to make generics suck more
19:25:05 <oklopol> yeah he's sneaky like that
19:25:15 <cheater99> i hadn't changed my nick in months
19:25:24 <oklopol> cheater99: sure, but you'll have to pay me
19:25:27 <elliott> coppro: I am tired of tables that have all "Y"s on the product in question.
19:25:36 <elliott> coppro: could there be a more obvious example of bias?
19:25:42 <cheater99> and yet alise keeps on taking me off ignore and then pretending she forgot to add some bit.
19:25:42 <elliott> Put some shit you don't have yet in there!
19:25:47 <coppro> they could also have all Ns for the other stuff
19:25:57 <elliott> <coppro> oklopol: a-levels = high school exams, as I understand it
19:26:13 <elliott> <coppro> elliott: the server originally just retained the whole thing in memory
19:26:16 <elliott> who plays minecraft on servers
19:26:23 <elliott> it's paged out to disk now :p
19:26:35 <elliott> <coppro> basically it stores the generator, the seed, and a diff
19:26:35 <elliott> <coppro> so simply venturing out into infinity leaves no trace in memory
19:26:41 <elliott> it now saves the chunks to disk
19:26:52 <elliott> coppro: plz see http://www.minecraftwiki.net/wiki/Chunks
19:27:17 <coppro> it does that chunking thing
19:27:20 <oklopol> cheater99: WOW that's where i draw the line!
19:27:21 <coppro> but if you leave a chunk
19:27:26 <coppro> that chucnk stays around
19:28:51 <pikhq> It amuses me greatly that fuel efficiency can be measured in square meters.
19:28:52 <elliott> Sgeo: absolutely very insightful; cheater99: you too
19:29:04 <pikhq> (1L/100km = 0.01m^2)
19:29:50 <Vorpal> <elliott> Vorpal: please list the things you can do in minecraft to nooga
19:29:56 <pikhq> Hooray, base SI units.
19:30:02 <Vorpal> elliott, I delegate this to you and to fizzie
19:30:12 <cheater99> elliott is msging me with cusswords
19:30:26 <elliott> cheater99: i took you off ignore. nice try
19:30:33 <oklopol> so, if you took a cylinder containing the fuel of the length you can drive, it'd be the cylinder's thingie area
19:30:51 <Vorpal> <coppro> My understanding is that optimizations have been made, but this is still fundamentally a problem <-- I believe it will unload chunks that are some distance away
19:31:00 <oklopol> so actually i don't think it's at all funny anymore
19:31:01 <nooga> coppro: is netcraft closed-source?
19:31:28 <oklopol> because that's what's being measured, "how much fuel you need to move one differential forward"
19:31:32 <coppro> nooga: source will be released when it's ready for first release, is my understanding
19:31:41 <Vorpal> <coppro> netcraft's approach could still be exploited, but is far less dumb <-- netcraft?
19:31:42 <cheater99> has anyone ever run a turing machine on a GOL?
19:31:49 <elliott> Vorpal: a clone of minecraft
19:31:53 <elliott> coppro: good to know they're, uh, being creative
19:31:56 <elliott> coppro: even in their naming
19:32:14 <cheater99> i've seen some GOL's that had the name "turing machine" etc, of course the proof eluded me. i'm not much into GOL.
19:32:17 <Vorpal> elliott, oh. googling netcraft just gives netcraft.com results. Which are not very relevant here
19:32:20 <elliott> personally i couldn't have thought of "Let's make a game that's like Minecraft... and... yes."
19:32:36 <Vorpal> elliott, got a link to netcraft?
19:32:49 <elliott> it's people at coppro's uni
19:32:52 <elliott> demonstrating how creative they are
19:33:25 <oklopol> cheater99: they were turing machines
19:33:42 <pikhq> Erm, wrong order of magnitude.
19:33:58 <Vorpal> <elliott> demonstrating how creative they are <-- XD
19:34:10 <cheater99> oklopol: have you ever looked into the problem of running turing machines on GOL's more in-depth than just looking at GOL presets?
19:34:20 <elliott> Vorpal: well could you have thought of the idea "it's like minecraft, but we take the mine bit off and add net"?!
19:34:50 <Vorpal> elliott, so no mining?
19:35:23 <elliott> Vorpal: no you just net things up
19:35:31 <elliott> and then craft them into hideous corpse art
19:35:39 <oklopol> i don't know JCGOL, same rules?
19:35:47 <Vorpal> elliott, like the zombie pigmen ;P
19:35:50 <pikhq> ... Bweheheh. That is, of course, the integral of V(x) from 0 to distance travelled, with respect to x. It makes some amount of sense, too!
19:35:58 <oklopol> cheater99: what else is there?
19:37:11 <cheater99> it does seem like fairly useful research
19:37:16 <oklopol> i might if there was a high-level explanation, but most of it consists of guns etc so actually you could probably just look at it
19:37:37 <cheater99> but the thing is the single cells are so very very simple
19:37:46 <cheater99> they're even simpler than transistors you'd think
19:38:30 <cheater99> they usually just need connection to some neighbour cells
19:38:51 <cheater99> which could mean, you make a cpu which just has those simple cells, and that's that
19:39:03 <oklopol> surely it's interesting, but proofs that things can simulate turing machines aren't usually considered very important
19:39:16 <oklopol> because they can always do that
19:39:41 <cheater99> a rubik's cube can't simulate a turing machine.
19:39:58 <cheater99> i think GOL could be an interesting processor architecture
19:40:09 <cheater99> sure, you need several cells to signify a single bit
19:40:15 <cheater99> but those cells could be so much smaller
19:40:28 <cheater99> like, almost single molecules probably
19:40:37 <oklopol> you're doing a lot of useless stuff, guns and all that, and still the rule is complicated
19:40:57 <oklopol> even wireworld would be better
19:41:33 <cheater99> but you could probably come up with something crazy like say growing peptide crystals where each molecule strand is neatly crafted into the network and represents a single GOL cell.
19:41:41 -!- Zuu has joined.
19:41:56 <cheater99> this is just pure sci-fi, but i wouldn't be surprised if it were possible
19:42:07 <oklopol> in gol, you have signal crossing *everywhere*
19:42:25 <oklopol> the neighborhood already forces it
19:43:08 <Vorpal> cheater99, ... transistors don't need address lines...
19:43:11 <oklopol> well, you know, you can't put down tiles on the plane so that they are gol-connected
19:44:01 <oklopol> because a's connection to its bottom-right crosses a's right neighbor's connection to bottom-left
19:44:02 <elliott> Vorpal: minecraft just got updated
19:44:06 <elliott> Vorpal: wanna bet f4 no longer works?
19:44:24 <Vorpal> elliott, well I have a copy, anyway I have heaps of obsidian
19:44:35 <elliott> Vorpal: no, newer than that
19:44:37 <cheater99> Vorpal: i was thinking of FPGAs there for a second
19:44:48 <Vorpal> elliott, I have a copy of the one where f4 works I mean
19:44:50 <cheater99> Vorpal: since normal CPUs are not as programmable as a GOL board
19:45:12 <Vorpal> cheater99, sure they are. But only during design
19:45:22 <elliott> Vorpal: wow, there's *two* minepedias...
19:45:24 <elliott> http://minepedia.net/index.php?title=Main_Page
19:45:34 <madbr> what's a GOL board?
19:45:42 <Vorpal> elliott, that one I haven't seen
19:46:10 <Vorpal> elliott, anyway I get connection error to minecraft.net
19:46:21 <madbr> what, someone did a game of life simulator with lots of parallelism on some custom arch?
19:46:21 -!- ais523 has joined.
19:46:55 <cheater99> madbr: we're considering gol AS the arch
19:49:38 <elliott> "here is the kicker, lava produces heat, energy at a certain frequency that is harmful to us, just as we can see light because we perceive things through this 4x box, of 3 dimensions of space, and 1 dimension of time.
19:49:38 <elliott> Quantum, we understand that waves and particles are one in the same, that particles are waves. So lets just say that monsters are made of of particles that are exist in a certain frequency that is not effected by the same frequency as heat."
19:49:42 <elliott> (comment on minecraft blog)
19:51:06 <elliott> "Now, you might think that is silly, because it is heat. I mean, every burns at a high enough temperature, right? wrong. Since heat is bound by the existence of time, thus the bound by the speed of light and all the energy required by it. As things get hotter and hotter, eventually they start producing gamma rays and such, etc, but it gets to a point where the energy required to make more heat goes off to infinity when you get temperatures that h
19:51:06 <elliott> ave energies that have frequencies near the speed of light, so thus, things that are made up of particles that are smaller than light can perceive and such, can be non effected by heat, as heat will have to break out of the existence of time before it can effect these particles... but if it breaks out of time, then it is not longer heat, per say."
19:51:12 <elliott> "So lets just say that monsters are made up of atomic, subatom, etheric, and subetheric particles that exist on a frequency where, quantumly, they are visible by light, and can effect physical objects, but their physical existence is outside the bounds of our reality and heat do not effect them.
19:51:12 <elliott> I am not saying this is the reason, I am just saying that when you look into things from a quantum mechanical level, everything becomes waves and probabilities. everything is possible. =)
19:51:14 <elliott> enjoy, minecraft is awesome."
19:51:38 <elliott> i have a better explanation: you can punch trees, therefore lava doesn't harm ghasts Q.E.D.
