00:00:02 <nooga> avarything should be a pointer
00:00:13 <nooga> like in a goddamn C# or something
00:01:03 -!- augur has joined.
00:01:05 <Sgeo> Why is Factor more popular than Newspeak?
00:01:15 <Sgeo> Why am I comparing apples and oranges?
00:01:23 <Sgeo> <insert Gregor reference here>
00:01:33 <elliott> ... so anyway, ok seriously, what's the easiest way to become nobody in a shell script
00:01:46 <Vorpal> elliott, su nobody <command>?
00:01:59 <Sgeo> Wouldn't that require root access?
00:02:03 <elliott> Vorpal: require's nobody's password. it, of course, has none.
00:02:13 <Vorpal> elliott, root shouldn't need it
00:02:15 <elliott> Vorpal: sudo works for any user, of course, but requires the user's password
00:02:18 <elliott> Vorpal: indeed, i'm not root
00:02:26 <Vorpal> elliott, then you can't change user
00:02:35 <Vorpal> elliott, and you can use sudo
00:02:38 <elliott> [sudo] password for elliott:
00:02:40 <Vorpal> elliott, to do it without password
00:02:43 <elliott> as i said, pointlessly requires the user's password
00:02:52 <elliott> also, i'd rather not depend on sudo
00:02:53 <Vorpal> elliott, yes indeed sudo is suid root
00:02:56 <Vorpal> that is why it can do it
00:03:03 <elliott> you still haven't answered my question
00:03:24 <Sgeo> Stop trying to hack into nobody's account! nobody runs web servers!
00:03:24 <Vorpal> elliott, but you can't do it without having root or *effectively having root thanks to being suid root*
00:03:33 <Sgeo> (Or, well, should. Regular users shouldn't)
00:03:42 <elliott> Vorpal: the latter is slightly acceptable.
00:03:45 <Vorpal> Sgeo, no it shouldn't. A special www user should
00:03:52 <elliott> Vorpal: any user on a system should be able to build a package
00:03:56 <Vorpal> elliott, the latter wouldn't work on the shell script
00:04:08 <elliott> Vorpal: i could just write an asnobody.c
00:04:25 <Vorpal> elliott, that is what you would have to do
00:04:26 <elliott> i don't really like strictly more setuid programs than necessary :)
00:04:34 <Vorpal> elliott, or use sudo with NOPASSWD
00:04:54 <Vorpal> I think is is a bad idea though in general
00:04:56 <elliott> Vorpal: (I wonder why there isn't an asnobody already, all it can do is reduce privileges...)
00:05:16 <Vorpal> elliott, it feels insecure. Could they mess up for other users also building packages?
00:05:39 <Vorpal> elliott, what else on the system uses nobody?
00:05:56 <elliott> i guess they could rm -rf it
00:06:05 <elliott> what we really need is asnewtemporaryuser :)
00:06:28 <elliott> Vorpal: debian-specific, and WAY overblown for this
00:06:35 <elliott> Vorpal: it's as simple as setuid(rand()) :P
00:06:44 <elliott> Vorpal: it's as simple as setuid(max_uid_in_etc_passwd+rand()) :P
00:07:43 <elliott> Vorpal: i might actually just use su here, it may end up that you need to run it as root anyway
00:07:48 <elliott> Vorpal: due to busybox tar not having --owner=
00:08:37 <Vorpal> elliott, why do you use busybox?
00:08:53 <Vorpal> elliott, also if you are packaging projects what about stuff that needs to install as separate uses
00:09:00 <Vorpal> elliott, just look at qmail for example
00:09:05 <Vorpal> several different users
00:09:20 <Vorpal> qmailq, qmails and so on iirc
00:09:21 <elliott> Vorpal: --owner just changes the owners of the files
00:09:47 <Vorpal> elliott, yes and they need to be different owners. Not all should be changed to root
00:09:54 <elliott> Vorpal: postinstall script :P
00:10:01 <elliott> Vorpal: it's because packages aren't built as root for obvious reasons
00:10:05 <elliott> i mean there's little other option really
00:10:18 <elliott> postinstall script for the rare such package is probably the easiest way
00:12:04 <Ilari> APNIC: 3.67 /8s in RIR Pool... And APNIC is extremely likely to get the last 2x/8s.
00:14:24 <Sgeo> There's apparently work on NS3
00:14:35 -!- oerjan has quit (Quit: Good night).
00:20:34 <elliott> pikhq: Question. Should I bother including cc in the build dependencies for any package?
00:20:51 <fizzie> fakeroot is always an option. :p
00:20:53 <pikhq> elliott: I say "yes".
00:21:02 <elliott> fizzie: requires dynamic linker.
00:21:16 <elliott> pikhq: Yeah, but... who tries to compile a package without them? :P
00:21:24 <pikhq> Though most distros just say "If it's in the base system, it's not marked as a dependency of anything."
00:22:22 <pikhq> Yeah, that's because Debian doesn't fuck around.
00:23:50 <elliott> pikhq: I most definitely fuck around!
00:23:52 -!- Mathnerd314 has joined.
00:23:56 <fizzie> Well, fakeroot with some sort of ptrace-hooked syscalls, then. The elegant choice!
00:23:58 <elliott> My package manager is like a piece of string!
00:24:00 <Ilari> Then there's less known problem of ASN depletion.
00:24:07 <elliott> pikhq: The string is FLIMSY!
00:24:16 <elliott> pikhq: Hey, an Actual Runtime Dependency for vi.
00:24:28 <elliott> (Got a better name for that?)
00:24:53 <elliott> # The preferred choice for ex on Linux distributions, other systems that
00:24:53 <elliott> # provide a good termcap file, or when setting the TERMCAP environment
00:24:53 <elliott> # variable is deemed sufficient, is the included 2.11BSD termcap library.
00:25:07 <pikhq> Ilari: ... ASN depletion.
00:25:33 <elliott> pikhq: Oi oi oi we're naming packages here! Focus on the important stuff!
00:26:30 <Ilari> There's of course an upgraded spec that solves it. But you need upgraded systems to peer with AS with extended ASN...
00:26:31 <pikhq> Oh, it was extended to 32 bits a few years ago. I hope that the BGP routers actually have been updated.
00:26:53 <elliott> pikhq: So what should I call the termcap db. :p
00:27:06 <elliott> (Or is it all terminfo these days? As quoted, vi uses 2.11BSD termcap.)
00:27:06 <pikhq> elliott: termcap-db
00:27:58 <elliott> Is there any other kind of terminal database other than termcap, really?
00:28:54 <elliott> pikhq: Can't that read termcap files?
00:29:03 <Ilari> At least ASN upgrade should be less of a hassle than IP upgrade, since only systems that peer with extended ASNs need to be upgraded.
00:29:39 <elliott> pikhq: I mean, I really don't want to fuck around with terminfo because it uses binary files and crap.
00:30:06 <elliott> pikhq: Please validate me :P
00:30:08 <fizzie> You can convert to/fro termcap file / terminfo database, at least up to some extent.
00:30:09 <pikhq> elliott: Pretty sure they're incompatible.
00:30:18 <elliott> I'll just build everything with termcap and hope for the best.
00:30:28 <elliott> pikhq: I'll offer both termcap and terminfo in the same package.
00:31:24 <pikhq> Hmm. Seems that some programs explicitly call out to termcap or terminfo...
00:31:34 -!- elliott has quit (Quit: Leaving).
00:32:10 <fizzie> I don't seem to have termcap files anywhere any more.
00:36:14 -!- sftp has quit (Remote host closed the connection).
00:37:33 <fizzie> Debian seems to have thrown out their old "termcap-compat" package. (They just ship a ncurses "libtermcap" that actually reads terminfo; termcap-compat was for old code you couldn't for some reason or another recompile. And/or those people with custom termcap reading code.)
00:40:38 <fizzie> Already five years ago, in fact.
00:43:28 -!- Mathnerd314 has quit (Disconnected by services).
00:43:32 -!- Sasha has joined.
00:43:50 -!- Mathnerd314_ has joined.
00:44:26 -!- Mathnerd314_ has changed nick to Mathnerd314.
00:44:27 -!- oklofok has joined.
00:44:51 -!- Sasha2 has quit (Ping timeout: 245 seconds).
00:47:25 -!- oklopol has quit (Ping timeout: 245 seconds).
00:48:46 <Sgeo> "Files are extremely important in current computing experience. Much too important. Files should be put in their place; they should be put away."
00:50:10 <Sgeo> "Ultimately, it is about control: If you dont have a file system, it becomes harder for you to download content from unauthorized sources. This is also good for security, and in a perverse way, for the user experience. And its also good for software service providers." suddenly, I feel ill
00:51:19 <Sgeo> I don't think this person actually supports that as a reason to get rid of file systems
00:53:17 <oklofok> everything implies a true proposition
00:54:21 <oklofok> (that file systems should be gotten rid of)
00:55:08 <oklofok> i don't really understand what he's saying
00:55:32 <Sgeo> http://gbracha.blogspot.com/2010/02/nail-files.html
00:57:15 <Gregor> Spamusers fails at sodomy.
01:00:08 <oklofok> i still don't see what the security thing was
01:00:36 <oklofok> does he assume that with objects comes some sort of not being able to download arbitrary objects and do whatever the fuck you want with them
01:00:52 -!- madbr has joined.
01:19:09 -!- Sasha2 has joined.
01:19:10 -!- Sasha has quit (Read error: Connection reset by peer).
01:21:12 -!- augur has quit (Remote host closed the connection).
01:22:14 -!- Sasha has joined.
01:23:22 -!- Sasha2 has quit (Ping timeout: 240 seconds).
01:34:36 -!- FireFly has quit (Quit: swatted to death).
01:46:23 <Ilari> "This proposal should be considered an emergency proposal. IANA
01:46:36 <Ilari> exhaustion is likely to occur prior to the next ARIN meeting."
01:46:37 -!- BeholdMyGlory has quit (Remote host closed the connection).
01:47:04 <Ilari> ... Looks like the depletion is expected to occur very soon...
01:47:13 -!- GreaseMonkey has quit (Ping timeout: 272 seconds).
01:48:53 <Ilari> Looking at when APNIC could justify allocation, depleting the pool completely... The IANA depletion could occur at any moment.
01:49:04 <pikhq> The next ARIN meeting is apparently early April.
01:49:13 <pikhq> Wait, the IANA depletion could occur *any moment*?
01:50:04 -!- GreaseMonkey has joined.
01:50:37 <Ilari> Formally, the depletion estimates have not changed because APNIC allocates last, but in practice, the date may have changed...
01:52:17 <Ilari> An that "at any moment" comes from global policies. APNIC probably could justify allocation even now.
01:52:19 <pikhq> So. D-day is basically any time in the next 5 months or so.
01:54:18 <Ilari> I think it is in next 3 and half months...
01:54:56 <pikhq> I'm allowing for *extreme* optimism.
01:57:26 <pikhq> Some bastard is going around suggesting that companies think about IPv6 migration "in the next 2-5 years".
01:57:40 <pikhq> By which time the Internet will have been full for several years.
01:58:02 <Ilari> Even Houston model, which seems optimistic predicts 04-Mar-2011 ... That's sightly over 3 months away...
01:59:19 <Ilari> Now there's new predictions about X-day this year.
02:00:25 <Sgeo> So, celebrations in 3 months?
02:01:30 <Ilari> Prepare for having to do it sooner... A lot sooner...
02:03:05 <Ilari> Well, no official model predictions... But informal ones...
02:10:01 -!- Mathnerd314 has quit (Ping timeout: 255 seconds).
02:12:05 <Ilari> Wonder when next IPv6 allocation will occur (bringing the pool down to 505 blocks, which is under 99%)...
02:13:20 <pikhq> It amazes me that even 1% has been used. IPv6 is just so very, very large...
02:14:30 <Ilari> 1 various block and 5 RIR blocks... IIRC, the actual amount of delegations is 0.027% or so...
02:15:06 <Ilari> That is, those allocations are at about 2.7 or so...
02:15:10 <pikhq> It's being pieced up from the IANA in ridiculously large blocks, then.
02:15:35 <Ilari> /12s ... 1M * 2^32 networks.
02:15:58 <pikhq> Yeah, that's a pretty gigantic block.
02:21:53 <Ilari> Oh, and then there are 3456 blocks in undefined address ranges...
02:22:23 <Ilari> IPv6 address space is just gigantic.
02:24:20 <pikhq> Oh, right. Currently the IPv6 space is *only defined* in 2000::/3
02:27:50 <Ilari> Global unicast space, that is. There are also some other blocks, such as ULA space and broadcast space.
02:28:17 <pikhq> Wow. The "NFL International Series". A scheme whereby the NFL plays a regular season game in London.
02:28:37 <pikhq> ... I didn't know American football had any fans at all outside of North America.
02:31:29 -!- Zuu has quit (Read error: Connection reset by peer).
02:34:06 <Ilari> Heh... The "realtime" IPv4 depletion counter that uses the Houston model is down.
02:34:13 <pikhq> Hmm. There's an American football World Cup. The US didn't play in the first two. XD
02:37:04 -!- madbr has left (?).
02:38:38 -!- Zuu has joined.
02:38:44 <Sgeo> Although I almost thought... meh, no spoilers here
02:47:20 <Sgeo> To a character: "In case you forgot, [spoiler]"
02:47:33 <Sgeo> I think the audience forgot
03:00:40 -!- augur has joined.
03:47:51 -!- guny has joined.
03:48:07 <guny> hi watch my music video!!! http://www.youtube.com/watch?v=A_Q6U2O-qx4
03:48:09 <guny> hi watch my music video!!! http://www.youtube.com/watch?v=A_Q6U2O-qx4
03:48:10 <guny> hi watch my music video!!! http://www.youtube.com/watch?v=A_Q6U2O-qx4
03:48:12 <guny> hi watch my music video!!! http://www.youtube.com/watch?v=A_Q6U2O-qx4
03:48:14 -!- guny has quit (Client Quit).
03:49:20 -!- zeotrope_ has quit (Ping timeout: 265 seconds).
03:51:10 -!- zeotrope has joined.
03:51:36 -!- zeotrope has changed nick to Guest75069.
04:08:39 -!- augur has quit (Remote host closed the connection).
04:11:16 -!- wareya has quit (Read error: Connection reset by peer).
04:12:35 -!- wareya has joined.
04:47:03 -!- GreaseMonkey has quit (Quit: ilua).
04:59:37 <Goosey> I hate being a polite programmer
04:59:44 <Goosey> Why do I have to say please :(
05:20:14 -!- yorick has quit (Read error: Operation timed out).
05:27:00 <pikhq> INTERCAL, I presume?
05:37:30 -!- augur has joined.
05:51:25 -!- wareya has quit (Ping timeout: 255 seconds).
05:52:08 -!- wareya has joined.
05:53:06 -!- yorick has joined.
06:02:30 -!- adu has joined.
06:02:53 -!- wareya has quit (Read error: Connection reset by peer).
06:03:28 -!- wareya has joined.
06:08:19 -!- wareya has quit (Ping timeout: 264 seconds).
06:09:17 <Sgeo> "And when Simon Peyton Jones, one of the designers of Haskell, was asked why Haskell has only such a basic module system, he said that they didn't feel they were smart enough to design a real one. Let that sink in ... The designers of Haskell. Not smart."
06:09:21 <Sgeo> http://stackoverflow.com/questions/4057973/osgi-like-modularity-in-other-programming-languages
06:09:25 <Sgeo> Is... that a real thing?
06:15:22 <adu> Sgeo: i've heard something similar
06:15:38 <adu> but SPJ didn't use the word "smart" in the version i read
06:16:01 <adu> he made it sound as tho it was "sufficient" the way it was
06:19:30 <Sgeo> "The fact that one rarely needs more than one window is one of the things I really like about Hopscotch. Theres no need for a docking bar, or tabs for that matter. Tabs are popular these days, but they dont scale: they occupy valuable screen real estate, and beyond half a dozen or so become disorienting and unmanageable."
06:19:37 <Sgeo> http://gbracha.blogspot.com/2008/07/debugging-visual-metaphors.html
06:19:44 <Sgeo> So much for a tabbed Newspeak browser
06:24:51 <adu> wow OSGi sounds interesting
06:28:07 * Sgeo attempts to redirect adu to Newspeak
06:28:33 <adu> what's Newspeak
06:28:58 <Sgeo> http://newspeaklanguage.org/
06:29:10 <Sgeo> It's the language fawned over in the answer to that SO question
06:31:17 -!- Goosey has quit (Remote host closed the connection).
06:35:25 -!- Guest75069 has quit (Ping timeout: 260 seconds).
06:38:22 -!- yorick has quit (Read error: Operation timed out).
06:46:20 <adu> Sgeo: Newspeak sounds a lot like io
06:46:37 * Sgeo has only heard of IO and knows nothing about it
06:47:15 <adu> There are several languages I wish could have babies
06:47:34 <adu> Go + Io + Prolog + Haskell
06:47:55 <adu> that would be a cute kid
06:55:57 -!- yorick has joined.
07:02:41 -!- FireFly has joined.
07:10:16 -!- Sgeo has quit (Ping timeout: 245 seconds).
07:16:56 -!- wareya has joined.
07:18:08 <Vorpal> <adu> Go + Io + Prolog + Haskell <-- how would that work?
07:19:02 <Vorpal> adu, oh? you think so?
07:19:23 -!- Sgeo has joined.
07:19:28 <Vorpal> adu, I can't see how you could reconcile the imperative and purely functional aspects there
07:20:04 <Vorpal> adu, and if you think it would work so well, why not implement it!
07:20:28 <adu> everything Go does would be of type IO () and everything Prolog does would be at the typeclass level
07:20:34 <adu> Io doesn't really fit
07:20:51 <Vorpal> I don't know much about Io so I can't say anything about that
07:21:55 <Vorpal> have to leave now, cya
07:22:02 <adu> There's pretty much a 1-to-1 mapping between (almost) anything imperative and Haskell's IO ()
07:56:56 -!- perdito has quit (Ping timeout: 245 seconds).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:05:30 -!- adu has quit (Quit: adu).
08:14:26 -!- Sgeo has quit (Read error: Connection reset by peer).
08:28:45 -!- nooga has quit (Ping timeout: 272 seconds).
08:28:49 -!- nooga has joined.
08:38:36 <ineiros> fizzie: http://open.spotify.com/album/29dWA4uMn07qxfEAGO3wSh
09:06:01 -!- MigoMipo has joined.
09:08:18 -!- FireFly has quit (Quit: swatted to death).
09:18:15 -!- FireFly|n900 has quit (Ping timeout: 245 seconds).
09:20:18 -!- augur_ has joined.
09:20:19 -!- augur has quit (Read error: Connection reset by peer).
09:23:19 -!- atrapado has joined.
