00:00:03 <elliott> Vorpal: gparted can't shrink jfs partitions. nothing can.
00:00:13 <elliott> it's one of the deep mysteries of the universe
00:00:14 <Vorpal> elliott, indeed, but who needs that
00:00:31 <elliott> Vorpal: anyone who can't make perfect partition size estimates ahead of time?
00:00:42 <Vorpal> elliott, just make a smaller one and grow as needed?
00:00:54 <elliott> Vorpal: anyone who sets up a single-OS machine and then wants to installer another OS?
00:01:17 <Vorpal> elliott, well, look, my volume group has over 600 GB unallocated!
00:01:19 <elliott> Vorpal: what if you expand it to do some disk-heavy work, then rm the data, and later you want to shrink it to install another OS?
00:01:41 <elliott> Vorpal: btw, lvm won't be supported in the installer :P
00:01:42 <Vorpal> elliott, anyway you create a throw away partition for that
00:02:01 <Vorpal> elliott, dude, I done it several times
00:02:01 <elliott> Vorpal: I don't use LVM and nor do many others.
00:02:07 <Vorpal> elliott, well, your own issue
00:02:20 <elliott> Vorpal: Our own lack of a headache...
00:02:56 <Vorpal> elliott, it is *easy*. Takes a few hours to learn
00:03:07 <elliott> Vorpal: I understand it, it is just a headache.
00:03:14 <Vorpal> then all you use is lvcreate, lvextend and lvremove
00:03:22 <elliott> I want nothing of it. Also, I do not appreciate its Linux-specificity at all. Linux vendor lockin FTL.
00:03:27 <Vorpal> elliott, initial setup then you just need 3 commands
00:03:32 <elliott> Vorpal: I understand it, it is just a headache.
00:03:32 <elliott> I want nothing of it. Also, I do not appreciate its Linux-specificity at all. Linux vendor lockin FTL.
00:04:03 -!- sftp has quit (Remote host closed the connection).
00:04:18 <Vorpal> elliott, well you could port it to other OSes
00:04:29 <Vorpal> elliott, it just uses the device mapper
00:04:33 <elliott> Vorpal: Would you justify using Microsoft software like that?
00:04:37 <Vorpal> All of the LVM logic is user space
00:04:52 <Vorpal> elliott, well it is closed source, lvm is open
00:05:18 <elliott> Vorpal: "Software X is unacceptable because it only runs on platform Y. We don't want to lock ourselves into Y and may even look into Z in the future." "Well, you could just PORT THE FUCKING COMPLEX SOFTWARE."
00:05:33 <Vorpal> elliott, you need a basic "map these byte ranges from this partition and that partition into this pseudo drive"
00:05:41 <Vorpal> elliott, that is all you need from the kernel for basic LVM stuff
00:05:48 <Vorpal> the rest is user space
00:05:57 <Vorpal> okay there is the snapshot stuff too, but who uses that?
00:06:03 <elliott> Yeah -- I DOUBT the user space shit will compile on BSD.
00:06:03 <pikhq> elliott: Few months ago.
00:06:19 <elliott> pikhq: Please explain to Vorpal why just because a piece of software is portable to another OS, does not mean it does not lock you in to a certain OS.
00:06:28 <pikhq> elliott: When I realised I had made my Gentoo root FS too small.
00:06:32 <Vorpal> elliott, I'm saying that porting it would be viable
00:06:36 <Vorpal> and would be a nice project
00:06:37 <elliott> pikhq: It can't shrink, but it can grow.
00:06:51 <pikhq> elliott: Oh, it's unshrinkable is all?
00:06:53 <elliott> Vorpal: Yes, okay. Hey, you should install FreeBSD. What's that, you can't?
00:06:53 <Sgeo> Honey, I Shrunk the Partition
00:06:53 <pikhq> elliott: Acceptable.
00:07:02 <elliott> pikhq: (And it's a rather fundamental issue, I gather; shrinking will never happen.)
00:07:06 <elliott> pikhq: But it's so awesome so fuck it, JFS.
00:07:30 <pikhq> elliott: Shrinking a partition basically occurs when: you want to install a new OS dual-boot.
00:07:35 <Vorpal> elliott, well. If you are going to switch OS anyway (which in a production system would be a major PITA), you probably won't do so live on that system
00:07:40 <Vorpal> you would migrate anyway
00:07:46 <Vorpal> elliott, so your argument is void
00:07:53 <elliott> pikhq: And who would ever want to use anything other than Kitten, right?
00:08:16 <Sgeo> I guess if I make the Kitten partition too large, I could always destoy Kitten and its maker
00:08:20 <elliott> 7611392 kilobytes total disk space.
00:08:23 <pikhq> elliott: On a UNIX system, you're *unlikely* to want a dual boot with another UNIX system, and Windows *cannot* be installed after any other OSes.
00:08:36 <elliott> pikhq: (Yes it can. You just need to reinstall the boot loader.)
00:08:41 <pikhq> Well, unlikely to want to do so retroäctively.
00:09:07 <Sgeo> I once tried to have a dual-boot with Freespire and Ubunt
00:09:25 <Sgeo> broke the boot-loader somehow, couldn't get into Ubuntu
00:09:30 <Sgeo> Was stuck with Freespire
00:09:37 <Sgeo> For about 3.5 months
00:09:55 <pikhq> Vorpal: Being capable of being ported but not actually *being* ported is no different from being impossible to port until you're willing to spend a gigantic swath of time on the porting.
00:10:41 <pikhq> Vorpal: *Inevitably*, a single-OS program is going to make a lot of assumptions about the underlying system. A lot of which nobody actually thinks about at all.
00:10:42 <Vorpal> pikhq, I suspect that the user space would be fairly easy to port. And a device mapper equiv would be nice for lots of other stuff anyway.
00:11:23 <pikhq> How likely are you to realise that you should avoid *fork* if you want it to ever run on Windows without pain and agony, for instance?
00:11:51 <Vorpal> pikhq, but that is a windows bug
00:12:10 <pikhq> Basically, porting is *insanely* difficult.
00:12:20 <pikhq> Well, when you try to leave the confines of UNIX.
00:12:24 <pikhq> (oh so comforting)
00:12:27 <Vorpal> pikhq, I'm not suggesting that
00:12:32 <Vorpal> pikhq, I'm suggesting to a *bsd
00:12:52 <pikhq> Vorpal: That still takes quite some effort.
00:12:55 -!- Decarabia has quit (Ping timeout: 240 seconds).
00:13:25 <pikhq> It's much much easier than leaving UNIX, but... You *do* realise why Autocrap was made, right?
00:13:48 <Sgeo> http://www.dangermouse.net/esoteric/hanoisort.html
00:13:55 <Sgeo> I think the time complexity is technically wrong
00:14:11 <elliott> so that programs could pretend to be portable to nongnu?
00:14:11 <Sgeo> It doesn't account for time to determine the correct solution, does it?
00:14:19 <Vorpal> pikhq, yes, but thankfully device-mapper is quite a general concept. Sure it needs to be implemented. But compared to the full lvm2 it is fairly simple. You could skip everything but the basic model to begin with (that is, skip snapshots and such, which are compile time kernel options on linux).
00:14:26 <Vorpal> the entire lvm logic is in userspace
00:14:42 <Vorpal> sure the API against the kernel would change. Probably some udev code need to be replaced
00:14:44 <pikhq> Vorpal: ... Waitwaitwait, porting stuff like *device-mapper*?
00:14:50 <pikhq> That is deep fucking magic. No.
00:15:06 <Vorpal> pikhq, I'm suggesting implementing a equivalent feature
00:15:23 <pikhq> Would you like a pony?
00:15:48 <Vorpal> pikhq, basically it is after all this: [(device,start-offset,stop-offset), ...] -> fake-device
00:16:15 <Vorpal> pikhq, that is what device-mapper without snapshot stuff and so on is
00:16:29 <pikhq> Vorpal wins the oversimplification award of the day.
00:16:59 <olsner> Vorpal: if you think it is easy you don't know the problem very well... this applies to all problems
00:17:13 <Vorpal> olsner, I don't think it is easy. But it think it is possible.
00:17:25 <Vorpal> olsner, writing *any* kernel code is hard
00:17:33 <pikhq> The device mapper is awesome, but fuck trying to get that on any non-Linux OS.
00:17:51 <pikhq> I'd sooner implement LVM via FUSE.
00:17:55 <Vorpal> pikhq, well fuck trying to get the snapshot stuff, the crypt target and so on yes
00:18:02 <Sgeo> olsner, I think 1+1 is easy
00:18:04 <pikhq> (which is actually... Entirely practical.)
00:18:28 <Vorpal> pikhq, that would just mean writing a device mapper in user space targeting fuse!
00:18:50 <pikhq> Sgeo: Calculate it from ZFC. (you will need to start by defining 1 and +)
00:19:28 <Sgeo> (Yeah, not entirely useful without all the rest of the context)
00:19:29 -!- Decarabia has joined.
00:19:49 <Vorpal> elliott, I would presume {} is the empty set?
00:19:51 <pikhq> (okay now I'm just being an ass.)
00:20:02 <Vorpal> pikhq, indeed you are a donkey
00:20:35 <Sgeo> How are negatives defined?
00:20:41 <Sgeo> How are rationals defined? Reals?
00:20:42 <pikhq> elliott: Yes, I know, it's a notational thing, and well outside of the semantics of a formal system.
00:21:24 <pikhq> Sgeo: The integers are defined by a mapping to the naturals, the rationals are defined as an ordered pair of two integers, the reals are FUCK YOU THATS WHAT.
00:21:55 <elliott> DEDEKIND CUTS FVUCXK YEHAAAAHHH
00:22:30 <Sgeo> "such that all"
00:22:33 <elliott> rational is Z*N not Z^2 pikhq
00:22:38 <Sgeo> FFFFFFFUUUUUUUUUU Google
00:22:48 <Sgeo> And Wikipedia for being slow
00:23:05 <Vorpal> pikhq, what do you mean with that statement about the reals?
00:23:24 <pikhq> Vorpal: Harder to derive from memory.
00:24:22 <Vorpal> why start with ZFC and not peano?
00:24:37 <Vorpal> if we are looking at working with integers and so on
00:24:46 -!- Decarabia has quit (Remote host closed the connection).
00:24:52 <pikhq> ZFC is significantly more general is all.
00:25:10 <elliott> can't do reals or anything in peano...
00:25:56 <Sgeo> Why are we assuming C?
00:26:26 <Vorpal> Sgeo, are we assuming complex numbers?
00:26:41 <Sgeo> Vorpal, wrong C
00:26:53 <Vorpal> Sgeo, then what C are you talking about
00:27:05 <Vorpal> The server at en.wikipedia.org is taking too long to respond. <-- wtf
00:27:27 <Vorpal> pikhq, I was checking something else at wikipedia
00:27:29 <pikhq> I can't remember why that's included as an axiom.
00:28:35 <pikhq> Eh, whatever. Just one choice of axioms.
00:29:01 -!- madbr has joined.
00:29:38 -!- Decarabia has joined.
00:30:51 <wareya_> How is measurement dealt with in non-integer dimensional space?
00:32:35 <wareya_> I mean, how can I define the location of something in a space where the number of dimensions is fractional or imaginary?
00:34:15 -!- Decarabia has quit (Ping timeout: 255 seconds).
00:35:09 <elliott> pikhq: axiom of choice is an axiom because it's truw
00:35:41 <pikhq> elliott: No, I'm just asking why it was chosen as an axiom in the ZFC axiomatic system. I know that it's a god-damned axiom.
00:35:51 <Sgeo> And it's true because it's an axiom!
00:36:04 <elliott> pikhq: Because we started with ZF and then we realised the Axiom of Choice was true.
00:36:16 <pikhq> elliott: Nooot how axioms work.
00:36:21 <elliott> pikhq: (ZF~C is quite screwy, although as consistent as ZFC.)
00:36:23 <elliott> pikhq: it really is, though
00:36:35 <elliott> pikhq: ~C has implications we don't like, there are no real objections to C
00:36:41 <elliott> pikhq: and a LOT of theorems are based on it
00:36:52 <elliott> (pick one that you know. good probability it relies on choice)
00:36:56 <pikhq> elliott: So, we take C as an axiom because our intuition suggests it *ought* to be true. Got it.
00:37:10 <Sgeo> pikhq, what other reason is there for ch... what elliott said
00:37:35 <pikhq> Except when we dont'.
00:37:35 -!- Decarabia has joined.
00:37:49 <pikhq> (let's take ~C as an axiom!)
00:37:51 <elliott> pikhq: If only the Bible contained an axiomatic system -- a good proportion of the population wouldn't have to argue :P
00:38:06 <elliott> (extra points for demonstration of the bible's inconsistency in that axiomatic system)
00:38:26 <Sgeo> The Bible's axiomatic system is complete!
00:38:38 <Sgeo> [And contains arithmatic, but that ruins the brevity of the joke]
00:38:49 <elliott> pikhq: "But WHY do we accept those axioms?" "'cuz God says so." "Oh. Okay then."
00:39:04 <elliott> pikhq: Infinite points for proving that the axiomatic system is inconsistent and thus contradicting the Bible's infallibility.
00:39:41 <pikhq> elliott: And from that we demonstrate that God can produce a stone too heavy for him to lift, and thus by being omnipotent he isn't omnipotent.
00:40:14 <elliott> pikhq: I think you're going to Hell for that.
00:40:43 <Sgeo> Only if that's blaspheming the Holy Spirit
00:40:46 <elliott> pikhq: Man, the Bible should have come with a controversy-resolving appendix.
00:40:56 <Sgeo> Does blaspheming "God" count as blaspheming the Holy Spirit?
00:41:00 <elliott> pikhq: Thou shalt not kill.[1]
00:41:04 <elliott> pikhq: 1. Killing for God counts as killing.
00:41:53 <elliott> (lie with another man abomination blah blah)[1]
00:42:01 -!- Decarabia has quit (Ping timeout: 240 seconds).
00:42:26 <elliott> 1. But lesbians are DEFINITELY okay.
00:44:06 <pikhq> elliott: If only God had left some notes in the margin.
00:44:20 * Sgeo wonders why his link karma is increasing
00:44:30 <pikhq> "Thou shalt not kill. No, seriously. No killing at all, that's a total dick move."
00:44:34 <Sgeo> Last I checked the link I submitted, it was doing badly (and arguably rightfully so0
00:44:57 <elliott> pikhq: And verily, Jesus spoketh unto the men: "Additionally, thou'st polynomial time art not the same as thou'st polynomial time that be deterministic not."
00:45:58 <elliott> pikhq: It would be SO AWESOME if people became atheists because they thought P=NP after that.
00:46:58 <pikhq> elliott: Man. With a time machine you could troll people insanely well.
00:47:04 <pikhq> The Gospel of Elliott.
00:47:32 <elliott> I'd go stab all the authors and yell "DO YOU KNOW WHAT CONSEQUENCES THIS WILL HAVE GODDAMMIT"
00:47:51 <elliott> pikhq: Theory: Most of the Bible is Bible fanfiction.
00:48:07 <elliott> pikhq: Later on, people didn't realise this, as it was stored in the same place, and combined them.
00:48:11 <elliott> Thus the inconsistencies, bullshit, etc.
00:48:34 <Vorpal> elliott, who is then the Mary Sue?
00:48:46 -!- Decarabia has joined.
00:49:02 <pikhq> elliott: That is essentially the only way to actually take the Bible in *any* way seriously without being completely retarded.
00:49:13 <elliott> Vorpal: Either God or Jesus or both.
00:49:23 <elliott> Jesus if they were a hippie, God if they were a fucking asshole.
00:49:24 <pikhq> elliott: Same, I'm sure, goes for most any other rather old religious body of work...
00:49:30 <elliott> (God killed way too many people in the Bible.)
00:49:54 <elliott> pikhq: Isn't figuring out what parts of the Bible to believe too much damn work? Atheism is so much simpler...