19:51:42 <ais523> elliott: you know, if he'd just said that monsters were transparent to infrared, it would have been a lot simpler
19:52:05 <ais523> even though that reasoning doesn't really deal with conducted or convected heat...
19:52:53 <elliott> ais523: it doesn't help his quantum pseudo-reasoning that you can bash them with swords, either :)
19:53:40 -!- augur has joined.
19:53:48 <oklopol> that was the biggest bundle of crazy nonsensicality i've seen this week
19:53:55 <oklopol> i've been ircing too little :(
19:54:02 <elliott> Vorpal: HAHA -- one of the possible names for Nether was "Norway"
19:55:13 <elliott> Vorpal: http://www.reddit.com/r/Minecraft/comments/dx88i/notch_if_you_read_this_please_leave_the_fire/c13mm12?context=1
19:56:37 <elliott> "I think you should call is Steve." "That would be the funniest damn thing he could do. Construct menacing portal...enter menacing portal....'You have entered....STEVE""
19:56:49 <nooga> THIS IS AN ULTIMATE GAME
19:57:08 <nooga> http://www.youtube.com/watch?v=gSOQGazo7Oo
19:58:16 <ais523> I'm still on just-online mode, when I try to do about 8 things at once and so does my computer
19:58:52 <elliott> ais523: YOUR COMPUTER WILL RUN FASTER WITH KITTEN
19:59:14 <elliott> ais523: It's got static linking. That's what plants crave!
19:59:15 <Gregor> `translate Hey guys, mimi fasta kutafsiri amri.
19:59:16 <HackEgo> Hey guys, I fixed the translate command.
19:59:29 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
19:59:35 <elliott> Gregor: Translate is LAME :P
19:59:47 <Gregor> `translateto sv smorgasbord
20:00:06 <Gregor> `translateto sv elliott's translation method is so rife with lame.
20:00:09 <HackEgo> Elliotts översättning metoden är så fulla av lama.
20:00:12 <elliott> `translateto it Dirty Hungarian Phrasebook
20:00:17 <elliott> `wl en it Dirty Hungarian Phrasebook
20:00:18 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
20:00:27 <ais523> elliott: is static linking actually faster?
20:00:28 <elliott> `wl en it Dirty_Hungarian_Phrasebook
20:00:29 <oklopol> computer translation is getting pretty scary, because they are making less and less errors in grammar, but just... completely misunderstand everything
20:00:38 <Gregor> Anyway, translatefromto actually uses the Google APIs now :P
20:00:39 <elliott> Yours doesn't translate "dirty"
20:00:39 <oklopol> so it's like reading the translation of a native retard
20:00:42 <Gregor> No more trying to parse HTML.
20:00:51 <elliott> ais523: the linker is a lot simpler, so it's faster, and starting executables is much faster
20:01:01 <elliott> ais523: (since it doesn't have to load all the libraries)
20:01:01 <ais523> I'd expect it to be slower because it couldn't keep libc in the cache
20:01:08 <ais523> although I suppose it depends on how often it context-switches
20:01:15 <elliott> ais523: Uhh... you do realise Linux does sharing of all the libcs?
20:01:15 <ais523> perhaps it starts faster and runs slower?
20:01:20 <nooga> `wl en pl dirty hungarian phrasebook
20:01:22 <HackEgo> You get NOTHING! You LOSE! Good DAY sir!
20:01:24 <elliott> You never have two copies of libc functions in RAM with static linking.
20:01:25 <ais523> elliott: indeed, but how can it do that with static linking?
20:01:39 <ais523> that's pretty ingenious, actually
20:01:47 <ais523> sort-of dynamic unstaticing
20:01:51 <elliott> ais523: I'm not sure at what granularity Linux does it, but it's good enough
20:02:19 <elliott> ais523: but yes, linker is a lot faster, starting executables is a lot faster, runtime is the same really
20:02:44 <Gregor> "You never have two copies of libc functions in RAM with static linking." Well this smells like total BS.
20:03:13 <elliott> well function may be false
20:03:17 <elliott> but there is most definitely sharing.
20:03:24 <elliott> I know that Plan 9 does pervasive sharing
20:03:32 <elliott> Aren’t statically linked executables consuming more memory?
20:03:33 <elliott> We believe that due to the small size of the base system the opposite will be the case. First of all, the kernel will load each static executable’s .rodata, .data, .text and .comment sections only once for all instances into memory. Second, because each static binary has only been linked with the object files necessary, it has already been optimised at linkage time for memory consumption. When loading it, we don’t require the kernel to map al
20:03:33 <elliott> l dependent dynamic libraries into memory from which our binary might only use 5% of the functions they provide. So, in reality, the memory footprint is becoming less, and the dead code hold in memory (or paged) reduces overall consumption. This is also true for programs, like surf, which don’t use all webkit/gtk/glib functions.
20:03:50 <elliott> ok, so less sharing than plan 9
20:04:01 <elliott> ais523: also, more will fit into cache as the libc is a lot smaller than glibc :p
20:04:17 <elliott> http://9fans.net/archive/2002/02/21 plan 9 sharing
20:04:25 <Gregor> Yes, it will share each INSTANCE of the SAME program, but it won't share bits of it across programs.
20:04:26 <ais523> elliott: the plan 9 statement there seems to imply that it shares .rodata and .text, etc., for two programs with the same executable, not any two programs
20:04:37 <ais523> sharing between two different programs would be rather harder
20:04:40 <elliott> i do recall systems with more sharing
20:04:58 <ais523> especially as, given static linking, the functions aren't going to be in the same relative positions in memory in the two programs
20:05:06 <elliott> ais523: anyway, i don't know of any reports saying static linking has slower runtime performance.
20:05:21 <elliott> ais523: and it seems unlikely to me, especially when your libc is *much* more compact than glibc
20:07:06 <elliott> maybe i should read linux from scratch :)
20:07:20 <ais523> hmm, really, there should be a two-tier libc
20:07:32 <ais523> one that does everything that C99 and POSIX require, the other for all the GNU extensions
20:08:18 <pikhq> elliott: Not much of LFS is actually special information for building a base Linux system, BTW.
20:08:26 <elliott> ais523: you can do the latter with gcc: just append "|| rm *.c" to the command line, without the quotes
20:08:30 <elliott> pikhq: I can't parse your sentence
20:08:41 <pikhq> elliott: Most of it is just "here's how you build foo".
20:09:09 <pikhq> elliott: Basically, after you've got the toolchain up and running you're just installing exceptionally common dependencies.
20:10:17 <elliott> pikhq: ISTR some part of recent perls depend on static linking :)
20:10:27 <ais523> elliott: I don't see how deleting all C files if the compile fails would help
20:10:33 <pikhq> elliott: And LFS has you build a conventional glibc system.
20:11:03 <elliott> ais523: it removes the GNU dependencies from all the .c files in the directory
20:11:14 <ais523> meanwhile, a sign of Slashdot strangeness: I've just metamoderated two of the comments I've just moderated
20:11:15 <elliott> (assuming your compile is failing because of missing gnu extensions)
20:11:32 <ais523> in other words, I'm currently keeping tabs on my own behaviour
20:11:41 <ais523> elliott: that's not what I want
20:11:47 <elliott> ais523: Who moderates the moderatormen?
20:11:52 <ais523> the GNU extensions are often useful, but they should be in a separate library
20:11:59 <ais523> that's different from arbitrarily deleting programs that use GNU extensions
20:12:06 <elliott> usually the gnu extensions are useful ideas, implemented terribly and with a terrible api :)
20:12:11 <elliott> nobody really uses them apart from gnu though.
20:12:17 <ais523> (especially as gcc is entirely capable of compiling programs that use them...)
20:12:29 <ais523> elliott: the course on C I'm teaching uses GNU extensions
20:12:35 <ais523> not by my choice, incidentally
20:12:42 <ais523> it makes a lot of other dubious decisions, too
20:12:56 <ais523> I think GNU have managed to successfully embrace-and-extend C
20:13:00 <pikhq> elliott: Linux's sharing works *precisely* the same as Plan 9's. The only stuff that isn't shared (and even that, I'm pretty sure, is COW) is stuff that's writable.
20:13:15 -!- Zuu has quit (Ping timeout: 260 seconds).
20:13:26 <elliott> ais523: too many people are complacent with bad design and tactics just because they're perceived as supporters of freedom
20:13:43 <elliott> who calls out GNU for encouraging people to go beyond the standards and write code that isn't portable to non-GNU systems?
20:13:55 <pikhq> And it functions just the same for dynamic libraries. The only gain dynamic libraries get you is that there's only a single copy of each symbol floating around, rather than a single copy per program using it.
20:14:03 <ais523> elliott: I'm not even sure /how/ to call them out
20:14:08 <elliott> pikhq: Right. But if the symbols are tiny, it doesn't make any sense.
20:14:15 <elliott> ais523: you can't, really; any anti-gnu site would just be dismissed as FUD
20:14:25 <elliott> ais523: I have become rather cynical towards GNU.
20:14:38 <elliott> pikhq: Right. But if the symbols are tiny, it doesn't make any difference.
20:14:56 <elliott> pikhq: For instance, a system running dynamically-linked glibc executables will use the same or more memory than one using statically-linked uClibc executables.
20:14:58 <pikhq> Which makes whether or not you get a benefit from dynamic or static linking, well, complex.