09:32:37 -!- FireFly|n900 has joined.
09:36:51 -!- MigoMipo has quit (Ping timeout: 265 seconds).
09:37:25 -!- FireFly|n900 has quit (Ping timeout: 245 seconds).
09:47:53 <fizzie> ineiros: "Uh." (Incidentally, I've never heard any of the in-game music.)
09:51:21 -!- FireFly|n900 has joined.
09:56:10 -!- FireFly|n900 has quit (Ping timeout: 245 seconds).
10:05:53 -!- perdito has joined.
10:10:36 -!- FireFly has joined.
10:20:34 -!- FireFly|n900 has joined.
10:25:01 <nooga> it appears that UK is paralyzed because of some minor snow
10:43:44 -!- MigoMipo has joined.
11:19:44 -!- myndzi has quit (Ping timeout: 250 seconds).
11:24:01 -!- nopseudoidea has joined.
11:40:28 -!- rodgort has quit (Quit: Coyote finally caught me).
11:40:39 -!- rodgort has joined.
11:41:54 -!- nopseudoidea has quit (Quit: Quitte).
11:52:54 -!- MigoMipo has quit (Quit: Quit).
11:55:45 -!- FireFly|n900 has quit (Ping timeout: 245 seconds).
12:10:02 -!- FireFly|n900 has joined.
12:20:55 -!- myndzi has joined.
12:22:59 -!- Sasha2 has joined.
12:24:59 -!- Sasha has quit (Ping timeout: 272 seconds).
12:42:44 -!- FireFly|n900 has quit (Ping timeout: 255 seconds).
12:42:44 -!- myndzi has quit (Read error: Connection reset by peer).
12:43:11 -!- myndzi has joined.
12:50:47 -!- FireFly|n900 has joined.
12:54:31 -!- myndzi has quit (Ping timeout: 264 seconds).
12:55:59 <nooga> my raytracer now generates modern art
13:04:21 -!- myndzi has joined.
13:20:51 -!- elliott has joined.
13:22:32 -!- Phantom_Hoover has joined.
13:24:21 -!- Leonidas has quit (Ping timeout: 240 seconds).
13:24:25 -!- Leonidas has joined.
13:25:52 -!- sftp has joined.
13:38:11 -!- Phantom_Hoover_ has joined.
13:41:05 -!- oerjan has joined.
13:41:06 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
13:45:29 <elliott> augur_: I find your recent statements about INTERCAL highly offensive
13:58:03 <elliott> http://www.reddit.com/r/programming/comments/eecvm/a_compiler_language_which_has_nothing_at_all_in/c17hc7w
13:58:22 <elliott> I think ais523 and oerjan, accomplished INTERCAL programmers, would take great objection to this slight.
14:02:50 -!- FireFly|n900 has quit (Ping timeout: 255 seconds).
14:08:09 <nooga> elliott: is your school closed?
14:08:34 <elliott> i am unsure why you are laughing
14:08:38 <nooga> like what... snow is toxic or what?
14:09:20 <elliott> more seriously, i don't suppose you comprehend the idea of driving in heavy snow that continues to fall being perhaps /dangerous/?
14:09:41 <nooga> i do this during whole winter
14:10:18 <elliott> nooga: i think we have firmly established that Poland has not quite come to grips with the concept of safety yet
14:10:42 <nooga> ask some secure swedes
14:10:48 <nooga> tehy've got shitloads of snow too
14:10:49 <elliott> guess what, we don't all drive Volvos.
14:11:54 <elliott> hmm is it known which s gives SHA-1(s) = 0?
14:12:12 -!- MigoMipo has joined.
14:12:14 <nooga> http://img.wiadomosci24.pl/g2/4b/cd/83/11082_1162885840_b968_p.jpeg
14:12:29 <nooga> + modern cars with ASR, BAS and whatever
14:13:08 <elliott> nooga: yes, we do indeed remove snow from roads. guess what! we have rural areas.
14:13:15 <elliott> guess what! it takes time to clear the snow from everywhere.
14:16:39 <Phantom_Hoover_> Even though the snow is considerably worse than yesterday.
14:16:55 <elliott> Phantom_Hoover_: They decided they were open, and upon looking out the window immediately decided that no, they're not open.
14:17:20 <elliott> But yeah, jesus christ, this snow.
14:17:27 <Phantom_Hoover_> elliott, nope, I went. Although we all went home at lunch.
14:17:37 <elliott> Phantom_Hoover_: I meant here.
14:18:03 <Phantom_Hoover_> It's because our cretin of a First Minister said that the schools should reopen and they did, in spite of the police saying the opposite.
14:19:52 <nooga> ppl could just walk to school
14:20:09 <elliott> nooga: i don't think you realise the kind of distances present in britain...
14:20:22 <Phantom_Hoover_> It's still stupid, especially since the previous two days were both snow days.
14:20:37 <nooga> even in poor Poland we've got schools max 3-4km from home
14:20:44 <Phantom_Hoover_> So apparently once the snow goes over a certain critical depth it ceases to matter.
14:20:50 <elliott> nooga: do you delight in being really fucking stupid?
14:21:02 <nooga> i delight annoying you :D
14:21:04 -!- nopseudoidea has joined.
14:23:14 <elliott> Phantom_Hoover_: Quick! What's the SHA-1 hash of a directory?
14:29:59 -!- FireFly|n900 has joined.
14:32:48 <elliott> Phantom_Hoover_: http://en.wikipedia.org/wiki/CubeHash
14:32:56 <elliott> Phantom_Hoover_: The SHA-3 competitor I'm rooting for, thanks djb.
14:33:04 <elliott> (It's not "rooting" is it?)
14:34:27 -!- BeholdMyGlory has joined.
14:35:01 <fizzie> I don't think CubeHash is all *that* simpler; the wikipedia doesn't bother listing the round transformation, which is he usual add-rotate-swap-style mess, just like the SHA-1 compression function.
14:35:19 <fizzie> It *is* nice to see something that's not the usual Merkle–Damgård construction though.
14:37:31 <elliott> fizzie: djb instantly simplifies everything he touches. duh.
14:37:57 <elliott> schneier's function has been cryptanalysed a bit cuz he's a luzr
14:38:11 <elliott> although I can't think of any :P
14:38:27 <elliott> Merkle–Damgård construction
14:38:28 <elliott> (Redirected from Merkle-Damgård construction)
14:38:55 <fizzie> It's a different sort of hyphen.
14:39:27 <elliott> the thing in the wikibox links to one that's redirected
14:39:41 <fizzie> And that someone could be you!
14:40:05 <elliott> "Although no proof has been constructed, Oozlybub and Murphy is thought to be Turing-complete if and only if Goldbach's Conjecture is true." --cpressey
14:40:24 <elliott> "Oozlybub and Murphy is a programming language. Despite appearances, this name refers to a single language.
14:40:24 <elliott> The majority of the language is named Oozlybub. The fact that the language is not entirely named Oozlybub is named Murphy.
14:40:24 <elliott> For the sake of providing an "olde tyme esoterickal de-sign", the language combines several unusual features,
14:40:26 <elliott> including multiple interleaved parse streams, infinitely long variable names, gratuitously strong typing, and only-conjectural Turing completeness."
14:40:28 <elliott> http://catseye.tc/projects/oozlybub-and-murphy/doc/website_oozlybub-and-murphy.html
14:41:18 <elliott> It even has Unicode support!
14:44:10 <nooga> cpressey is a monster
14:53:38 -!- Sasha2 has quit (Read error: Connection reset by peer).
14:53:45 -!- Sasha has joined.
14:53:55 <elliott> "That's true though. I invented DNA. I also invented the invention itself." --Peter Sunde, Pirate Bay co-founder
14:55:27 -!- augur_ has quit (Remote host closed the connection).
14:55:46 <elliott> Hahaha! Huckabee wants to execute Assange.
14:55:53 -!- augur has joined.
14:56:30 -!- augur has quit (Read error: Connection reset by peer).
14:56:36 -!- augur has joined.
14:57:01 <elliott> "Finally, Julian Assange is no hero. He is a twit. He should not be made into a liberal icon. He gives hackers a bad name. He and his organization are indeed enemies of the U.S. government and the people represented by that government; they should be stopped, and they richly deserve to be punished for this latest leak. And that goes double for the person or people in the U.S. government who leaked the documents in the first place. None of t
14:57:01 <elliott> hese people deserve your support any longer."
14:57:09 <elliott> -- Larry Sanger, solidifying his reputation as... a twit.
14:57:43 <elliott> (The rest of this "essay", if you can call it that -- http://www.larrysanger.org/wikileaks.html -- constitutes basically saying "Well, you see, they're MEANT to be private, because the government decided releasing them to the public would be a bad idea. Therefore it's dangerous to do so, and no government would ever abuse this! DUH.")
14:58:19 <Phantom_Hoover_> I have certain reservations about Wikileaks due to the whole "names of informants" thing, but I wouldn't go that far..
14:58:29 <elliott> Phantom_Hoover_: Um, they've been redacting names.
14:58:50 <Phantom_Hoover_> elliott, yes, but they didn't for that earlier military leak.
14:59:03 <elliott> Phantom_Hoover_: (They also asked the US government for *help* redacting information that could endanger people to minimise any risk. The US refused.)
14:59:09 <elliott> (for the most recent leak)
15:00:57 -!- Sasha2 has joined.
15:01:10 -!- augur has quit (Ping timeout: 260 seconds).
15:01:32 -!- ais523 has joined.
15:01:59 <elliott> I just love the chain of events there --
15:02:26 <elliott> Wikileaks: Hello, you know those documents we're going to release. We don't want anyone to get hurt; it's in your best interests to help us redact any information that could put people's lives at risk.
15:02:37 <elliott> Wikileaks: [releases documents]
15:02:52 <elliott> US govt., mass media: That's IRRESPONSIBLE TERRORISM! Think of the LIVES at risk!
15:03:18 -!- Sasha has quit (Ping timeout: 265 seconds).
15:04:15 <nooga> Wikileaks does good work
15:06:56 <elliott> Phantom_Hoover_: Really? How?
15:07:36 <elliott> Julian Assange is more a lightning rod than anything else, anyway -- you never hear about anyone but him and just about all you ever hear about him is negative; nobody else gets any shit. That's definitely intentional.
15:08:25 <elliott> Phantom_Hoover_: Semi-relatedly, have you seen his old blahhg? http://web.archive.org/web/20071020051936/http://iq.org/
15:08:29 <elliott> It was on reddit a while ago.
15:09:31 -!- nopseudoidea has quit (Quit: Quitte).
15:10:35 <elliott> ais523: in a file of "<sha1hash> <filename>", how would you denote a directory?
15:10:49 <elliott> I've been using an sha-1 length of zeroes, but it's conceivable that some string could actually hash to 0.
15:11:36 <elliott> Phantom_Hoover_: Umm... what the hell is the SHA-1 hash of a directory?
15:18:37 <Phantom_Hoover_> Make it "0", then append the / to the name so there's no chance of confusion,
15:19:53 <elliott> Phantom_Hoover_: Just 0? Not 40 0s?
15:20:44 <elliott> Phantom_Hoover_: Hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm.
15:23:14 <fizzie> Make it a random 40-digit hex string, and then change a single, random digit in there to a "g".
15:25:21 <fizzie> It was optimized for confusion, of course.
15:26:38 <elliott> No, I think the g is best.
15:26:53 <elliott> Change one "a" into the identical, Cryllic "a".
15:27:00 <elliott> (If there is no a in the string, regenerate it.)
15:28:12 <fizzie> Aren't crylls just palette-swapped trolls?
15:29:37 <elliott> You got 15 out of 19 Programming Language if You Know Their Creators.
15:30:01 <ais523> elliott: I'd just leave the hash out
15:30:09 <elliott> One mistake was really embarrassing... one or two of them I'm proud of... and the last one I just feel meh about.
15:30:18 <ais523> or hash the directory itself, they are technically speaking files, just you can't read them via normal methods
15:30:38 <elliott> http://www.sporcle.com/games/supreddit/prog_lang_wirth
15:30:39 <elliott> "Can you name the Programming Language if You Know Their Creators? (Niklaus Wirth Edition)?"
15:31:06 <elliott> ais523: "cat" will cat a directory on NetBSD :)
15:31:16 <ais523> hmm, is it decidable whether two regular expressions (actual regular expressions, without backreferences etc.) match the same set of strings?
15:31:32 <elliott> Differences between Oberon-07 and Oberon
15:31:32 <elliott> Niklaus Wirth, 8.8.2007 / 17.12.2007
15:31:32 <elliott> Oberon-07 is a revision of the original language Oberon as defined in 1988/1990.
15:31:35 <elliott> How is the man still alive...
15:31:38 <elliott> ais523: I don't /think/ so
15:31:48 <ais523> really? I was guessing yes
15:31:56 <elliott> ais523: I know that you can't minimise a regexp to its provably shortest form, IIRC
15:32:10 <ais523> for ordinary regular expressions, you can compile them into state machines
15:32:20 <ais523> then because there's a finite number of states, you can bruteforce
15:32:55 <oerjan> ais523: yes it is decidable
15:33:07 <ais523> yep, I feel a bit silly that I had to ask rather than figuring it out
15:33:16 <oerjan> you can do set difference and union
15:33:17 <ais523> it's a lot less obvious with backreferences involved
15:33:23 <elliott> "- Applied to values of type SET, the unary minus denotes the set complement, and the function
15:33:23 <elliott> ABS yields the number of elements of a set. The relations <= and >= denote set inclusion."
15:33:42 <oerjan> from which you can construct xor of two languages
15:33:42 <ais523> elliott: so you can do - on the empty set, and get a set of everything?
15:34:01 <ais523> oerjan: hmm, that's a different proof from the one I gave, and requires assumptions I didn't know
15:34:15 <elliott> I wonder why there is not more literature on I\Xi.
15:34:42 -!- MigoMipo has quit (Remote host closed the connection).
15:35:24 <elliott> ais523: can I have permission to troll the esolangs wiki with a language
15:37:04 <ais523> would the language be at all interesting in its own right?
15:37:14 <ais523> I suppose being created for the purpose of trolling makes a language an esolang
15:37:16 <elliott> ais523: a Turing-complete-and-no-greater language in which all programs nonetheless halt
15:37:18 <ais523> (/me coughs at LOLCODE)
15:37:21 <fizzie> ais523: A regular expression is a DFA, you can construct the minimal equivalent DFA pretty easily, and the minimal DFA (up to state naming) accepting a particular regular language is unique; so just construct the DFA (might have an exponential number of states, though) and minimize it for both regexps; if they are the same, they match the same set of strings.
15:37:35 <ais523> elliott: is that even theoretically possible?
15:37:37 <elliott> ais523: of course, such a language is only implementable on a machine with a Turing machine halting oracle; I will, therefore, provide an implementation. In Banana Scheme.
15:37:52 <ais523> fizzie: that was pretty close to my proof
15:37:56 <elliott> ais523: it's theoretically possible on a super-Turing machine; if halts(X) then run(X) else done
15:38:22 <elliott> hey, http://esolangs.org/wiki/Banana_Scheme has a serious error!
15:39:24 <ais523> elliott: I'm not entirely sure that such a language would technically be TC
15:39:42 <oerjan> plot plot plot plot plot plot plot plot banana scheme
15:39:44 <ais523> in fact, by the typical mathematical definition, it definitely wouldn't be
15:39:53 <elliott> ais523: that's why it's trolling!
15:39:56 -!- Sgeo has joined.
15:40:15 <elliott> ais523: you can run any Turing-complete-requiring computation you want to; even if it doesn't halt, you can run N steps for arbitrary finite N
15:40:18 <elliott> you just can't loop forever
15:40:41 <Sgeo> elliott, my new obsession is Newspeak
15:40:53 <elliott> ais523: ooh, now I want to make it so that you can write a Brainfuck program which halts if you put in "blah" and doesn't if you put in "bluh"
15:41:02 <elliott> ais523: and then make sure my implementation doesn't stop the program until you put in "bluh"
15:41:25 <elliott> hmm, so I have to do (H 0 `(assuming-we-get-the-character ,n)) for all 255 ns
15:41:39 <elliott> no, wait, much before every read
15:42:00 <nooga> how does Newspeak look? samples?
15:42:35 <Sgeo> Except a bit more syntax
15:42:47 <Sgeo> And slightly different conventions
15:42:57 <elliott> nooga: http://newspeaklanguage.org/. just ignore Sgeo, he has never known a language he didn't fall in love with and then reject for really stupid fucking reasons before ever using it
15:45:03 <ais523> hmm, 4 of the last 11 /8s were just allocated
15:45:20 * Sgeo would love to see Slava and Bracha collide
15:45:28 -!- augur has joined.
15:45:36 <ais523> when the next two are assigned, the RIRs get one each, so it's not long before there are no free /8s left
15:45:52 <elliott> 00:01:18 <fizzie> It's just that udev's better. :p
15:46:20 <fizzie> I don't remember the context at all.
15:48:16 <fizzie> Better than devfs, it seems.
15:48:36 <pikhq> ais523: Yeah, but people were calling it "2 of the last 9" because 2 of those went to ARIN, which was absolutely *going* to allocate yesterday.
15:48:40 <elliott> You're better than devfs. So there.
15:48:43 <pikhq> It's just RIPE's allocation is a surprise.
15:49:35 <fizzie> Well, you know us Europeans, we like to be surprising.
15:50:15 <pikhq> And APNIC could allocate any time in the next few months, thereby causing IPv4 depletion.
15:52:32 <oerjan> just a bit of friendly european surprise
15:52:52 <fizzie> Nobody expects the Norwegian swatquisition.
15:53:13 <fizzie> "Showing results for sw acquisition. Search instead for swatquisition"
15:57:07 <nooga> that should be a quote
15:58:11 <oklofok> "<ais523> hmm, is it decidable whether two regular expressions (actual regular expressions, without backreferences etc.) match the same set of strings?" <<< yes, obviously
15:58:44 <oklofok> "<ais523> for ordinary regular expressions, you can compile them into state machines" "<ais523> then because there's a finite number of states, you can bruteforce" <<< just be sure it's an algorithm and not a semialgorithm tho
16:00:37 <oklofok> "<elliott> ais523: a Turing-complete-and-no-greater language in which all programs nonetheless halt" <<< enumerate turing machines that halt, the program 1 runs the first one, 2 runs the second one etc
16:01:03 <elliott> oklofok: I'm trolling, remember? It's going to be brainfuck.
16:01:18 -!- augur has quit (Remote host closed the connection).
16:01:35 -!- pikhq has quit (Ping timeout: 265 seconds).
16:01:42 <Phantom_Hoover_> oklofok, while you're at it, biject to the computable reals.