00:49:57 <pikhq> (for more modern religious works, it's fairly easy to just demonstrate that the author was retarded.)
00:50:25 <elliott> Most of them are pretty similar to the Bible except N years later :P
00:50:26 <pikhq> (see: Scientology)
00:50:37 <coppro> pikhq: or possibly brilliant
00:50:45 <elliott> They only seem crazier because we've adopted so much Christianity in our culture.
00:50:49 <elliott> pikhq: OK, Scientology I'll grant.
00:51:01 <Vorpal> only Buddhism is anything like a halfway sane religion IMO.
00:51:03 <elliott> The space planes that look just like NASA ones is... a bit too much to even start to take seriously.
00:51:21 <elliott> Vorpal: Ahh, I see that a lot and, well, most people have an idealised perception of Buddhism that simply doesn't reflect modern practice.
00:51:34 <Vorpal> elliott, true, not all branches of it are sane
00:51:36 <pikhq> elliott: Yes, but it's really really *really* easy to see that more modern religions are genuinely bullshit. Whereas older ones, the evidence of making it up kinda died. What with not being popular in a time when the only way to perpetuate something was to be popular enough to get copied over and over again.
00:51:38 <Vorpal> elliott, which I never claimed
00:51:38 <elliott> Vorpal: Buddhists -- well, most western buddhists are just hippies, but non-western -- believe in gods, hell, etc.
00:51:54 <elliott> Vorpal: Sort of like how most of what Jesus said was pretty cool but nobody actually adheres to it.
00:52:18 <Vorpal> elliott, true for several branches indeed.
00:52:19 <elliott> pikhq: How are you still Christian? You have amazing powers of cognitive dissonance X-P
00:52:32 <Vorpal> wait, is pikhq christian?
00:52:47 <Vorpal> that is just.... *total mental reverse*
00:52:51 <pikhq> elliott: Actually, I'm more along the point of "Jesus was a pretty cool dude, and why the hell do people even care about the rest of that shlock."
00:52:56 <elliott> Vorpal: Well, that information is months and months old.
00:52:57 <pikhq> Erm, lines. Not point.
00:53:10 <elliott> Vorpal: I mean it might be outdated :P
00:53:20 <elliott> pikhq: Okay; you'd previously stated yourself to be Christian, which I kind of mentally interpret to include belief in supernatural things and all that.
00:53:58 <pikhq> elliott: I'm willing to accept a supernatural origin of the universe, but do I know? Not at all.
00:54:26 <elliott> Yeah the furthest I go speculating about the universe's origin is the mathematical universe hypothesis.
00:54:35 <Vorpal> elliott, wait, about ZFC above. Why would *you* want ZFC? You are constructivist, right?
00:54:37 <elliott> (tl;dr every self-consistent mathematical object exists)
00:54:52 <elliott> Vorpal: I was only ever half-serious about that.
00:54:54 <Sgeo> What would someone who has the idea "The idea of Jesus, whether or not he was real in any way, is a nice way to help unload some of my unhealthy guilt" be called?
00:55:15 <pikhq> Sgeo: That's a security blanket, dude.
00:55:29 <elliott> Vorpal: I like Coq and the like which use constructivist logics, and I do have a sneaking suspicion that ZFC is inconsistent, and I *do* instinctively distrust non-constructive things... but really, naw.
00:56:06 <pikhq> elliott: Oh, and I am and always have been absolutely refusing to buy the common idea that a loving deity would actually send people to hell for doing things that he declares somehow "wrong". I mean, just *what the hell*.
00:56:30 <elliott> pikhq: yeah hell is just beyond bullshit
00:56:32 <pikhq> Even if you buy the idea that a deity created everything, is personally involved in your life, and has the power to do that...
00:56:45 -!- Decarabia has quit (Ping timeout: 260 seconds).
00:56:49 <elliott> If there's a god that will punish me for not believing in him, fuck him. All the cool people will be in hell anyway.
00:56:51 <pikhq> Why in the world would a benevolent being sentence anyone to *eternal pain and agony*?
00:57:02 <Vorpal> heaven might be bullshit too
00:57:17 <elliott> Vorpal: I mean bullshit as in "a really terrible idea".
00:57:29 <elliott> Admittedly the Christian view of Heaven is incredibly naïve.
00:57:36 <Vorpal> pikhq, ah but it is the devil. Err wait. What happened to that omnipotence?
00:57:40 <elliott> It's white and pearly and awesome! Fuck you!
00:57:46 <Sgeo> I want someone to tell me "You're going to burn in hell, I really don't want you to" so I can tell them that they're nicer than God
00:57:56 <pikhq> You'd have to be genuinely malevolent to sentence people to eternal torment.
00:58:00 <elliott> Sgeo: OH THAT'LL LEARN THEM
00:58:10 <elliott> Sgeo: Next you'll be questioning their epistemological justification.
00:58:12 <Vorpal> elliott, I believe it is like minecraft. Only real and without the lag.
00:58:34 <elliott> Vorpal: So, like Prime Intellect. :P
00:58:39 -!- Decarabia has joined.
00:58:50 <elliott> Except you can do it automatically if you want.
00:58:53 <Vorpal> elliott, well I didn't actually mean like that, but you could interpret it that way.
00:59:19 <elliott> Vorpal: Prime Intellect is basically an advanced real-life Minecraft + automation tool :P
00:59:31 <Vorpal> elliott, I read that stuff
00:59:48 <elliott> Of course it's not very utopian.
01:00:03 <elliott> Vorpal: oh, you mean Creating Friendly AI?
01:00:06 <pikhq> I can buy the concept of a benevolent deity creating heaven. I can even just barely accept the idea of that same deity creating this shitty planet (you have to start by assuming he's not actually perfect, but just well-intentioned...). But *eternal pain and agony is not at all removed from taking a magnifying glass and burning ants to watch them squirm.*
01:00:07 <Vorpal> elliott, yeah. but not at 01:59
01:00:14 <elliott> Vorpal: I meant The Metamorphosis of Prime Intellect.
01:00:46 <elliott> pikhq: I DO THAT! ...I don't actually
01:01:01 <elliott> pikhq: At least ant sentience is far from certain :P
01:01:30 <Vorpal> elliott, an anthill as a collective might have a sentience
01:01:38 <Sgeo> When I was a kid, I.. kind of stepped on ants :/
01:02:03 <Vorpal> elliott, well, not sure about ants, for bees it seems somewhat plausible.
01:02:12 <elliott> "Man eats a pound of bacon a day for a month and loses weight, lowers his blood pressure."
01:02:17 <elliott> IT'S LIKE A DREAM COME TRUE
01:02:53 <Vorpal> elliott, and it would be nice and fit a story. Thus probably just a wishful dream (wrt ants and bees)
01:03:12 <elliott> Vorpal: Eh... collective sentience seems quite plausible to me.
01:03:38 <elliott> Vorpal: I think if you could somehow ask a really difficult question to everyone on the internet, every discussion forum, everything -- and make sure they listened -- you'd have it answered insanely quickly, insanely well.
01:03:39 <Vorpal> elliott, well, perhaps. But you have to be careful to not just think so because it fits a mental story.
01:03:52 <elliott> Vorpal: Of course the problem is that you can't; the Internet is probably sentient in some sense but you can't do anything with it.
01:03:57 <Vorpal> elliott, enough to ask here. Takes a bit longer of course.
01:04:00 <elliott> Vorpal: But you see... glimpses of it, every now and then.
01:04:45 <Vorpal> elliott, either someone in the channel is an expert or knows someone who is. This was established some time ago.
01:04:57 <elliott> this channel is rather awesome.
01:05:15 <elliott> Now if we could just stop anyone else joining :P
01:05:17 <elliott> Vorpal: can a normal person actually buy the POSIX standard?
01:05:18 <pikhq> It's even more cruel to believe in a God who sentences sinners to hell, *and has made a world whereby it is impossible for humans not to sin*.
01:05:33 <pikhq> You'd just have to be a complete asshole to be that God.
01:05:44 <Vorpal> elliott, why would you need to?
01:05:49 <elliott> http://code.google.com/p/unix-jun72/source/browse/#svn/trunk/tools/apout THIS PORTABLE UNIX PROGRAM, WHEN GIVEN A PDP-11 EXECUTABLE, WILL EMULATE IT.
01:05:49 <Vorpal> elliott, you can get it for free
01:06:00 <elliott> It will *see your filesystem*.
01:06:03 <pikhq> (it is commonly accepted Christian theology that all have sinned, and hence all are sentenced to hell.)
01:06:08 <elliott> $ /usr/local/bin/apout /work/unixv5/bin/ls -l
01:06:10 <elliott> And get a directory listing.
01:06:43 <elliott> pikhq: Calvinism has to be my favourite screwy theology.
01:06:56 <elliott> pikhq: The nice thing is that you can completely ignore it because your fate is predetermined. The not so nice thing is that you're probably fucked.
01:07:13 -!- Decarabia has quit (Ping timeout: 245 seconds).
01:07:19 <Vorpal> elliott, anyway you need to either get POSIX by IEEE or by opengroup. The latter is free, gratis. Just need to register your email, wait a bit (iirc it was manual activation, so took about a day), then download the pdf
01:07:40 <elliott> Vorpal: is that the official one, or the in-development one you can find via google?
01:07:44 <elliott> Vorpal: also, is it available as html?
01:08:00 <elliott> Vorpal: IEEE isn't an option as I'm trying to boycott them :)
01:08:22 <Vorpal> elliott, no idea about html. The older one, 2001/SuSv<whatever> is available as tarball of html though
01:08:41 <elliott> Older is fine, I just want to beat Microcosm's ass >:)
01:08:48 <elliott> We're talkin' user-mode POSIX here.
01:09:03 <Vorpal> elliott, well the issue with microcosm is supporting windows
01:09:17 <pikhq> elliott: Y'know, the world would be a much better place if Christians were in any way like Christ.
01:09:17 <Vorpal> elliott, which means a VFS
01:09:33 <elliott> > device add block 8G ./some_file
01:09:41 -!- Decarabia has joined.
01:09:44 <elliott> Vorpal: In this case, it'd manage all its "devices" itself.
01:09:53 <elliott> Vorpal: You'd just make various "devices" be parts of the host machine.
01:10:01 <Vorpal> elliott, the mtime/ctime/atime semantics
01:10:04 <elliott> It'd be just like a POSIX kernel running on bare hardware, except it's running on POSIX instead.
01:10:14 <elliott> Vorpal: That's a filesystem option, isn't it?
01:10:17 <elliott> I may have misinterpreted you.
01:10:41 -!- cal153 has quit (Ping timeout: 265 seconds).
01:10:42 <Vorpal> elliott, you need to do something for it though
01:10:51 <elliott> Vorpal: I don't understand.
01:11:36 <Vorpal> elliott, yes it is not available on all FSes. But what do you return for those cases? Some junk value?
01:11:49 <elliott> Vorpal: Well, it'd depend on your inner system.
01:11:56 <Vorpal> elliott, stat()s structure looks the way it does for *all* fses after all
01:11:58 <elliott> Vorpal: Remember that in this case, some_file is treated like /dev/sda.
01:12:10 <elliott> Vorpal: Except that I phrased it more generally; it'd be /dev/block1 or something on the inner system.
01:12:16 <elliott> Vorpal: You'd partition it and do "mkfs" on it.
01:12:28 <elliott> Vorpal: Hey, it's user-mode POSIX.
01:12:32 <Vorpal> elliott, that drops the point of it
01:12:36 <elliott> Vorpal: Not user-mode-oh-but-we-use-bits-of-the-host-system POSIX.
01:12:42 <elliott> Vorpal: It's a proper POSIX kernel implemented in user space!
01:13:15 <elliott> http://limerickdb.com/?380 ;; someone please explain quantum mechanics in limericks
01:14:12 -!- Decarabia has quit (Ping timeout: 250 seconds).
01:14:40 * Sgeo diffracts elliott
01:15:14 <Vorpal> elliott, so what do you think of string theory?
01:15:33 <elliott> It's not true, unless it is!
01:15:46 <Vorpal> elliott, and which do you suspect is more likely?
01:15:52 <elliott> I am not really sure it is even vaguely falsifiable. They need to propose tests, stat.
01:16:09 <elliott> Until they do, I can't really comment, especially as I have no physics education of that level.
01:16:19 <Sgeo> What does diffracted matter look like, actually?
01:16:30 <Vorpal> elliott, as far as I understand, it is kind of hard to come up with new tests though
01:16:41 <elliott> Vorpal: But I'll say that while it doesn't seem particularly likely to me, I don't see any promising unifications of quantum mechanics and relativity, either.
01:16:49 <elliott> Vorpal: Well, if their theory isn't falsifiable, they've already lost.
01:17:09 <Vorpal> elliott, hm maybe they *can't* be unified
01:17:15 <Sgeo> What if it's falsifiable in theory but not in practice?
01:17:16 <Vorpal> maybe there isn't a global single rule
01:17:36 <Vorpal> <Sgeo> What does diffracted matter look like, actually? <--- uh....
01:17:37 <elliott> hmm, what happens with -o gid,uid on a unix partition?
01:17:54 <elliott> i'd like to mount /dev/sda2 on /k to do kitten work, owned and writable by my user
01:17:55 <Vorpal> elliott, hopefully "unknown option"
01:17:58 <elliott> but have it save as files owned by root
01:17:59 <Sgeo> Vorpal, not asking for a picture
01:18:18 <elliott> elliott@dinky:~$ sudo mount -o uid=1000,gid=1000 /dev/sda2 /k
01:18:20 <Vorpal> elliott, mount it and chmod it?
01:18:33 <elliott> Vorpal: that'll save files as user 1000 / group 1000, though
01:18:38 <elliott> Vorpal: I want them to be stored as 0 / 0
01:18:41 <Sgeo> I would like a computer-generated image of such a thing though
01:18:45 <Vorpal> elliott, uh. Did that work?
01:18:55 <Vorpal> elliott, I mean, okay it mounted
01:18:58 <elliott> Vorpal: it ran but it probably didn't do what i want
01:19:06 <Vorpal> elliott, what are the semantics indeed
01:19:16 <elliott> elliott@dinky:~$ echo hi >/k/x
01:19:16 <elliott> elliott@dinky:~$ sudo umount /k
01:19:16 <elliott> elliott@dinky:~$ sudo mount /dev/sda2 /k
01:19:17 <elliott> -rw-r--r-- 1 elliott elliott 3 Nov 11 01:18 x
01:19:32 <elliott> Vorpal: probably just makes sure /k is writable by me or something silly like that
01:19:36 <pikhq> elliott: The LHC is actually testing *a* string theory.
01:19:43 <Vorpal> elliott, it is the *sanest* thing to do
01:20:16 -!- Mathnerd314 has quit (Ping timeout: 250 seconds).
01:21:37 -!- Decarabia has joined.
01:22:25 <pikhq> elliott: Uh, actually. Fek, this is hard to understand.
01:22:50 <pikhq> Okay, it's testing something to do with string theory that I postively do not understand at all.
01:23:28 <pikhq> And testing electroweakstrong unification. That I *do* understand.
01:24:26 <pikhq> And the extra dimensions predicted by string theories.
01:24:41 <elliott> pikhq: Say, with Gentoo I can make it compile all packages with a statically-linked uClibc if I want, right? X-P
01:24:47 <elliott> (note: /me is never going to use gentoo, don't worry)
01:24:58 <pikhq> elliott: Not been supported for 5 years.
01:25:10 <elliott> pikhq: What isn't? DOING THINGS YOUR OWN WAY?
01:25:57 <Sgeo> electroweakSTRONG unification?
01:26:03 <Sgeo> I've vaguely heard of electroweak
01:26:33 <Sgeo> Also, is this strong on the level of nucleons, or strong on the level of quarks and gluons?
01:26:53 <elliott> Strong on the level of MEN.