20:15:08 <ais523> elliott: the advertising clause in the GPLv2 is pretty bad (not nearly as bad as the one in the GFDL...)
20:15:19 <elliott> ais523: it has an *advertising clause*?
20:15:23 <elliott> pikhq: You get benefits, just maybe not memory ones.
20:15:37 <ais523> elliott: if you modify a non-interactive GPLv2 program to become interactive
20:15:43 <ais523> then it has to put up a GPL blurb when it loads
20:15:44 <pikhq> elliott: Well, yes. I was just concerning myself with the memory and disk usage for this.
20:15:50 <elliott> ais523: does v3 have that?
20:16:05 <elliott> pikhq: Disk space is definitely a win, measure static uClibc vs. dynamic glibc sometime. (Okay, browsers are bigger.)
20:16:38 <elliott> pikhq: But still, I have disk. And RAM. Do you?
20:16:39 <pikhq> elliott: If you've got a lot of static binaries, there *might* be a benefit from dynamic linking.
20:16:58 <pikhq> Using the precise same libraries, of course...
20:16:59 <cheater99> ais523: have you read the new chapters in "learn you a haskell"?
20:17:00 <ais523> elliott: yes to some extent, but it's been toned down from v2's
20:17:06 <ais523> cheater99: not recently; I read the one about zippers
20:17:23 <pikhq> Glibc ends up producing gigantic dynamic libraries compared with uClibc's static libraries, of course.
20:17:37 <ais523> <GPLv3> An interactive user interface displays "Appropriate Legal Notices" to the extent that it includes a convenient and prominently visible feature that (1) displays an appropriate copyright notice, and (2) tells the user that there is no warranty for the work (except to the extent that warranties are provided), that licensees may convey the work under this License, and how to view a copy of this License. If the interface presents a list of user
20:17:38 <ais523> commands or options, such as a menu, a prominent item in the list meets this criterion.
20:18:19 <ais523> if you're interested, that clause (in v2 and v3) has only triggered twice in me writing GPL programs: C-INTERCAL and jettyplay (both of which were based on previous GPL code)
20:18:58 <ais523> C-INTERCAL handles it with a license blurb when starting up the debugger; jettyplay has a comment about the GPL as the default status bar when it loads
20:19:10 <cheater99> ais523: how long did it take to write c-intercal?
20:19:18 <ais523> cheater99: I didn't write it from scratch
20:19:25 <ais523> I think esr wrote the original version in a weekend, but it didn't actually work
20:19:31 <ais523> I've spent years on it, but not continuously
20:19:35 <cheater99> yea but what you were doing with it
20:19:41 <pikhq> ais523: If you are the copyright holder, you can add an additional permission allowing for you to not do that with GPLv3.
20:19:53 <elliott> does anyone know if xchat can be set to ignore mentions of a name too?
20:19:59 <elliott> or at least addressed messages
20:20:03 <ais523> pikhq: I'm not the only copyright holder for either of the programs in question
20:20:12 <ais523> C-INTERCAL has loads of copyright holders, jettyplay has 3 IIRC
20:20:36 <pikhq> (the copyright holder on a GPLv3 work may add any number of additional permissions; these permissions can be removed by anyone at will.)
20:21:39 <elliott> 12:11:59 <ais523> that's different from arbitrarily deleting programs that use GNU extensions
20:21:39 <elliott> 12:12:17 <ais523> (especially as gcc is entirely capable of compiling programs that use them...)
20:21:46 <elliott> not library extensions, if you don't use glibc
20:21:48 <ais523> the bzip2 library was originally BSD, but I relicensed it as GPLv2 to simplify the license for the program as a whole
20:22:11 <ais523> elliott: not to mention, it deletes all C files in the current directory, rather than the files you were compiling, which might or might not be in the current directory
20:22:20 <ais523> also, the files you're compiling need not end with the .c extension
20:22:24 <elliott> ais523: sry will never joke ever :P
20:23:04 <ais523> incidentally, GPLv3 is /much/ saner than GPLv2, despite its many detractors
20:23:14 <ais523> I discovered this due to reading GPLv3 incessantly while testing azip
20:23:25 <ais523> (it's one of the two test files I'm using)
20:24:08 <elliott> ais523: CAN I HAZ AZIP LOLRZL
20:24:16 <ais523> elliott: oh right, I got it working again
20:24:20 <ais523> let me see if I got unazip working too
20:24:31 <pikhq> ais523: Yeah, the GPLv3 is genuinely a better license, FUD aside.
20:24:46 <madbr> are licenses that important
20:25:05 <ais523> madbr: they can be, it depends on what you're doing
20:25:07 <pikhq> madbr: Because we live in a society where copyright law has gone crazy: YES.
20:25:41 <elliott> There isn't much more important than freedom. (Have I really turned into a freetard? Oh well, at least I still dislike the GPL.)
20:25:46 <ais523> elliott: I'm just testing it atm, I'll send it to you if it's working, and otherwise check what's wrong
20:25:58 <elliott> ais523: i didn't expect you to actually send it, but sure :P
20:26:12 <ais523> anyway, I can't see much of a reason why you'd choose GPLv2 over GPLv3, except compatibility with other GPLv2 things, which is pretty large
20:26:19 <pikhq> A programmer is basically forced to be a bit of a lawyer to not get sued for all the money in this day and age.
20:26:21 <ais523> I can see why you'd choose, say, BSD3 over GPLv3
20:26:42 <ais523> what was it obsoleted by?
20:26:51 <ais523> I think I disagree, there
20:27:29 <elliott> I'd also say that BSD2 was obsoleted by ISC but there you go :P
20:27:36 <pikhq> The endorsement clause of BSD3 is a noöp.
20:27:45 <ais523> depends on which country you're in, I think
20:28:26 <elliott> ais523: Berne convention, I think
20:29:05 <ais523> elliott: http://pastebin.ca/1987512
20:29:16 <ais523> try using the -t option to see what it's doing
20:29:41 <pikhq> And nations that haven't signed that don't recognise foreign copyright at all, so it's a moot point.
20:29:43 <elliott> ais523: your typical copyright terms, I presume? (I can't do anything without asking you) -- just checking :P
20:29:52 <elliott> well, apart from modifying it locally of course...
20:29:59 <ais523> elliott: copyright currently undecided, thus it's all-rights-reserved by defaut
20:30:21 <ais523> if it actually turns out to be useful, I'll license it under something more permissive
20:30:36 <elliott> ais523: ugh, now I have to write a shell one-liner to split that into two files, as is my duty
20:30:38 <ais523> note: statistically speaking, it's likely to infringe at least 1000 US patents
20:30:51 <ais523> due to being a) software, b) a compression algorithm
20:31:15 <ais523> elliott: it was a shell one-liner to join them into one file
20:31:22 <ais523> I love using more for things like that
20:31:30 <elliott> ais523: you used *more* for that?
20:31:32 <ais523> I've done things like "more * | less" before now
20:31:48 <ais523> elliott: that's what more does if you give it multiple files and stdout isn't a terminal
20:33:18 <elliott> ais523: now I just need unmore!
20:33:29 <elliott> ais523: I know! I'll use sed to turn it into a shell script that outputs those two files!
20:33:37 <elliott> (what's the sed to merge the next line with this one again?)
20:33:40 <elliott> er, keeping the \n in the buffer
20:34:18 <elliott> ais523: so have you made an azip quine yet?
20:36:15 <elliott> ah, N is the command I wanted
20:39:28 <ais523> elliott: no, I haven't tried
20:39:31 <elliott> pikhq: gah, uClibc doesn't support leap second
20:39:39 <elliott> ais523: how can you make me code sed like this
20:39:54 <ais523> it'd be pretty awkward due to the way the encoding changes all the time
20:39:59 <ais523> as bad as trying to write a bzip2 quine
20:43:31 -!- Phantom_Hoover has joined.
20:44:36 <elliott> ais523: I give up; I'm extracting the files manually.
20:47:40 <elliott> ais523: filebin converts to DOS line endings, jesus christ
20:47:54 <Vorpal> elliott, wrt last highlight of me :P
20:48:06 <elliott> ais523: can't you just uuencode a tarball to sprunge.us? :)
20:48:11 <Vorpal> elliott, reddit thread
20:48:29 <elliott> Vorpal: specifically, the post and notch's reply
20:49:28 <elliott> Vorpal: btw, why do I get &e0 when i die without points?
20:49:30 <elliott> strange way of saying 0 :P
20:49:48 -!- digimunk has joined.
20:50:07 * elliott wonders whether anyone other than dreamhost employees ircs from dreamhost
20:50:17 <Vorpal> elliott, also uh points?
20:50:29 <elliott> Vorpal: umm yes you get told your # of points when you die
20:50:54 <digimunk> I am not an employee of dreamhost
20:50:55 <Vorpal> elliott, http://i.imgur.com/YDI84.png (no I didn't take this one, but abuse of F4!)
20:51:07 <elliott> digimunk: why would you irc from dreamhost?
20:51:19 <digimunk> I tunnel out of work through dreamhost
20:51:25 <elliott> Vorpal: http://i.imgur.com/yBRWd.png
20:51:42 <digimunk> but it is banned from a lot of servers ;)
20:51:49 <Vorpal> elliott, issue is they probably all go to the same portal at the other end
20:51:52 <Vorpal> and it is fairly useless
20:52:45 <nooga> http://i.imgur.com/YDI84.png < p0rtals?