16:02:37 <Sgeo> The thing that has an interpreter that's just a BF interpreter but those interpreters don't correctly report invalid programs
16:03:05 -!- pikhq has joined.
16:03:11 <Sgeo> elliott, your language that you're discussing
16:03:20 <elliott> <Phantom_Hoover_> oklofok, while you're at it, biject to the computable reals.
16:03:34 <elliott> difficult, since you'd have to consider programs under the equivalence relation of "same result as"
16:03:48 <elliott> otherwise you'd have countably infinite programs for each computable real
16:03:54 <elliott> Sgeo: no, the interpreter is written in banana scheme.
16:04:21 <Sgeo> ... so will the language count as implemented or unimplemented?
16:04:30 <oklofok> i heard today that if you assume the negation of aoc, you can prove there is a subset of R that doesn't have a countable subset, and i was surprised
16:04:37 <elliott> 18:28:17 <pikhq> Wow. The "NFL International Series". A scheme whereby the NFL plays a regular season game in London.
16:04:37 <elliott> 18:28:37 <pikhq> ... I didn't know American football had any fans at all outside of North America.
16:04:42 <elliott> Sgeo: implemented, just not compuatbly
16:04:50 <elliott> oklofok: i remember hearing that once... from wikipedia :P
16:05:08 <elliott> oklofok: http://en.wikipedia.org/wiki/Axiom_of_choice#Statements_consistent_with_the_negation_of_AC
16:05:15 -!- augur has joined.
16:05:44 <elliott> # There exists a model of ZF¬C in which real numbers are a countable union of countable sets.[11]
16:06:09 * Sgeo <baby noise that irritates elliott>
16:06:14 -!- Vorpal has quit (Read error: Operation timed out).
16:07:02 <elliott> 19:48:07 <guny> hi watch my music video!!! http://www.youtube.com/watch?v=A_Q6U2O-qx4
16:07:07 <oerjan> pikhq: i think there's even an american football team in Oslo
16:07:19 <oklofok> that was pretty trivial yeah
16:07:28 <Sgeo> She's going over bitwise operators
16:07:44 <Sgeo> She asked the class if anyone's classes delt with bitwise operators. Of course not
16:07:48 <Sgeo> I hate this school
16:08:02 <oklofok> well obviously they are too trivial to teach
16:09:01 <oklofok> a friend of mine used to play american football so it's here too i guess
16:09:02 <elliott> 22:47:34 <adu> Go + Io + Prolog + Haskell
16:09:03 <elliott> 22:47:55 <adu> that would be a cute kid
16:09:30 <oerjan> http://en.wikipedia.org/wiki/Norway_American_Football_Federation
16:09:35 <elliott> 01:47:53 <fizzie> ineiros: "Uh." (Incidentally, I've never heard any of the in-game music.)
16:09:44 <elliott> 00:38:36 <ineiros> fizzie: http://open.spotify.com/album/29dWA4uMn07qxfEAGO3wSh
16:11:57 -!- Vorpal has joined.
16:12:48 <Sgeo> Newspeak comes with late night wisdom
16:13:34 <Sgeo> http://ideone.com/mf4Lm
16:15:14 <Phantom_Hoover_> Sgeo, do you actually /program/, or do you just get incessant and short-lived obsessions with languages?
16:15:45 <Phantom_Hoover_> I actually think you've written less programs I know of than me.
16:16:20 <elliott> Phantom_Hoover_: HE WROTE PSOX
16:16:38 <Sgeo> I have a bunch of Python scripts to do random things on my old HD
16:17:15 <Sgeo> Including scraping a news site's sports section for photos from my home HS
16:17:24 <Sgeo> (This was when I was in HS. Don't get ideas)
16:18:51 <Sgeo> I have a few Haver clients lying around
16:19:22 <Sgeo> That's not my fault, is it?
16:20:42 <elliott> 09:23:05 <ihope> Hmm, what about a language where the dimensions are finite, but the number of them is infinite?
16:20:43 <elliott> 09:23:18 <ihope> You know: a 2x2x2x2x2x2x2x2x2x... universe.
16:21:00 <elliott> 09:27:27 <SimonRC> And, assuming integral coordinates in each case, is it clear by the diagonal argument that there is more space in that kind of universe than in an infinite univers with a finite number of dimensions
16:22:42 * Sgeo had a bit of trouble visualizing that diagonalization, but figured it out
16:23:00 <Sgeo> Actually, I may be visualizing it wrong still
16:23:19 <Sgeo> Phantom_Hoover_, I know what it looks like for decimals and... I was visualizing it for the 2x2x2x.... wrong
16:23:52 <oerjan> http://www.reddit.com/r/AskReddit/comments/ee95j/whats_the_most_internet_famous_thing_youve_done/c17guwl
16:23:53 <Sgeo> I was visualizing 2 rows, and only 0 and 1 were digits in both numbers
16:25:24 <Sgeo> Because I'm tired
16:25:34 <Sgeo> And wasn't thinking properly
16:25:59 <Sgeo> It still holds though
16:26:09 <Phantom_Hoover_> Moreover, the set of programs is logically equivalent to the set of subsets of the reals.
16:26:33 <Sgeo> Only need to swing down to the second number once in fact to make the nonstored num... it occurs to me that that is a bit overkill
16:26:42 <Sgeo> For the situation that I was envisioning
16:27:36 <elliott> 15:12:27 <fizzie> Mathematica is like unto a God:
16:27:36 <elliott> 15:12:28 <fizzie> In[1]:= Sum[1/(2^i), {i, 1, Infinity}]
16:27:36 <elliott> 15:12:28 <fizzie> Out[1]= 1
16:27:36 <elliott> 15:13:12 <SimonRC> yeah, but God got the answer by summing thw whole lot
16:27:36 <elliott> 15:13:37 <SimonRC> F*ck knows how Chuck Norris got his answer.
16:27:37 <elliott> 15:14:03 <fizzie> Who knows how Mathematica did it; might be magic!
16:29:05 <Phantom_Hoover_> For instance, are the accessible cells restricted to those indexed by a CR?
16:30:55 <Sgeo> http://ideone.com/to8J2
16:31:22 <elliott> 16:11:02 * SimonRC has worked out the definitive difference between scripting languages and "real" programming languages: in scripting languages, a simple string can be like 'foo' or "foo", but "real" languages only accept one of these (usually the former).
16:31:22 <elliott> 16:11:30 <lament> usually the altter
16:31:22 <elliott> 16:11:43 <SimonRC> erm, *latter
16:31:22 <elliott> 16:11:50 <SimonRC> yeah, thanks
16:31:25 <elliott> 16:11:54 <kipple> bah, real languages doesn't have strings...
16:31:33 <elliott> Phantom_Hoover_: It's not a language.
16:32:06 <Phantom_Hoover_> <elliott> 09:23:05 <ihope> Hmm, what about a language where the dimensions are finite, but the number of them is infinite?
16:32:23 <elliott> Phantom_Hoover_: Why not ask him?
16:33:30 <oerjan> he's on freenode as tswett
16:34:20 <oerjan> doesn't come here much
16:34:33 <Phantom_Hoover_> But not at the moment, and not that I can remember recently.
16:35:01 <oerjan> ...he's on the network now, says whois
16:35:11 <oerjan> he just doesn't come to this channel
16:35:34 <elliott> Gregor: OH MY GOD I HATE GHEXTRIS
16:36:14 <Vorpal> hum. I wonder what made my computer crash the moment I plugged in my mouse.
16:36:36 <Vorpal> works perfectly now, worked perfectly in my laptop today
16:37:11 <elliott> Phantom_Hoover_: sudo aptitude install ghextris
16:37:13 <Sgeo> "It's 3:00:00 am. Go get some rest!"
16:37:17 -!- perdito has quit (Ping timeout: 240 seconds).
16:37:25 <ais523> Vorpal: brownout, I wonder? it could be there's an intermittent short circuit in the mouse and it browned out your computer when you plugged it in
16:37:32 <Sgeo> ^^What Newspeak's IDE will tell you at the bottom if it's open at that time
16:38:13 <ais523> elliott: hmm, tetris on a hex grid?
16:38:32 <elliott> ais523: yes. and oh dear god it is impossible
16:38:59 * oerjan recalls playing hextris back in the day
16:39:07 <elliott> HEXAGONAL GEOMETRY MAKES NO SENSE
16:40:50 <Sgeo> She's now teaching the class hexadecimal
16:45:07 <elliott> 23:19:08 * _wildhalcyon_ blames linux for Kevin Federline's career.
16:49:14 <ais523> Phantom_Hoover_: being taught hex doesn't mean you don't know it already
16:49:20 <ais523> just that someone wanted to teach it to you
16:49:29 <ais523> there are some subjects I ended up learning over and over again
16:49:48 <ais523> boolean algebra is one I remember, I was taught it at least twice and new it already before then
16:50:00 <elliott> hmm... I want to revive ESO OS, now
16:50:23 <elliott> I will feel accomplished if I can get enough of an interpreter into a bootsector that there's enough space left to write the actual bootloader in the esolang
16:50:25 <ais523> I pretty much type by sound, for some reason, rather than by spelling
16:50:33 <ais523> so I often have to go back and correct a word into one of its homophones
16:50:37 <elliott> ais523: hooked on phonics eh
16:50:56 <ais523> elliott: some interps would fit into a bootsector just fine
16:51:02 <ais523> MiniMAX, for instance, or many other OISCs
16:51:23 <elliott> 23:36:47 <fizzie> (Of course there's _always_ a workaround: perhaps adding an evil library (to mangle the stdout in an __attribute__((constructor)) routine) to LD_PRELOAD might work, if stdout exists already when those are called.)
16:51:38 <elliott> ais523: yes, but you need to have enough space left in the boot sector to put the bootloader program in, written in the esolang
16:51:47 <elliott> ais523: can MiniMAX programs do "int 10h"?
16:52:10 <Sgeo> I should google that, so I will
16:52:21 <ais523> elliott: I wrote a couple of extensions for that
16:52:36 <elliott> ais523: are they on the wiki?
16:52:45 <ais523> there was a 32-byte interp that could not only do arbitrary DOS system calls, but also return values from them to a jump table
16:52:51 <ais523> but I never ran it and am not entirely sure if it works
16:52:53 <elliott> also, I hereby proclaim the OS to be the ESO OS, edition E
16:53:01 <elliott> wait, that isn't quite a palindrome
16:53:05 * Sgeo videos elliott
16:53:11 <elliott> also, I hereby proclaim the OS to be the ESOS (ESO OS), edition E
16:53:31 <Vorpal> <ais523> Vorpal: brownout, I wonder? it could be there's an intermittent short circuit in the mouse and it browned out your computer when you plugged it in <-- maybe
16:53:42 <Vorpal> ais523, how could I check this?
16:53:59 <elliott> ais523: hmm... the problem is, I don't even fail vaguely like writing a minimax program :)
16:54:05 <ais523> elliott: I haven't put the extensions on the wiki, mostly because I'm not sure whether they work, and can no longer remember how they work
16:54:16 <elliott> ais523: maybe I'll base it on Underload
16:54:16 <ais523> Vorpal: intermittent shorts aren't at all easy to check
16:54:26 <elliott> somehow, extending Underload seems less awful than extending brainfuck
16:54:26 <ais523> and some methods of checking for them have a tendency to set things on fire
16:54:31 <elliott> although I'm not sure how I'd do mutating a register
16:54:40 <Vorpal> ais523, I wouldn't want that
16:54:46 <Sgeo> She's now talking about swapping variables with tempoary variables
16:54:49 <Sgeo> This is Perl...
16:54:50 <ais523> it's what happened to my shorting power supply, though
16:55:01 <ais523> Sgeo: ($a, $b) = ($b, $a)
16:55:06 <ais523> elliott: how did you type that so quickly?
16:55:12 <ais523> also, we both missed the semicolon
16:55:14 <elliott> ais523: because I type really quickly
16:55:21 <elliott> also, don't need it if it's in {} :)
16:55:23 <Sgeo> Her response: "It depends on the language"
16:55:24 <Vorpal> ais523, anyway it could be the computer. I very rarely hotplug anything except the mouse to my desktop
16:55:29 <Sgeo> Which is correct, obviously
16:55:36 <ais523> Vorpal: I take it this is a USB mouse?
16:55:41 <elliott> Phantom_Hoover_: The non-computable reals?
16:55:50 <ais523> PS/2 mice aren't hotpluggable, and have been known to break motherboards when people try
16:55:56 <ais523> but that's a rather old technology
16:56:10 <Vorpal> ais523, I'm aware. Which is why I would never hotplug this keyboard
16:56:18 <elliott> Phantom_Hoover_: For any constructivist Coq work, including the computable reals, use http://c-corn.cs.ru.nl/.
16:56:18 <ais523> (serial mice are hotpluggable, but you have to be someone like me to ever have used one)
16:56:28 * oerjan read that as the hotpluggable reals
16:56:29 <Vorpal> ais523, I have *heard* of them that is all
16:56:43 <ais523> haven't used it in a while, though
16:56:51 <elliott> Phantom_Hoover_: I *think* the Reals section in http://coq.inria.fr/stdlib/ is non-constructive, by using axioms instead of constructions.
16:56:56 <Vorpal> ais523, anyway this mouse is technically serial when you expand the abbrev. USB
16:56:56 <elliott> Phantom_Hoover_: But C-CoRN is awesome.
16:57:03 <ais523> actually, I think I technically own three, but only one works nowadays
16:57:04 <elliott> Phantom_Hoover_: Russell O'Connor runs it.
16:57:17 <elliott> ais523: any suggestions to someone feeling like extending underload? apart from "don't"?
16:57:31 <ais523> elliott: remove the S statement and replace it with something more appropriate
16:57:39 <ais523> then you can compile it without having to keep the source around as well
16:57:43 <elliott> ais523: i wasn't planning to support S anyway
16:57:47 <ais523> Underload's actually pretty extensible
16:58:04 <ais523> hmm, now I'm wondering what an Underload interp in Forth would look like
16:58:09 <elliott> ais523: I was thinking about adding a command % which takes one value off the stack and "interpcalls" it
16:58:10 <Vorpal> so some other way of output? Or just having the final program state as output?
16:58:22 <ais523> I was just wondering how to do a minimal Underload interp, and what I thought up was very like the way Forth works
16:58:23 <elliott> ais523: the value would be expected to be an underload church numeral
16:58:50 <elliott> ais523: so e.g. (::**)% would be "syscall 3", and the remaining things on the stack could be arguments
16:58:50 <ais523> elliott: church numerals aren't massively space-efficient
16:58:59 <elliott> ais523: well, no, and this is also a rather boring extension
16:59:01 <Vorpal> elliott, ah so a system call like mechanism?
16:59:09 <ais523> more or less the best you could do would be along the lines of (:*:*:*:*:*)% for a high-numbered syscall
16:59:17 <elliott> ais523: I just want to be able to read a floppy using the BIOS in an extended esolang :)
16:59:27 <elliott> that I can write an interpreter/compiler for short enough to fit in a bootsector with enough space left to have the program there
16:59:30 <Vorpal> no more efficient representation of numbers in underlambda?
16:59:41 <ais523> Vorpal: they both represent numbers the same way
17:00:02 <ais523> and in underlambda, at least, where you don't have S to worry out, I'm assuming that most decent interps will recognise numbers and optimise them internally
17:00:19 <Vorpal> ais523, but can there not be any general number representation that is more space efficient in underload?
17:00:23 <Sgeo> Wait, there's really an underlambda?
17:00:38 <ais523> you can't recognise /all/ numbers as that's an uncomputable issue, but you can identify most common ways to construct them
17:00:38 <Vorpal> with general I mean "not special cased to a finite range"
17:00:44 <ais523> Sgeo: yes, it's semi-vaporware
17:01:00 <Sgeo> http://eurekamag.com/keyword/u/026/underlambda.php
17:01:21 <ais523> Sgeo: hmm, I somehow doubt that's the meaning I use the word for
17:01:32 <elliott> To differentiate their respective functions, oligonucleotide-directed site-specific mutagenesis was used to change the ATG start codon of the.vphi.X174 A* gene, previously cloned into pCQV2 under.lambda. Repressor control, into a TAG stop codon. The altered A* gene was then inserted back into.vphi.X replicative form DNA to produce an amber mutant,.vphi.XamA*. Two different Escherichia coli amber suppressor strains infected with this mutant produc
17:01:32 <elliott> ed viable progeny phage with only a slight reduction in yield. In Su+ cells infected with.vphi.XamA*,.vphi.X gene A protein, altered at one amino acid, was synthesized at normal levels; A* protein was not detectable. Prophage integration occurs at different chromosomal sites, including lacY and malB, but not at attB All.lambda.cam112 prophages are excised from the chromosome after induction but with various efficiencies for different locations. H
17:01:33 <elliott> eteroduplex analysis of.lambda.placZ transducing phages isolated from a lacY::.lambda.cam112 prophage reveals an insertion sequence 1 element at the joint of viral and chromosomal DNA Two lines of evidence indicate that.lambda
17:01:56 <Vorpal> ais523, hm what about compression? On very large church numerals you could presumably apply a compression algorithm to get a smaller one. (The usual caveats of compression applying here of course. You will get larger values sometimes)
17:01:58 <ais523> the word doesn't appear anywhere there but the title
17:02:19 <Sgeo> ais523, under.lambda
17:02:22 <ais523> Vorpal: :*:*:*:*-style compression isn't /that/ bad, really, it's O(log n)
17:02:34 <oerjan> Vorpal: you could obviously convert a church numeral to some binary representation
17:02:48 <Phantom_Hoover_> ais523, Underlambda was the proof-by-isomorphism language, wasn't it?
17:03:31 -!- tswett has joined.
17:03:37 <elliott> Uh oh, what did Phantom_Hoover_ do.
17:03:43 <elliott> (Or maybe tswett reads logs compulsively.)
17:04:00 <tswett> I receive an SMS every time someone says something in this channel.
17:04:06 <Vorpal> <oerjan> Vorpal: you could obviously convert a church numeral to some binary representation <-- well yes
17:04:14 <tswett> They cost me 25 cents apiece, so you'd better not say anything unless it's important.
17:04:20 <Vorpal> oerjan, but I meant in underload as the source is written
17:04:30 <ais523> tswett: that is probably a bad idea
17:04:59 <elliott> i think ais523 has a broken joke lobe
17:05:03 <Vorpal> <elliott> Uh oh, what did Phantom_Hoover_ do. <elliott> (Or maybe tswett reads logs compulsively.) <-- there is some context missing here?