01:26:54 * Sgeo guesses the latter
01:27:10 * Sgeo still has no idea what the weak force DOES, exactly
01:27:36 <pikhq> Sgeo: Electroweak unification is confirmed.
01:28:20 * Sgeo still needs a handle on what the weak nuclear force is
01:28:26 <Sgeo> Particles changing themselves?
01:28:49 <Sgeo> Oh, some radioactive stff
01:29:25 <Sgeo> Also, only force affecting neutrinos, besides gravity
01:29:27 <elliott> pikhq: So what isn't supported? :p
01:29:28 <Sgeo> That's awesome
01:29:53 <pikhq> elliott: uclibc. At all.
01:30:17 <Sgeo> neutron -> proton + electron
01:30:27 <Sgeo> That's a simpler thing than I was expecting
01:30:42 <Sgeo> But why does a conversion like that need messenger particles/
01:30:58 <pikhq> That is not a snark, it's the actual reasoning.
01:31:00 <elliott> pikhq: Gentoo: Freedom from choice!
01:31:28 <Sgeo> Oh, the page on Wikipedia has a nice diagram. The heavy thingy decays
01:32:00 * Sgeo wants to learn all this stuff without formal schooling
01:32:43 <pikhq> Quantum mechanics is both easier and harder than you expect.
01:33:00 -!- Decarabia has quit (Remote host closed the connection).
01:33:02 <coppro> quantum mechanics is /awesome/
01:33:06 <elliott> Sgeo: CHANGE YOUR MAJOR TO PHYSICS! On second thoughts, DON'T.
01:33:12 <coppro> also quantum information theory
01:33:21 <elliott> coppro: you said "elliott: register"
01:33:41 <pikhq> elliott: What are the odds of a physics undergrad covering all the awesome stuff in physics, rather than just classical mechanics and relativity?
01:33:47 <Sgeo> elliott, I've been somewhat offput from physics courses due to my HS physics being macrolevel stuff
01:33:59 <pikhq> ... Okay, I'll grant that there's some awesome stuff in relativity as well.
01:34:06 <Sgeo> I'm pretty sure we didn't cover relativity
01:34:16 <pikhq> Mass-energy equivalence, for instance, is pretty sweet.
01:34:38 <pikhq> Sgeo: "Undergrad". As in "an undergraduate physics program".
01:34:40 <Sgeo> In elementary school, I used to fantasize about that
01:34:57 <pikhq> And the units work!
01:35:02 <elliott> pikhq: I should probably mount kitten at /mnt/kitten, not /k, shouldn't I.
01:35:02 <pikhq> (hooray, SI base units)
01:35:11 <elliott> pikhq: And just set K=/mnt/kitten.
01:35:30 <coppro> pikhq: mine certainly covers that stuff
01:35:31 <elliott> <Sgeo> Why are we assuming C?
01:35:31 <elliott> <Vorpal> Sgeo, are we assuming complex numbers?
01:35:31 <elliott> <Vorpal> Sgeo, then what C are you talking about
01:35:31 * Sgeo feeds elliott a banana
01:35:33 <elliott> <Vorpal> The server at en.wikipedia.org is ta
01:35:34 -!- Decarabia has joined.
01:35:48 <coppro> elliott: p. sure comex != coppro
01:36:16 <Sgeo> elliott, why the paste?
01:36:22 <pikhq> coppro: So, you go beyond classical and relativistic mechanics out into quantum?
01:36:34 <coppro> pikhq: Quantum Mechanics 1 is a second-year course
01:36:39 <coppro> (which I fully intend on taking!)
01:36:54 <pikhq> coppro: ... Dammit, now I think I'm in the wrong major.
01:37:19 * Sgeo wonders if there is any OpenCourseWare on quantum mechanics, and if it's decent
01:37:39 <coppro> pikhq: Admittedly, my school has a lot of focus on quantum because of quantum computing
01:37:41 * Sgeo goes looking for the answer to the first part of that question
01:38:00 <elliott> coppro: Just tell me nobody buys into the D-Wave crap.
01:38:10 <elliott> pikhq: WHY SHOULDN'T I MOUNT IT AS /k :P
01:38:28 <Sgeo> elliott, because I'll keep feeding you bananas
01:38:28 -!- wareya_ has quit (Read error: Connection reset by peer).
01:38:30 <elliott> Sgeo: Either morons or unscrupulous cons that claim to have quantum computers far beyond the state of the art.
01:38:35 <elliott> The media tends to believe them.
01:38:44 <coppro> pikhq: also the Perimter Institute. That place is epic.
01:38:57 -!- wareya has joined.
01:39:20 <elliott> pikhq: See, you have NO arguments against my awesome path.
01:39:54 <elliott> Sgeo: "In the absence of evidence from D-Wave that their 16 qubits are coherent, scientists are understandably skeptical. If D-Wave’s qubits are not coherent, as many scientists suspect, their computer would be classical, not quantum. This would still be consistent with the results of the demo, since the decohering qubits would act like classical random bits, and the adiabatic computer would act like
01:39:54 <elliott> a classical computer implementing simulated annealing, which would be quite fast for a small 16 bit Ising problem." etc.
01:40:04 <pikhq> elliott: Unless they mean "computers that rely on quantum mechanics to function", they're full of shit. (gotta love the tunnelling diode)
01:40:34 <elliott> [[During this time, D-Wave also claimed that by the end of 2008 they would have a 128-qubit system, a claim that was widely doubted due to both the lack of evidence that they are capable of creating coherent qubits at all, and that in November 2007 they only had a 28-qubit system. However, on December 19, 2008, they announced a "128 qubit" chip.]]
01:40:51 -!- Decarabia has quit (Ping timeout: 255 seconds).
01:40:56 <Sgeo> Has anyone gotten their hands on one of these?
01:41:08 <elliott> Sgeo: Uhh, I think they gave a university department a computer.
01:41:10 <elliott> So, avoid that university.
01:41:16 <elliott> http://en.wikipedia.org/wiki/D-Wave_Systems#Criticism
01:41:28 <Sgeo> Surely that university is trying it out to see if they can do anything interesting
01:41:52 <elliott> It is almost impossible to tell whether the processors are in fact quantum really.
01:41:59 <elliott> Without seeing their manufacturing process.
01:42:00 -!- Decarabia has joined.
01:42:18 <pikhq> And if you believe that I've got a bridge to sell you.
01:42:20 <elliott> pikhq: Best part: [[ "Their claimed speedup over classical algorithms appears to be based on a misunderstanding of a paper my colleagues van Dam, Mosca and I wrote on “The power of adiabatic quantum computing”. That speed up unfortunately does not hold in the setting at hand, and therefore D-Wave’s “quantum computer” even if it turns out to be a true quantum computer, and even if it can be
01:42:20 <elliott> scaled to thousands of qubits, would likely not be more powerful than a cell phone."]]
01:42:36 <elliott> pikhq: The worst part is that their blog: http://dwave.wordpress.com/ doesn't look strange or suspicious at all.
01:42:48 <elliott> Until you get to them offhandedly mentioning their 839572398472389472389 qubit computers.
01:43:07 <elliott> pikhq: Note: Number is a JOKE.
01:43:14 -!- Decarabia has quit (Read error: Connection reset by peer).
01:43:14 <elliott> [[Vern Brownell is the founder of Egenera and was the CTO of Goldman Sachs for 11 years. His most important job to date, however, is his current one which is being CEO of D-Wave.]]
01:43:17 <elliott> Low opinion of Goldman Sachs.
01:44:02 <pikhq> elliott: 74,5690 pebiqubits would be quite astounding.
01:44:27 <pikhq> Heck, 74,5690 pebibits would be awesome RAM.
01:44:45 <elliott> pikhq: No, that figure is QUBIT DISK.
01:45:11 <pikhq> RAID-1, I suppose. You can afford to halve the space.
01:46:19 -!- Decarabia has joined.
01:47:16 <elliott> pikhq: Linux From Scratch appears to assume you're root all the way through. Nice...
01:48:17 <elliott> pikhq: I think that Kitten will start out as x86-64 only, but with a high likelihood of supporting i386, especially if disk space isn't a problem.
01:49:14 <elliott> http://usu-shaft.com/2010/my-bishop-masturbation-leads-to-homosexuality/ That conversation log... uhh, wow @ Mormonism.
01:51:01 <Sgeo> "I come from a sexual family, sorry to say. I got it from both parents."
01:51:18 <Sgeo> Maybe I shouldn't be laughing at this
01:52:17 * Sgeo ponders the possibility of societal pressures forcing two asexuals to get together and have a family
01:52:25 -!- aloril has quit (Read error: Operation timed out).
01:52:33 <elliott> pikhq: Please tell me what /media vs. /mnt is meant to mean. /media is external media and /mnt is internal hard drives?
01:52:38 <elliott> pikhq: (Kitten will only have /mnt.)
01:52:48 <elliott> Sgeo: You do realise that asexual people can still love?
01:53:21 <elliott> Sgeo: Hell, if they don't have an aversion to sex and just don't enjoy it, they could easily make a baby the old fashioned way.
01:53:23 <pikhq> elliott: /media is managed automatically.
01:53:32 <elliott> pikhq: Is that the sole reason for its existence?
01:53:42 <Sgeo> Do heterosexual men tend to fall in love with men?
01:53:52 <elliott> Sgeo: Asexual != aromantic.
01:54:13 <elliott> Sgeo: For non-asexual people, love is generally quite strongly linked to sexual desire.
01:54:21 <elliott> Almost certainly not always.
01:54:25 <elliott> For asexual people this is of course not true.
01:54:37 -!- augur has quit (Remote host closed the connection).
01:54:47 <elliott> Wikipedia cites http://www.asexuality.org/home/relationship.html and http://www.gayline.gen.nz/asexual.htm for asexual people having relationships.
01:55:29 <elliott> pikhq: Who thought atime was a good idea?
01:56:10 * Sgeo suddenly ponders
01:56:32 <Sgeo> I have the same feeling that I had when I read Wicked
01:56:38 <pikhq> elliott: People who think writing is awesome.
01:57:09 <Sgeo> Wicked opened my eyes to the fact that "aspirituality" is in, some sense, an option. Doesn't mean I believed it because it's in a fiction book, but just.. the notion
01:57:20 <elliott> pikhq: Sweet! Kitten's installer will put noatime everywhere.
01:57:27 <Sgeo> Here, the idea of sexuality and romantic feelings being disconnected is ... something I can relate with, I think
01:57:52 <elliott> pikhq: (Note: Until my laziness ceases, Kitten's installer will be an xterm and a keyboard operated by an intelligent user reading the partitioning, package installation and bootloader set-up manual.)
01:57:59 <Sgeo> Until 12th grade or so, for me, there was always a very distinct ... difference between "wanting to see naked" and "having a crush on"
01:58:06 <Sgeo> This is getting TMIy
01:58:19 <elliott> YOU SAID THE WORD NAKED ZOMG
01:58:58 <elliott> pikhq: I can't believe the joking feline nickname has stuck X-P
02:00:53 * Sgeo also wonders if it's possible to want to see nudity, but be ... I'm pretty certain it is possible, but I have no idea if that's the case with me
02:01:58 <elliott> That... makes basically no sense.
02:02:10 <Sgeo> but be uninterested in having sex itself
02:02:32 <elliott> I suppose one could be too scarred from some traumatic incident to have actual sexual contact, but still be privately sexual.
02:02:40 <elliott> But that's not a sexual preference, that's just trauma.
02:03:08 <Sgeo> Then again, the only metric I currently have to measure my interest is my imagination
02:03:56 <elliott> Sgeo: You could just be rationalising a perceived inability to get laid :P
02:05:31 <Sgeo> In my mind's eye, I'd love to cuddle with a girl. I'd love to see her naked. But actual sex... well, considering that I don't know what it's like.. (not that I really know what those other things are li... that's technically not true, but I'd prefer to pretend)
02:06:11 <Sgeo> That sounds wrong. Gives a wrong impression
02:06:24 <Sgeo> There's no trauma that I know of involving any physical contact
02:06:45 <elliott> "Oh, my penis was ripped off when I was 3 -- I JUST FORGOT ABOUT IT."
02:06:51 <elliott> (note: repressed memories are unscientific)
02:07:04 <Sgeo> What about forgotten memories?
02:07:14 <Sgeo> Although I guess it's difficult to forget some things
02:07:26 <elliott> If you forget abuse, it either wasn't abuse or you have chronic amnesia.
02:08:54 <elliott> Sgeo: I think it is "relatively" common to shy away from actual sex due to some worry about harming the partner...
02:09:01 <elliott> At least I'm pretty sure I've seen it.
02:09:38 -!- aloril has joined.
02:12:18 <elliott> pikhq: Btw, you should say Unix, not UNIX.
02:12:50 <elliott> pikhq: The original papers typeset it as "Unix", smallcapsed. UNIX is just the obnoxious trademark failure of a rendering :P
02:12:59 <elliott> Gregor: SHUT UP I'M SPREADING THE WORD (and pikhq is far more annoyingly pedantic than I)
02:14:41 <Gregor> EUNUCHS EUNUCHS EUNUCHS
02:15:47 <pikhq> elliott: |_||\|1><
02:15:50 <elliott> pikhq: if you don't want me to use a big ol' static dev now is the time to convince me
02:16:02 <pikhq> elliott: Arguments against static dev: hotplugging.
02:16:15 <elliott> pikhq: Just have the devices you hotplug already there :P
02:16:26 <elliott> Sure, they won't *work* if you don't have the device plugged in, but...
02:16:52 <pikhq> But what if you don't know what devices you might plug in?
02:17:23 <pikhq> Anyways, arguments for static dev: no magic, it just freaking works, and it's easy to understand.
02:17:32 <elliott> pikhq: Well, if you have sd[abcdefg][0-9], that covers all USB drives and the like.
02:17:42 <elliott> pikhq: Admittedly that's a lot of files, but whatever.
02:18:14 <elliott> pikhq: I don't particularly *mind* udev from what I've seen, but it has a library, a daemon, an administration command...
02:18:28 <elliott> And I can't think of any good arguments against static dev other than "well, it's a lot of inodes!".
02:18:43 <pikhq> elliott: You could also pull out Busybox's udev-alike.
02:19:05 <elliott> pikhq: (It's possible that I'll use BusyBox for coreutils, unless you link me to a decent coreutils that isn't Heirloom in three seconds.)
02:19:49 <pikhq> Busybox is workable as a coreutils. Though make it so that replacing individual programs in it is easy, please?
02:20:22 <elliott> pikhq: Ugh, if I can get away with it I'll make it build every program separately.
02:20:43 <elliott> pikhq: I guess you don't know of any alternatives, then. :
02:21:22 <pikhq> elliott: Heirloom or porting or writing your own.
02:21:28 <elliott> pikhq: Incidentally, have you *seen* BusyBox? It has its own gzip/bzip2/xz implementation. Its own tar. Its own *dpkg*.
02:21:52 <pikhq> And it still sucks less than many other things.
02:22:16 <pikhq> How do you feel about Minix?
02:22:25 <Sgeo> busybox includes dpkg? Why?
02:22:41 <elliott> Sgeo: No. It includes its *own* mini version of dpkg.
02:22:51 <elliott> pikhq: What license is Minix again?
02:22:59 <Sgeo> Why does it include its own mini version of dpkg?
02:23:11 <elliott> http://www.minix3.org/license.html
02:23:14 <pikhq> elliott: Minix 3 is BSD.
02:23:15 <Sgeo> Does it include its own mini version of rpm?
02:23:26 <elliott> pikhq: [[# Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.]]
02:23:33 <elliott> pikhq: Isn't that, y'know, an advertising clause?
02:23:52 <Sgeo> Wasn't advertising clause the thing that caused issues with BSD licenses?
02:23:59 <pikhq> elliott: No, that isn't an advertising clause.
02:24:31 <Sgeo> elliott, so why would an advertising clause imply compatibility with a BSD license, is what I was asking
02:24:39 <pikhq> elliott: That's in 2-clause BSD, BTW.