20:53:01 <elliott> ais523: do you know which of -O3/-Os is faster?
20:53:09 <elliott> nooga: "p0rtals"? seriously? you replace o with 0? in 2010?
20:53:15 <elliott> digimunk: more than a bit :)
20:53:53 <elliott> Vorpal: http://i.imgur.com/F8mY0.png this guy did what i did but moreso and, uh, creepers
20:55:21 <Vorpal> elliott, anyway creating portals from the real world seems to like to end up in the same place. Need to create all but the first portal from within nether, or it doesn't work well
20:55:24 <Vorpal> which is very backwards
20:55:36 <Vorpal> since distances are shorter in nether, not the other way around
20:55:46 <elliott> Vorpal: yes, but one nether pixel = 8 world pixels
20:55:49 <Vorpal> elliott, and quite far from each other
20:55:57 <elliott> so if you create nearby portals in the real world
20:56:02 <elliott> they can all map to a single nether point
20:56:04 <Vorpal> like 100 blocks in nether
20:56:11 <elliott> Vorpal: http://i.imgur.com/Q3z6K.jpg portal bridge
20:56:16 <Vorpal> elliott, dude I walked about 300 tiles in the real world
20:56:32 <Vorpal> elliott, I doubt that bridge is very walkable!
20:57:06 <elliott> Hmm, is there an easy way to stop Minecraft updating apart from disconnecting?
20:57:28 <Vorpal> elliott, entering wrong username?
20:58:18 <Phantom_Hoover> I just saw someone say elsewhere that providing free education for all is selfish elitism.
20:58:21 <nooga> i mean what is it in Minecraft
20:58:59 <Vorpal> nooga, it is a hell like word in minecraft
20:59:06 <Vorpal> nooga, elliott did answer you
20:59:25 <nooga> and why 1 nether pixel = 8 rw pixels?
20:59:38 <oklopol> Phantom_Hoover: why is it selfish elitism?
20:59:55 <Phantom_Hoover> oklopol, because the stupid should be allowed to go to university too!
21:00:27 <oklopol> and if they can get there for free, they can't go?
21:00:33 <oklopol> i'm not sure i'm following the logic
21:00:41 -!- Sgeo has joined.
21:00:57 <Phantom_Hoover> oklopol, no, the person said something idiotic about the "less intellectually able subsidising the educated elit
21:00:58 <Vorpal> elliott, isn't it more fun to do the portals the proper way anyway?
21:01:32 <Phantom_Hoover> The fact that they were saying this on the internet was lost on them.
21:01:46 <Vorpal> elliott, btw I think the posting times on twitter are way off... http://twitter.com/notch the top one said "less than 20 seconds" half an hour ago for me...
21:01:54 <Vorpal> I cleared cache and such...
21:02:00 <oklopol> Phantom_Hoover: lost on me too
21:02:17 <elliott> Vorpal: server-side cache presumably
21:02:45 <elliott> Phantom_Hoover: where was it
21:02:53 <Vorpal> <nooga> and why 1 nether pixel = 8 rw pixels? <-- well blocks, not pixels. And it is to use it for fast travel...
21:02:58 <elliott> Vorpal: what editor have you been using to look at the level?
21:03:03 <oklopol> Phantom_Hoover: i don't get how it's ironic to say that on the internet
21:03:04 <elliott> Phantom_Hoover: that's not an answer
21:03:16 <Vorpal> elliott, you mean the map I posted before? was a map viewer, not an editor
21:03:25 <oklopol> Phantom_Hoover: people who paid for their uni courses?
21:03:46 <oklopol> in any case i don't think that's all that ironic
21:03:52 <Vorpal> elliott, it is kind of hugely slow once the world becomes larger than about 20 MB
21:04:01 <Vorpal> as in, taking minutes to load
21:04:25 <Vorpal> elliott, anyway, it was "Minutor" (see http://www.minecraftwiki.net/wiki/Programs_and_Editors#Map_Viewers)
21:04:32 <Sgeo> I was more asking if there was a name for the size of the smallest regex, I guess
21:04:47 <Vorpal> elliott, I tried "Minecraft X-Ray" too, but kind of useless
21:04:52 -!- Zuu has joined.
21:05:04 <Sgeo> Well, smallest regexes..
21:05:17 <Vorpal> elliott, using a dedicated mapper (generates images basically) might work better
21:05:17 <Sgeo> The min() of all the sizes of all relevent regexes
21:05:24 <oklopol> (ab)*a and a(ba)* for instance
21:05:38 <oklopol> Sgeo: yeah no, i don't see how that would be very useful anyway
21:05:46 <elliott> Vorpal: heh only linux-compatible alpha *editor* is shareware
21:05:49 <oklopol> really it's rather coincidental which ones are minimal
21:05:50 <elliott> well, more like donateware but whatever
21:05:59 <Vorpal> elliott, indeed, and requires python 2.6
21:06:15 <Vorpal> elliott, it has *.pyo files
21:06:16 <Sgeo> Illustrating the difference between, say, elliott and eliott vs elliott and ehird
21:06:25 <Vorpal> elliott, no, there are no .py
21:06:36 <elliott> Vorpal: what a load of crap.
21:06:38 <Vorpal> elliott, anyway, my system has 2.7 so...
21:07:01 <Sgeo> Although the size of the regex varies directly with the size of the strings, so
21:07:05 <elliott> wonder if anyone's written docs on the format
21:07:17 <Vorpal> elliott, there is on the wiki iirc
21:07:20 <Sgeo> I guess I sort of wanted a string difference like thing that works on more than two strings to give a single number
21:07:30 <Vorpal> elliott, wine + a windows editor might work
21:07:39 <elliott> i don't think i care enough to do that
21:07:42 <Vorpal> elliott, I'm not really interested in editors though
21:07:54 <elliott> Vorpal: Bet it'd expect it to be in the Minecraft folder for Windows.
21:08:10 <elliott> I only want an editor to make the game engine suffer :)
21:08:27 <Vorpal> elliott, uh I think it goes into Application Data/.minecraft on windows or something
21:08:39 <Vorpal> elliott, presumably you could figure out where wine puts that and then symlink
21:08:57 <Vorpal> elliott, anyway, saw it listed somewhere on the wiki
21:10:57 <nooga> i have shit tons of work to do and i'm just sitting here and staring at the terminal
21:11:13 <Vorpal> just close the irc client
21:11:27 <elliott> Vorpal does not understand akrasia
21:11:28 <nooga> i'm procrastinating
21:12:03 <Vorpal> elliott, I'm sure I will once I read the wikipedia article!
21:12:20 <elliott> Vorpal: [[Akrasia is the state of acting against one's better judgment. Examples of akrasia include procrastination and inability to form strong cooperating communities.]]
21:12:27 <elliott> http://wiki.lesswrong.com/wiki/Akrasia
21:12:35 <nooga> if i close the irc client i will move to facebook or some other shit
21:12:55 <elliott> i'd laugh even before the upgrade, but seriously?
21:13:05 <elliott> digg = corporate spam site
21:13:11 <nooga> i've never been there
21:13:34 <elliott> Phantom_Hoover: yeah the update ... fucked things up
21:13:38 <elliott> and caused a mass digg->reddit exodus
21:13:47 <elliott> Phantom_Hoover: basically stuff is no longer posted by users, it's posted by sites.
21:13:54 <elliott> that's not strictly true for all cases, but
21:14:30 <Vorpal> elliott, of course, reddit will go down the bad path at some point too. It is kind of built in in that type of site I think...
21:14:33 <Phantom_Hoover> It smacks a little of the inability of Facebook users to deal with the change of a single pixel...
21:15:01 <elliott> Phantom_Hoover: the fact that some complaints are bullshit
21:15:06 <elliott> does not imply that all complaints are bullshit, ok?
21:15:14 <elliott> i remember the whole last.fm upgrade bullshit. it wasn't like that
21:15:18 <elliott> digg is, quite literally, a graveyard.
21:15:54 <elliott> Phantom_Hoover: let's put it this way: a few days after the Digg upgrade, the majority of front page links linked to *reddit threads* instead of the link directly. that's literally how much it had fallen, the top links were all people linking to reddit
21:16:13 <nooga> now my gf asks me out
21:16:17 <Phantom_Hoover> elliott, OK, I just want to know what made people leave/
21:16:30 <nooga> how could I resist
21:16:44 <elliott> Phantom_Hoover: i can't find a summary of the changes, and don't have the patience to blab about it
21:16:49 <elliott> as i've never been a digg user
21:17:15 <Phantom_Hoover> I only thought it was trivial since that's how the changes seemed when I looked it up.
21:18:02 <elliott> the whole submission structure was changed
21:18:13 <elliott> Vorpal: i have pumpkins everywhere. wat
21:19:27 <elliott> Vorpal: btw realworld->nether translation isn't literal; it won't spawn you on lava, for instance
21:20:35 <elliott> Vorpal: or on water, the other way
21:21:12 <Vorpal> <elliott> Vorpal: or on water, the other way <-- on ice though
21:21:40 <Vorpal> elliott, what it should do is spawn you on a small platform
21:22:10 <Vorpal> with the portal in the middle
21:22:21 <Vorpal> and a few tiles above the lava
21:22:25 <Vorpal> or at sea-level (if water)
21:23:10 <elliott> Vorpal: Fun thing to do: Get to a place where F4 will spawn you into a portal. Do so. Step back a bit so you don't get sucked in. Press F4 a lot. You get stuck in block and lose health rapidly, but then jumping somehow evaporates it and you end up in a box of portals.