17:05:13 <elliott> Vorpal: yes, it's called read your damn scrollback :)
17:05:30 <Vorpal> elliott, my scrollback goes back to when I joined after the crash half an hour ago
17:05:38 <ais523> elliott: my typical reaction to identifying jokes is to act like they're serious and try to drive the conversation into the absurd
17:05:47 <ais523> unfortunately, this means I act much the same way whether I miss a joke or not
17:05:49 <elliott> ais523: you could do it more interestingly :)
17:06:03 <oerjan> Vorpal: well that :*:*:* style _is_ essentially binary, adding :* multiplies a number by 2
17:06:10 <Vorpal> ais523, hm I actually do the same sometimes.
17:06:16 <elliott> tswett: you're lucky, this is the first time we've talked about esolangs in months
17:06:37 <elliott> 23:49:05 <GregorR> I love how devfs survived for like a year :-P
17:06:37 <elliott> 23:49:34 <calamari> yeah, what was wrong with it? worked for me
17:06:37 <elliott> 23:50:04 <GregorR> I really don't know.
17:06:37 <elliott> 23:50:08 <GregorR> Always worked great for me.
17:06:39 <elliott> It's not object XML enough.
17:06:44 <ais523> oerjan: that's what I was thinking; it has *2 and +1 operations
17:06:46 <elliott> (Is devfs in the newest kernels?)
17:06:52 <ais523> so it's a sort of generalised binary that allows arbitrary digits
17:06:57 <elliott> Removed in 2.6.13; glorious ...
17:06:57 <Vorpal> elliott, I don't think udev uses XML?
17:07:04 <elliott> Vorpal: No, but it might as well. :)
17:07:06 <oerjan> oh and adding 1 is putting : and * _around_ the number, isn't it?
17:07:28 <Vorpal> elliott, no xml would be worse
17:07:49 <elliott> Vorpal: I'm saying that it's so overcomplex that making it use XML would hardly change a thing.
17:07:49 <Sgeo> She wants us to install MySQL Query Browser
17:07:58 <elliott> Maybe I will see if devfs works on current kernels.
17:08:07 <elliott> Or just use static /dev like I was planning to.
17:08:15 <Vorpal> elliott, there is that other one there was some talk about. Don't remember details
17:08:32 <elliott> I wonder if devtmpfs can work without udev.
17:08:58 <Vorpal> elliott, anyway, why do you want to move stuff into a monolithic kernel?
17:09:15 <elliott> Vorpal: udev would be no better if it was in the kernel
17:09:41 <Vorpal> elliott, it is overcomplicated yes to some degree.
17:09:43 <elliott> yeah. when was the last time you were using devfs and thought --
17:09:48 <elliott> "I wish I could rename my hard drive device file."
17:10:04 <Vorpal> elliott, but then I doubt devfs allowed you to run a script that loaded joystick calibration values when it was connected
17:10:10 <elliott> I'm all for flexibility, it's just that when you add too much flexibility and try and modularise *Unix*, everything fucks up.
17:10:31 <elliott> See Hurd for the extreme of this.
17:11:05 <elliott> http://lwn.net/Articles/330985/ Yay -- devtmpfs looks like a proper devfs.
17:11:10 <elliott> "/dev will be fully populated and dynamic, and always reflect the current
17:11:10 <elliott> device state of the kernel."
17:11:22 <elliott> And apparently it makes init=/bin/sh work perfectly.
17:11:32 <Phantom_Hoover_> THINGS THAT ANNOY ME: people who equate "real" with "floating point".
17:12:09 <Vorpal> elliott, hm that is nice
17:13:13 <Vorpal> Phantom_Hoover_, things that annoy me: people who seem to take fortran or pascal seriously today.
17:13:42 <fizzie> "Removed /home (sorry!)" -- now that's interesting.
17:13:58 <Phantom_Hoover_> That of putting a stupid amount of store with floating point!
17:13:59 <fizzie> That's what it says on the latest blog post.
17:14:09 <fizzie> http://notch.tumblr.com/
17:14:25 <Vorpal> what? it removed /home?
17:14:36 <elliott> Back up .minecraft now, y'allz.
17:14:47 <elliott> $ cp -R .minecraft .minecraft_has_home_for_fucks_sake
17:14:58 -!- Sgeo has quit (Ping timeout: 255 seconds).
17:14:59 <elliott> Let's see if it's going to force me to upgrade!
17:15:00 <Vorpal> oh not /home on the fs
17:15:03 <fizzie> It might be a server-side change, though. I don't quite know how commands work.
17:15:27 <Vorpal> why would it remove /home
17:15:39 <Vorpal> elliott, also it says server upgrade is not mandatory
17:15:55 <fizzie> Yes, but also that there is a "rather bad memory leak bug in the server".
17:16:10 <elliott> ineiros: Instead of updating, why not try restarting the server every now and then?
17:16:13 <elliott> Oh, look, I died before quitting.
17:19:14 <elliott> Phantom_Hoover_: Recursively remove everything in /, duh.
17:19:19 <elliott> Or refuse to, if it's GNU rm.
17:19:27 <elliott> fizzie: http://imgur.com/gseyK.png http://imgur.com/ZZWCa.png http://imgur.com/xqBXT.png
17:19:30 <elliott> fizzie: Stop that, it's silley.
17:20:06 <fizzie> A daily stone-bath does wonders for your textures.
17:20:10 <elliott> Phantom_Hoover_: Um, bad things.
17:22:47 <elliott> fizzie: http://imgur.com/tzQ07.png http://imgur.com/1wqWD.png
17:23:18 <fizzie> Are my feet sticking out through the bridge?
17:25:07 <oklofok> http://esolangs.org/wiki/Banana_Scheme <<< erm, so, what's banana scheme?
17:25:45 <oklofok> "hey i invented a new language called X, the language called Y is defined as follows"
17:26:00 <elliott> oklofok: they are the Banana Schemes, collectively
17:26:04 <elliott> it's terminology from some irc log, see talk page
17:31:43 <elliott> Phantom_Hoover_: You breakin'?
17:32:08 <elliott> Phantom_Hoover_: Minecrafty.
17:32:46 <elliott> Phantom_Hoover_: Have you been putting those extra steps on the skyway?
17:32:57 <Phantom_Hoover_> I think I need a break. Also, this torrent is doing awful things to my connection.
17:33:11 <elliott> Phantom_Hoover_: Has he realised that you need three, not one?
17:35:12 <oerjan> http://politics.usnews.com/usnews/php/galleries/image.php/162/46/46.jpg
17:35:13 <fizzie> I did some of the widening too.
17:35:35 <elliott> oerjan: that was /almost/ funny
17:35:42 <fizzie> Anyway, it's already easier to walk like that, I don't really see why you'd need an actual barrier around.
17:38:01 <elliott> fizzie: Nobody ever told me you could ascend water without drowning!
17:38:31 <fizzie> There, now you've been told.
17:38:44 <Vorpal> <Phantom_Hoover_> And that was Vorpal.
17:39:07 <elliott> "My home is like a 20 minute walk from spawn!!!!!!!!" "Isn't that what portals help you with." yeah cuz portals work in MP
17:40:07 <elliott> Getting stuck in stone is now unfixable.
17:40:13 <elliott> You are stuck there. Forever. Until the server /tps you out.
17:40:32 <elliott> "You guys are missing the point. /home was not a feature implemented by Notch. Therefor he doesn't want it implemented. Like bureau.nic says, Perhaps it causes a huge bug in the code. Either way, notch has been coding and developing for a VERY long time. If he removes a feature that he knows people love (obvious he does or he wouldn't have said sorry) I'm sure he knows best and has a damn good reason for it. It IS his code game after all. So, why
17:40:32 <elliott> don't you sit down like a good alpha tester, and play the game like it's meant to be played."
17:40:34 <fizzie> Mostly you're supposed to suffocate, but I guess you still could easily get stuck.
17:40:56 <elliott> fizzie: He added code to suffocate rather than fixing the bug? Srsly?
17:41:06 <elliott> Also, you have to lose your inventory because of a bug? Niice.
17:41:28 <fizzie> No, I think you've always supposed to have been suffocated if you ended up inside a non-air block somehow.
17:41:33 <elliott> Also, what's all these comments talking about /spawn? Our server seems not to have that.
17:41:47 <elliott> "the key is to not get stuck in a mine. thats the point of the game."
17:42:57 <fizzie> I like the "play the game like it's meant to be played" comment. I suppose same sort of reasoning could be used to logicalize why any sort of unauthorized modding is ethically worng.
17:43:25 <elliott> "How could a player teleportation command cause bugs or issues? If it did, why weren't the others removed?"
17:43:28 <elliott> "idk ibm. Are you a programmer? Maybe learn a bit about that before you ask ignorant questions."
17:43:32 <elliott> As a programmer, #2 is full of shit.
17:43:57 <elliott> fizzie: why is everyone acting like /spawn does something, it doesn't for me
17:44:14 <fizzie> I don't really know, I've been wondering about spawn-point-setting too.
17:44:44 <fizzie> It seems that hMod already has a "enable-health" server option.
17:44:59 -!- nopseudoidea has joined.
17:45:32 <elliott> fizzie: Yes, but does disabling it work?
17:45:41 <fizzie> Well, according to some comments, yes.
17:46:10 <elliott> fizzie: Now tell me what the heck /spawn is meant to do >_<
17:47:16 <fizzie> http://wiki.hey0.net/index.php/Commands#.2Fspawn maybe?
17:47:20 <fizzie> (That's the hMod command list.)
17:47:24 <elliott> "We are not implementing anything crazy here like devfs did, including the later versions - there is no modprobe behind your back, no lookup hooks, no stupid new naming scheme, no new filesystem type to register."
17:47:44 <elliott> fizzie: Nice... so people are saying "just use /spawn!" when mods are meant to be BADHORRIBLE according to Notch.
17:47:53 <elliott> hMod does seem rather nice.
17:48:03 <elliott> Hey, it has support for "kits".
17:48:16 <elliott> Cool, you can set your own /home.
17:48:21 <elliott> ineiros: Totally requestin' hMod
17:49:02 <elliott> "Sievers outlines the differences between devtmpfs and Adam Richter's proposal from 2003. It mostly boils down to complexity; devtmpfs is a much simpler scheme, which really adds very little to the kernel. The implementation is around 300 lines of code, in comparison to roughly 3600 for devfs and 600 for an early version of Richter's mini-devfs."
17:49:36 <fizzie> So the patch notes say; we haven't actually tried out the new server version yet.
17:49:54 <elliott> And we're not going to! Well, not without hMod yet. On pain of ineiros being murdered.
17:50:16 <elliott> fizzie: Also, I love the idea that there can be a "proper" way to play a sandbox game.
17:51:30 <Vorpal> <elliott> Phantom_Hoover_: Has he realised that you need three, not one? <-- first it was fizzie who made it like that. Second 3 is wide enough to walk on without barrier with minimal risk of falling off
17:52:49 <fizzie> Phantom_Hoover_: http://notch.tumblr.com/
17:52:54 <fizzie> Phantom_Hoover_: Shouldn't you know that address already?
17:54:33 <Vorpal> <elliott> fizzie: Nice... so people are saying "just use /spawn!" when mods are meant to be BADHORRIBLE according to Notch.
17:54:41 <Vorpal> that only applies to client mods iirc
17:54:54 <elliott> I doubt he approves of deobfuscating, modifying, reobfuscating and then distributing the server code.
17:55:01 <Vorpal> from what I remember he said that since you don't need to pay for server anyway he minds much less for server mods
17:55:29 <elliott> "Forge seems to be working in SMP now, aside from the animation. Also, is health supposed to regenerate? If so, is there a setting?"
17:55:30 <elliott> "turning off monsters currently sets the server to peaceful mode. It will be an option."
17:55:32 <Vorpal> elliott, iirc he said that quite a bit of the code is shared between client and server (which seems sensible)
17:55:36 <elliott> How about making it an option before breaking it...
17:55:46 <elliott> Vorpal: Yes, but he's still too stupid to make singleplayer = connect to local server.
17:56:08 <elliott> fizzie: @notch Why did you remove the /home command? Bugs?
17:56:09 <elliott> @PHLAK it's not supposed to be there!
17:56:11 <elliott> tl;dr it's not coming back
17:56:19 <Vorpal> elliott, I wouldn't be able to play single player then since then it would be far above the memory available on here
17:56:23 <fizzie> Pretty impressive if two lines is tl;dr.
17:56:42 <elliott> Vorpal: That's just because the server code sucks.
17:56:59 <Vorpal> elliott, possibly. Go implement your own one?
17:57:09 <elliott> Vorpal: Others already are.
17:57:13 <elliott> Anyway I don't feel like figuring out the protocol.
17:57:21 <Vorpal> I mean, complaining here doesn't really give any result :P
17:57:49 <fizzie> It's not just figuring it out, it seems to be a moving target.
17:57:55 <elliott> LOL: http://www.pcgamer.com/2010/12/01/the-minecraft-experiment-day-7-when-you-are-engulfed-in-flames/
17:58:27 <fizzie> "/home was basically a way to get yourself out of impossible places, but now that you can kill yourself more easily, it's purpose has been fulfilled. Maybe he'll bring it back in another way."
17:58:52 <elliott> fizzie: Yeah, but he clearly won't since it seems it wasn't "meant" to be in there.
17:58:55 <elliott> Maybe he only noticed now.
17:59:12 <elliott> "I ignored the game entirely when it was a purely creative toy – yeah, yeah, people made amazing stuff. I’m amazed. I’ve been amazed so often now that I’m in a permanent state of maze, and it would take someone building a working time machine in Zuma Deluxe to un- and subsequently re-maze me. But when it added a health bar, suddenly I was interested. I can die? I love to die! I’m there."
17:59:41 <ais523> elliott: I think the idea is that it's basically doing an end-run around the "difficulty" of the game
17:59:51 <ais523> imagine adding a /home command to a typical roguelike
18:00:10 <Phantom_Hoover_> That guy's Galactic Civilisation II playthroughs have to be seen to be believed.
18:00:22 <elliott> ais523: yes, but we don't play it as survival
18:00:24 -!- atrapado has quit (Quit: Abandonando).
18:00:26 <elliott> apart from maybe Vorpal but nobody cares about him
18:00:37 <elliott> ais523: we're essentially playing it as creative, but with minecarts and circuits and all that fun modern stuff
18:00:46 <Vorpal> elliott, I don't really either indeed
18:00:51 <fizzie> elliott: You're playing it wrong!
18:00:56 <elliott> ais523: which worked, right up until notch decided that making it into survival like it's meant to be without even adding switches to turn it back into peaceful was a good idea
18:01:00 <elliott> ais523: which is, of course, rather insulting.
18:01:00 <Phantom_Hoover_> elliott, well, Creative mode will be implemented for the current version at /some/ point.
18:01:10 <elliott> ais523: (it would not be as bad if updates weren't automatic and involuntary, and you couldn't connect to old servers)
18:01:19 <elliott> Phantom_Hoover_: Yeah... at /some/ point. Just like adventure mode.
18:01:22 <elliott> /home isn't coming back though.
18:01:30 <elliott> And /home is damn useful if you're not playing to survive and just want to explore and create.
18:01:34 <Vorpal> elliott, we can survive without /home
18:01:42 <elliott> Vorpal: I can survive without Minecraft too.
18:02:07 <elliott> It has those kits, which are like automated multi-/gives.
18:02:18 <elliott> And you can set a personal home that you can teleport to.
18:02:32 <elliott> And ops can set arbitrary predefined teleportation points.
18:02:38 <elliott> And you can disable health. :p
18:03:07 <Vorpal> wait, connection refused?
18:03:18 <Vorpal> It said on the blog it was an optional server upgrade
18:03:47 <fizzie> Maybe it crashed because of all the memory leaks. :p
18:03:55 <elliott> It's been down for a lil' while.
18:04:20 <fizzie> Doesn't seem that hMod has yet been updated to 0.2.7, but it's probably just a matter of (short) time.
18:04:37 <Vorpal> the remember server bit fails
18:04:42 <elliott> And you can write plugins for it!
18:04:45 <Vorpal> since it doesn't remember port number
18:04:46 <fizzie> Yes, it drops the port number.
18:04:55 <Vorpal> fizzie, it doesn't in the config
18:04:56 <elliott> ineiros can finally have his /donateacreeperto.
18:05:04 <elliott> Put the server in, click cancel.
18:05:07 <elliott> It will remember the port forevermore.
18:05:16 <Vorpal> elliott, no it won't after restarting the client
18:05:23 <elliott> What, Minecraft just updated again I think. (Did it?)
18:05:29 <Vorpal> the config is: key:<value> which means key:value:port
18:05:37 <Vorpal> which I guess is why it fails to parse it properly
18:05:40 <Vorpal> it writes the right thing
18:06:06 <Vorpal> elliott, it is a sane config format. But with xml it wouldn't have failed :P
18:06:16 <elliott> "When dawn finally does break, I climb out of my awkward hole and look around. There’s something different about this mountain today. I don’t know if it’s the grass, the earth, the rocks, the walking pillars of flame – hm, were there walking pillars of flame yesterday?"
18:06:16 <Vorpal> ineiros, connection failure to your server
18:06:32 <Vorpal> ineiros, it simply times out
18:07:05 <elliott> ineiros: Hi. Please install hMod at your earliest convenience. Not only can we set a "home" point to teleport to as well as teleporting to spawn point -- the newest server update just removed /home -- but you can create "kits" like diamondtools that people can award themselves willynilly; you can also disable health.
18:07:09 <ineiros> Ah, apparently not a voluntary update, this.
18:07:10 <ineiros> 2010-12-01 19:52:39 [SEVERE] Unexpected exception
18:07:12 <elliott> ineiros: Oh, and you can set arbitrary named spawn points as an op.
18:07:17 <elliott> But the old server has a memory leak.
18:07:21 <elliott> I think you ran out of memory.
18:07:24 <elliott> ineiros: Anyway, see above :P
18:07:32 <Vorpal> elliott, what was the name of that blog again?
18:07:34 <elliott> hMod doesn't run on the newest server yet, but "only a matter of time".
18:07:43 <elliott> Vorpal: towardsdawns.blogspot.com; but I'm not quoting from that.
18:07:51 <Vorpal> elliott, oh where was it from then?
18:08:05 <elliott> Vorpal: http://www.pcgamer.com/2010/11/20/the-minecraft-experiment-day-1-chasing-waterfalls/
18:08:09 <elliott> Vorpal: In which he deletes his world whenever he dies.
18:08:36 <Phantom_Hoover_> Also worth reading: both his Galactic Civilisations II playthroughs.
18:09:12 <elliott> "In landscape gardening – bear with me here – this is called a ha-ha. A drop that acts like a wall in one direction, but is almost invisible to the fops and dandies sipping tea on their manicured lawn above. I mention this in part to explain why I thought it would be a good idea to stand on top of it, laughing at the creatures and punching them in the face."