02:24:42 <Sgeo> incompatibility dammit
02:24:51 -!- augur has joined.
02:24:51 <elliott> Sgeo: BSD4 is no longer used at all.
02:24:59 <elliott> Sgeo: Clause 4 is the advertising clause.
02:25:10 <elliott> Sgeo: And besides, it's a custom license whether it's compatible or not; it is not the BSD license.
02:25:18 <elliott> pikhq: For *binary* redistributions? Really?
02:25:27 <pikhq> elliott: "All advertising materials mentioning features or use of this software must display the following acknowledgement: This product includes software developed by the <organization>."
02:25:32 <pikhq> elliott: That is the advertising clause.
02:25:35 <elliott> pikhq: No. I don't see huge lists of BSD-licensed programs in Debian listed in any manual.
02:25:59 <Sgeo> pikhq, thank you for encouraging my laziness
02:26:13 <pikhq> elliott: It's in the /usr/share/doc/... for each BSD-licensed program.
02:26:41 <elliott> pikhq: LAWL, remind me not to use BSD2.
02:27:05 <pikhq> elliott: Something similar is also in the GPL.
02:27:07 <elliott> pikhq: Well. I suppose it depends on the interpretation.
02:27:30 <elliott> pikhq: Of course the notice should be retained with source distributions. But I don't see why manuals for random software should have to have one of them for EVERY BIT OF BSD SOFTWARE USED.
02:27:52 <pikhq> Anyways, it appears that MINIX possesses a NetBSD coreutils that's not BSD-specific.
02:28:01 <pikhq> Might be worth a shot.
02:28:55 <Sgeo> I have managed to recognize bad fiction as bad1
02:28:56 <elliott> pikhq: Bah, even the Fair License has that clause:
02:28:58 <elliott> [[Usage of the works is permitted provided that this instrument is retained with the works, so that any entity that uses the works is notified of this instrument.
02:28:58 <elliott> DISCLAIMER: THE WORKS ARE WITHOUT WARRANTY.]]
02:28:58 <Sgeo> It's a miracle!
02:29:07 <Sgeo> (Then again, it's deliberately bad)
02:29:15 <Sgeo> http://marigoldfarmer.tumblr.com/ [NSFWText]
02:30:02 <Sgeo> http://questionablecontent.net/
02:30:14 <elliott> Sgeo: Where, exactly, is the NSFWness in the text?
02:30:34 <Sgeo> Actually, now that I actually read a bit more.. I don't thiink there is
02:31:15 <elliott> pikhq: Okay, TODO: Try and build NetBSD coreutils. Then: Minix.
02:31:20 <Sgeo> "So they totally had sex, and it was the most amazing sex in the history of wizarding or muggles or even Space Wizard"
02:31:27 <elliott> Sgeo: ...that's not even profane..
02:31:37 <elliott> pikhq: Even though I don't really want NetBSD coreutils; cp(1) doesn't even have -a!
02:31:50 <Sgeo> elliott, I see that
02:31:55 <Sgeo> When I linked, I didn't
02:32:44 -!- madbr has left (?).
02:34:26 <elliott> Makefile:8: *** missing separator. Stop.
02:34:46 <elliott> All of the Makefiles are like that.
02:35:00 * elliott checks out the full sources to try and find that
02:35:43 <pikhq> Makes everything so much harder.
02:35:58 -!- TLUL has joined.
02:36:54 -!- TLUL has quit (Client Quit).
02:37:13 -!- TLUL has joined.
02:37:57 <elliott> Fuck that, src/ is huge. I'll get the files I need manually.
02:38:08 <elliott> pikhq: The BSD guys should really just reimplement gmake :P
02:38:40 <elliott> [[I transfer a prop from Sgeo, who I can't recall has ever submitted a productive proposal]] --coppro
02:39:00 <elliott> coppro: all my proposals have been productive!! i don't actually recall what i've proposed :P
02:39:27 <Sgeo> I changed some FINE stuff, didn't I!
02:39:31 <Sgeo> Which.. didn't really last
02:39:32 <elliott> pikhq: csh. I'm so reassured.
02:40:31 <elliott> pikhq: Believe this: NetBSD echo(1) does not compile with stock gcc. Not joking.
02:40:41 <elliott> Because I need sys/cdefs.h.
02:41:11 <elliott> pikhq: Why has everyone FORGOTTEN WHAT PORTABILITY IS.
02:41:46 <Sgeo> I got four proposals adopted!
02:42:09 <elliott> pikhq: elliott@dinky:~/nc/src/bin/echo$ cc -D'__COPYRIGHT(x)=' -D'__RCSID(x)=' echo.c -o echo
02:42:09 <elliott> /tmp/ccaV89zK.o: In function `main':
02:42:09 <elliott> echo.c:(.text+0x1f): undefined reference to `setprogname'
02:43:40 <elliott> boot etc kernel LICENSE man test
02:43:40 <elliott> drivers include lib Makefile servers tools
02:43:42 <Sgeo> coppro, you really believe all four of my successful proposals were unproductive/
02:43:45 <elliott> pikhq: Err... where are you seeing their coreutils?
02:43:47 * Sgeo is taking this a bit too seriously
02:43:53 <Sgeo> How do you define "productive"?
02:44:02 <elliott> Sgeo: haven't all your proposals been useless hardening against nonexistent threats
02:44:16 <Sgeo> elliott, only 3
02:44:40 <Sgeo> http://zenith.homelinux.net/assessor/list.php?author=Sgeo
02:44:52 <elliott> elliott@dinky:~/minix/src/tools$ ls
02:44:52 <elliott> chrootmake.sh Makefile release revision
02:44:52 <elliott> issue.install mkboot release.sh tell_config
02:45:00 <pikhq> elliott: ... Or not.
02:45:07 <pikhq> elliott: I had it pulled up earlier.
02:45:13 <elliott> pikhq: that's from the source tarball. did you look at svn?
02:45:22 <elliott> Here be dragons! Finding unexpected bugs and incompatabilities is "par for the course" when tracking current. By and large, you are on your own if anything goes wrong; although, you always can try to ask for help in the newsgroup. Good bug reports, however -- if it comes to that -- are very welcome.
02:45:26 <Sgeo> Internment, Senator-Only Elections, and Indefinite Emergencies are the invasion ones
02:45:28 <elliott> at least coreutils will hopefully not change much
02:45:30 <pikhq> Ah, it's in commands/ on their SVN.
02:45:35 <Sgeo> *Indefinite-length Emergencies
02:46:13 <elliott> $ svn --username anonymous checkout https://gforge.cs.vu.nl/svn/minix/trunk/src/commands
02:46:27 <elliott> Hope it doesn't depend on other files...
02:46:40 <elliott> pikhq: You didn't tell me it was a heap of commands not sorted into core vs. extra :P
02:47:05 <Sgeo> elliott, the prop doesn't transfer
02:47:11 <elliott> cal.c seems relatively sane.
02:47:28 <elliott> pikhq: and cal.c doesn't compile
02:47:29 <Sgeo> Unless it's anothe Sgeo-situation with prop transfers
02:47:41 <Sgeo> coppro sent it to a-d
02:49:18 <pikhq> elliott: Busybox, then?
02:49:39 <elliott> pikhq: bleh. the build system didn't even LET me split it out last time
02:49:45 <elliott> i might have missed something.
02:52:25 <elliott> pikhq: I KNOW! LET'S WRITE OUR OWN COREUTILS
02:53:44 <elliott> [[Work is underway on new options such as "make standalone" to build separate binaries for each applet, and a "libbb.so" to make the busybox common code available as a shared library. Neither is ready yet at the time of this writing.]]
02:54:37 <coppro> elliott: I want a busy beaver
02:54:53 <pikhq> Give me a beaver and I'll put it to work.
02:55:12 <coppro> pikhq: uncomputably so?
02:55:20 <elliott> `addquote <pikhq> Give me a beaver and I'll put it to work.
02:55:24 <elliott> ITS FUNNY BECAUSE INNUENDO
02:55:39 <HackEgo> 258|<pikhq> Give me a beaver and I'll put it to work.
02:58:03 <pikhq> coppro: Well, I certainly don't know how much wood a woodchuck would chuck if a woodchuck could chuck wood.
02:58:30 <coppro> pikhq: more than wood^chuck?
02:58:31 <pikhq> coppro: And I extrapolate this same lack of knowledge to beavers.
02:59:15 <elliott> pikhq: OK, standalone = not implemented at all :P
02:59:55 <Sgeo> coppro, how have I not submitted productive proposals?
03:00:10 <coppro> Sgeo: they were concerned about some random invasion fears
03:00:23 <Sgeo> coppro, only three of them!
03:00:35 <elliott> pikhq: Boot loader: http://busybox.net/~vda/mboot/README.txt :P
03:00:49 <elliott> coppro: i never knew Sgeo was so funny when he was offended
03:00:57 <elliott> tempted to transfer a prop to coppro
03:02:21 <Sgeo> /too obscure for #esoteric
03:03:30 <Sgeo> You know how in Pharo, items on the menus have the letter that you alt-press to do something?
03:04:30 <Sgeo> I wouldn't imagine that all Smalltalk systems have the same UI conventions
03:04:40 <Sgeo> I'd imagine Pharo and Squeak being similar, ofc
03:05:10 <elliott> Sgeo: Squeak has heritage directly from Smalltalk-80.
03:05:20 <elliott> Its UI is -- minus the kiddy colours and the like -- Smalltalk's canonical UI.
03:05:38 <elliott> Sgeo: (Squeak is, literally, Smalltalk-80, thirty years on. Same codebase.)
03:08:35 <elliott> i should totally name kitten releases after kitty emoticons
03:09:06 <elliott> except, rolling release :(
03:10:48 -!- TLUL has changed nick to TLUL|afk.
03:12:10 <elliott> pikhq: Dynamically-linked BusyBox with EVERYTHING except SELinux turned on:
03:12:10 <elliott> -rwxr-xr-x 1 elliott elliott 1.5M Nov 11 03:11 busybox
03:12:26 <elliott> pikhq: Compare to size of gnu equivalents, weep.
03:12:41 <elliott> (dpkg and rpm in one binary -- is this legal? Moral?)
03:13:08 <elliott> pikhq: Look at my god damn commands: http://sprunge.us/XcBf
03:13:18 -!- zzo38 has joined.
03:13:18 <Sgeo> My opinion would be awesome
03:13:25 <Sgeo> But then again, I have atrocious taste
03:14:31 <Sgeo> dpkg and rpm in one... Oh!
03:14:38 <Sgeo> You meant as in busybox
03:14:39 <pikhq> elliott: You turned on *everything* and got a 1.5M busybox.
03:14:54 <Sgeo> I thought you meant you were making a file that was both .deb and .rpm
03:14:55 <pikhq> Moral of the story, Busybox is amazing.
03:15:02 <elliott> pikhq: Technically, I disabled SELinux. Also, I had to manually link it dynamically, since it tried to do it statically and libpam wanted dlopen and friends.
03:15:07 <Sgeo> I was wondering "How is it possible, but if it's possible, it's awesome!"
03:15:21 <pikhq> And a Linux distro in and of itself.
03:15:29 <elliott> pikhq: If you linked that statically to even uClibc, that thing would be a few megabytes more, I'd bet.
03:15:34 <elliott> Then again, libc.a isn't very big, even with uClibc.
03:16:04 <elliott> pikhq: Also, indeed. All you need is a filesystem with /linux, /busybox and a link /sh to /busybox.
03:16:16 <elliott> pikhq: You could even link /init to /busybox instead if you wanted something BLOATED like that.
03:16:21 <elliott> pikhq: (But you'd need more files, I think.)
03:16:44 <elliott> pikhq: Note that a few of the options in busybox may be to *disable* some stuff :P
03:17:00 <pikhq> elliott: Add a cc and a make and you have a build chain.
03:17:02 <elliott> pikhq: Hey, I think that allyesconfig would enable the "tools look for busybox before execing", meaning that /sh would actually work with all the tools, you wouldn't have to type "./busybox".
03:17:09 <elliott> pikhq: Although... it looks at /proc to find itself.
03:18:18 <pikhq> You could IRC with that sucker.
03:20:35 <elliott> pikhq: Although it's an evil, smelly, evil non-Hobbit nc.
03:20:47 <elliott> pikhq: In 2040 I will still be using Hobbit netcat.
03:23:08 <zzo38> I do think netcat is a very good program to include in any system that has internet access, as well as ping and network configuration display. These are the minimum, although an FTP client and a very simple email client should be included too.
03:23:42 <elliott> pikhq: Hey. I should see what toolchain Mastodon uses.
03:24:22 <zzo38> (You can easily use netcat, together with I/O redirection/pipe to download a file from Gopher protocol as well as headerless HTTP.)
03:28:43 <elliott> zzo38: ...headerless HTTP? You mean just the query line?
03:31:00 <zzo38> elliott: Yes I mean just the query line. Headerless HTTP is accessed by omitting the version from the request.
03:31:37 <elliott> zzo38: That doesn't work for e.g. google at the very least./
03:32:36 <zzo38> For Google it still produces a header on output. For my server (Apache) it produces no header on output if that is done.
03:34:02 <zzo38> And I have seen some servers that wait for a header anyways.
03:34:05 <elliott> zzo38: for google it just returns a redirect loop
03:34:39 <zzo38> But this is all wrong; if no version number is given, it should wait for no header and output no header. At least, this is my opinion of what it should do.
03:35:20 <zzo38> (Now; it isn't quite as important for Google; the Google homepage is useless to download outside of a web-browser anyways)
03:35:26 * Sgeo remembers cleaning up Wikipedia vandalism that claimed.. something
03:36:14 <Sgeo> Oh, right, Wikipedia's sucking today
03:39:03 <zzo38> But if you have software packages on there, you ought to support headerless HTTP. (Another alternative is to set up a Gopher server, that is an even simpler protocol than headerless HTTP)
03:39:06 <Sgeo> Dear browser, stop fucking up
03:40:30 <Sgeo> http://en.wikipedia.orrg/w/index.php?title=List_of_HTTP_status_codes&oldid=200015439
03:42:39 <zzo38> Why did you write "orrg" instead of "org"?
03:43:09 <Sgeo> Because I typoed
03:43:20 <Sgeo> I was using my phone's browser instead of my laptop's browser
03:43:59 <Sgeo> My laptop was acting crappily
03:44:39 <elliott> pikhq: maybe i'll just use plan9port :)
03:46:39 -!- wareya_ has joined.
03:49:32 -!- wareya has quit (Ping timeout: 240 seconds).
03:59:03 <pikhq> zzo38: Mmm, HTTP 0.9.
04:01:29 <elliott> pikhq: Now add virtual hosts.
04:03:10 <Sgeo> Isn't that a 1.1 thing? So surely, it wouldn;t work with 1.0 either
04:03:35 <elliott> it's also nearly mandatory
04:05:32 * Sgeo remembers wondering as a kid why is it that putting the IP address as the host of the URL wasn't the same as putting the domain name in some circumstances
04:07:13 <elliott> pikhq: there is a multibyte patch for st but it uses wchat :-P
04:09:27 <pikhq> elliott: Screw virtual hosts.
04:09:32 <pikhq> elliott: IPv6 and virtual machines.
04:10:05 <elliott> pikhq: i don't even have an ipv6 link :)
04:22:29 <Sgeo> elliott, is Ruby strongly or weakly typed? I'm arguing strongly
04:22:59 <elliott> Sgeo: Strongly with a handful of unimportant type coercions.