21:24:11 <Vorpal> elliott, I don't want to mess up my map I told you
21:24:31 <elliott> Vorpal: So make a new world :P
21:25:12 -!- TLUL has joined.
21:26:16 <elliott> Vorpal: It's awesome because you can run quickly through the portals, but if you end up at a dead end... NETHER
21:26:34 <Vorpal> elliott, screenshot I told you
21:26:38 <Vorpal> elliott, also nether is no dead end
21:26:52 <elliott> screenshotted but nowhere to put them, feel free to offer scp
21:26:53 <Vorpal> elliott, with a pickaxe you can get elsewhere and make a portal and go back
21:27:20 -!- Zuu has quit (Ping timeout: 260 seconds).
21:27:25 <Vorpal> elliott, what about http://imageshack.us/ ?
21:27:45 <elliott> btw holding F3 shows stats
21:28:00 <Vorpal> elliott, also I can offer scp but only if you set up an ipsec tunnel with X.509 certs to me. Send me your local root CA cert ;P
21:28:00 <elliott> Vorpal: i use imgur.com, which is like imageshack.us but not run by an asshole. but it's tedious to upload multiple images
21:28:17 <Vorpal> <elliott> btw holding F3 shows stats <-- well known
21:28:45 <Vorpal> elliott, you can lock it on by pressing alt after iirc
21:28:48 <elliott> nether transition should be smoother :(
21:28:59 <pikhq> elliott: Fortunately, it seems very likely that the leap second will be abolished.
21:30:59 <pikhq> elliott: The IERS vote on the issue will be finished in 2011. If 70% of member nations agree, then the leap second will cease to be.
21:31:13 <elliott> pikhq: I *like* the leap second.
21:31:41 <pikhq> It's more pain than it's worth, IMO.
21:31:57 <pikhq> The main problem with it is that it's not algorithmically computable.
21:32:44 <pikhq> Instead, you just have about 6 months notice about whether or not there will be one.
21:33:56 -!- madbr has quit (Ping timeout: 240 seconds).
21:36:00 <pikhq> Phantom_Hoover: Impractical without a very good means of measuring DUT1.
21:36:07 <pikhq> (the difference between UT1 and UTC)
21:36:16 <elliott> pikhq: simulate the universe
21:36:19 <elliott> see what they say in that universe
21:36:36 <pikhq> elliott: No, it's much easier than that...
21:37:53 <pikhq> elliott: You need to simply be able to observe UT1 (the mean solar time at the meridian through Greenwich), and see if the difference between UT1 and UTC would go above or below 0.9s in the next 6 months.
21:40:23 <Sgeo> Suppose we get rid of the leap second. How much time until it becomes an issue? I suppose technically it's not determinable given our lack of ability to actually determine long enough in advance when leap seconds should be, but an approximation
21:42:20 <elliott> pikhq: If Greenwich ever gets nuked we are so fucked.
21:42:30 <elliott> pikhq: Congratulations! You are now the Kitten toolchain consultant.
21:43:09 <elliott> pikhq: Your job is to get a working pcc/uClibc where the uClibc was compiled with pcc and the pcc was compiled with pcc and statically-linked with uClibc -- on Linux.
21:43:16 <pikhq> Actually, the Prime Meridian no longer goes directly through the Greenwich observatory.
21:44:00 -!- madbr has joined.
21:48:42 <pikhq> The Prime Meridian is 5.31 arcseconds east of the meridian going through the transit circle at Greenwich, rather than going straight through.
21:48:53 <pikhq> Because of an accident when setting up GPS.
21:49:30 <pikhq> And they chose to change the definition of the Prime Meridian rather than change the satellites.
21:50:06 <Sgeo> Is that within walking distance?
21:50:08 <pikhq> Ah, sorry, when setting up TRANSIT, which was the very first satellite navigation system.
21:50:31 <pikhq> Sgeo: It's 102.5 meters at the latitude of the Royal Observatory.
21:52:09 <elliott> pikhq: what's your rate for getting insane linux toolchains to work
21:54:16 <pikhq> elliott: 10g of ¹⁹⁷Au per hour.
21:54:45 <pikhq> I doubt you can pay that.
21:54:59 <elliott> pikhq: Sheesh, I might charge for Kitten if nobody else does any drudge work at all :P
21:56:11 <Sgeo> In that case, I should charge for PSOX!
21:57:38 <Sgeo> pikhq, any reason you don't want the other isotope in your gold?
21:58:30 * Sgeo double-checks something
21:59:57 * Sgeo fails to find it online
22:01:02 <elliott> pikhq: and THEN what would you do!
22:01:06 -!- elliott has left (?).
22:01:59 -!- elliott has joined.
22:02:09 <elliott> so can gparted resize partitions mounted on / yet?
22:02:15 <elliott> or is linux still lagging behind os x in that aspect :)
22:03:48 <elliott> pikhq: and if you say use LVM i'll kill myself
22:05:05 <elliott> 7 gigs should be enough for a kitten base install methinks
22:05:06 -!- Mathnerd314 has joined.
22:05:44 <elliott> Phantom_Hoover: That's not a question.
22:05:53 <elliott> Phantom_Hoover: Well, you have to do drudge work, then.
22:06:24 <elliott> Phantom_Hoover does not get Kitten for free!
22:06:31 <elliott> Phantom_Hoover: Not successfully. :P
22:06:59 <elliott> pikhq: NOW IS YOUR CHANCE TO CHANGE THE HISTORY FUTURE: JFS or btrfs?
22:07:16 <Sgeo> Would I be able to do any of the druge work?
22:07:21 <Sgeo> I mean, if I wasn't ignored?
22:07:28 <elliott> Phantom_Hoover: pinging people like that is irritating and will get you ignored
22:07:57 <elliott> Sgeo: maybe, have you ever compiled anything ever?
22:08:15 <Sgeo> Make that a yes
22:08:40 <elliott> Sgeo: do you have linux installed?
22:08:57 <Sgeo> Not at this very moment, no, but I do have access to my school's Linux system
22:09:33 <elliott> Sgeo: Have you ever debugged problems with the C stdlib initialisation files (crt1.o) etc. on Linux?
22:09:38 <elliott> Phantom_Hoover: but it doesn't run active worlds!
22:10:09 <Sgeo> elliott, no :(
22:10:10 <elliott> Phantom_Hoover: You won't do drudge work.
22:10:23 <elliott> Phantom_Hoover: (Drudge work = compiling uClibc and pcc twice or so, sorting out problems.)
22:10:31 <elliott> Phantom_Hoover: (This involves recompiling pcc with pcc and uClibc.)
22:10:41 <elliott> Phantom_Hoover: (Also probably compiling a cross-compiling gcc, so that uClibc works.)
22:10:47 -!- Sasha2 has joined.
22:11:58 <elliott> Sgeo: Then I can't think of anything you can do except test it, which would involve... installing it.
22:12:54 <Phantom_Hoover> Since it was for Microcosm, and apparently despite the fact that *he had said the ABI was x86-64 UNIX* he claimed that the ABI hadn't been picket.
22:13:13 <elliott> Microcosm is just a lame version of my user-space POSIX project, anyway. :P
22:13:40 <Sgeo> MAKE WORDS WITH TTIASACRMMMILLHTMITEHIT
22:14:21 <Sgeo> (Disclosure: I'm doing this for Reddit stuff)
22:15:20 <elliott> Sgeo: Too wimpy to install it? :p
22:15:34 <Sgeo> elliott, maybe over the weekend when I have time
22:15:53 <Sgeo> Also, can I install in user-space?
22:16:04 <Phantom_Hoover> True; terrible Italians always scrape along crevasses. Remember, multi-meters militantly inject legumes litigiously, having tried mayonnaise. I tried eating hare, I think.
22:18:25 <elliott> Sgeo: ...how the hell would you be able to install in user-space?
22:18:47 <Sgeo> Depends on what exactly is being installed.
22:18:54 <Sgeo> OS stuff, obviously not
22:18:59 <elliott> Sgeo: Actually you'll probably only be able to install from Linux to start with... as I haven't written an installer.
22:19:31 <elliott> Sgeo: You realise that Kitten is an OS?
22:20:35 <elliott> Sgeo: ... it's a Linux distro.
22:22:31 <elliott> Sgeo: So, seems I have to look elsewhere for a drudger :p
22:22:41 <elliott> Phantom_Hoover: So want to do the drudge-work?! That isn't actually drudgy?!
22:22:44 <Sgeo> Or just wait until I have some free time
22:23:06 <Phantom_Hoover> elliott, the maximum amount of work I'm doing is "sh drudge.sh".
22:23:22 <elliott> Phantom_Hoover: Oh come on, compiling stuff is fun!
22:23:41 <pikhq> ザルゴイットカームズ!ヒーフーウェイツビハインドザワルズ!
22:24:08 -!- kar8nga has quit (Remote host closed the connection).
22:24:49 <elliott> pikhq: You get to compile, then!
22:25:43 <pikhq> elliott: sàrukò i'to kâmusù? hî hû uēitu hìhaintò sà warusù?