18:09:34 <elliott> http://media.pcgamer.com/files/2010/11/Minecraft-Diary-Creeper-Tail-Closest.jpg AAAAAAAAAAAAAAH
18:10:28 <elliott> "I know this seems slow to you, but I’m pretty sure it took the human race longer than this to invent tools, and I’m probably going to discover fire before teatime."
18:10:44 <Phantom_Hoover_> http://www.computerandvideogames.com/article.php?id=161570&site=pcg
18:11:00 <Phantom_Hoover_> [[ I couldn't do it. The Spectres bow to no-one, plea for no quarter. Engraved on the seal at the base of a mile-high statue of their leader, Paul Davies Mutilator of Worldsblood, are the words "Bring it the fuck on." In Latin.]]
18:11:02 <fizzie> elliott: The "Painterly" texpack's default moon has three crater-looking creeper faces on it.
18:11:34 <elliott> Phantom_Hoover_: STOP IT STOP IT STOP IT STOP IT
18:12:08 <elliott> "At some point my staircase hits an earthy patch, and I can hear running water. An underground river! The best thing possible!"
18:12:47 <elliott> "It turns out I already have metal, about 16 blocks of it – it’s those lumps of Caramac I’ve been finding in the stone. I thought it might be, but I couldn’t figure out a way to turn them into something I can craft with."
18:12:52 <elliott> Interesting way to describe Caramac.
18:13:34 <elliott> "As soon as I do, I strike legs. A rich vein of purest legs. I wasn’t mining for legs, I am not trained in leg extraction, but legs I have found. In quantity."
18:14:36 <elliott> ineiros: HOLY SHIT IS THE SERVER ODWON;
18:14:42 <elliott> odwon, the purest state of being.
18:14:57 <elliott> "I don’t know if reaching through zombie legs to loot treasure chests is dangerous, but it feels dangerous."
18:15:03 <Vorpal> hm "I will start looking at server-side inventory soon in an attempt to reach beta soon."
18:15:05 <ineiros> 2010-12-01 20:14:29 [INFO] Connected players: fizzief, ehird
18:15:09 <Vorpal> where are the inventories now
18:15:19 <elliott> "Eggs! For that souffle I’ve always dreamed of!
18:15:20 <elliott> Gunpowder! For that gunpowder souffle I’ve always dreamed of!"
18:15:27 <Vorpal> ineiros, it fails to connect to it still.
18:15:42 <elliott> "When I stop dancing, I get back to digging my tunnel. And before long, I hit another type of block I’ve never seen before. I have a split second to identify it as ‘lava’ before it floods into my face."
18:15:53 <elliott> ineiros: SO WILL YOU USE HMOD :|
18:16:18 <Vorpal> elliott, hmod hasn't been upgraded yet (at least fizzie said that above)
18:16:30 <Vorpal> elliott, and I doubt he would want the mem leak
18:16:41 <Vorpal> indeed that is the question
18:17:40 <elliott> http://kerneltrap.org/node/4893
18:17:44 <elliott> "I make a single boot+root floppy disk in minix file format and lilo boot loader." --2005
18:18:25 <Vorpal> ineiros, restarting it sure is slow
18:18:38 <fizzie> Vorpal: You should've paid for a better SLA.
18:19:04 -!- poiuy_qwert has joined.
18:19:13 <Vorpal> fizzie, I blame notch not ineiros for this though
18:19:31 <elliott> ais523: So, have you ever written an actual minimax program?
18:20:04 <Vorpal> elliott, you mean as in the minmax algorithm for game solving?
18:20:17 <ineiros> Vorpal: Sorry, I'm rebooting the whole server, since I've avoided that for a few kernel updates already.
18:20:30 <elliott> ineiros: Is it running on Kitten?
18:20:52 <Vorpal> elliott: where would he have downloaded kitten?
18:20:54 <elliott> ineiros: So yay or nay for hMod? :p
18:22:21 <ineiros> elliott: I'll look into it when I have time. Definite maybe. :)
18:22:42 <elliott> ineiros: You do realise it's just a few .classes to replace? :p
18:22:53 <Vorpal> I like this flashlight. LED, 60 lumen, made of Aluminium. Feels very solid and durable. And still compact (just two AAA batteries)
18:23:17 <elliott> ineiros: Hmm, in fact, it's a batch file that downloads the server and does its magic. Hmmmmmmmmmmmm.
18:23:28 <Vorpal> elliott, not ported yet
18:23:28 <elliott> ineiros: Oh, there's an .sh too.
18:23:39 <elliott> ineiros: And it supports MYSQL! Note: Don't use MySQL.
18:24:00 <Vorpal> elliott, out of morbid curiosity: what does it use mysql for?
18:24:12 <elliott> Vorpal: You can make it use MySQL as opposed to flatfiles.
18:24:15 <elliott> Presumably that's quicker.
18:24:27 <Vorpal> elliott, for the game world?
18:24:55 <fizzie> I would guess just the hMod-specific warp-points and kits and such.
18:25:33 <Vorpal> Phantom_Hoover_, well duh
18:25:33 <elliott> This is not an uncommon position :P
18:25:43 <Vorpal> also oracle now. Even worse
18:25:59 <Vorpal> ineiros, rebooting atm?
18:26:04 <Vorpal> ineiros, or what point is it at
18:26:13 <elliott> IT'S DONE WHEN HE SAYS IT'S DONE
18:26:14 <Vorpal> elliott, just wondering how far it got
18:26:47 <fizzie> "IRC <--> Minecraft chat relay and advanced administration bot". The most sensible thing evar. (Perusing the hMod plugin list.)
18:27:55 <elliott> fizzie: I would like that, just to annoy people who are playing.
18:28:22 <Vorpal> ineiros, are you logged in atm? Or do we get ghosts of logged in people now?
18:28:37 <Vorpal> since dwarf in there just stands around doing nothing
18:30:37 <Phantom_Hoover_> [[All other secondary schools (23 in total) will be open for 4th-6th year pupils tomorrow]]
18:31:12 <elliott> IT'S FOR YOUR EDUMCACAITNO
18:33:19 <Phantom_Hoover_> No, it's because we have a poseur and idiot for a First Minister.
18:33:27 <elliott> fizzie: ineiros: Vorpal: I am on fire near the spawnpoint. I refuse to stop being on fire.
18:38:14 <elliott> Gregor: 14:53:14 * GregorR-W doesn't even know what tldr means :P
18:38:14 <elliott> 14:53:51 <kipple> heh, I had to look that one up too
18:38:14 <elliott> 14:54:36 <GregorR-W> Too Long Didn't Read?
18:40:32 <oklofok> okay destroying the world whenever you die sounds like an awesome idea
18:41:32 <oklofok> i mean say in a mmorpg, there could be an infinite amount of worlds, and when you die in one, you're forever blocked from reentering it
18:41:53 <oklofok> your friends could all kill themselves to join you in the new world, but
18:42:27 -!- poiuy_qwert has quit (Ping timeout: 272 seconds).
18:42:30 <elliott> fizzie: Vorpal: Phantom_Hoover_: http://imgur.com/IwLKB.png
18:43:08 -!- poiuy_qwert has joined.
18:43:13 -!- cal153 has quit (Ping timeout: 265 seconds).
18:43:46 <Vorpal> elliott, everyone seen that
18:43:50 <Vorpal> elliott, not as foggy though
18:43:56 <Vorpal> elliott, do you use tiny or short distance?
18:44:09 <elliott> Vorpal: I have more impressive ones:
18:44:17 <Vorpal> elliott, I seen such on far before
18:44:22 <Vorpal> elliott, it is nothing new
18:44:29 <elliott> Vorpal: That was looking *downwards on top of the sea*.
18:44:47 <Vorpal> elliott, I been on top of holes with seeing caverns below too
18:44:50 <elliott> http://imgur.com/L6ouQ.png http://imgur.com/WZqgE.png http://imgur.com/vkQoS.png These are really nice.
18:46:01 <Phantom_Hoover_> <Vorpal> elliott, I been on top of holes with seeing caverns below too
18:47:07 -!- Sgeo has joined.
18:49:44 <Phantom_Hoover_> Ah, Sgeo. Have any liaisons with Newspeak planned for tonight?
18:50:34 <Sgeo> I plan on actually trying to write something
18:51:06 <oklofok> think about it, really the only thing mankind is lacking in games is true fear of death
18:51:31 <Phantom_Hoover_> You'll probably abandon it in 3 days like you did with Smalltalk, Factor, Scala, ...
18:51:45 <Sgeo> Did I abandon Factor in 3 days?
18:51:49 <oklofok> yeah Sgeo you're a real fuckface and everyone hates you
18:52:12 * oklofok is just trying to be popular
18:53:08 <Sgeo> Now, let's see if I can avoid doing the same with real relationships
18:53:26 <oklofok> abandoning people in 3 days that is
18:53:37 <Sgeo> I don't... think I would
18:54:14 <oklofok> you haven't abandoned me yet
18:54:21 <elliott> oklofok: he never truly loved you
18:54:23 <oklofok> and we've been in an ircual relationship for years
18:54:30 <elliott> ineiros: http://wiki.nexua.org/Plugin:iStick wat
18:54:40 <elliott> oklofok: well it's well known that Sgeo is extremely gay of course
18:54:59 <oklofok> there's nothing gay about true online friendship
18:55:33 <elliott> oklofok: indeed; there's nothing faggy about a deep loving relationship with another man!
18:56:13 <Sgeo> <insert opposite of typical Sgeo joke here>
18:56:34 <oklofok> i just realized newborns would probably be rather homicidal
18:56:41 <elliott> `addquote <oklofok> i just realized newborns would probably be rather homicidal
18:56:57 <oklofok> context being the mmorph with death
18:56:57 <Vorpal> fizzie, ended up in stone thanks to the issue, had to /home
18:57:06 -!- poiuy_qwert has quit (Quit: This computer has gone to sleep).
18:57:09 <Vorpal> fizzie, and just after I saw spawn I timed out
18:57:56 <Vorpal> ineiros, I noticed a pattern in your connection issues: off peak hours there are fewer drops. If fewer are playing there are fewer drops
18:57:58 <fizzie> Ended up somewhere underground too.
18:58:01 <elliott> "@notch instead of being an ass why don't you downgrade the server so we can play on smp you jeesh we're not your fing beta testers"
18:58:03 <Sgeo> Minecraft is buggier than AW. [Yes, I know that that's not saying anything at all]
18:58:04 <elliott> "@warlordv1 actually, you're alpha testers.."
18:58:08 <Vorpal> ineiros, maybe it is related to traffic?
18:58:13 -!- poiuy_qwert has joined.
18:58:58 <Vorpal> fizzie, down for quite some time now
18:59:36 <Vorpal> Phantom_Hoover_, I never said anything like that
18:59:37 <elliott> Even more subtle when you point it out.
19:00:22 <Phantom_Hoover_> elliott, I remembered too late that Vorpal is about as subtle as a brick.
19:00:56 <elliott> ais523: aren't you going to stand up for the Micro?!
19:00:57 <Vorpal> Phantom_Hoover_, I don't believe in subtlety
19:01:54 <elliott> He's actually a BASIC program.
19:02:41 <tswett> I've seen photos of ais523.
19:02:50 <elliott> Yeah -- he's quite the pixel painter.
19:03:07 * tswett looks on his Wikipedia page.
19:03:18 <tswett> Oh, he no longer has a Wikipedia page.
19:03:30 <tswett> ais523 is now a redirect to "Wolfram's 2-state 3-symbol Turing machine".
19:03:36 -!- cheater99 has quit (Ping timeout: 245 seconds).
19:03:40 <Vorpal> Phantom_Hoover_, why is it ridiculous?
19:04:02 <Vorpal> Phantom_Hoover_, why is the beard insane?
19:04:22 -!- cal153 has joined.
19:04:46 <Vorpal> Phantom_Hoover_, I mean, who doesn't have a beard...
19:04:54 -!- oerjan has quit (Remote host closed the connection).
19:04:54 <Phantom_Hoover_> http://www.wolframscience.com/prizes/tm23/images/alex_smith_wolfram_turing.jpg
19:05:21 <Vorpal> Phantom_Hoover_, okay in the subset men
19:05:29 <Vorpal> Phantom_Hoover_, I don't see why it is strange that he has a beard
19:05:35 <Vorpal> or why it is ridiculous
19:05:40 <tswett> I'm going to refer to ais523 as Wolfram's 2-state 3-symbol Turing machine today.
19:05:46 <fizzie> I don't know, for some reason the photo reminds me of a soccer player.
19:05:47 -!- oerjan has joined.
19:05:55 <Vorpal> Phantom_Hoover_, could you please enlighten me why the beard is ridiculous?
19:06:32 <ineiros> elliott, Vorpal, fizzie, Phantom_Hoover_: Are you able to log in now?
19:06:36 <tswett> But I'm going to have to wait for wolfram's 2-state 3-symbol Turing machine to return before the fun begins.
19:06:55 <Phantom_Hoover_> Vorpal, look at it and use your goddamn in-built ridiculousness sensor.
19:06:59 <Vorpal> ineiros, just times out
19:07:12 <tswett> (Since "wolfram's 3-state 3-symbol Turing machine" is so long, every time I type "wolfram's 3state 3-sombol Turing machine" incorrectly, I'm going to replicate the typo from then on so that it will get shorter.)
19:07:18 <Vorpal> Phantom_Hoover_, I have a beard too. Not the same model
19:07:31 * tswett stops saying stuff that doesn't contribute to the conversation at all.
19:07:33 <Vorpal> still it looks perfectly normal
19:08:52 <Phantom_Hoover_> Vorpal, I don't really want to think about how ridiculous you look full stop.
19:09:15 <Vorpal> Phantom_Hoover_, there is NOTHING silly or ridiculous with how ais523 looks on that photo
19:09:59 <oerjan> tswett: i note you already introduced a typo
19:10:44 <Vorpal> Phantom_Hoover_, well it was you who criticised him in the first place so that seems a bit hypocritical to say.
19:10:52 <Vorpal> ineiros, still timeout
19:10:52 <tswett> I typo "wolfram's 3state 3-sombol Turing machine" every time I type "wolfram's 3state 30-sombol Turing machine". Except not one of those times.
19:11:03 <tswett> It got longer. I'm going to throw out typos that make it longer.
19:11:20 <elliott> tswett is hill-climbing towards the empty string.
19:12:26 <Vorpal> ineiros, downess is still a property exhibted by the server
19:13:46 -!- Sgeo has quit (Ping timeout: 255 seconds).
19:14:14 <elliott> tswett: By the way, Vorpal is AnMaster if you've been gone long enough not to know that.
19:14:29 <elliott> tswett: I'm not going to tell you who I am because it's bloody obvious.
19:14:59 <Vorpal> elliott, I don't know him from before
19:15:17 <oerjan> very forgettable guy, Vorpal
19:15:26 <Vorpal> elliott, yes I checked whois after
19:15:30 <Vorpal> elliott, before that I didn't
19:15:40 <Vorpal> oerjan, reading failure
19:16:00 <elliott> yeah unfortunately nobody can ever forget Vorpal.
19:16:18 <oerjan> he vorps into your mind
19:17:57 <Vorpal> ineiros, anyway: very much down
19:18:01 -!- cheater99 has joined.
19:24:11 -!- Sasha2_ has joined.
19:24:13 -!- nopseudoidea has quit (Quit: Quitte).
19:24:45 <tswett> Vorpal is AnMaster. You're Mr. Hird.
19:24:56 <tswett> Or do you have a different title now?
19:26:06 -!- Sasha2 has quit (Ping timeout: 245 seconds).
19:26:20 <tswett> wolmfra's 3state 3-synmob Turing machine. womfra's 3state 3-symbon Turing machine. womfra's 2state 30synbom Truing machine. womfra's 3state 20symobn Truing machine.
19:27:43 <fizzie> Wombat's 3-stack 20-sylph Turning machete.
19:29:00 <Vorpal> tswett, what about "W.'s 2,3-TM"? or such
19:32:12 <tswett> womfra's 3state 29synb Truying macihemn. womfra's 3state 29smb Truing maicne. womfr'as 3state 29mws Truign amicne. womfr'as 2state 29mws Truign amicne. womfra's 2state 29mwd Truign acmien.
19:32:28 <elliott> I need to make an esolang called "womfra's 2state 29mwd Truign acmien".
19:33:14 <tswett> womfra's state 29mswd Trign amcien. womfra's state 29mws Tirng amcien. womfr'as state 29mws Tirng amcien. womfra's state 29sms Tirng amcien. womra's state 29sms Ting amcien. wom's state 29sms Ting amcien. woms' state 219sms Ting amcien.
19:33:15 <oerjan> it will be a fncutonal language
19:34:18 <tswett> woms' state 2s9sms Ting amcien. wom's state 29sms Ting amcien. woms' state 29sms Ting amcien. woms' state 29sms Ting amcien. woms' state 29sms Ting amcien. woms's tate 29sms Ting amcien. woms' tate 29sms Ting amcien. woms' tate 29sms Ting amcien. woms' tate 29sms ting amcine.
19:34:42 <Vorpal> elliott, also I don't think it is hill climbing. It seems more like simulated annealing.
19:35:11 <elliott> Vorpal: T' = mutate(T); if (fit(T') > fit(T)) T = T'; repeat
19:35:21 <tswett> wome' state 29sms ting amcine. wome' tate 29sms ting amcine. wome' stst 29sms ting amcine. wome stst 29sms ting amcine. wome stst 29sms ting amcine. wome stt 29sms ting ancine. wome stt 29sms ting ancine. wome stt 29mss ting ancine. wome stt 29mdd ting ancine. wome stt 29mdd ting ancine.
19:35:27 <elliott> In this case, fit(s) = -length(s) and mutate(s) = try and write s carelessly.
19:35:27 <Vorpal> elliott, yes but does he check all the neighbours in each step?
19:36:23 <Vorpal> elliott, thus the closer fit to simulated annealing. With a slow temperature change
19:36:27 <oerjan> when in wome, do as the womans
19:36:34 <fizzie> I don't think hill-climbing in general means you'd check all neighbours.
19:36:47 <fizzie> "In simple hill climbing, the first closer node is chosen, whereas in steepest ascent hill climbing all successors are compared and the closest to the solution is chosen."
19:36:54 <fizzie> See, it's the first alternative there.
19:37:12 <tswett> wome stt 29mdd ting ancine. wome stt 29mdmd ting ancine. wome stt 29emje ting ancine. wome stt 29mdd ting ancine. wome xstt 29mdd ting ancine. wome stt 29mdd ting ancine. wome stt 29mdd ting ancine. wome stt 29mdd ting ancine. wome stff 29dndd ting ancine. wome stt 29mdd ting ancine.
19:37:17 <Vorpal> fizzie, oh I didn't know about "simple hill climbing" I only heard about the second one
19:37:19 <tswett> Yeah, it's pretty much stabilized now.