04:23:26 <elliott> coppro: of course there are so many different aspects
04:23:29 <elliott> strong vs. weak, dynamics vs. static
04:23:35 <elliott> ruby is dynamically strongly typed
04:23:46 <elliott> coppro: if Ruby was weakly typed
04:24:35 <Sgeo> <Sgeo> Strongly typed is orthogonal from statically typed
04:24:35 <Sgeo> <xternal> that's a bit of a religious debate ;)
04:24:53 <coppro> elliott: not necessarily
04:24:55 <Sgeo> (He was arguing that duck typing meant that Ruby is weakly typed)
04:25:09 <coppro> strong is definitely orghogonal from static
04:25:15 <elliott> why do you argue with people like that Sgeo :)
04:25:25 <elliott> coppro: i should write down my definitions sometime, i don't even remember them myself
04:25:32 <Sgeo> Until right about then, he had my respect when it comes to programming
04:28:00 <elliott> (1) he's an idiot but (2) your respect is too easily gained and too easily broken
04:28:29 <pikhq> Sgeo: ^ This. So much.
04:29:36 <elliott> sometimes in acme i move the columns around just to laugh at how nice it is
04:30:14 <pikhq> Yes, but not for that.
04:30:37 <elliott> pikhq: i love pointer warping. am i crazy?
04:31:38 <elliott> pikhq: Software moving the pointer to a location of its choosing without user action.
04:32:02 <elliott> pikhq: For instance, if you move a column in acme, acme warps the pointer to the layout square. On the other hand, you just grabbed and dragged the layout square, so this isn't disorienting.
04:32:22 <Sgeo> Is 2^sizeof(void*) the maximum amount of heap in bytes a C program can use?
04:32:33 <Sgeo> Wait no that makes no sense
04:32:36 <pikhq> That is either a violation of law of least surprise or absolutely amazing.
04:32:43 -!- coppro has quit (Changing host).
04:32:43 -!- coppro has joined.
04:32:46 <elliott> Sgeo: 2^(8*sizeof(void *))
04:32:54 <elliott> Sgeo: although, I think T * can be differently-sized to void *, technically
04:33:10 <elliott> pikhq: Yeah, it has been... abused in the past. (Old X programs used to warp programs to newly created dialogues.)
04:33:18 <pikhq> Sgeo: Also, the above is the maximum amount of memory space in units of char that a C program can use.
04:33:24 <Sgeo> What is sizeof(void*)?... Oh, for some reason, I had it in my mind as 1
04:33:25 <elliott> Sgeo: That actually includes the maximum stack, too.
04:33:30 <pikhq> elliott: Yes, but they must all be castable to void*.
04:33:35 <Sgeo> It's something like 4, isn't it?
04:33:35 <elliott> int x; /* &x is an (int *) */
04:34:03 <elliott> Sgeo: if sizeof(x) = y, then char foo[y] can store exactly one x. basically.
04:34:11 <pikhq> Sgeo: The only restriction is that it's a multiple of char.
04:34:14 <elliott> sizeof(void *) = 8 on 64-bit architectures because 64 bits = 8 bytes
04:34:26 <pikhq> Sorry, a finite natural multiple of char.
04:34:35 <elliott> pikhq: that can fit into a size_t
04:35:04 <pikhq> elliott: Okay, the only restriction is that it's a multiple of char that can fit into a size_t.
04:35:30 <pikhq> And a size_t must have a finite maximum and be integral.
04:35:55 <pikhq> I'm not sure they actually *forbid* signed size_t.
04:36:06 <pikhq> Okay, then natural.
04:36:14 <elliott> pikhq: That is why the signed ssize_t exists on some systems.
04:36:25 <elliott> pikhq: "The value of the result is implementation-defined, and its type (an unsigned integer type)
04:36:26 <elliott> is size_t, defined in <stddef.h> (and other headers)." --C99
04:36:40 <Sgeo> "That is why the signed ssize_t exists on some systems."
04:37:18 <elliott> pikhq: Say... I think standard C without variable arguments might be TC.
04:37:24 <elliott> pikhq: Well, wait, no. With variable arguments.
04:37:37 <elliott> pikhq: Does anything say that stack frames must be addressable?
04:37:48 <elliott> pikhq: Store values by *recursing*.
04:37:51 <pikhq> elliott: All the values on the stack must be addressable.
04:37:57 <elliott> pikhq: To get the value on the top of the stack, return NEXT_FUNC;
04:38:06 <elliott> pikhq: The function underneath will then call NEXT_FUNC(0) or NEXT_FUNC(1).
04:38:17 <elliott> pikhq: Yes, but if you don't use any variables in these functions...
04:38:20 <zzo38> It was also my idea before perhaps C can be Turing-complete if you are not allowed to convert pointers to numbers.
04:38:32 <elliott> pikhq: Then I think it would obey C to have no part of these stack frames be addressable.
04:38:37 <pikhq> ... I don't know if return addresses have to be addressable in straight-up ISO C.
04:38:56 <pikhq> (though it is necessary in POSIX)
04:39:22 <elliott> pikhq: POSIX is TC I think because of the filesystem functions.
04:39:33 <elliott> pikhq: Although, wait, no, char * has necessarily limited range.
04:39:47 <pikhq> elliott: POSIX has bounded file sizes.
04:40:05 <Sgeo> I think I just made a fool of myself
04:40:06 <Sgeo> <Sgeo> C isn't turing-complet
04:40:06 <Sgeo> <Sgeo> complete
04:40:09 <pikhq> elliott: And bounded file names.
04:40:11 <Sgeo> <Sgeo> [stuff]
04:40:15 <pikhq> elliott: And bounded directory tree depth.
04:40:18 <Sgeo> <fando> so? you can push things out of memory and load new things
04:40:18 <Sgeo> <fando> that's your infinite tape
04:40:36 <pikhq> elliott: ... Waait. Chroot.
04:40:41 <pikhq> It's TC because of chroot.
04:40:44 <elliott> Sgeo: Copy this line to him:
04:41:11 <elliott> Sgeo: Yes, but fopen() and friends take a (char *) which necessarily has limited range, thus providing finite page-out space.
04:41:11 <pikhq> You can chroot infinitely.
04:41:24 <elliott> Sgeo: also, where the hell is this?
04:43:15 <zzo38> elliott: But it was my idea you can make a C variant which is Turing-complete by making it that pointers and numbers are not interchangeable with each other. If you do this, it should be possible to have infinite memory? Another way is to use standard C but add two CPU instructions, next bank and previous bank.
04:43:41 <pikhq> zzo38: This would, I presume, make it so that pointers can be unbounded.
04:43:59 <Sgeo> How do you physically do unbounded pointers?
04:44:11 <pikhq> Sgeo: With a Turing machine.
04:44:42 <zzo38> Sgeo: Of course Turing-complete computers are actually impossible, but you can make a programming language specifications which makes a programming language with is Turing-complete.
04:45:00 <elliott> pikhq: I am now attempting to write some C code to simulate some sort of infinite memory with the call stack.
04:45:03 <Sgeo> I meant, the layout of the pointer in memory
04:45:27 <Sgeo> It's a bignum obvi.. well, I just answered my question
04:45:34 <Sgeo> And I don't know how bignums work >.>
04:46:21 <elliott> Sgeo: Consider a list of base 10 digits, that you don't know how long it is. Let's say that "10" means "there are at least two more values".
04:46:32 <elliott> Sgeo: Let's start by assuming there's one value.
04:46:42 <elliott> Sgeo: Now obviously 0-9 are base 9 values now.
04:46:52 <elliott> Sgeo: So 10 1 2 = whatever "12" is in base 9.
04:46:58 <elliott> Sgeo: 10 10 1 2 3 = "123" in base 9.
04:47:03 <elliott> Sgeo: 10 1 10 2 3 = the same.
04:47:07 <elliott> Sgeo: Bignums are like this except way, way smarter.
04:49:14 <elliott> pikhq: Gah, just realised that you basically need arguments to do it.
04:49:31 <elliott> pikhq: Feel free to write an infinite tape/stack/whatever with argumentless, variableless recursing functions :P
04:49:36 <elliott> pikhq: Globals are okay of course.
04:53:01 <zzo38> I read the log file it says GraphicsMagick can use MIFF files. If it can, then GF-Magick should be able to work with GraphicsMagick. There are four ways to do so: [1] Modify "gfmagick.w" to support GraphicsMagick. [2] Make a ".ch" file for supporting GraphicsMagick.
04:53:36 <zzo38> [3] Add 'special "convert"' immediately after 'def init_settings =' in the "gfmagick.mf" file. [4] Write a weapper script that calls GraphicsMagick.
04:59:07 <zzo38> Which if these four ways do you prefer?
05:00:09 <elliott> pikhq: It's irritating that my main problems with getting the Kitten toolchain to work are all because of GNU lock-in.
05:01:11 <elliott> comex: you're either (1) being sarcastic (2) about to be be repeatedly beaten over the head by me
05:01:26 <comex> well, some of the GNU lock-in features are pretty nice features :p
05:01:28 <comex> like linker scripts
05:01:46 * Sgeo wonders if he can make an argument that Haskell isn't strongly typed
05:01:47 <comex> you don't get anything close to the power of those on OS X
05:01:47 <elliott> comex: in this case, the lock-in features are gcc being a bitch with uclibc
05:01:57 <elliott> comex: OH YEAH I'M TOTALLY SUGGESTING OS X AS A REPLACEMENT
05:02:50 <elliott> comex: note: i am not suggesting os x as a replacement
05:02:54 <Sgeo> Haskell is statically-typed. However, data carries no type information with it. Without the static check, data could be misinterpreted as the wrong type
05:03:29 <zzo38> What problems are you having with gcc and uclibc?
05:03:35 <elliott> comex: anyway i don't care about gnucrap as long as programs don't *purport* to be portable and then not work on anything that isn't gnu'd up. but i *do* hate how, starting from an existing, gnuified toolchain, it is very difficult to get anything else due to rampant incompatibility
05:03:38 <comex> clang? that's basically OS X
05:03:48 <elliott> os != toolchain != compiler
05:04:00 <elliott> "clang is sooo much better than windows, it runs all my games"
05:04:05 * comex is only wondering what you prefer to GNU lock-in
05:04:27 <elliott> comex: well in this case it's irrelevant, i'm not objecting to gnu features, i'm objecting to the difficulty in using gcc to bootstrap a non-gnu toolchain
05:04:53 <elliott> zzo38: uClibc-linked executables segfault. works with the gcc that buildroot builds, not sure why at all.
05:07:29 * elliott emails DKS asking about Symbolics shipping
05:08:03 -!- poiuy_qwert has quit (Quit: Leaving).
05:08:48 * Sgeo wants an imperative language with a Haskell-like type system
05:09:34 * Sgeo should relearn Scala
05:09:50 <elliott> Sgeo: that language is Haskell
05:10:15 <elliott> you put do before every function, and in "if" clauses. also, if you do an IO operation that gives a result, you must assign it to a variable before using it in other expressions
05:10:45 <elliott> Vorpal: do you still have that OpenGenera on Linux tutorial?
05:11:01 <zzo38> Is it possible to make a implementation of C standard library that uses assembly language codes and only compiles in the functions that are used by your program?
05:11:37 * Sgeo should sleep soon
05:11:55 <elliott> zzo38: The latter is easily accomplishable, and is done by most C libraries designed for static linking. You simply compile each function (or small group of functions) into its own .o object file, and then pack them all into a resulting .a.
05:12:08 <elliott> zzo38: C libraries not designed for this will often have very big .o files, and thus programs will pull in symbols they do not need.
05:12:16 <elliott> zzo38: The former would be possible but pointless.
05:13:59 <zzo38> I think it is sensible to make at least some of the C library functions written in assembly language. You can have them written in C as well, for using on other computers. And some functions might be written only in C, although someone can write them in assembly language too later on if they want to.
05:17:18 <zzo38> Some of the simple C library functions can be written using whatever instructions are available on the processor, if needed you can also write them separately depending on the optimization setting of the program you are compiling.
05:18:06 <zzo38> Some group of functions might go together so you would then write them in the same module with the entry points inside of each other if necessary.
05:18:18 <elliott> zzo38: But why not write the libc in C?
05:18:59 -!- Sgeo has quit (Read error: Connection reset by peer).
05:19:40 <comex> because C is slow ;)
05:19:47 <zzo38> elliott: I am not saying don't write it in C, I am just saying that you can also write some of them in assembly language for specific computers, too.
05:19:50 <comex> also, I don't know what you guys are talking about
05:19:55 <comex> of course any sane standard library will have assembly functions
05:19:58 <comex> especially for things like memcpy
05:20:19 <zzo38> You can also write all of the function in C for use with computers that the library has not been ported to yet.
05:21:36 <elliott> ;; Leave these last in case server is down!
05:21:36 <elliott> "my birthday" "the day before my birthday"
05:21:36 <elliott> "1 hour before dlw's birthday"
05:21:41 <elliott> --PARSE-UNIVERSAL-TIME, OpenGenera
05:21:47 <zzo38> Will any C compilers compile strings into the executable file such that the following expression is true: ("abcdef"+2=="cdef")
05:22:21 <comex> because you need the null terminator
05:23:16 <elliott> zzo38: hypothetically yes.
05:23:41 <comex> I'm surprised that gcc doesn't
05:24:21 <elliott> pikhq: symbolics is so dormant:
05:24:23 <elliott> [[To be fair to DKS, he is not the owner of Symbolics' property. He is pretty much an independent contractor who is handling folks who need refurbished hardware and hardware maintenance on existing systems.
05:24:23 <elliott> He is not in a position to make decisions regarding Symbolics' intellectual property.
05:24:23 <elliott> The person who could make such decisions, AFAIK, is Andrew Topping who passed away, leaving Symbolics assets tied up in probate.]]
05:24:37 <elliott> let's hope dks will respond to my shipping question :)
05:24:55 <zzo38> Compiling constant strings this way would make the executable file smaller and take up less memory but it should not affect the speed of the running program.
05:25:11 <comex> on the other hand, it's unlikely to crop up very often in real programs
05:25:18 <elliott> I may offer telnet/ssh access to the box if I can get that working (perhaps for a small fee, since it's a single-user architecture and I don't think there's any security -- i doubt someone i know would give me money to trash the box :))
05:25:35 <elliott> hell, X forwarding, let's just go KER-AAAAAAAZY
05:25:44 <elliott> remote X access to something not based on X, it can only go well
05:27:38 <elliott> Voting begins: Sat, 28 Jun 2008 05:53:03 -0700
05:27:38 <elliott> Voting ends : Sat, 5 Jul 2008 05:53:03 -0700
05:27:38 <elliott> x5577 O1 1.0 omd I don't deserve Scamster!
05:27:41 <elliott> comex: were you omd in 2008?
05:29:09 <elliott> "As I recall, the skeleton of Symbolics makes a living through maintenance contracts for ancient government-owned systems."
05:29:27 <comex> also, Hillary Rodham Clinton
05:30:13 <comex> at least Murphy had the grace not to change the caller of http://zenith.homelinux.net/cotc/viewcase.php?cfj=2180 to me
05:32:21 <elliott> comex: your MOM is hillary rodham clinton.
05:35:35 <elliott> http://www.sts.tu-harburg.de/~r.f.moeller/symbolics-info/color.html symbolics symbolics symbolics symbolics
05:46:01 <elliott> comex: symb motherfucking olics
05:51:12 <elliott> Phantom_Hoover encouraged oerjan to commit suicide.
05:52:01 <elliott> (Note: Former statement is 99.9999999999999999% bullshit, 0.zeroes1% gross misrepresentation.)
05:55:01 -!- sftp has joined.
06:02:35 -!- sftp has quit (Remote host closed the connection).
06:28:54 <elliott> "Key Laboratory of Modern Chinese Medicines
06:28:54 <elliott> Department of Traditional Chinese Medicine
06:28:54 <elliott> China Pharmaceutical University"
06:41:48 <pikhq> comex: zzo38: Regarding "abcdef"+2=="cdef". That will not be true for any complying C compiler.
06:42:40 <pikhq> At least, I don't think so.
06:42:52 <coppro> I think constant folding is allowed
06:43:03 -!- Zuu has joined.