22:26:11 <elliott> pikhq: So, JFS or btrfs, which do you want to see?
22:26:22 <pikhq> elliott: Definitely ZFS.
22:26:41 <Phantom_Hoover> pikhq, also, your vowel extension lines are MISPLACED!
22:26:52 <elliott> Phantom_Hoover: They're not lines.
22:26:59 <elliott> They're pikhq's own romanisation scheme.
22:27:05 <elliott> pikhq: You... do realise that the native ZFS port is not yet released?
22:27:29 <elliott> pikhq: Oh look: "This project solves the licensing issue by distributing ZFS as a separate kernel module users will have to download and build for themselves."
22:27:35 <elliott> pikhq: So no, I cannot use ZFS. Pick again.
22:27:48 <pikhq> elliott: I'm aware, I was being unhelpful.
22:27:51 <elliott> pikhq: [[There's still some major work to be done, so this is not production-ready code. The ZFS Posix Layer has not been implemented yet, therefore mounting file systems is not yet possible; direct database access, however, is. Supposedly, KQ Infotech is working on this, but it has been rather quiet around those parts for a while now.]]
22:28:04 <elliott> pikhq: I won't even make you work on it! Just offer opinions, dammit :p
22:28:24 <pikhq> elliott: Btrfs seems significantly more awesome.
22:28:53 <elliott> pikhq: btrfs is also still labelled as "WILL DESTROY YOUR DATA FUCKING FUCK FUCKSHIT", and is not part of any official stable kernel release.
22:29:07 <elliott> pikhq: Also, btrfs needs a separate /boot.
22:29:55 <elliott> pikhq: Whereas JFS is insanely stable, "fsck" takes less than a second if there's nothing wrong and a second or two for excellent recovery if there is, has many of the advanced XFS features without the unreliability, and is generally awesome. The *single* downside to JFS is that it doesn't support resizing on Linux. IIRC it does on OS/2.
22:30:22 <pikhq> Phantom_Hoover: ... Loogoid? What?
22:30:26 <pikhq> Phantom_Hoover: That was Zalgo.
22:30:40 <elliott> pikhq: Oh, and btrfs is under Oracle control.
22:30:45 <pikhq> Phantom_Hoover: You suck at reading kana, apparently.
22:30:52 <pikhq> elliott: Wait, seriously? Fuck Oracle.
22:31:07 <elliott> pikhq: Yes; it is an Oracle project. The lead developer works on it at Oracle.
22:31:10 <pikhq> Phantom_Hoover: Then... He sucks. :P
22:31:19 <pikhq> Phantom_Hoover: Or just doesn't handle English-in-kana well.
22:31:27 <pikhq> (which, granted, is completely screw.)
22:31:32 <elliott> pikhq: You will now whine about how JFS doesn't support resizing.
22:32:16 <pikhq> elliott: Well, then JFS isn't an option either.
22:33:05 <elliott> pikhq: Personally the lack of resizing doesn't bother me too much. But obviously it's a deal-breaker for many. So, tell me: what do you suggest? :P
22:33:27 <pikhq> elliott: Resizing is a very nice thing.
22:33:48 <pikhq> Sufficiently so that I consider filesystems not supporting it a non-option.
22:34:01 <elliott> Phantom_Hoover: Just that.
22:34:09 <elliott> Being able to resize a partition.
22:34:17 <elliott> pikhq: Come on then, I'm waiting :)
22:34:46 <elliott> pikhq: ReiserFS isn't an option for obvious reasons. XFS is a bit prone to data loss, apparently.
22:35:05 <elliott> pikhq: Apparently Linux supports the FFS BSD filesystem, but really now, no :P
22:35:36 <pikhq> elliott: Namesys is somehow still functioning.
22:35:49 <elliott> pikhq: The future of the company fell into doubt after Reiser was found guilty of murder and announced plans to sell the company to pay for his legal defense.[2] Their website has not been accessible since November 2007. Edward Shishkin, a Namesys employee, was quoted in a January 2008 CNET article as saying that "commercial activity of Namesys has stopped".[3]
22:35:59 <pikhq> elliott: They get funding from DARPA.
22:36:00 <elliott> pikhq: Namesys is dead, dead, dead and ReiserFS with it.
22:36:15 <elliott> pikhq: ReiserFS 4 development is still stopped.
22:36:21 <elliott> pikhq: And I don't want to use ReiserFS, so there :P
22:36:22 <pikhq> elliott: Why Hans hasn't sold it is beyond me.
22:36:33 <elliott> pikhq: ...can he? He's in jail, you recall.
22:36:38 <pikhq> elliott: And they're trying to get Reiser4 into the kernel...
22:36:42 <elliott> pikhq: Also, he's hardly a sane man.
22:36:51 <pikhq> elliott: Yes, you can still sell stock ownership in jail.
22:37:01 <elliott> pikhq: See my note about his sanity :P
22:37:05 <Sgeo> If he was found guilty, what's he hoping for wrt the legal defense? Less jail time or avoiding of capital punishment?
22:37:13 <elliott> Sgeo: ...he's already in jail.
22:37:29 <elliott> On September 5, 2008, Hans Reiser arrived at San Quentin State Prison to begin serving his sentence. Reiser tried to appeal his second-degree murder conviction on October 30, 2008. The request was denied by Judge Larry Goodman on November 13, 2008.[50][51] On January 10, 2009, it was reported that Reiser was recovering after having been beaten by several prisoners.[52][53] On January 28, 2009, he was transferred to Mule Creek State Prison.[54]
22:37:31 <Sgeo> So what can he be hoping for in terms of legal defense?
22:37:42 <elliott> Sgeo: ...why would he be hoping for any legal defence at this point?
22:37:47 <elliott> He confessed, he's in jail, 15 years to life.
22:37:57 <Sgeo> "The future of the company fell into doubt after Reiser was found guilty of murder and announced plans to sell the company to pay for his legal defense"
22:38:21 <pikhq> "Reiser cannot appeal his conviction or sentence as a result of his plea bargain."
22:38:25 <elliott> Before he confessed, Sgeo.
22:38:41 <elliott> pikhq: Picked a filesystem yet? :P
22:39:07 <pikhq> elliott: Fuck filesystems.
22:39:20 <elliott> pikhq: WELL THAT SOLVES THE PROBLEM
22:40:12 <elliott> The ext series' obsolescence is planned; Ted Ts'o says that ext4 is a stopgap measure, and the future is btrfs.
22:40:28 <elliott> Therefore ext5 will not exist; ext4 is the last version, before a transition to btrfs.
22:40:37 <olsner> rename it ext5, plobrem sloved
22:41:05 <Sgeo> So not only will we have to time travel, but we'll have to go to a different universe
22:41:07 <pikhq> olsner: Except that all the other ext filesystems are related.
22:41:08 <Sgeo> Problem solved
22:41:23 -!- cheater99 has quit (Ping timeout: 245 seconds).
22:41:29 <madbr> why are there so many linux filesystems :D
22:41:51 <pikhq> Most of the filesystems Linux supports were designed for other OSes...
22:41:52 <elliott> madbr: there aren't that many.
22:42:17 -!- Phantom_Hoover has quit (Remote host closed the connection).
22:42:30 <elliott> pikhq: Maybe I'll use venti. :P
22:42:52 <pikhq> Ones for Linux in particular are: extfs, btrfs, ReiserFS. That I can think of.
22:42:56 <elliott> pikhq: Maybe I'll use fossil, I mean. :P
22:43:06 <elliott> pikhq: You'd need a ported fossil server in your initramfs, and 9P support in the kernel.
22:43:18 <elliott> pikhq: ext is an extended version of the Minix filesystem.
22:43:37 <elliott> pikhq: You're right, ext is Linux-specific.
22:43:53 <pikhq> And the other ext filesystems are patches on that.
22:43:57 <elliott> pikhq: Maybe I'll use http://en.wikipedia.org/wiki/Xiafs, ext2's competitor.
22:44:08 <elliott> "The maximum size of a file was 64 MiB and the maximum size of a partition was 2 GiB."
22:44:34 <pikhq> ... Wow, I'm pretty sure FAT16's better than that.
22:44:44 <elliott> pikhq: Apparently that's *better* than ext.
22:44:47 <elliott> ext must have been terrible.
22:44:56 <pikhq> No, FAT16 is precisely that.
22:45:06 <elliott> pikhq: ext -- worse than FAT16!
22:45:19 <elliott> (Or, I guess, it might be the same as ext and have improved other things. Maybe.)
22:45:35 <elliott> pikhq: If you like specific bootloaders, now's the time to demand one.
22:45:43 <elliott> (As part of the default package set, that is.)
22:45:50 <madbr> fat16 handles large files no?
22:45:52 <pikhq> elliott: Eh, Grub2's not bad it seems.
22:46:04 <elliott> pikhq: Pick again. (I already know what my favourite is.)
22:46:11 <pikhq> elliott: I don't care that much, to be honest.
22:46:23 <elliott> pikhq: Correct! The answer is: LILO.
22:46:41 <pikhq> elliott: I mean, a bootloader runs for all of a few hundred clock cycles. Doesn't matter in the slightest.
22:46:50 <elliott> LIIIIIIIIIIIIILOOOOOOOOOOOOOOOO
22:47:26 <pikhq> madbr: FAT16 handles up to 4GB files.