19:37:31 <elliott> wome stt 29 mdd thing machine.
19:37:36 <elliott> wome stt 2w9mdd ting acine.
19:37:43 <elliott> woms tt 2wm89dnd ting acin
19:37:48 <elliott> woms tt w2mws989dnd ting aicn
19:37:54 <Vorpal> wome tt 28 thingachine
19:37:57 <elliott> wms tti 28jndmndnd ting acin]#
19:38:03 <elliott> wms tti28m89dns tninag gnc
19:38:19 <elliott> tswett: Try "wms tti 2m89dnd ting acni"; that's the shortest I've got so far.
19:38:25 <elliott> tswett: Shortening that middle block will be hard.
19:38:48 -!- Phantom_Hoover_ has changed nick to Phantom_Hoover.
19:39:07 <tswett> wome stt 29mdd ting ancine. wome stt 29mdd tinvg acine. wome stt 29mdd ting acine. wome stt 29mdd ting acine. wome stt 2mdd ting acine. wome stt 29mdd ting acine. wome stt 2mdd ting acine. wome stt 2mdsd ting acine. wome stt 32mm ting acine. wome stt 3mm ting acine. wome stt 3mm ting acine. wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome stt 3mm ting aicne.
19:39:18 <Vorpal> tswett, still longer than the one I got to
19:39:33 <Vorpal> well lets implement a genetic algorithm on this
19:39:41 <fizzie> Conclusion => he types better.
19:39:55 <Vorpal> fizzie, who does "he" refer to here
19:40:00 <tswett> wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome stt 3mm ting aicne. wome st 3mm tine acine. wome sst 3mm tine acine. some st 3mm tine acine. some st 3mm tine acine. some tt 3mm tine acine. some tt 3mm tine acine. some ttt 3mm tine acine.
19:40:14 <Vorpal> somet tt3 m m tine inac
19:40:31 <fizzie> Some thingie oh-my titanic.
19:41:08 <tswett> some tt 3mm tine acine. some tt 2m3m tine acine. some tt 3mm tine acine. some tt 3mm tine acine. some tt 2mm tine acine. some ti 3mm tine acine. some ti 3mm tine caine. some ti 3mm tine caine. some ti 3mm tine caine. some ti 3mm tine caine. some ti 3mm tine caine. some ti 3mm tine ciane.
19:41:25 <Vorpal> elliott, I got shorter :P
19:41:36 <elliott> tswett: http://www.mastodon.biz/
19:41:55 <elliott> let's go back the other way
19:41:58 <elliott> let's try and typo it larger
19:42:01 <elliott> to get back to the original
19:42:14 <elliott> I'M SURE THIS PROCESS IS REVERSIBLE YOU GUYS
19:42:14 <Vorpal> elliott, THIS TIME! a genetic algorithim
19:42:18 <tswett> some ti 3mm tine caine. some ti 3mm tine caine. some ti emm tine caine. some ti emm tine caine. some ti emm tine caine. some ti emm tine caine. some ti emm tine caine.
19:42:35 <tswett> Now that it contains "ti emm", I'm going to replace those sounds with the letters they name.
19:43:05 <Vorpal> elliott, you can do multiple on same line like tswett does
19:43:11 <elliott> Vorpal: this is to keep my honset
19:43:39 <tswett> some TM tine caine. some TM tine caine. some tM tine caine. some tM tine caine. some tM tine caine. some tM tine caine. some tM tine caine. some Tm tine caine. some Tm tine caine. some Tm cint caine. some Tm cint caine. some Tm cint caine. some Tm cint caine. some Tm cint caine. some Tm cint caine. some Tm cint caint. some Tm cint taine. some Tm cint taine. some Tm cint taine. some Tm tcint taint. some Tm cint taint. some Tm cin
19:44:00 <Vorpal> ineiros, still very very down
19:44:04 <Vorpal> ineiros, just checked again
19:44:23 <fizzie> It's not just down, or very down; it is in fact very very down.
19:44:37 <fizzie> You'll need to start it thrice to get it all the way back up.
19:44:46 <tswett> some Tm cint taint. some tm cint that. some tm cint that. some tm cint that. some tm cint that. some tm cint that. some tm cint that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that. some t cnt that.
19:45:15 <tswett> Eh, I've gone far enough for my taste.
19:45:31 <tswett> ais523: your name is now "some Tm cint taint".
19:46:29 <Vorpal> fizzie, port is still open says nmap
19:47:00 <fizzie> Yes, it wouldn't stay in the "logging in" mode if it wasn't, I guess.
19:50:34 <elliott> "You may have thought the latest Postoffice beta release was a fairly trivial one. But no, that honor is reserved for discount, which has been pushed up to version 2.0.3 by simply updating the markdown(1) and markdown(3) manpages to correctly describe the thicket of MKD_flags available in the 2.x version of the published interface."
19:51:28 <elliott> Phantom_Hoover: Just David Parsons' insane software ramblings.
19:53:54 <elliott> I wonder whether postoffice is less pain than qmail.
19:55:26 <elliott> Vorpal: ls -A | xargs -d'\n' find \
19:55:33 <elliott> Vorpal: I'm bugging you today --
19:55:44 <elliott> Vorpal: How could one write this in a way that works? As it is, it gives the path last to find, which Does Not Work.
19:55:50 <elliott> (No, I can't use GNU xargs's -I.)
19:57:32 <Vorpal> elliott, waiting for rest of command :P
19:57:37 <Vorpal> elliott, there is a \ on the end
19:57:43 <elliott> Vorpal: The rest is irrelevant (just find expressions).
19:57:48 <elliott> The issue is that you can't give find paths after expressions.
19:57:57 <Vorpal> elliott, indeed you can't
19:58:06 <elliott> Vorpal: So my xargs doesn't work there.
19:58:08 <Vorpal> elliott, if you use ls -A I guess no space
19:58:30 <Vorpal> elliott, the paths obviously are sanely delimited?
19:58:41 <elliott> Vorpal: So the issue is, how can I do this?
19:58:42 <Vorpal> elliott, MAAAYBE mess with IFS
19:58:48 <Vorpal> elliott, but I don't know
19:58:55 <elliott> Vorpal: All I want to do is have find not put "./" before everything.
19:58:58 <Vorpal> elliott, use a saner language than shell script
19:59:19 <Vorpal> elliott, find ... | sed 's/^\.\///' ?
19:59:22 <elliott> find is the Right Thing here.
19:59:23 <fizzie> Can't you just postprocess the find output to sensib...
19:59:32 <elliott> \( -type l -printf '0000000000000000000000000000000000000000 %p\n' \) -o \
19:59:32 <elliott> \( -type d -printf '0000000000000000000000000000000000000000 %p/\n' \) -o \
19:59:32 <elliott> \( -type f -exec sha1sum '{}' \; \)
19:59:36 <elliott> "Maybe, but not with that script."
19:59:54 <Vorpal> elliott, I suggest that you post process it indeed
19:59:56 <elliott> Oh, wait; BusyBox find doesn't even have printf.
20:00:17 <Vorpal> elliott, why are you limiting yourself to busybox here?
20:00:56 <elliott> Vorpal: because Kitten uses BusyBox.
20:01:03 <elliott> BusyBox isn't exactly lacking in features.
20:01:09 <elliott> (Take a look at "busybox ls --help" sometime.)
20:01:30 <elliott> Vorpal: If you know of any coreutils replacement that's even vaguely as complete, I'd love to hear about it...
20:01:41 <Vorpal> elliott, whatever *BSD uses?
20:01:50 <Vorpal> elliott, well freebsd mostly
20:01:59 <elliott> Vorpal: *BSD have coreutils in their source tree. They are not portable.
20:02:01 <Vorpal> openbsd is quite a lot worse
20:02:12 <Vorpal> elliott, oh? what sort of BSD specific functions?
20:02:16 <elliott> They do not compile on anything other than BSD; perhaps not even anything other than *that* BSD.
20:02:20 <elliott> Vorpal: Mostly headers, actually.
20:02:30 <elliott> Vorpal: But the Makefiles also only work with BSD make and require the BSD make includes (bsd.prog.mk and the like).
20:02:34 <elliott> So I pretty much just gave up at that point.
20:03:01 <elliott> I tried Minix but their Makefile system is similarly tangly.
20:03:10 <ineiros> I think I have to apply the update.
20:03:47 <ineiros> The server seems to keep doing those unexpected exceptions.
20:03:59 <ineiros> Apparently it doesn't work well with the updated client.
20:04:45 <Vorpal> I wonder what sending it specially crafted (no pun intended) packets would do?
20:04:57 <Vorpal> presumably easy to crash it
20:05:42 <elliott> Hmph. Maybe I'll store the type of the file before the hash.
20:05:52 <elliott> Writing a useful manifest is irritating...
20:06:21 -!- mtve has quit (Ping timeout: 265 seconds).
20:06:29 <Vorpal> elliott, why not sha2?
20:06:50 <elliott> Vorpal: Because the sole thing this is used for is for checking whether the user has changed a given file.
20:06:55 <elliott> Vorpal: I could even get away with CRC32.
20:07:01 <Vorpal> elliott, ah not for download integrity then
20:07:11 <elliott> Vorpal: Yeah, rsync will handle that.
20:07:38 <elliott> Vorpal: Basically, when you uninstall, it'll remove all the files whose hashes match, and print out the names of all the files that don't.
20:07:43 <elliott> So you can remove configuration files manually if you want.
20:07:49 <Vorpal> elliott, ah interesting
20:07:59 <elliott> This means I don't have to keep track of what's a package file and what's a user file. :p
20:08:26 <Vorpal> elliott, btw I thought about the owner thing. Pretty much every daemon that installs files into /var and that can get away with not running as root will have files that shouldn't be owned by root
20:09:09 <elliott> Vorpal: Well, I could write something that goes and chmods every file owned by the user building the package.
20:09:15 <elliott> But it seems easier just to use a postinst script.
20:09:44 <Vorpal> elliott, also stuff like nethack that is sgid games
20:09:56 <elliott> Vorpal: Some distros run NetHack as root. :-)
20:09:59 <elliott> (Of course, I'm not that stupid.)
20:10:09 <Vorpal> elliott, that way there can only be one player?
20:10:25 <Vorpal> elliott, uh. if you run as root you are root?
20:10:31 <Vorpal> the player you are I meant
20:10:42 <elliott> Vorpal: Um, I think it checks the different UID.
20:10:44 <elliott> The one that doesn't change.
20:10:51 <elliott> (But there *was* a security flaw in NetHack that left such distros vulnerable.)
20:11:36 <elliott> │ If you want the uClibc math library to contain the full set C99 │
20:11:36 <elliott> │ math library features, then answer Y. If you leave this set to │
20:11:36 <elliott> │ N the math library will contain only the math functions that were │
20:11:36 <elliott> │ listed as part of the traditional POSIX/IEEE 1003.1b-1993 standard. │
20:11:36 <elliott> │ Leaving this option set to N will save around 35k on an x86 system. │
20:11:41 <elliott> I love how specific some of these options are.
20:12:00 <elliott> │ j0, j1, jn - Bessel functions of the first kind │
20:12:00 <elliott> │ y0, y1, yn - Bessel functions of the second kind │
20:12:03 <elliott> I... don't think I need those.
20:13:11 <elliott> │ The kernel source you use to compile with should be the same │
20:13:11 <elliott> │ as the Linux kernel you run your apps on.
20:13:22 <elliott> That's... really irritating.
20:14:27 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
20:15:23 <elliott> In fact, I'm not even sure how to solve that.
20:15:28 <elliott> Upgrade my kernel manually on Debian?
20:16:18 -!- poiuy_qwert has joined.
20:17:03 <Phantom_Hoover> elliott, what if you need to compute Bessel functions!
20:17:19 <elliott> PIC doesn't work with static libraries, right? :p
20:17:35 <Vorpal> elliott, I'd assume PIE would
20:17:50 <Vorpal> elliott, doubtful on 32-bit x86 though
20:18:05 <elliott> Vorpal: i.e. should I bother telling uClibc to use -fPIC if I'm doing all static linking
20:18:11 <elliott> fizzie: (oblig. http://www.weebls-stuff.com/wab/pie/)
20:18:16 <elliott> fizzie: Saved you the effort of Vorpal's questioning!
20:18:25 <Vorpal> fizzie, know of a 3x3 shaft with spiral stair next to your house?
20:18:26 <elliott> <Vorpal> But I don't HAVE Flash!
20:18:33 <elliott> Vorpal: Yes, tell it to use -fPIC? Really?
20:18:33 <Vorpal> fizzie, does all the way down and all the way up
20:18:50 <Vorpal> elliott, sure you can't do ASLR without it
20:19:16 <elliott> Vorpal: Which doesn't work with static linking...
20:19:43 <Vorpal> elliott, sure? I thought it did
20:19:51 <Vorpal> elliott, since it could move around heap and stack and such
20:19:58 <Vorpal> elliott, and also where the binary is loaded
20:22:26 <elliott> Security measures like load address randomization cannot be used. With statically linked applications, only the stack and heap address can be randomized. All text has a fixed address in all invocations. With dynamically linked applications, the kernel has the ability to load all DSOs at arbitrary addresses, independent from each other. In case the application is built as a position independent executable (PIE) even this code can be loaded at rand
20:22:26 <elliott> om addresses. Fixed addresses (or even only fixed offsets) are the dreams of attackers. And no, it is not possible in general to generate PIEs with static linking. On IA-32 it is possible to use code compiled without -fpic and -fpie in PIEs (although with a cost) but this is not true for other architectures, including x86-64."
20:22:52 <elliott> (ASLR is really not that useful, though; it's a bit of a niche exploit to cover for.)
20:25:09 <elliott> Phantom_Hoover: They are rather good.
20:26:17 <elliott> Phantom_Hoover: The ending is quite happy, really... apart from that one thing.
20:27:36 -!- Quadrescence has quit (Quit: omghaahhahaohwow).
20:27:40 <elliott> Phantom_Hoover: I have lost my PDF copy of it!
20:28:29 <elliott> Phantom_Hoover: Have you read the epilogue? http://qntm.org/free
20:28:40 <elliott> I included it in the PDF after a few blank pages.
20:32:01 <elliott> Phantom_Hoover: I think it's sort of meant to take place outside of any actual timeline.
20:33:49 <elliott> │ bcmp, bcopy, bzero, index, rindex, ftime, │
20:33:49 <elliott> │ bsd_signal, (ecvt), (fcvt), gcvt, (getcontext), │
20:33:49 <elliott> │ (getwd), (makecontext), │
20:33:49 <elliott> │ mktemp, (pthread_attr_getstackaddr), (pthread_attr_setstackaddr), │
20:33:49 <elliott> │ scalb, (setcontext), (swapcontext), ualarm, usleep, │
20:33:54 <elliott> Do I need these, I wonder.
20:34:20 <elliott> Ha! There's an option not to include gets solely because it's obsolete.
20:34:31 -!- ais523 has quit (Remote host closed the connection).
20:34:35 <pikhq> elliott: Ulrich Drepper there is blatantly lying.
20:35:20 <pikhq> elliott: Position independent executables are entirely feasible with static linking. However, you get fixed offsets in the binaries.
20:35:21 <Vorpal> ineiros, are you torrenting or something? it is very laggy
20:35:24 <Vorpal> ineiros, but not timing out
20:35:33 <elliott> ineiros: You must feel so scrutinised.
20:35:37 <elliott> Torrent away; Vorpal needs a break.
20:35:48 <elliott> pikhq: Mm. I think I won't bother though.
20:35:49 <pikhq> elliott: Making it much less *useful* than with dynamic linking, where each library can be loaded at a random address.
20:35:57 <elliott> pikhq: Address space randomisation is... well... fairly pointless.
20:37:09 <elliott> │ Answer Y to enable repeated reading of the '/etc/TZ' file even after │
20:37:09 <elliott> │ a valid value has been read. This incurs the overhead of an │
20:37:09 <elliott> │ open/read/close for each tzset() call (explicit or implied). However, │
20:37:09 <elliott> │ setting this will allow applications to update their timezone │
20:37:09 <elliott> │ information if the contents of the file change. │
20:37:12 <pikhq> It makes a certain class of attacks somewhat harder, but in that class of attacks you're already pretty well fucked.
20:37:13 <elliott> Eh, it's not much overhead.
20:37:22 <elliott> /etc/TZ is not a nice name for the file though.
20:38:59 <elliott> pikhq: Is it evil to have a file named /etc/timezone with completely different syntax to glibc's?
20:41:59 <Phantom_Hoover> elliott, FWIW, I had a little niggle at the back of my head when the energy virus was introduced by analogy with Life.
20:44:38 <elliott> Phantom_Hoover: What was it this time. :p
20:45:32 <Vorpal> fizzie, looked like an optimised design
20:45:38 <Vorpal> fizzie, what is the obsidian risk in it?
20:45:45 <Vorpal> fizzie, and how much goes back into the lava?
20:45:51 <Phantom_Hoover> elliott, there's no such pattern known in Life, and I'd put large quantities of money on it not existing.
20:45:55 <Vorpal> elliott, that z is so american
20:46:27 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
20:46:43 <oklofok> what kind of pattern doesn't exist in life?
20:46:44 <elliott> Phantom_Hoover: What, something that can eat everything?
20:47:05 <elliott> Phantom_Hoover: I'm not so sure...
20:47:22 <elliott> Phantom_Hoover: Well, "almost" anything anyway.
20:47:26 <Phantom_Hoover> elliott, show me a spaceship that can survive a collision with a blinker.
20:47:29 <oklofok> i have to think about this
20:47:40 <elliott> Phantom_Hoover: who says it's necessarily a spaceship
20:48:02 <elliott> Phantom_Hoover: have some eaty things on the end, and have a special kind of spacefiller in the middle that, when it collides with the eaty things from *behind*, "pushes" them further forward
20:48:14 <quintopia> | <-- how does one type this character?
20:48:18 <elliott> although I think it's unlikely
20:48:18 -!- poiuy_qwert has joined.
20:48:27 <elliott> quintopia: altgr-` on uk keyboards
20:48:51 <Phantom_Hoover> Energy virus: self-propagating pattern into *everything*.
20:48:53 <oklofok> so basically, we're asking, for a given CA, if there exists a finite pattern such that given any configuration with that pattern in the middle, the orbit converges to 0
20:48:53 <quintopia> i didn't output the character i meant to because apparently it is unicode
20:49:05 <oklofok> orbit of the configuration i mean
20:49:24 -!- augur has quit (Remote host closed the connection).
20:49:25 <Phantom_Hoover> oklofok, such patterns exist in several CAs, and some unconventional formulations of Life.