06:44:18 <coppro> "cde" == "cde" is certainly unspecified
06:44:33 <elliott> pikhq: it's easy to imagine a C compiler caching strings
06:44:41 <elliott> pikhq: so when it sees "cdef", it looks for that in its current Big Heap O' String
06:44:57 <elliott> finds it (coincidentally, two bytes past where we decided "abcdef" was when we stored it)
06:45:05 <coppro> pikhq: I would veer on the side of suffix reuse is ok
06:45:26 <elliott> bleh, i'm not sure I trust myself to halve the contents of capsules reliably
06:45:43 <elliott> (the only capsules of melatonin this site cells are 3mg and i want to take 1.5mg (I can't swallow pills))
06:45:49 <coppro> elliott: the tricky compiler can merge "a\0b\0c", "a", "b", and "c"
06:46:21 <elliott> apparently one of those has one drop = 1/4 mg, so if i took 6 drops that'd be the dosage i want
06:46:35 <elliott> coppro: but why would you want to? :)
06:47:12 <zzo38> I know it is not required that ("cde"=="cde") or that ("abcde"+2=="cde") but nothing should go wrong if it compiles that it is true
06:48:29 <elliott> zzo38: the c standard is twisty in more ways than you or i can imagine
06:48:37 <elliott> (and in more ways than the committee could!)
06:48:59 <zzo38> And like coppro mentioned, a compiler could possibly even allow ("a\0b\0c"=="a")
06:50:44 <elliott> coppro: in fact... I think you could even have "a"=="b"
06:51:02 <elliott> coppro: consider, the compiler changes some memory-read-semantics register before dereferencing the pointer
06:51:10 <elliott> coppro: although i suppose that should be part of the C pointer...
06:51:53 <elliott> has anyone (apart from Sgeo) taken melatonin here?
06:55:42 <coppro> elliott: hmm... I suppose that could happen
06:56:02 <elliott> coppro: yeah, but I bet you'd trip over other C semantics by not including the register in the pointer
06:56:13 <elliott> coppro: still, if not... HA HA fuck you everyone who thinks that a==b implies !strcmp(a,b)!
06:56:22 <elliott> now to implement this perverse CPU and compiler
06:56:40 -!- Zuu has quit (Ping timeout: 260 seconds).
07:00:05 -!- TLUL|afk has quit (Quit: *disappears in a puff of purple smoke*).
07:10:24 -!- Sasha2 has quit (Read error: Connection reset by peer).
07:16:29 <elliott> "YES THAT IS RIGHT, WE ARE RELEASING GEOCITIES ON A TORRENT." --http://ascii.textfiles.com/archives/2720
07:16:56 <elliott> http://thepiratebay.org/torrent/5923737/Geocities_-_The_Torrent
07:17:03 <elliott> pikhq: You got 688614781452 bytes free?
07:17:11 <elliott> pikhq: Dedicate those 6 and a half gigs to GEOCITIES.
07:19:14 <elliott> pikhq: Technically 909 gigs uncompressed.
07:19:55 <elliott> damn damn damn i want to make a cluster grep this
07:20:08 <elliott> if we could do that in 1996 google would have never been created
07:24:50 <zzo38> What is your opinion of this logo? http://sprunge.us/VPHV
07:26:26 <elliott> I would totally comment except I don't have a single piece of the software needed to display it.
07:29:22 <zzo38> http://zzo38computer.cjb.net/black_associates_logo.png
07:29:57 <pikhq> http://en.wikipedia.org/wiki/United_States_federal_government_shutdown_of_1995
07:30:08 <pikhq> God fucking dammit I hate the Republican party more with each passing day.
07:30:30 <pikhq> It's like they're not even *trying* to do anything but evil pricks.
07:30:47 <pikhq> Though they may also be doing evil pricks, what with the homophobia and all.
07:30:50 <zzo38> elliott: Now can you see the picture?
07:31:05 <elliott> zzo38: Indeed. It's, uh, a very good logo.
07:32:00 <pikhq> Why the hell do my choices keep being between ineffectual and maniacal. Just why.
07:32:31 <zzo38> I had previously made that logo in SVG, but you couldn't see through the holes. This new program is better.
07:33:48 <zzo38> It is better for other reasons too, not just that, though.
07:34:11 <pikhq> elliott: It's likely the GOP will play the same freaking game again this coming session of Congress.
07:34:30 <pikhq> elliott: Only with the debt ceiling instead of the budget.
07:34:31 <elliott> pikhq: Enjoy your meltdown.
07:34:53 <pikhq> elliott: Enjoy the US defaulting on its debt thereby melting down the world economy.
07:35:55 <zzo38> If you want to suggest changes to this logo you can, it is not quite official yet, it is just a draft so far. But the trademark for it still belongs to Black Associates.
07:35:59 <elliott> pikhq: i forget how many predicted meltdowns i've lived through so far, i'm still here :)
07:36:05 <elliott> zzo38: Did you register it as a trademark?
07:36:09 <elliott> zzo38: If not, it's not a trademark.
07:37:12 <zzo38> I think there is both registered and unregistered trademarks, with different rules. (The circled R is for register trademark and TM is unregistered trademarks, I think I read it somewhere)
07:37:22 <pikhq> elliott: I... don't think you realise the consequences of the US defaulting on its debt.
07:37:37 <elliott> pikhq: i predict an event will occur that is not the US defaulting on its debt
07:37:54 <pikhq> elliott: That is dependent on the GOP being sane.
07:38:10 <elliott> pikhq: even the republicans are smart enough to realise that completely, utterly and instantly destroying the US would be bad for their corporatist goals
07:38:25 <pikhq> They think Palin is a nifty candidate.
07:38:29 <elliott> pikhq: and the dems are ~compromising~ enough already, they'll have no problem agreeing to GOP stupidity to keep the US from meltdown.
07:38:48 <pikhq> Okay, true, the Democrats are still spineless bastards.
07:38:49 <elliott> pikhq: you do realise that the actual GOP is not so much *stupid* as *misguided* and selfish?
07:38:55 <elliott> the supporters, sure, stupid. (most dems too)
07:38:56 <zzo38> But I don't know any trademark lawyers though
07:39:54 <pikhq> elliott: It still worries me that a bunch of selfish, sociopathic pricks have the power to cause every currency to look like the Zimbabwe dollar.
07:40:25 <elliott> pikhq: i think you underestimate the power of the EU to sustain itself temporarily
07:40:45 <elliott> pikhq: the euro is backed by many strong countries (ok, some weak ones too) and it'd be very hard to destroy it overnight.
07:41:45 <elliott> pikhq: meanwhile, take a look at the bullshit Facebook does to stop people deactivating their account: http://collison.ie/pics/60b1637dffc83eeb2597be5afe89fe5c.png
07:41:57 <pikhq> elliott: It's partially backed by the US debt.
07:42:03 <elliott> pikhq: "But if you deactivate all these people will be sad? LOOK - here's PICTURES of them! Now you MUST tell us WHY we're deactivating, or we won't let you deactivate."
07:42:23 <pikhq> Hmm. Actually, China would be most fucked outside of the US.
07:42:38 <pikhq> What with having almost a trillion dollars of the US debt backing their currency.
07:42:40 <elliott> pikhq: I don't think the US will default on its debt soon.
07:42:49 <pikhq> ... And being pegged to the dollar.
07:43:19 <pikhq> elliott: All it takes is one Senator.
07:43:59 <pikhq> One completely and utterly crazy Senator.
07:44:21 <elliott> pikhq: now you must of course realise that the consequences of the us defaulting on its debt provide absolutely *no* benefit to *any* US senator
07:44:44 <pikhq> That's why they'd have to be completely and utterly crazy, and not merely stupid.
07:44:51 <elliott> and that while most senators of untrustworthy corporatist douchebags, they're not literally-actually-psych-ward insane in the sense of random acts of arson.
07:45:14 <elliott> and the probability of one suddenly becoming one quick enough to do this is... ~0
07:45:24 <pikhq> Yeah, that's only their supporters.
07:46:00 <pikhq> If we actually got the tea party in charge of things, well. I'd find a nuclear bunker and wait out the next several centuries there.
07:46:15 <elliott> pikhq: the us will probably implode sometime, but i don't think they'll default on their debt :)
07:46:44 <pikhq> elliott: It's just one of the many ways that the US could go poof, and one of the more dramatic.
07:46:53 <elliott> pikhq: after all, what about every other senator and their beloved corporatocracy? i doubt they'd let a single senator jeopardise that
07:47:14 <elliott> pikhq: fake offers of insane amounts of money, threats, coercion of various forms, making sure they're not able to vote... you name it
07:47:22 <elliott> i wouldn't put it past 99% of 'em
07:47:44 <pikhq> More likely ways involve all the corporations just taking their ball and going to $country...
07:48:45 <pikhq> Ooor Americans just growing up.
07:48:49 <pikhq> (hah, like that'll happen.)
07:49:11 <elliott> pikhq: give it some years. i doubt the us can continue to exist in its current state for the rest of your natural life.
07:49:27 <elliott> pikhq: you might want to consider moving elsewhere before the meltdown :)
07:50:10 <elliott> pikhq: So should I buy a Symbolics Lisp Machine?
07:50:19 <pikhq> In its current state? Definitely not. Something has to give.
07:50:25 <elliott> pikhq: (An actual one, not one of those Lisp Machines on a card plugged into a 68k Mac, or an Alpha machine running OpenGenera.)
07:50:49 <pikhq> It's certainly not in a *stable* state right now.
07:51:29 <elliott> pikhq: They are actually remarkably cheap. The cheapest Lisp Machine is $675 including the CRT, keyboard and everything.
07:51:33 <elliott> pikhq: (Yes, *the* Symbolics keyboard.)
07:53:07 <elliott> pikhq: Of course the postage to the UK will be hideously expensive (it weighs 70 pounds, and that's just the computer). So while I'm at it, might as well add some upgrades: $150 would double my disk space to 1.5 gigs, $200 would max out the memory to 8 megawords ($50/megaword), and the 19" premium monitor is $300. (Premium in this case probably means "even vaguely reliable"; 80s CRTs are... fun things.
07:53:07 <elliott> The stock monitor is 17". Both are monochrome.)
07:53:20 <elliott> pikhq: I have emailed DKS asking about shipping to the UK.
07:53:44 <elliott> [[We provide a 90 day Return-To-Factory warranty on all equipment we sell.
07:53:44 <elliott> You are responsible for all shipping costs and there is a $50 crating fee
07:53:44 <elliott> for the XL1200 and 36xx machines, unless you pick up the machine from our
07:53:44 <elliott> location in Burke, VA. Payment is required in advance. We can accept
07:53:44 <elliott> credit card payments through the Payment Pal system with a 3% surcharge (4%
07:53:50 <elliott> pikhq: Sweet, so postage is $50 + postage :P
07:53:55 <elliott> pikhq: (lol @ "Payment Pal")
07:55:09 <elliott> Now will Vorpal wake up so he can link me to that guide?
07:55:12 <Vorpal> <elliott> Vorpal: do you still have that OpenGenera on Linux tutorial? <-- uh, maybe. On a different computer then, which I don't have time to mess with before leaving for university. grep the logs for url if you want it before late this evening
07:55:29 <elliott> Vorpal: wasn't it in /msg?
07:55:46 <Vorpal> elliott, it was probably a sprunge link
07:55:55 <elliott> it was on sporksirc i thin
07:56:07 <Vorpal> elliott, I doubt it. I could ssh in there and check now though
07:56:22 <elliott> Vorpal: that would be great :)
07:56:43 <Vorpal> the only place it would have been does not contain it
07:56:48 <elliott> <Vorpal> hm I'd love to just be able to pull up an inspector on anything, just like in genera, or squeak. Genera more so iirc.
07:56:54 <elliott> Vorpal: ElliottOS does that!
07:57:07 <elliott> $ egrep -r 'Vorpal>.*(open)?genera\b' 10*
07:57:11 <elliott> returns a lot fewer results than i'd expect
07:57:24 <Vorpal> elliott, try adding -i too
07:57:40 <elliott> Vorpal: -i is much slower and returns the same results
07:57:44 <elliott> this is why I used [Oo] [Gg] :P
07:57:46 <Vorpal> elliott, wrt your OS: sure. But well I'd also love if it was available today...
07:58:02 <Vorpal> elliott, I doubt I used openGeNERA
07:58:05 <elliott> yeah, yeah, you can't perfectionify in a day :)
07:58:21 <elliott> Vorpal: oh you may have been Vorpal_
07:58:40 <elliott> $ egrep -r '(Vorpal|AnMaster).*>.*([Oo]pen)?[Gg]enera\b' 10*
07:58:46 <Vorpal> elliott, well, I agree wrt perfectionify. But then it isn't a viable option *today* as such.
07:59:02 <elliott> Vorpal: bleh, didn't you make the guide as vorpal?
07:59:10 <Vorpal> elliott, I have no idea
07:59:20 <elliott> did you ever try opengenera over some X forwarding?
07:59:29 <Vorpal> anyway, must prepare lunch box. bbl
07:59:49 <elliott> 10.07.30:08:26:36 <AnMaster> alise, did you get opengenera to work nicely? I wrote up a somewhat more up-to-date guide for it. Also includes how to install the symbolics X fonts (otherwise small text like in Show Keyboard Layout output is unreadable).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:01:18 <elliott> Vorpal: if you get back, i sure hope debian works instead of ubuntu
08:02:16 <elliott> now what xorg did 7.10 use...
08:06:42 <elliott> Vorpal: don't suppose you can seed opengenera again? i lost the file :p
08:08:59 <elliott> [ ]opengenera2.tar.bz219-Nov-2008 23:01 197M
08:09:08 <elliott> fuck yeah http :P but actually my download just sped up so never mind
08:09:29 <Vorpal> elliott, um. Not on this computer
08:09:42 <elliott> yeah it's ok it sped up :P
08:14:22 -!- atrapado has joined.
08:16:23 <Vorpal> elliott, I just realised that light pollution is a major issue in minecraft XD
08:17:09 <Vorpal> elliott, all that lit outdoors. Not directed only downwards but omnidirectional light sources...
08:17:48 <elliott> it's the SUN! it can't be light pollution!
08:18:00 <Vorpal> elliott, well what about the torches?!
08:18:40 <Vorpal> elliott, I mean the roof is lit to prevent stuff spawning on top at night when I switch away from peaceful
08:18:59 <elliott> oh, so that's how you stop the FUCKING HISSING.
08:19:12 <elliott> in future my shelters are under-fucking-ground.
08:19:45 <Vorpal> elliott, just don't use earth for the roof. lit earth (maybe only with grass, not sure) will spawn cows, pigs, chickens and so on
08:20:10 <Vorpal> well, the sound of them is annoying
08:21:23 <Vorpal> elliott, anyway, just build the roof high enough so stuff can't jump up and make sure it is lit so stuff can't spawn on top
08:21:59 <Vorpal> elliott, monsters during the day too in other words
08:22:13 * Vorpal watches elliott's reaction
08:23:15 <elliott> Vorpal: do you recall what x version 7.10 used
08:24:24 <Vorpal> elliott, no and I can't check without missing the bus :P
08:26:27 <Vorpal> elliott, must be 64-bit
08:26:34 -!- zzo38 has quit (Remote host closed the connection).
08:26:39 <elliott> Vorpal: oh shit i'll have to use qemu
08:26:43 <elliott> Vorpal: well it'll be faster than a 3620!
08:27:12 <elliott> Vorpal: is the setup workable without sudo or should i just install sudo to avoid pain
08:27:34 <elliott> hope that virtualbox bug doesn't apply to qemu too
08:27:57 <Vorpal> elliott, uh, I needed to run the stuff as root. Presumably you don't have to if you figure out a way to let a normal user mess with tun interfaces
08:28:11 <elliott> Vorpal: i just mean, you say to make sure the user is sudoable
08:28:21 <elliott> doesn't the emulator run as root anyway?
08:28:30 <Vorpal> elliott, well, it was on ubuntu. su would have been more work there
08:28:37 <elliott> [[Extract the Open Genera tarball somewhere, you should have a directory og2 from it.