22:47:33 <pikhq> madbr: But only up to 2GB filesystems.
22:48:11 <pikhq> Likewise for FAT12, which handles up to 4GB files, but only up to 32MB filesystems.
22:48:19 <pikhq> elliott: The LInux LOader certainly works.
22:48:32 <elliott> pikhq: It can also chainload! It can even read JFS partitions!
22:48:34 <pikhq> And doesn't do much more than that.
22:48:40 <Sgeo> Surely it's just an accidental feature that it theoretically handles files up to that size?
22:48:45 * Sgeo should read the specs
22:49:15 <Sgeo> What's a good first filesystem to read about?
22:49:16 <elliott> pikhq: http://i.cbsi.com.au/story_media/339303370/slackware-131_7.jpg Doesn't look bad if you ask me.
22:49:39 <elliott> pikhq: Different design with multiple OSes: http://www.designlegion.com/linux/screenshots/lilobmp.jpg
22:49:53 <pikhq> Sgeo: The file size is stored as 4 bytes.
22:50:06 <pikhq> Sgeo: On all FAT filesystems.
22:51:08 <elliott> pikhq: Question. Does any filesystem split files into multiple inodes?
22:51:59 <pikhq> elliott: Why would it? An inode is just metadata.
22:52:00 <Sgeo> So what happens if a Windows user tries to ... oh, FAT
22:52:13 <Sgeo> Does anyone still use FAT?
22:52:23 <pikhq> Sgeo: Most flash drives.
22:52:26 <madbr> on my win98 box, yeah
22:52:36 <pikhq> Sgeo: And a lot of consumer electronics. A *lot*.
22:52:37 <elliott> pikhq: If two files share the same N byte chunk... would be nice to share their storage? Eh, I dunno.
22:52:57 <madbr> but is the added complexity of that worth it?
22:52:59 <Sgeo> elliott, Amiga Fast File System?
22:53:00 <pikhq> elliott: So, you're asking about deduplicative storage.
22:53:02 <Sgeo> Or something else?
22:53:07 <elliott> Sgeo: Unix/BSD Fast File System.
22:53:26 <Sgeo> http://en.wikipedia.org/wiki/Berkeley_FFS ?
22:53:46 <elliott> Sgeo: Read it all, the history is quite informative.
22:53:58 <elliott> pikhq: Well, yes. A lookup tree from some hash to where it's stored. Only problem there is that you have to not only do a lookup but then a compare of N bytes to make sure...
22:54:30 -!- cheater99 has joined.
22:55:06 <elliott> pikhq: How big are the chunks? Configurable? (Default?)
22:55:13 <olsner> (I want some kind of citation before I believe that 4GB file-size limit for FAT12)
22:55:30 <elliott> olsner: see http://en.wikipedia.org/wiki/File_Allocation_Table
22:55:36 <elliott> Max file size 4 GB minus 1 byte (or block size if smaller)
22:55:39 <elliott> for FAT-12, FAT-16, FAT-32
22:55:41 <pikhq> elliott: Filesystem block level.
22:55:48 <elliott> pikhq: how big are those typically again? 4k?
22:55:55 <elliott> olsner: so FAT-12 can store 4 GB files but only 32 MB volumes :D
22:56:10 <olsner> elliott: yeah, I saw that, but I want proof that it's also true
22:56:17 <elliott> olsner: well, imagine if it only used two bytes
22:56:23 <elliott> olsner: and a non-power-of-two is ... iffy
22:57:11 <olsner> a ... non-power-of-two-times-eight power of two
22:57:16 <pikhq> elliott: Buut by default it just trusts the hash.
22:57:42 <Sgeo> olsner is questioning what's so iffy about 3 bytes
22:57:46 -!- TLUL has left (?).
22:57:55 <Sgeo> For the size thing
22:57:56 <olsner> if you mean it goes 2 -> 4 -> 8 bytes, then you have 2^(8*2^n)
22:58:13 <elliott> olsner: 2^1 = 2. 2^2 = 4. 2^3 = 8
22:58:21 <pikhq> elliott: Though deduplication is not enabled by default.
22:58:35 <pikhq> elliott: And you can tell it to verify a lack of collision with a filesystem option.
22:58:40 <elliott> pikhq: I can't put my faith in an filesystem that I can't trust to always work if my hardware is OK.
22:58:47 <elliott> And verification sounds slow.
22:59:03 <Vorpal> elliott, see "pig carting" on http://www.minecraftwiki.net/wiki/Saddle
22:59:39 <pikhq> elliott: Eh, involves reading or looking in the filesystem cache to verify.
22:59:53 <olsner> elliott: that's the storage size in bytes, what I referred to was the maximum size you can store
23:00:09 <elliott> olsner: err, surely it just gives a start pointer and file size
23:00:12 <elliott> i don't see why it wouldn't
23:00:24 <pikhq> elliott: And it'd have to do a disk access anyways; with deduplication your blocks are reference-counted.
23:00:37 <Sgeo> Learn You A Haskell now has a chapter on zippers!
23:00:51 <olsner> elliott: also, I'm still talking about the formulas for non-iffy sizes
23:00:56 <olsner> not about FAT12 anymore
23:01:20 <elliott> olsner: the maximum file size in these is controlled entirely by the file size field
23:01:26 <elliott> which any sane person would size as a power of two
23:01:42 <elliott> olsner: a power of two number of bytes, that is
23:01:46 <elliott> olsner: what's the C type for 3 bytes?
23:02:48 <pikhq> elliott: But in FAT-12 the actual filesystem pointers are 12 bits. I think they were mad.
23:02:55 <Vorpal> elliott, there isn't one?
23:03:00 -!- FireFly has quit (Quit: swatted to death).
23:03:06 <Vorpal> elliott, why do you need one?
23:03:07 <olsner> I'm just saying that if the field size grows as 2^n, the file sizes you can store in that field grows as 2^(8*2^n)
23:03:15 <elliott> Vorpal: READ CONTEXT OR DON'T REPLY AT ALL FFFFFFFFFFFFFFFFF
23:03:29 <Vorpal> elliott, calm down, after all I did for you
23:03:48 <elliott> olsner: or, 2^(2^(n+3)) :P
23:03:57 <Vorpal> <pikhq> elliott: But in FAT-12 the actual filesystem pointers are 12 bits. I think they were mad. <-- um, do they use the remaining 4 bytes up to the 16 byte boundary for anything else?
23:04:11 <elliott> Vorpal: THE FIRST FEW BYTES OF THE FILE CLEARLY
23:04:49 <Vorpal> elliott, well it could be used for flags or some such
23:04:52 <olsner> Vorpal: 12 bits plus 4 bytes? that's 44 bits
23:04:54 <Vorpal> elliott, like "is read only"
23:04:59 <Vorpal> olsner, 4 bits I meant
23:05:18 <madbr> vorpal: they pack them together
23:05:20 <pikhq> Vorpal: Erm. Well, yeah, it's not 0-filled after the 12.
23:05:28 <Vorpal> pikhq, ah, what is it used for?
23:05:44 <pikhq> Vorpal: Depends on where in the structs...
23:05:44 <madbr> so it codes 2 entries over 3 bytes
23:06:17 <Vorpal> elliott, actually you can do this in C. bitfield
23:06:22 <madbr> floppies are nasty
23:06:25 <elliott> Vorpal: not packing though
23:06:26 -!- ais523 has quit (Remote host closed the connection).
23:06:32 <pikhq> Vorpal: All on-disk file pointers are 12 bits, and the whole thing is a bunch of linked lists...
23:06:42 <madbr> if you use 25% less space for the FAT, you're saving on the reading speed
23:06:42 <Sgeo> This looks vaguely like it would benefit from being a monad
23:07:03 <Vorpal> pikhq, as long as they don't cdr-encode XD
23:07:25 <elliott> pikhq: I am so tempted just to use ext3.
23:07:43 <madbr> most floppy drives are ridiculously slow
23:08:45 <Vorpal> elliott, anything wrong with ext4?
23:09:10 <elliott> Vorpal: The curmudgeon inside me sees no point to extents.
23:09:19 <elliott> Vorpal: Or this fancy allocation tricky thingamajig.
23:09:29 <pikhq> elliott: It has benefits.
23:09:36 <elliott> http://upload.wikimedia.org/wikipedia/en/c/c5/E2fsck-uninit.png
23:09:50 <elliott> Keep talkin', keep gettin' my shit.
23:10:48 <elliott> UFSv2 uses 16k blocks by default cuz its MODERN
23:10:56 <pikhq> elliott: Okay, eventually we'll have you only supporting FAT12.
23:10:57 <Vorpal> elliott, well it is useful when you have huge files
23:11:07 <elliott> (space or time). The file system can either be instructed to try
23:11:07 <elliott> to minimize the time spent allocating blocks, or to try to mini‐
23:11:07 <elliott> mize the space fragmentation on the disk. If the value of min‐
23:11:07 <elliott> free (see above) is less than 8%, the default is to optimize for
23:11:07 <elliott> space; if the value of minfree is greater than or equal to 8%,
23:11:09 <elliott> the default is to optimize for time. See tunefs(8) for more
23:11:10 <Vorpal> elliott, like virtualisation disk images
23:11:11 <elliott> details on how to set this option.
23:11:32 <elliott> $ sudo mkfs.ufs -J /dev/sda2
23:11:33 <Vorpal> elliott, from where is that?