20:49:26 <oklofok> in english, bigger and bigger balls are filled with 0
20:49:36 <quintopia> elliott: the vertical bar in you messages at 15:33 and 15:36. what unicode value is it?
20:49:51 <elliott> quintopia: i don't know, ask python
20:50:03 <oklofok> Phantom_Hoover: well yes, obviously you can make a CA with such a pattern
20:50:16 <elliott> on windows it's a broken |
20:50:17 <Phantom_Hoover> oklofok, also, for this purpose I'll go for chaos for some value of "chaos".
20:50:17 <oklofok> you can even make it nilpotent, in which case every pattern has that property
20:50:49 -!- kar8nga has joined.
20:51:14 <pikhq> elliott: If you ever want to run a glibc program, absolutely TERRIBLE.
20:51:23 <pikhq> elliott: Otherwise, not at all.
20:51:40 <elliott> pikhq: doesn't glibc look at $TZ before /etc/timezone
20:51:53 <Phantom_Hoover> And the blank universe isn't really distinguished from the infinitely striped universe.
20:51:54 <oklofok> Phantom_Hoover: i don't get your chaos comment
20:51:59 <elliott> pikhq: i do want to be able to run quake ii, which is circa-1997 gcc/glibc static linking :)
20:52:07 <pikhq> elliott: You're asking about how brain-damaged Glibc is.
20:52:12 <pikhq> elliott: The answer is "very:".
20:52:16 <pikhq> Any further questions?
20:52:29 <elliott> pikhq: No! Okay, name it then. Basically, writing to this file is the same as setting $TZ to the contents.
20:52:34 <elliott> pikhq: /etc/TZ is the default but that is *ugly*.
20:52:53 <Phantom_Hoover> oklofok, it needn't be 0 within the bubble, just that the bubble overwrites all preëxisting structures and is almost impossible to stop.
20:53:06 <oklofok> and yes obviously we can define identicles w.r.t. every configuration (i liked your typo so much i'd like to name this concept that, even though it probably has a name already, and identicle doesn't make any sense)
20:53:24 <pikhq> elliott: /etc/org.sun.xml.config.time.zone.xml
20:53:38 <oklofok> well you just typoed identical "identicle"
20:54:00 <oklofok> but so erm what does almost impossible mean
20:54:23 <oklofok> i guess measure theoretical or topological duh
20:54:48 <Phantom_Hoover> oklofok, it's hard to define formally, but that almost all possible patterns will be destroyed would do.
20:55:36 <oklofok> we could just take the topological denseness and measure theoretical full measureness, and hope they make sense in this context too
20:55:41 <Phantom_Hoover> elliott, incidentally, engineering isn't a very good solution to durability in Life.
20:55:47 <pikhq> Phantom_Hoover: Perhaps you should commit sehų'ku instead.
20:56:02 <pikhq> Erm, sorry. se'hųku
20:56:10 <Phantom_Hoover> pikhq, MY GUTS ARE SPILLING OUT YOUR ROMANISATION IS NOT IMPORTANT
20:56:22 <pikhq> Phantom_Hoover: Your romanisation was spelled wrong anyways.
20:56:27 <pikhq> Phantom_Hoover: "Seppuku".
20:56:51 <elliott> fizzie: This is just wrong: http://imgur.com/yMEWy.png
20:56:59 <pikhq> "Sepukku" comes out as something like "back poo phrase"
20:57:12 <oerjan> Phantom_Hoover: hey if pikhq hadn't, i would
20:58:25 <fizzie> Speaking of the sparkles, http://zem.fi/~fis/current-hird.png
20:58:28 <oklofok> oerjan: is there an identicle for gol?
20:59:39 -!- augur has joined.
20:59:59 <oerjan> well what if you have two of them colliding?
21:00:30 <oklofok> what's problematic about that
21:00:58 <pikhq> http://cablegate.wikileaks.org/cable/2008/01/08OTTAWA136.html
21:01:00 <oklofok> you have two growing balls of zeroes, those would just both have to keep growing
21:01:28 <pikhq> The US... Actively monitors Canadian TV for unAmerican things on prime time‽
21:01:38 <oklofok> Phantom_Hoover: a finite pattern P such that, given any configuration C with P in the middle, T^n C --> 0 where T is the CA rule
21:01:45 <oklofok> and 0 is the all zero configuration
21:01:56 <oklofok> and limits are taken w.r.t. product topology
21:02:21 <elliott> pikhq: RR RUEHAG RUEHAST RUEHDA RUEHDF RUEHFL RUEHGA RUEHHA RUEHIK RUEHKW
21:02:21 <elliott> RUEHLA RUEHLN RUEHLZ RUEHPOD RUEHQU RUEHROV RUEHSR RUEHVC RUEHVK
21:02:21 <elliott> DE RUEHOT #0136/01 0252315
21:02:24 <Phantom_Hoover> oklofok, there does have to be a wavefront made of live cells at the edge of the identicle.
21:02:43 <Phantom_Hoover> Moreover, there are rules about propagation into vacuum.
21:02:51 <elliott> pikhq: [["THE BORDER" -CANADA'S ANSWER TO 24, W/O THAT SUTHERLAND GUY]] LOL
21:03:04 <pikhq> "We need to do everything we can to make it more difficult for Canadians to fall into the trap of seeing all U.S. policies as the result of nefarious faceless U.S. bureaucrats anxious to squeeze their northern neighbor."
21:03:14 <pikhq> Hmm. How's about you stop doing stupid shit. I think that'd do it.
21:03:27 <oklofok> surely there is a wavefront, yes
21:03:57 <oklofok> "rules about propagation into vacuum, bounding boxes and things"?
21:04:09 <elliott> [[GIVE US YOUR WATER; OH WHAT THE HECK WE'LL TAKE YOUR COUNTRY TOO]]
21:04:11 <elliott> pikhq: these headlines are amazing
21:04:36 <pikhq> And this was *never once classified*.
21:04:38 <Phantom_Hoover> oklofok, OK, and the absolute, unbreakable rule about propagation into the vacuum is that it must stay within a diagonal box expanding at less than c/2.
21:04:43 <oklofok> there should be a wavefront?
21:04:48 <oerjan> limits in product topology is a pretty lenient requirement, the zero region can expand as slowly as it wants
21:05:11 <oklofok> oerjan: imo that's the correct definition
21:05:18 <elliott> fizzie: Now watch as I fail to disappear.
21:05:25 <Phantom_Hoover> That's restricted further to less than c/4, but I'm not sure.
21:05:26 <Phantom_Hoover> oklofok, if there's one in Life it *must* have a wavefront.
21:05:34 <oerjan> well i'm just pointing out that it doesn't conflict with slow propagation
21:05:59 <oklofok> c/2 is rather arbitrary, if you halve a rule, make it slower that is, it can suddenly have no identicle, even though it's essentially the same rule
21:06:41 <oklofok> Phantom_Hoover: what must stay within a diagonal box exp...?
21:07:10 <Phantom_Hoover> Any configuration of live cells cannot expand beyond the box.
21:07:19 <oklofok> "<Phantom_Hoover> oklofok, if there's one in Life it *must* have a wavefront." <<< yes and i'm sure we can prove the existence of a kind of wavefront in general, in non-nilpotent rules
21:07:53 <Phantom_Hoover> Also, movement at >= c/2 diagonally basically writes "THIS RULE EXPLODES" on the starting grid.
21:07:56 <oklofok> what is propagation into vacuum even?
21:09:05 <oklofok> Phantom_Hoover: i find that a less beautiful definition than mine though, i would prefer say calling certain identicles explosive, or even better, defining their explosion speeds
21:09:27 <Phantom_Hoover> oklofok, turning on of live cells into an area filled with state 9.
21:10:39 -!- Sgeo has joined.
21:12:01 <Phantom_Hoover> oklofok, incidentally, is your T^n thing indicating that T is a transition function operating n times?
21:12:20 <oklofok> so erm, can you reiterate, what are the things you'd like in the definition?
21:12:57 * Sgeo wants a Truing machine
21:13:09 <Sgeo> Makes false statements true by changing reality
21:13:33 <Sgeo> ^^would have been funnier without those last three words, I think
21:13:58 <Phantom_Hoover> I'm just poorly expressing why I think no such configuration exists.
21:13:59 <oklofok> i think it's nice and pure
21:14:30 <oklofok> oh you were? i'm not really interested in gol in particular, i just thought there might be some neat properties you could prove for identicles in general
21:16:06 <Phantom_Hoover> Hmm. "In general" is pretty boring in CAs, since there's very little common ground other than the discreteness.
21:16:14 <oklofok> basically what an identicle is a neighborhood such that all points of it converge to zero
21:16:31 <oklofok> so in fact there is a very natural way to express this concept for a general dynamical system
21:17:20 <oklofok> Phantom_Hoover: in general is not boring in ca's, and when it happens to be, you assume a property from the ca's, you don't take one particular one, that's ugly
21:18:04 <oklofok> that's like talking about the properties of the number 8, who gives a shit
21:18:24 <oklofok> maybe not exactly, but close :D
21:20:00 <oklofok> so basically, given a point in a CA, an identicle is an open set such that blah blah, this concept is probably very much connected with attracting sets or something
21:20:31 <oklofok> Phantom_Hoover: i usually hope it ends in a theorem/-y
21:21:14 <oklofok> where did i know maths you don't
21:21:25 <oklofok> if you're talking about attracting sets, i basically just said i don't know anything about those :D
21:21:39 <oklofok> actually i didn't say the part where i know nothing about them
21:21:44 <oklofok> you don't know what open sets are?
21:22:06 <oklofok> in this case or in no case?
21:22:15 <Phantom_Hoover> Depends on whether it's in terms of intervals or topology and stuff.
21:22:31 <oklofok> it's the same thing, except w.r.t. the metric on the space of all configurations
21:22:58 <oklofok> and not intervals but balls
21:23:09 <oklofok> an open set is a set U such that for each x \in U, there is an open ball B(x, r) \subset U
21:23:31 <Deewiant> Intervals are one-dimensional balls
21:24:25 <Phantom_Hoover> Incidentally, if you're considering lifeoid CAs there are some laws you can work out in terms of speed limits in some configurations.
21:24:31 <oklofok> and the metric says roughly that if you take the ball B(x, r), given a configuration x, you actually take the pattern P in x in the ball around origin, of size 2^(-r), and you take all configurations y that have that pattern in the middle
21:24:56 <oklofok> so basically, two points are close if they are the same around the origin
21:25:06 <oklofok> pretty much anyway you define that formally gives you the same topology
21:25:24 <Phantom_Hoover> For instance, for anything to move diagonally at c/2 or greater through an area with no live cells, the rule basically has to be explosive.
21:25:41 <oklofok> oh there's a definition for explosiveness?
21:26:26 <oklofok> because i thought that was your definition of explosiveness
21:26:52 <oerjan> of course gol is particularly interesting for this question precisely because that CA apparently makes it impossible to create a structure that can survive contact with arbitrary chaos
21:29:40 <oklofok> "<Phantom_Hoover> Depends on whether it's in terms of intervals or topology and stuff." <<< i don't think open sets refer to anything but the topology kind
21:29:40 <oklofok> but i contain an open ball around all of my points for corrections ofc
21:29:42 <elliott> Phantom_Hoover: Care to hear about my INSANE MINECRAFT FORT PLANS?
21:29:42 <oklofok> erm right, i actually slightly misparsed you
21:29:42 <oklofok> or misunderstood more like
21:29:43 <oklofok> elliott: let's talk about identicles instead
21:29:43 <Phantom_Hoover> oklofok, explosive rules are those in which patterns which expand without bound are very common.
21:30:08 <oklofok> oerjan: i didn't know it was known for it's ability to that
21:30:57 <oerjan> well i mean, i've never heard of anyone inventing a gol structure that can survive such contact
21:33:09 <oklofok> i would certainly like to know what given value
21:33:11 <oklofok> good to know, i thought he just thought hey this is neat.
21:33:32 <Phantom_Hoover> I think he tried multiple Lifelike CAs and went for the one he found that supported long-term dynamic behaviour without everything filling the board with chaos.
21:34:27 <Phantom_Hoover> There is a most resilient known structure, and it's not very resilient.
21:35:03 <elliott> Phantom_Hoover: hm but can you build a spacefiller that works by replicating itself?
21:35:04 <elliott> as in, sure, tons of them may die
21:35:13 -!- Sasha2_ has quit (*.net *.split).
21:35:14 -!- Leonidas has quit (*.net *.split).
21:35:14 -!- nooga has quit (*.net *.split).
21:35:14 -!- sshc has quit (*.net *.split).
21:35:14 <elliott> but they're replicating fast enough that eventually, they will end up destroying the debris
21:35:47 <Phantom_Hoover> elliott, all complex structures in life are *really really fragile*.
21:36:06 <Phantom_Hoover> As in, "whoops I hit it with a glider and it's dead" fragile.
21:36:21 -!- sshc has joined.
21:36:21 <elliott> Phantom_Hoover: So make the structure simple.
21:36:33 <elliott> Phantom_Hoover: Rather than trying to fill space, just make a simple thing that makes little baby space-fillers.
21:36:37 <Phantom_Hoover> The Gemini replicator would completely fail if a single glider in its instruction tape was changed.
21:36:41 <elliott> Basically, instead of trying to grow to fill the space...
21:36:46 <elliott> Just try and fill the universe with paperclips.
21:36:46 <Phantom_Hoover> elliott, a spacefilling replicator with simple structure?
21:37:01 <elliott> Phantom_Hoover: Well, OK, I would try it in a Life-like rule with simple replicators first.
21:37:12 <elliott> But the basic idea is: Keep replicating further away from yourself, constantly.
21:37:19 <Phantom_Hoover> Spacefillers leave no room; they expand outwards at maximum speed in all directions.
21:37:29 <elliott> Phantom_Hoover: I didn't mean a regular kind of spacefiller.
21:37:39 <elliott> I just meant something that eventually gets rid of everything else on the plane.
21:37:55 -!- Sasha2_ has joined.
21:37:55 -!- Leonidas has joined.
21:37:55 -!- nooga has joined.
21:38:00 <elliott> Phantom_Hoover: Sure. Except that instead of one breeder, each unit is a breeder in itself.
21:38:17 <elliott> If one of them gets sucked up by some debris, no problem; another will end up being bred to fill its space.
21:39:03 <Phantom_Hoover> And all of them will die when they attempt to replicate over each other.
21:41:14 <elliott> Phantom_Hoover: Well, the trick is to make two colliding replicators result in one replicator. :p
21:41:33 <Vorpal> elliott, I'm next to a deeep pit
21:41:46 <Vorpal> elliott, from 20 below surface to lava lake
21:42:06 <elliott> Vorpal: Phantom_Hoover knows a deeper pit.
21:42:06 <Vorpal> elliott, 1x1 wide almost all the way
21:42:15 <Vorpal> elliott, you mean the one you fell in?
21:42:35 <elliott> Vorpal: I mean a *really* deep pit.
21:42:39 <Vorpal> elliott, argh an even deeper one over here
21:42:54 <elliott> Vorpal: Not as deep as Phantom_Hoover's.
21:43:02 <Phantom_Hoover> Vorpal, I'm assuming he means the hole in the bedrock I found under the Mt. Hoover tunnel.
21:43:06 <Vorpal> elliott, this one goes down to the minecart tracks
21:43:10 <Vorpal> Phantom_Hoover, ah right
21:43:15 <Vorpal> elliott, I saw him mention that
21:43:20 <Vorpal> I didn't consider it a pit
21:43:22 <oklofok> so without any assumptions on the rule, is it true that, for each patterns P and block Q, where P is an identicle, there exists a constant k such that for all configurations C with P in the middle, Q is filled with 0 from T^k onwards? i don't think the function from configurations to "first everzeroings" is continuous, but it still seems like this should be true
21:43:40 <oklofok> *is necessarily continuous
21:43:43 <Vorpal> Phantom_Hoover, this pit ends up above the cobble-water barrier next to the minecart tracks
21:44:29 <elliott> pikhq: How to set a timezone in kitten: "echo CST6CDT >/etc/tz".
21:44:58 <elliott> pikhq: Although you should probably also add "export TZ=CST6DT" to /etc/profile to avoid file accesses.
21:45:07 <elliott> pikhq: In fact, I'll probably add "export TZ=$(cat /etc/tz)" to /etc/profile.
21:45:09 <Vorpal> heh it goes up to surface too
21:45:18 -!- Mathnerd314 has joined.
21:45:39 <oklofok> hmm i guess there's still a direct compactness argument
21:49:05 <elliott> Vorpal: BTW, cfunge will get bug reports if it doesn't run on Kitten.
21:49:15 -!- oerjan has quit (Quit: Good night).
21:49:31 <elliott> Vorpal: So be prepared to support Linux/x86-64/pcc/uClibc :)
21:49:50 <Vorpal> elliott, I depend on POSIX 2001
21:49:58 <Vorpal> if you don't do that then fuck you
21:50:07 <elliott> Vorpal: I do include XSI functions.
21:50:13 <elliott> I don't know how 2001y uClibc is but it should be good.
21:50:22 <elliott> Vorpal: Math functions are slightly lacking but I doubt you use the full extent of POSIX math.
21:50:48 <Vorpal> elliott, I use sinl and such
21:50:54 <elliott> Vorpal: Those are of course included.
21:51:04 <elliott> Vorpal: I somehow doubt cfunge will work with pcc, though.
21:51:14 <Vorpal> elliott, also mmap extension
21:51:22 -!- poiuy_qwert has left (?).
21:51:31 <Vorpal> elliott, it works with gcc, icc and clang. tcc except that tcc had somewhat limited C99
21:51:37 <elliott> http://www.noradsanta.org/js/data.js NORAD have leaked Santa Clause's Christmas flight path!! Somebody submit this to Wikileaks!
21:53:23 <elliott> Vorpal: http://en.wikipedia.org/wiki/NORAD_Tracks_Santa
21:53:51 <elliott> [[According to NORAD's official web page on the NORAD Tracks Santa program, the service began on December 24, 1955. A Sears department store placed an advertisement in a Colorado Springs newspaper. The advertisement told children that they could telephone Santa Claus and included a number for them to call. However, the telephone number printed was incorrect and calls instead came through to Colorado Spring's Continental Air Defense Command (CONAD
21:53:51 <elliott> ) Center. Colonel Shoup, who was on duty that night, told his staff to give all children that called in a "current location" for Santa Claus. A tradition began which continued when the North American Aerospace Defence Command (NORAD) replaced CONAD in 1958]]
21:53:57 <elliott> MOST AWESOME ORIGIN STORY *EVER*
21:54:11 <elliott> Hey! "Many volunteers are employees at Cheyenne Mountain and Peterson Air Force Base."