08:28:37 <elliott> It should contain a directory sys.sct, note down the path to it, we will use it
08:28:44 <elliott> WOO! i get to figure out how to move it from qemu to there
08:28:54 <elliott> -fda opengenera.tar.bz2, then dd if=/dev/fd0
08:29:03 <elliott> Vorpal: i have the opengenera2.tar.bz2 file locally
08:29:06 <elliott> Vorpal: i need to get it into qemu
08:29:07 <Vorpal> elliott, also what was the virtualbox bug? I don't remember
08:29:10 <elliott> possibly by pretending it's a floppy.
08:29:16 <elliott> If your NAT the network to the guest, like by default in VirtualBox, then
08:29:16 <elliott> make sure it does not use 10.* for the subnet. VirtualBox uses this by default.
08:29:17 <elliott> There is a bug in VirtualBox 3.2.6 that causes an issue here, see
08:29:17 <elliott> http://www.virtualbox.org/ticket/6176
08:29:17 <elliott> $ VBoxManage setextradata OpenGenera "VBoxInternal/Devices/e1000/0/LUN#0/Config/Network" "172.23.24/24"
08:29:19 <Vorpal> elliott, bloody large floppy
08:29:19 <elliott> Where OpenGenera is the name of the VM.
08:29:21 <elliott> Do this while the VM is powered off.
08:29:27 <elliott> Vorpal: yeah well, second hard drive them
08:29:47 <Vorpal> presumably last virtualbox doesn't either
08:30:25 <elliott> Vorpal: how much disk do i need roughly?
08:30:31 <Vorpal> elliott, using numa for opengenera, already a somewhat brittle setup, is tempting fate
08:30:42 <Vorpal> elliott, don't remember
08:30:43 <elliott> Vorpal: i'm not THAT crazy
08:30:55 <Vorpal> elliott, I probably used 10 GB, dynamically growing
08:31:12 <Vorpal> elliott, but opengenera takes a bit of space
08:31:21 <Vorpal> elliott, I have no idea how much debian takes :P
08:31:27 <elliott> Vorpal: debian base install is small :P
08:31:35 <Vorpal> elliott, well you need X too
08:32:12 <elliott> $ qemu -vga vmware -cdrom debian-40r8-amd64-netinst.iso genera.qcow2
08:32:26 <Vorpal> bbl (not leaving yet, but has to check that I have everything I need in my backpack)
08:34:20 <elliott> Vorpal: hmm, it's probably a bad idea to try and use a non-US locale, right?
08:34:21 * Vorpal secretly replaces qemu with a wrapper script calling bochs on elliott's computer
08:35:43 <elliott> Vorpal: no utf-8 support eh :P
08:45:39 -!- ais523 has joined.
08:48:31 <elliott> ais523: if you want to stop me making a bad purchase that will leave me with an obsolete physical object that i really have no practical, real-world use for, now's the time
08:48:42 <elliott> also, the object takes up space, and by now i mean any time in the next few weeks
08:49:12 <Vorpal> elliott, a pitty there were no "lisptops" XD
08:49:43 <Vorpal> lisptop would be a lisp-laptop
08:50:03 <elliott> ais523: Also, estimates on how much it would cost to ship a large object over 70 pounds in weight from the US to the UK would be greatly appreciated and also help to dispel this ridiculous notion of me buying a Lisp Machine.
08:50:23 <elliott> (Specifically the machine weighs 70 pounds and the CRT weighs something extra. There is also the keyboard and stuff.)
08:51:06 <elliott> "back, i have a degree now"
08:57:14 <elliott> haha, swap partition for a vm
08:57:26 <elliott> wait that actually makes sense.
09:10:39 -!- oklopol has quit (Ping timeout: 264 seconds).
09:12:28 -!- dbc has quit (Ping timeout: 240 seconds).
09:12:56 -!- dbc has joined.
09:14:18 -!- oklopol has joined.
09:20:02 <elliott> ah stanislav does indeed own a 3620
09:20:09 <elliott> http://www.loper-os.org/wp-content/bolix2.png
09:20:11 <elliott> http://www.loper-os.org/wp-content/bolix3.png
09:28:29 -!- Deewiant has quit (Remote host closed the connection).
09:28:34 -!- Deewiant has joined.
09:39:08 <elliott> http://en.wikipedia.org/wiki/Single-level_store
10:06:30 -!- choochter has joined.
10:23:04 -!- Wamanuz2 has joined.
10:23:08 -!- Wamanuz has quit (Ping timeout: 240 seconds).
10:27:48 <cheater99> http://www.youtube.com/watch?v=4iPlhcSZmvg&feature=player_embedded
12:34:04 -!- FireFly has joined.
14:07:30 -!- ais523 has quit (Remote host closed the connection).
14:08:42 -!- MigoMipo has joined.
14:21:33 -!- mycrofti1 has changed nick to mycroftiv.
14:25:35 -!- sftp has joined.
14:46:56 -!- wareya_ has quit (Ping timeout: 240 seconds).
14:54:29 -!- digimunk has joined.
14:55:12 -!- yorick has joined.
15:09:59 -!- ais523 has joined.
15:11:15 <ais523> hmm, Motorola sued Microsoft
15:11:32 <ais523> patent mutually assured destruction isn't just theory now, but actual practice
15:14:02 <ais523> I'm amazed that none of the people involved are lobbying to get software patents repealed
15:14:10 <ais523> surely it must be in the interest of at least one of them?
15:22:10 -!- Sasha2 has joined.
15:28:51 -!- wareya has joined.
15:38:00 -!- nooga has quit (Ping timeout: 245 seconds).
15:47:00 <pikhq> ais523: The lawyers would have to tell management that.
15:47:06 <pikhq> Do you think they're giving up the gravy train?
16:04:23 -!- Phantom_Hoover has joined.
16:07:23 -!- sftp has quit (Remote host closed the connection).
16:23:55 -!- Zuu has joined.
16:45:40 -!- sftp has joined.
16:48:23 <Phantom_Hoover> Am I the only person who thinks MathML is an abomination?
16:48:45 <Phantom_Hoover> There already *is* a markup language for mathematical notation.
16:50:42 -!- Zuu has quit (Read error: Connection reset by peer).
16:50:58 <Phantom_Hoover> 19 lines for ax^2+bx+c, while in LaTeX it's more or less that string.
16:55:22 -!- Zuu has joined.
16:57:11 -!- atrapado has quit (Quit: Abandonando).
16:57:18 -!- augur has quit (Remote host closed the connection).
17:00:17 -!- sftp has quit (Remote host closed the connection).
17:00:45 -!- sftp has joined.
17:22:13 -!- jcp has quit (Ping timeout: 245 seconds).
17:27:12 -!- jcp has joined.
17:29:09 -!- augur has joined.
17:32:48 <cheater99> Phantom_Hoover: don't get me started. all tag-based languages are fucking abominations and need to die.
17:33:57 <cheater99> and besides, have you ever viewed source on a web page?
17:34:16 <cheater99> the text:markup ratio is like 1:10-1:20
17:35:56 <Phantom_Hoover> I assume Berners-Lee had a reason for it, but I'm not sure what it was...
17:48:20 <Vorpal> Phantom_Hoover, it wasn't designed with the modern web in mind
17:49:02 <Vorpal> Phantom_Hoover, it was designed with something like <html><head><title></title></head><body><h1>foo</h1> a few paragraphs of pure text here</body></html>
17:49:10 <Vorpal> in which case it seems somewhat saner
17:49:27 <Vorpal> the head stuff was probably just to future proof it
17:50:02 <Vorpal> Phantom_Hoover, if you mean that it grow out of it's original intentions: well... did you never use netscape 1 or 2?
17:50:03 <Phantom_Hoover> Well, more stuff got lumped onto a framework that wasn't designed to support it.
17:50:28 <Vorpal> Phantom_Hoover, and they reused an existing format. SGML
17:50:43 <Vorpal> SGML was designed for something else
17:54:39 -!- cal153 has joined.
17:57:39 <Phantom_Hoover> http://en.wikipedia.org/wiki/MathML#Example_and_comparison_to_other_formats
17:59:24 <Phantom_Hoover> It gives 4 formats in which the quadratic formula takes a single line of markup and then gives the Presentation MathML code, which takes up most of my browser's vertical space, and the Content version, which requires me to scroll down halfway again just to read the whole thing.
18:00:00 <Phantom_Hoover> You could probably derive the entire formula from scratch in that space with a sane markup language.
18:20:46 -!- cal153 has quit (Ping timeout: 250 seconds).
18:26:30 -!- cal153 has joined.
18:33:11 -!- kar8nga has joined.
18:53:16 -!- kar8nga has quit (Remote host closed the connection).
19:09:23 -!- zzo38 has joined.
19:18:53 <elliott> <Phantom_Hoover> Huh. CLISP is a dependency for AUCTeX.
19:18:58 <elliott> Install SBCL and it won't install CLISP.
19:50:23 <zzo38> After I finish reading the book and make a few more plans, I can make a program implementing TeX (or rather, implementing something similar to TeX), I should call it YeX (where the "Y" is really supposed to be a Greek letter Upsilon). And I shall use Enhanced CWEB to write it. And using fixed-point numbers for glue set values instead of floating-point
19:50:32 <zzo38> What is your opinion?
19:52:32 <zzo38> I have no intention it should pass the TRIP test (I could make it check if the input is the TRIP test and if so, make the output of the TRIP test, but that would be cheating!) but it will pass another test called TRIP'' test (which I have written and tested on MiKTeX already)
20:04:05 <Phantom_Hoover> zzo38, well, if it doesn't use FP senselessly I approve of it.
20:06:55 <zzo38> Phantom_Hoover: What do you mean by "use FP senselessly"?
20:07:55 <Phantom_Hoover> Where you a) don't need very fast calculations and b) need exact precision in your calculations.
20:08:17 <Phantom_Hoover> i.e. most of the time you're doing calculations with subdivided integers.
20:10:10 <zzo38> I intend that identical output cna be produced on different computers. (With TeX, the glue set values are floating point but dimensions are in fixed point.)
20:13:26 <zzo38> s/ cna be / can be /
20:16:20 <Phantom_Hoover> Nor should I have mentioned that O(log_32 n) is the same as O(ln n).
20:19:09 <Phantom_Hoover> Because log_32 n = ln n/ln 32, and O(kf(x)) is the same as O(f(x)).
20:21:50 <Phantom_Hoover> <Chousuke> Phantom_Hoover: but O(ln n) is misleading since it's much faster :P
20:22:48 <Phantom_Hoover> He eventually remembered that big O notation is a measure of algorithmic complexity, not time taken, although not nearly as intelligently as that.
20:25:27 <Phantom_Hoover> Someone else in the channel didn't know what logs were, but I'll give him the benefit of the doubt.
20:26:39 <Phantom_Hoover> More highlights: <amalloy> Phantom_Hoover: in a mathematical perfect world, O(log32 n) would be "the same" as O(log2 n)
20:26:56 <Phantom_Hoover> The fact that this "mathematical perfect world" is the real one escaped him.
20:27:23 <Phantom_Hoover> He then said something I still haven't grasped about O(1) and O(ln n).
20:27:33 <Phantom_Hoover> <Chousuke> it is the same, but not when you execute the algorithm on a machine ;P
20:29:15 -!- Zuu has quit (Ping timeout: 260 seconds).
20:29:20 <Phantom_Hoover> Conclusion: don't let people who think "Lisp in Java" is a good idea near maths.
20:45:00 <Vorpal> fizzie, where is fungot?
20:49:22 -!- fungot has joined.
20:49:28 <fungot> fizzie: nor even by srfi 1, or n-k as ecraven suggested? or is this going to be redefining chord, rather than 7.7.1 ( which is in a language like that *used* to make me right now
20:50:19 <pikhq> Phantom_Hoover: Conclusion, people fail hard at big-O.
20:50:42 <Phantom_Hoover> [["I am not racist, I've been friendly with an Indian for 30 years. I've also been to a Muslim wedding where it was explained to me that alcohol would not be served and I respected that.]] — a Daily Mail quotee, doing what people quoted in the Mail do best.
20:50:55 <pikhq> *Especially* people who believe that Lisp in Java is a good idea.
20:51:55 <pikhq> Also, if you follow up "I am not racist" with "I have a $RACE friend", you might just be a racist.
20:52:21 <elliott> racist is an adjective fffffffffffffffffffffff
20:53:54 <pikhq> elliott: Many many many adjectives are also nouns, interpreted as "an $adjective person"... Sorry.
20:54:41 <elliott> it tends to be used in a prejudicial way, i find
20:54:43 <Phantom_Hoover> Still, not as bad as ais523's condemnation of the use of "damn" as an intransitive verb.
20:56:57 <pikhq> I'm pretty sure that's just a property of pretty much all prejudice being based on arbitrary classifications with simple adjectives.
20:57:25 <pikhq> I doubt there's much prejudice against disestablishmentarians.
20:57:58 <pikhq> Or antidisestablishmentarians.
20:58:39 <olsner> antidisestablishmentarianism is quite hard to type
20:59:44 <Phantom_Hoover> My password was once "ultraneocontraantidisestablishmentarianistically".
20:59:57 <Gregor> <Phantom_Hoover> Whoops, still is!
21:00:22 -!- digimunk has left (?).
21:00:46 <olsner> you'd be a bloody fast typist then!
21:00:52 <Phantom_Hoover> It was for a school computer with an obnoxious password policy.
21:01:39 <Phantom_Hoover> When you have a 48-character password, you get it right *first time*.
21:03:46 <olsner> you kind of missed the point there, but whatever
21:04:00 <pikhq> Phantom_Hoover: Failure. "Ultraneocontraäntidisestablishmentarianistically". You sir are a failure of a human being.
21:04:17 <pikhq> </far_over_the_top_spelling_nazi>
21:05:33 <olsner> damn, writing an os/kernel is a wide project, there are about a hundred of different things I could start implementing now
21:06:04 <olsner> it boots, goes into long mode, can jump between kernel and user mode
21:07:06 <olsner> yeah, it would be... having something that can use that console would also be good :P
21:08:06 <pikhq> I suggest getting a hosted ISO C environment up and running.
21:08:12 <pikhq> (bare minimum for a useful OS and all that)
21:08:33 <pikhq> Or, yes, you could just copulate with C and call it a day.
21:09:27 <olsner> well, as I see it that's all wrapped in a libc that would eventually result in system calls (those system calls are obviously not implemented yet), so really I'm at the stage of getting a hosted assembly environment up and running
21:09:38 <olsner> although not even at that stage
21:10:43 <elliott> (if you do want that, pdclib, anyway)
21:11:17 <pikhq> What, you mean anyone could want anything *but* yet another Unix-alike?
21:12:04 <olsner> regardless of that, the libc will need kernel support
21:12:12 <elliott> even if olsner does writing libc is part of the fun
21:12:32 <olsner> even if magic would be convenient it would ruin the fun
21:12:34 <pikhq> elliott: Newlib's pretty nice for getting things up and running really quickly, though.
21:13:18 <Ilari> Unexpected: AFRINIC apparently got a /8...
21:14:26 <olsner> I think a console driver, and some messages from the kernel, would be a pretty good start... then I can add syscalls for printing/reading from it (at least until I wrap it in whatever my stream/file/object api will look like)
21:15:49 <pikhq> Ilari: Huh. I thought their usage was such that they wouldn't need any more addresses until after IPv4 doomsday...
21:15:51 <pikhq> Phantom_Hoover: Yup, we've got a few more months of IPv4 address space left.
21:16:15 <Ilari> Yeah. 6 more /8s to allocate (plus the 5 /8s to distribute after those 6 are depleted).
21:16:28 <Phantom_Hoover> pikhq, what are the chances of total civilisational collapse afterwards?
21:16:45 <pikhq> Phantom_Hoover: Quite low.
21:17:21 <Phantom_Hoover> It would have been an awesome story to tell to the children around the fire in the ruins of our cities.