23:11:36 <elliott> pikhq: Just made your damn decision for you.
23:11:46 <elliott> pikhq: We're using the motherfucking BSD filesystem.
23:11:48 <Vorpal> elliott, ah *bsd again
23:11:56 <elliott> Vorpal: I'm doing this on Linux though :)
23:12:04 <elliott> Vorpal: UFS is fully supported in Linux.
23:12:08 <Vorpal> elliott, ufs support on linux is read-only iirc
23:12:11 <elliott> And mkfs.ufs is just the BSD newfs.
23:12:20 <elliott> Vorpal: Not that I know of. Hell, I just created one in Linux, I should hope it can write it.
23:12:38 <Vorpal> elliott, uh, I thought it was badly broken as well. Hm
23:12:58 <Vorpal> elliott, and not UFS2 support of course
23:13:02 <elliott> Vorpal: Yes, UFS2 support.
23:13:07 <elliott> Or at least, I just created a UFSv2 file system.
23:13:20 <elliott> [31980.209605] ufs was compiled with read-only support, can't be mounted as read-write
23:13:25 <elliott> Vorpal: Nope, it does UFSv2.
23:13:31 <elliott> Vorpal: It was just COMPILED not to.
23:13:45 <Vorpal> elliott, the kernel option says "Read write UFS support (DANGEROUS BROKEN!)" or some such
23:14:04 <Vorpal> elliott, I told you it wouldn't work...
23:14:16 <elliott> Vorpal: Interestingly I don't take your statements as certain fact.
23:14:23 <elliott> Especially when you put "I thought" in front of them.
23:14:39 <elliott> pikhq: I might just use XFS and never upgrade it.
23:14:43 <Vorpal> elliott, my word is of course law ;P
23:15:16 <olsner> I think ZFS is probably the coolest file system right now
23:15:37 <elliott> olsner: ZFS is (1) under Oracle control and (2) unsupported under Linux
23:15:46 <olsner> yes, unfortunate accidents
23:15:57 -!- digimunk has left (?).
23:15:59 <elliott> olsner: Not only does the Linux native port lack things such as "mounting a filesystem" (POSIX layer hasn't been implemented), I can never *ship* a kernel with it as that would be a license violation.
23:16:10 <elliott> olsner: btrfs is also out the window because of Oracle control.
23:16:34 <olsner> btrfs is oracle-filth?
23:17:05 <elliott> olsner: yeah -- it's an Oracle "product" and the lead developer works on it at Oracle
23:17:16 <olsner> oh joy, "Developer: Oracle Corporation" says wikipedia
23:17:31 <elliott> olsner: even if they don't corrupt it, I'm uninterested in using anything Oracle puts out.
23:18:48 <pikhq> http://www.dangermouse.net/esoteric/intelligentdesignsort.html
23:19:23 <Sgeo> Well, that didn't teach me stuff I didn't know
23:22:32 <pikhq> You are not permitted to learn.
23:22:52 <olsner> "My fear that Oracle would buy Sun only to let it die are becoming reality. I can't help but envision the corpse of Sun lying inert while a cloven-hoofed Larry Ellison dances around it, cackling -- such a tragedy."
23:24:36 <elliott> olsner: ...quite the vision
23:25:24 <olsner> collecting and piling the corpses, dancing around, cackling
23:25:39 <elliott> olsner: SO WHICH FISLESYSTEM SHOULD I USE
23:25:55 <elliott> Sgeo: i'm so sad you won't test kitten sniff sniffle sniffle
23:26:07 <Sgeo> elliott, if I had time I would
23:26:17 <Sgeo> Let me get VirtualBox working
23:26:18 <elliott> Sgeo: You spend an awful lot of time IRCing :P
23:26:23 <elliott> Sgeo: No, VirtualBox testing isn't useful.
23:26:26 <elliott> I have VirtualBox myself :P
23:26:26 <olsner> elliott: btw, if something starts boiling again, remind me that it's supposed to
23:26:32 <elliott> What I don't have is various different hardware configurations.
23:26:46 -!- madbr has quit (Ping timeout: 265 seconds).
23:26:49 <Sgeo> So, you want me to install some OS you made?
23:27:01 <elliott> You can easily put it on another partition.
23:27:07 <elliott> Although doing that might actually require Linux. Ho hum.
23:27:11 <Sgeo> You could still be doing malicous things
23:27:26 * Sgeo can't spell today
23:27:26 <elliott> Sgeo: I really have better things to do than try and fuck up your computer.
23:27:46 * Sgeo doesn't have time right now. Maybe tomorrow
23:27:55 <elliott> Sgeo: Hell, it's not ready today.
23:28:19 <elliott> Sgeo: Try a few weeks to a month :P
23:29:30 <elliott> pikhq: " The U.S. Justice Department accused Oracle Corp. of defrauding the federal government on a software contract that involved more than $1 billion in sales."
23:31:10 -!- Sasha2_ has joined.
23:31:47 <Sgeo> Hey, lame story about me being related to some quasi-famous thing in some way
23:32:04 * Sgeo has become self-aware!
23:32:23 -!- Sasha2 has quit (Read error: Connection reset by peer).
23:33:07 -!- Sasha2_ has changed nick to Sasha2.
23:34:01 <Vorpal> elliott, .... ... http://www.worldofminecraft.com/
23:35:13 * Sgeo hits Vorpal with a High-Tech Hand
23:35:39 <elliott> Vorpal: I assume you've seen:
23:35:43 <elliott> Vorpal: http://www.penny-arcade.com/comic/2010/9/17/
23:35:44 <elliott> Vorpal: http://www.penny-arcade.com/comic/2010/9/20/
23:35:45 <Vorpal> Sgeo, what do you mean?
23:36:00 <Vorpal> elliott, no I haven't. *reads*
23:36:02 <Sgeo> There was a Creatures fansite called The High-Tech Hand
23:36:14 <Vorpal> Sgeo, uh? creatures as in?
23:36:29 <Sgeo> (Well, series of games)
23:36:46 <olsner> elliott: what can you turn sticks into?
23:36:59 <elliott> olsner: ALL SORTS OF THINGS (when paired with wood)
23:37:06 <Vorpal> Sgeo, and what has that got to do with anything?
23:37:18 <Sgeo> Vorpal, the weird name for a site
23:37:30 <olsner> literally every sort of thing?
23:37:49 <Vorpal> olsner, nah. Just a huge number of things
23:38:15 <Vorpal> olsner, various tools, torches, weapons, fishing rod, lots more
23:38:34 <Vorpal> and also not just with wood
23:38:51 <Vorpal> well that would be for low quality tools/weapons
23:38:56 <Vorpal> you want rock later on
23:39:16 <Vorpal> perhaps, if you are lucky and find some, diamond
23:39:18 <Sgeo> elliott, have you been reading Sam Hughes's NaNoWriMo stuff?
23:40:11 <olsner> oh no... there was a plot twist but I missed it, now I have to backtrack to find out when it twisted and why
23:40:41 <olsner> also, don't distract me :)
23:41:36 -!- elliott has quit (Ping timeout: 255 seconds).
23:42:35 <Vorpal> Sgeo, isn't he still ignoring you?
23:43:24 <olsner> oh, now I get it, the cuts that looked like they were from a parallel story was actually the cuts that explained the plot twist
23:43:47 <Sgeo> Vorpal, um, no?
23:43:56 <Sgeo> Unless he was obsessively reading logs
23:53:40 -!- elliott has joined.
23:53:43 <elliott> Vorpal: Write a filesystem. (You'd be good with drudge work like that, I feel.)
23:57:29 <Vorpal> elliott, ouch. I'd hate to do that
23:57:42 <Vorpal> elliott, I'd just use zfs except for it's license
23:57:56 <elliott> Ugh, but then I don't want to use ext4, because fscking is so slow compared to jfs...
23:58:02 <Vorpal> elliott, yeah but with a saner license it wouldn't be in such control
23:58:05 <elliott> pikhq: When was the last time you resized a partition?
23:58:12 <elliott> Vorpal: it'd still need a fork for me to trust it.
23:58:15 <elliott> Vorpal: (btrfs has the same problem)
23:58:22 <Vorpal> elliott, to trust jfs?
23:58:23 <elliott> Vorpal: what do you mean "jfs?"
23:58:29 <Vorpal> elliott, I suggest using jfs!
23:58:41 <elliott> Vorpal: Absolutely perfect of course, except that it doesn't support resizing partitions.
23:58:43 <Sgeo> elliott, Sam Hughes NaNoWriMo
23:58:48 <Sgeo> Have you been reading it?
23:58:49 <Vorpal> elliott, it supports growing iirc
23:58:58 <Vorpal> elliott, not shrinking though
23:59:00 <elliott> Vorpal: not that I know of
23:59:10 <elliott> JFS includes a rather unusual partition-resizing ability: It's built into the kernel's JFS driver. You can use this feature to increase, but not to decrease, the size of the filesystem. As with most other partition-resizing tools, you must modify the partition size first by using fdisk to delete the partition and then recreate it with a larger size. After you've done this, you should mount the partition as you normally do and then issue the follo
23:59:26 <Sgeo> Just wondering. Also, there's at least one story that you'd have some shnaudenfraud over imagining me in it
23:59:45 * Sgeo is aware of the fail
23:59:47 <Vorpal> elliott, I suggest just using gparted