21:54:20 <elliott> I wonder if Daniel Jackson is there, or has he gone and ascended again?
21:55:17 <elliott> pikhq: Should I bother supporting locales? I know, I know, I'm a bad person if I don't...
21:56:08 <pikhq> elliott: 他の言語があるぜ。
21:56:24 <elliott> pikhq: Supporting locales != supporting other languages
21:56:42 <elliott> pikhq: Supporting locales == supporting output and UI text in other languages even though there are a ton of untranslated programs /anyway/
21:56:50 <elliott> I doubt, for instance, BusyBox has many transaltions.
21:57:08 <Mathnerd314> elliott: I did that last year... stayed up until midnight answering phone calls :-)
21:57:18 <pikhq> Support locales so that it's possible and/or easy to actually *do* translations.
21:57:38 <elliott> pikhq: But, but, it bloats things up!
21:57:39 <pikhq> Without locale support you literally have no choice in the matter.
21:57:55 <elliott> pikhq: Also, not true; pretty sure you can statically select a gettext translation at compile-time to be the default.
21:58:17 <pikhq> elliott: Your precious metric system is only supported via locales.
21:58:42 <elliott> pikhq: Either what you're saying is true but irrelevant or trivially false...
21:59:23 <pikhq> Programs that display units will look at the locale for which units to use.
21:59:23 <fizzie> Date formatting! Different decimal point separators! Proper sorting order for alphabets! It's not just translations.
21:59:42 <pikhq> And, like most everything else, default to US standard.
22:00:52 <elliott> fizzie: You'd say that; you're a Finn.
22:01:03 <elliott> pikhq: BUT I WARN YOU THAT THE CTYPE.H FUNCTIONS WON'T USE TABLES
22:01:06 <Mathnerd314> pikhq: but how does it work if it's distributed? you don't want some Chinese server giving you results in Chinese if you live in France
22:01:21 <elliott> │ this option will make uClibc much larger.
22:01:28 <elliott> │ Enabling UCLIBC_HAS_LOCALE with the default set of supported locales │
22:01:28 <elliott> │ (169 UTF-8 locales, and 144 locales for other codesets) will enlarge │
22:01:28 <elliott> │ uClibc by around 300k. You can reduce this size by building your own │
22:01:28 <elliott> │ custom set of locate data (see extra/locale/LOCALES for details). │
22:01:28 <Vorpal> elliott, then it will not be locale-aware
22:01:35 <Vorpal> elliott, no locales: I won't use kitten ever
22:01:46 <elliott> Vorpal: Don't worry, I really don't care if you use it or not. But I probably am including locales.
22:01:47 <pikhq> Mathnerd314: It's based on the locale environment variables.
22:01:57 <elliott> pikhq: I wonder how much of that 300k is included in a typical program?
22:02:06 <pikhq> elliott: Oh, and it'll default to USD for currency!
22:02:18 <elliott> pikhq: If you can't write a simple program and get a 10K executable or less it's broken.
22:02:19 <pikhq> elliott: And our MM/DD/YY date display!
22:02:23 <elliott> <elliott> pikhq: I wonder how much of that 300k is included in a typical program?
22:02:57 <pikhq> Very little of that should be included...
22:03:30 <Mathnerd314> pikhq: but what happens if it's a long-running daemon? changing the environment variables will have no effect.
22:04:57 <pikhq> Mathnerd314: ... Yes...
22:05:07 <elliott> Mathnerd314 does not quite understand, methinks.
22:07:28 <elliott> pikhq: Ugh. I don't really want to include every single locale; list all the languages that matter. :p
22:07:40 <elliott> pikhq: (The one advantage of dynamic linking: I could just build a uClibc based on whatever locales the user wants.)
22:07:42 <Mathnerd314> pikhq: so... there seems to be no way for pain-free locale support
22:08:01 <elliott> Mathnerd314: you do realise it's implemented in just about every existing linux system?
22:08:05 <pikhq> Mathnerd314: You seem not to understand what locales do.
22:08:07 <elliott> Mathnerd314: and your complaint about servers makes no sense?
22:08:42 <Mathnerd314> ok... my complaint is that you can't change the locale while a program is running
22:09:13 <elliott> Mathnerd314: why is that an issue?
22:09:29 <pikhq> Mathnerd314: Better than Windows, which doesn't allow changing the locales without *rebooting*.
22:09:43 <elliott> Mathnerd314: Servers shouldn't care what locale you're using.
22:09:49 <elliott> It only matters for normal programs and client programs.
22:09:56 <Mathnerd314> elliott: it's an issue because programs never stop
22:10:02 <elliott> Mathnerd314: Servers shouldn't care what locale you're using.
22:10:34 <Sgeo> The only bad thing about Newspeak that I see is its lack of libraries at the moment
22:10:49 <elliott> Mathnerd314: Anyway, if somehow you really want to change, say, the language that server logs are written in, and it is ABSOLUTELY VITAL that the server NEVER, EVER go down, not even for a second, then just support setting the locale as part of the server's control console.
22:10:54 <Sgeo> And all the features still not implemented or fully designed
22:10:55 <Mathnerd314> elliott: on my home PC, I have had uptimes of months
22:11:11 <elliott> Mathnerd314: Which, of course, you do not need, being that it is a home PC.
22:11:29 <pikhq> Mathnerd314: And how often do you want your program to switch from French to Chinese UI and back?
22:11:31 <elliott> pikhq: Will you hate me if I only build in a subset of locales?
22:11:34 <elliott> <pikhq> Mathnerd314: And how often do you want your program to switch from French to Chinese UI and back?
22:11:37 <elliott> not even relevant, we're talking servers here
22:11:54 <pikhq> elliott: Test the effects of having all the locales first?
22:11:58 -!- kar8nga has quit (Remote host closed the connection).
22:12:02 <Mathnerd314> pikhq: fairly often; I try to learn new languages
22:12:11 <elliott> Mathnerd314: We're talking about servers.
22:12:12 -!- mtve has joined.
22:12:22 <elliott> Mathnerd314: Please tell me why you would want to change a server's language while it's running.
22:12:26 <elliott> hi mtve! haven't seen you for... ages
22:14:20 <elliott> Mathnerd314: that is not a valid complaint
22:14:32 <elliott> Mathnerd314: you have yet to show why that is even vaguely desirable; i suspect you do not understand what locales are used for
22:14:49 <elliott> in what use-case would you want to change the locale of a server process? what would you hope to accomplish by doing so?
22:15:00 <Mathnerd314> elliott: programs should support features, and this is a feature. it doesn't matter if nobody uses it.
22:15:18 <elliott> Mathnerd314: wait, let me check --
22:15:29 <elliott> Mathnerd314: do you think every program should support every feature possible, regardless of whether it is even vaguely useful?
22:15:39 <elliott> Mathnerd314: If so, you're a complete and utter moron, stop wasting my time.
22:15:56 <Vorpal> Phantom_Hoover, want some more cobble?
22:16:01 <Vorpal> Phantom_Hoover, I can put some in that chest
22:16:08 <Mathnerd314> elliott: I think that it should be able to.
22:16:20 <elliott> <elliott> Mathnerd314: do you think every program should support every feature possible, regardless of whether it is even vaguely useful?
22:16:23 <elliott> Mathnerd314: please answer this question
22:16:35 <Vorpal> Phantom_Hoover, ah okay
22:17:13 <Vorpal> Phantom_Hoover, if you want a cobblestone generator I can build one, but I need your help since /home no longer works
22:17:27 <Vorpal> Phantom_Hoover, so one person must keep door to castle of doom open while the other fetch lava
22:17:33 <Mathnerd314> elliott: at any given time, a program will not support all possible features. but somebody should be able to support it in an extensible manner.
22:17:47 <elliott> Mathnerd314: OK, but that is not an actual yes or no answer to my question.
22:18:37 <Mathnerd314> I'd have to say yes. but one particular feature it should support is removing features.
22:18:50 <elliott> Mathnerd314: You sound suspiciously like zzo38.
22:18:59 <elliott> Mathnerd314: But hey, take heart -- you could, one day, be a GNU coreutils maintainer.
22:19:09 <elliott> They like your sort, and this is evidenced by the man page for ls(1).
22:20:37 <Vorpal> elliott, does busybox ls colour code the output?
22:20:44 <Vorpal> elliott, if not: oh well, won't use it
22:20:47 <Mathnerd314> elliott: but once it supports enough features, it might as well be an operating system
22:21:16 <elliott> Vorpal: yes with --color=auto, presumably some env variable could make this default
22:21:46 <Vorpal> elliott, alias ls='ls --color=auto'
22:21:49 <elliott> Vorpal: anyway you have to realise that I *really* don't give a shit whether you use Kitten or not; I will provide support if you do and mock you mercilessly if you run into problems Kitten doesn't have, but fundamentally you can use whatever you like, and if one of your requirements conflicts with one of my requirements, mine take priority.
22:23:16 <Vorpal> elliott, that mocking would be annoying. I assume you can turn that off
22:23:41 <elliott> Vorpal: It's the risk everyone takes by not using Kitten.
22:24:18 <elliott> Vorpal: Oh, and anyone who tells me that they'd "use Kitten, if only it had a proper installer rather than a guide to partitioning and installing the basic packages using a host Linux system" will get, I don't know, IP-banned from the package repository or something.
22:24:22 <elliott> (Or just mocked. Mercilessly.)
22:24:40 <elliott> (Note: I totally will make an installation program, I just don't want to be bugged about making one.)
22:25:01 <Vorpal> elliott, well I done gentoo. And LFS. I'm no stranger to lack of installers
22:25:12 <Vorpal> elliott, heck I even done cross-lfs
22:25:21 <Vorpal> elliott, not canadian-cross lfs though
22:25:38 <elliott> It shouldn't be very painful, anyway; due to static linking, all you really have to do is manually unpack the package manager and its dependencies, tell the package manager to install everything, and edit a few config files.
22:25:46 <elliott> Vorpal: Canadian-cross LFS. Oh man. Has anyone done that?
22:26:07 <Vorpal> elliott, I don't know if it CAN be done even
22:27:16 <elliott> Hey, I just realised that there's probably no TZ value that uClibc supports that handles automatic DST. Heh.
22:27:21 <elliott> "Oh well; file under solve later."
22:27:22 <Vorpal> elliott, I assume it will support software RAID?
22:27:36 <Vorpal> elliott, it can't read zoneinfo files?
22:28:01 <pikhq> elliott: So, should be only about as hard as installing Debian via debootstrap.
22:28:42 <pikhq> (which I have totally done before.)
22:29:45 <Vorpal> Phantom_Hoover, I figured out a way to get lava without going into temple of doom
22:30:42 <Vorpal> I refuse to tell anyone since the admin would fix it
22:30:47 <Vorpal> if he knew what I was doing
22:31:18 <elliott> <Vorpal> elliott, I assume it will support software RAID?
22:31:24 <elliott> Uh, if it works then yes. Otherwise no.
22:31:27 <elliott> <Vorpal> elliott, it can't read zoneinfo files?
22:31:33 <elliott> No; it has no /etc/timezone.
22:31:52 <elliott> You put a string like CST6CDT in /etc/tz and /etc/profile sets TZ to that.
22:32:33 <Vorpal> Phantom_Hoover, where do you want the generator?
22:32:55 <Vorpal> Phantom_Hoover, where in it
22:32:59 <Vorpal> Phantom_Hoover, and which model?
22:33:04 <Vorpal> Phantom_Hoover, fizzie made another one
22:33:44 <pikhq> elliott: You'll have to recompile the distro next time some country decides to fuck with their time zones.
22:33:59 <pikhq> elliott: And countries are positively in *love* with the idea. Fuckers.
22:34:06 <Vorpal> Phantom_Hoover, unknown as of currently
22:34:26 <elliott> Phantom_Hoover: fizzie's leaves you with ghosts of yourself in the pool.
22:34:39 <elliott> pikhq: Well, it's not even something I can configure. Anyway, "CST6CDT" implies to me that you can put offsets there.
22:34:54 <pikhq> (yet another reason that the time zone EVERYWHERE should be a rounded-to-the-hour approximation of the offset from GMT!)
22:35:01 <Vorpal> Phantom_Hoover, mine is the largest one
22:35:12 <Vorpal> Phantom_Hoover, I doubt that ghost was related to the issue in any way
22:35:21 <Vorpal> to the generator that was
22:35:33 <pikhq> (NOBODY gets to fuck with their time zones!)
22:35:46 <Phantom_Hoover> Build it in the air if it's somehow too large for the bottom of the hull.
22:35:47 <Vorpal> Phantom_Hoover, where in the bottom of the hull?
22:36:35 <Vorpal> Phantom_Hoover, it will almost the whole width
22:36:41 <Vorpal> Phantom_Hoover, of the base layer
22:36:51 <Vorpal> Phantom_Hoover, all but two colums if you put it along a side
22:36:56 <Vorpal> in the middle just one column
22:37:21 <Vorpal> Phantom_Hoover, lag atm...
22:37:47 <elliott> I just jumped down the Temple of Doom's lava and survived all the way. :p
22:37:58 <Vorpal> elliott, did you crash the server or something?
22:38:07 <Vorpal> oh no it is just lagged to hell
22:38:53 -!- Sasha2 has joined.
22:39:11 <Vorpal> Phantom_Hoover, it needs to go down 1 into the floor, I have to raise the thing I realised
22:39:12 <elliott> │ Set this to compile all sources at once into an object (IMA). │
22:39:12 <elliott> │ This mode of compilation uses alot of memory but may produce │
22:40:06 -!- Sasha2_ has quit (Ping timeout: 240 seconds).
22:40:36 <Phantom_Hoover> Vorpal, even better, build it in the air somewhere and connect it to the axes.
22:40:54 <Vorpal> Phantom_Hoover, gah already got quite far
22:41:02 <pikhq> elliott: Awesomeness.
22:41:16 <elliott> │ Note that you need a very recent GCC for this to work, like │
22:41:17 <elliott> │ gcc >= 4.3 plus eventually some patches. │
22:41:21 <pikhq> elliott: You want that option. It is awesomeness.
22:41:29 <elliott> pikhq: But I want to build this with pcc eventually! :-)
22:41:53 <elliott> make: *** [extra/locale/c8tables.h] Error 1
22:41:58 <elliott> GEN extra/locale/c8tables.h
22:41:58 <elliott> make: *** [extra/locale/c8tables.h] Error 1
22:42:08 <elliott> could not find a UTF8 locale ... please enable en_US.UTF-8
22:43:46 -!- oklopol has joined.
22:44:44 -!- Sasha2_ has joined.
22:45:41 -!- oklofok has quit (Ping timeout: 245 seconds).
22:46:34 -!- Sasha2 has quit (Ping timeout: 250 seconds).
22:49:33 <Phantom_Hoover> elliott, so wait, you've still not moved past The Story So Far in Fine Structure?
22:50:18 <Sasha2_> quit using the cordless phone, parents
22:50:25 <Sasha2_> it knocks out the wireless signal
22:53:08 <Sasha2_> it's a phone without a cord
22:53:11 <oklopol> yeah yeah i know what it yes
22:54:54 -!- Sasha2_ has changed nick to SashaPrime.
22:56:54 <Vorpal> Phantom_Hoover, it is done
22:57:31 -!- SashaPrime has quit (Read error: Connection reset by peer).
22:57:51 -!- Sasha2 has joined.
22:58:38 <pikhq> It takes an astounding moron to make a cordless phone that hogs the 2.4 GHz space like that.
22:58:51 <pikhq> "But what person would ever want two cordless phones?"
22:58:59 <Sasha2> spams the spectrum with noise
22:59:41 <Sasha2> even though they haven't got a cord to keep them in that room
22:59:48 <Sasha2> they gravitate towards the sofa
23:00:54 <pikhq> ... One for each room?
23:00:57 <pikhq> Your parents are morons.
23:01:28 <Sasha2> well, my phone, the one by this computer, is corded
23:02:18 <Sasha2> it flashes instead of rings, no screen, uses tone-dialing, and has a mute switch.
23:02:54 <pikhq> The entire POTS is such a freaking archaic joke.
23:02:58 <pikhq> As is the cell network.
23:03:04 <pikhq> And the cable system.
23:03:23 <Sasha2> totally keeping this phone though
23:03:36 <Sasha2> it plugs into the phone jack and sits there
23:03:45 <pikhq> In fact, every single telecom system that's not an Internet link is just awful.
23:03:53 <Sasha2> I don't really want a cell phone
23:04:28 <pikhq> Maintain several completely distinct high-bandwidth telecommunication systems? Such a stupid idea.
23:08:27 -!- Phantom_Hoover has quit (Remote host closed the connection).
23:11:29 * pikhq imagines a world where the entire radio spectrum is in use for Internet. And is ecstatic.
23:14:44 * Sasha2 imagines an alien race that sees that spectrum wiping us out because of light pollution
23:15:08 <pikhq> Sasha2: Eh, we'd already be fucked.
23:15:17 <pikhq> Pretty much every chunk of spectrum that can be used, is used.
23:25:24 <quintopia> thank goodness for dark sky communities...and eventually orbital telescope arrays
23:29:11 -!- perdito has joined.
23:29:58 -!- kresnicka has joined.
23:30:25 -!- Goosey has joined.
23:37:22 -!- kresnicka has quit (Quit: Leaving).
23:44:11 <elliott> <quintopia> thank goodness for dark sky communities...and eventually orbital telescope arrays
23:44:14 <elliott> i value internet more than the former
23:44:24 <elliott> even though that barely makes any sense
23:44:27 <elliott> your message didn't anyway
23:48:57 <elliott> [[The worldwide Haskell community met up over beers today to celebrate their unprecedented discovery of an industry programmer who gives a shit about Haskell.
23:48:57 <elliott> On Wednesday, researchers issued a press release revealing that 27-year-old Seth Briars of North Carolina, a Java programmer at Blackwater accounting firm Ross and Fordham, actually gives a shit about Haskell.]]
23:48:58 <elliott> --http://steve-yegge.blogspot.com/2010/12/haskell-researchers-announce-discovery.html
23:51:11 <Sgeo> It is, of course, physically impossible to Google for criticism of Newspeak
23:51:22 <elliott> [["I'm kind of surprised I'm the only person on earth who gives a shit about it," Briars continued. "I'd have thought there would be more people following the press releases closely and then not using Haskell. But they all just skip the press releases and go straight to the not using it part."]]
23:54:23 -!- Quadrescence has joined.
23:56:48 <Deewiant> [[I'm really disappointed that more programmers don't get actively involved in reading endless threads about how to subvert Haskell's type system to accomplish basic shit you can do in other languages.]]
23:57:15 * Sgeo learns of Ioke
23:58:56 -!- wareya has quit (Read error: Connection reset by peer).
23:59:24 -!- wareya has joined.