21:17:44 <pikhq> Decent chance of being a bit of a painful punch to the economy though.
21:18:02 <pikhq> "What do you mean, we can't take any more clients until we upgrade our 15 year old tech?"
21:18:03 <Phantom_Hoover> "So, granda, how did everything go wrong?" "We ran out of space on the internet."
21:18:27 <Ilari> Basically, there will be 3 more IPv4 allocations from IANA (2 /8s each). Then the 6 are exhausted and rest 5 are allocated immediately.
21:18:37 <Phantom_Hoover> DISCLAIMER: the old man didn't know exactly how IPs worked.
21:19:06 <olsner> ipv6 = the postapocalyptic internet
21:19:55 <Ilari> Each RIR will get one of those 5 blocks. And the previous three allocations would seem to go to ARIN, APNIC and RIPE.
21:20:48 <pikhq> elliott: IANA allocates to the RIRs, and each RIR allocates to entities in their region.
21:21:06 <pikhq> elliott: So, the actual end of IPv4 allocations will be a few months *after* IANA runs out.
21:21:16 <Ilari> Appraently in order: APNIC, RIPE, ARIN, all in first half of next year (unless something unexpected happens).
21:22:49 -!- zzo38 has quit (Quit: zzo38).
21:23:32 <Ilari> This model is estimating that when IANA runs out and the resulting dust settles, there are 22.82 blocks still held by RIRs (plus 3.56 blocks set aside).
21:24:35 <pikhq> And when does the first RIR run out?
21:25:09 <cheater99> olsner: http://www.dangermouse.net/esoteric/intelligentdesignsort.html
21:25:16 <Ilari> In about one year (unless there is run on the bank scenario... If that happens, could be a lot sooner).
21:25:49 <pikhq> So, about one year if we have essentially the same allocation trends...
21:26:06 <pikhq> And we're still not IPv6 ready.
21:26:11 -!- wareya has quit (Read error: Connection reset by peer).
21:26:47 -!- wareya has joined.
21:26:47 <Vorpal> elliott, yellow text (minecraft): "Absolutely dragon-free!"
21:26:49 <Ilari> As said, possibility of "run on the bank" scenario is still there. If that happens, all bets are off.
21:28:21 <pikhq> If that happens, well, maybe some bean counters will start having a heartattack once they realise they cannot grow at all until they do massive upgrades.
21:28:30 -!- wareya has quit (Read error: Connection reset by peer).
21:28:37 <pikhq> Actually, no, they almost certainly will, it's just a matter of when they realise that...
21:28:56 <pikhq> Phantom_Hoover: "Oh shit we're about to run out of IPv4. EVERYONE GET SOME!"
21:29:53 <Phantom_Hoover> I don't see why the heart attacks would be a bad thing.
21:30:42 <pikhq> Everybody should have finished upgrading years ago.
21:31:03 <Phantom_Hoover> Ah, but anything that thins out the accountants is good.
21:31:03 <pikhq> As it is, people aren't likely to start upgrading until, oh, 2012.
21:31:46 -!- wareya has joined.
21:33:04 <Ilari> And there's no telling what is going to happen after IANA announces it has allocated the last block...
21:33:33 <Phantom_Hoover> Fire raining from the skies, cats and dogs living together!
21:34:24 <Ilari> http://www.codinghorror.com/blog/2009/12/our-disaster-recovery-plan.png
21:34:32 <pikhq> Well, "The Internet is full, we need to use a new one" is sure to be a kick in the balls to investors.
21:40:10 <Phantom_Hoover> pikhq, that... that is the best description of the problem ever.
21:45:46 -!- MigoMipo has quit (Ping timeout: 240 seconds).
21:47:16 -!- augur has quit (Remote host closed the connection).
21:48:01 -!- augur has joined.
21:52:17 -!- Mathnerd314 has joined.
21:56:46 -!- jkenj999 has joined.
22:02:36 <olsner> oh right, flat address space, no segment override required
22:04:10 <olsner> tried to write stuff into what will become my kernel data area as [gs:edi] but realized I can just write to the address of it directly
22:04:52 -!- Sgeo has joined.
22:06:08 <Sgeo> Phantom_Hoover!
22:06:14 <Sgeo> They're the same
22:06:20 <Sgeo> Oh wait, that's a 0
22:07:05 <Sgeo> It's like asking which is faster, O(2) or O(1)
22:07:19 <Phantom_Hoover> Yes, you are correct. The people who _wrote_ Clojure got that wrong.
22:07:45 <Phantom_Hoover> Although they thought "Lisp in Java" was a good idea, so...
22:09:06 <olsner> hmm, I need to meditate further on the nature of the kernel stack
22:10:17 <Phantom_Hoover> Lisp as intermediate representation: as stupid as it sounds?
22:11:11 -!- poiuy_qwert has joined.
22:11:21 <olsner> I mean: if I have exactly one kernel stack [per cpu], will that simply work? or would I need a per-process kernel stack?
22:11:29 <Sgeo> I remember discussion on this channel about how s-expressions were originally intended as an intermediate representation
22:11:46 <olsner> I guess if I don't preempt I'll just never leave the kernel with a non-empty stack
22:13:02 <olsner> eventually though, this sucker should be smp capable, just because it's the right thing to do... and that's a bit mindbending
22:13:29 <Phantom_Hoover> SEXPS are actually basically used as an intermediate representation in cl-ppcre.
22:14:24 <olsner> eugh, interrupts get RSP loaded from the TSS, so obviously I need separate "kernel" and "interrupt" stacks
22:15:00 <Phantom_Hoover> This actually means that PPCRE is actually frequently *faster* than Perl's own regexes.
22:15:48 <elliott> Phantom_Hoover: er, cl-ppcre is just a binding, no?
22:16:02 <elliott> wait PCRE = perl compatible regexes wonder what the extra p is for
22:17:02 <Phantom_Hoover> IIRC it has a facility to compile regexes into some Lisp code that does what the regex would do.
22:17:18 -!- Zuu has joined.
22:17:21 <Phantom_Hoover> This can be compiled into native code, while Perl's library just interprets.
22:23:30 <Sgeo> This is the first time I was utterly confused by a Sam Hughes NaNoWriMo
22:24:35 -!- augur has quit (Remote host closed the connection).
22:24:59 <Sgeo> It was... difficult to read, I guess
22:25:20 <Sgeo> For the most part, I had little idea what was going on
22:26:48 <Sgeo> http://qntm.org/puggle
22:27:18 <Sgeo> Um.. according to Sam, something about destroying the world since it didn't end on Dec. 22, 2012 like it should have
22:27:26 <Sgeo> And that saves it
22:30:59 <elliott> Sam Hughes cannot possibly write anything more confusing than "Unbelievable scenes", so shut up.
22:33:56 <elliott> Phantom_Hoover: Now try http://qntm.org/unbelievable.
22:34:10 <elliott> Phantom_Hoover: The best part is, after that one, you can read the REST of Fine Structure! Unlike me, who hasn't read past The Story So Far!
22:35:20 <elliott> Iain M. Banks needs to write a novel with Sam Hughes. That is what I have decided.
22:37:47 -!- cheater99 has quit (Read error: Connection reset by peer).
22:38:15 <elliott> Phantom_Hoover: The latter, yes; Ed stories and most of his standalone fiction, plus I read a good chunk of his old webcomic and god knows what else.
22:38:35 <elliott> Fine Structure, his second novel -- although the Ed stories only became a novel near the end, really -- I have yet to complete.
22:38:47 <elliott> Phantom_Hoover: But, you know, http://qntm.org/unbelievable is the first chapter, so (unless you already have) go read it.
22:39:06 <Phantom_Hoover> Well, I've read everything by Iain M Banks, so I'll tackle Hughes later.
22:39:11 <elliott> Phantom_Hoover: (And no, it isn't just a series of disconnected vignettes.)
22:39:20 <elliott> Phantom_Hoover: http://qntm.org/unbelievable YOU CANNOT RESIST
22:39:24 <Sgeo> elliott, he has a webcomic?
22:39:43 <elliott> Phantom_Hoover: Also, the first two sentences of the second chapter will convince you for sure. (Read the first chapter first, of course.)
22:39:48 <elliott> Phantom_Hoover: (A chapter is a page.)
22:40:34 <elliott> Sgeo: http://qntm.org/files/stickmanstickman/index.php
22:40:54 <elliott> It appears that the actual comic is now down.
22:42:28 <elliott> http://qntm.org/files/stickmanstickman/comics.php But here's some source code.
22:43:16 <elliott> Phantom_Hoover: He definitely has his own (awesome) style, though. SECOND CHAPTER GO GO GO
22:43:29 <elliott> Phantom_Hoover: (Resistance... is... useless!)
22:44:17 <elliott> I hereby appoint him #esoteric's Poet Laureate.
22:44:31 <elliott> Phantom_Hoover: It's only 10:44. Man the fuck up.
22:44:39 <elliott> Phantom_Hoover: (Or read it TOMORROW.)
22:44:48 <elliott> Phantom_Hoover: (Note: It will take multiple days to release.)
22:44:51 -!- cheater99 has joined.
22:44:54 <elliott> Phantom_Hoover: (Note: It will take multiple days to read.)
22:45:23 <elliott> Phantom_Hoover: It is a *long* novel.
22:45:54 <elliott> Phantom_Hoover: http://qntm.org/structure is a TOC. (Some of the chapter titles are vaguely spoilery, so just glance, and then realise that all the ones marked (subdirectory) are multiple full-length chapters in one.)
22:47:01 <Phantom_Hoover> elliott, does he have Banks' habit of leaving you with /that much/ book left and the whole plot at a point where it seems a resolution would take an entire sequel?
22:47:44 <elliott> Phantom_Hoover: Not sure about Fine Structure. I'm find it hard working out whether the Ed stories do that.
22:48:15 <elliott> Phantom_Hoover: Actually, to optimise your enjoyment, you should probably read the Ed stories first, as it's not quite as grand-scale as Fine Structure and if you do it the other way around it'll seem anticlimatic.
22:48:26 <elliott> (http://qntm.org/ed; and yes, the writing does get incomprehensibly better.)
22:49:08 <elliott> Phantom_Hoover: What, even http://qntm.org/robot1? :P
22:49:27 <elliott> IN A WORLD... THAT MAKES SENSE ONLY TO CRAZY SCI-FI BUFFS... [[...so in the end, we - this is me and Ed - said "to hell with moisture detectors" and built the giant robot instead.]]
22:49:46 <elliott> Phantom_Hoover: I'm not sure that's a universal truth, there.
22:51:31 <Phantom_Hoover> elliott, incidentally, thanks for telling me about quicklisp.
22:51:46 <elliott> You're welcome! I do whatever xach tells me to.
22:51:54 <pikhq> http://www.aclu.org/national-security_technology-and-liberty/are-you-living-constitution-free-zone *sigh*
22:52:26 <elliott> pikhq: It would be AWESOME if the Constitution actually said that it applied to all of America except that border.
22:53:03 <elliott> Phantom_Hoover: You know, when *I* used Common Lisp, there wasn't even clbuild.
22:53:11 <elliott> Phantom_Hoover: Do you know what we had? We had ASDF-INSTALL. That was it!
22:53:23 <elliott> Phantom_Hoover: It complained about EVERY SINGLE PACKAGE not GPG-validating.
22:53:33 <elliott> Phantom_Hoover: And it spit out SO MUCH USELESS OUTPUT that you could NEVER tell what to do if it failed.
22:53:35 <pikhq> elliott: It's not that the Constitution says that, it's that the federal government is convinced it can ignore that.
22:53:43 <elliott> Phantom_Hoover: why would you do that
22:54:18 <elliott> pikhq: I'm not sure what you guys like so much about the constitution anyway. The left, anyway; it doesn't exactly provide for much socialism.
22:54:33 <elliott> Okay, violating it isn't good. But still, it's a bit of a crap constitution if you ask me.
22:54:45 <elliott> Phantom_Hoover: * Installing FOURPLAY...
22:54:56 <elliott> Phantom_Hoover: But, uhh, yeah, welcome to McCLIM.
22:55:04 <elliott> Phantom_Hoover: You won't be using it in a day or two.
22:55:24 <pikhq> elliott: Well, it's more that people worship the damned thing while ignoring every single provision of it.
22:55:26 <elliott> http://common-lisp.net/project/climacs/images/screenshots/jpegsinclimacs.png Apparently Lena is "aroused at the new functionality!".
22:55:27 <pikhq> elliott: In particular the bit about how the federal government has *no powers* that aren't granted to it by the Constitution...
22:55:50 <pikhq> De jure, the federal government is incapable of doing almost everything it de facto does.
22:55:55 <elliott> Phantom_Hoover: http://common-lisp.net/project/climacs/images/screenshots/completions.png Look at that modernity!
22:56:05 <elliott> pikhq: I think you guys should replace the Constitution with something saner.
22:56:22 <elliott> pikhq: Like, something that actually lets the government do helpful things. Maybe then people wouldn't get so used to violating it.
22:56:27 <elliott> Phantom_Hoover: Indeed. Lena is UGLY.
22:56:29 <pikhq> elliott: Heck, it'd just be saner if it were just followed and then the state governments actually did things...
22:56:46 -!- Phantom_Hoover has quit (Remote host closed the connection).
22:56:46 <elliott> pikhq: what is the benefit of the US being one country, again?
22:57:02 <pikhq> elliott: Not a whole lot.
22:57:29 <pikhq> It could just be the North American Union.
22:57:35 <pikhq> With the Amero as its currency.
22:58:10 <elliott> pikhq: Or, you know, separate currencies.
22:58:28 <pikhq> elliott: Y'know what the truly awesome thing is about this?
22:58:51 <pikhq> elliott: We could let the hicks suffer under their self-imposed Christian Sharia, and not have to deal with their bullshit any more.
22:59:05 <elliott> Yeah 'cuz they're totally localised :P
22:59:43 <elliott> pikhq: Anyway, separate currencies exist for a reason, even if they are dumbtarded.
22:59:48 <pikhq> elliott: http://upload.wikimedia.org/wikipedia/en/a/ac/PurpleNation.PNG Let the solid red states suffer.
23:00:08 <elliott> pikhq: It doesn't make sense to unify strong economies under one currency; only to unify multiple weak economies, plus a few strong economies. (The Euro is a large-scale example of this.)
23:00:36 <pikhq> elliott: Economic systems need some major refactoring.
23:00:50 <elliott> politics is so fucking boring
23:00:54 <elliott> we should be less boring here
23:01:30 <elliott> pikhq: so did you know that qemu-system-x86_64 is really slow
23:01:36 <Ilari> Yeah... Disturb economic systems enough and they don't bend, they collapse... With bad results.
23:02:08 <pikhq> Yeah, our economic systems are really really fragile.
23:04:22 <pikhq> elliott: Laik, use a better CPU
23:04:39 <elliott> pikhq: laik, intel have decided that laptop cpus don't deserve virtualisation support
23:04:44 <elliott> pikhq: because they're fuckwads
23:04:58 <elliott> my next processor will not be intel :)
23:05:02 <elliott> pikhq: anyway this is just for a lisp machine!
23:05:37 <elliott> pikhq: x86-64 Debian testing emulating x86-64 Ubuntu 7.10 emulating Alpha Tru64 UNIX basically-emulating a Lisp Machine running Genera.
23:13:12 <elliott> Vorpal: night, GENERA LOSER
23:13:15 <elliott> Vorpal: does genera do tcp
23:16:09 -!- Hiant has joined.
23:22:39 -!- Hiant has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]).
23:25:59 -!- augur has joined.
23:27:38 <elliott> pikhq: I am writing Kitten's service manager / process 1 replacement now.
23:27:52 <elliott> pikhq: It is, in fact, perfectly possible to manage it entirely with the filesystem.
23:28:19 <elliott> pikhq: Things like starting and stopping services and other such stuff equates to deleting/creating (blank) files, changing permissions, etc.
23:28:33 <elliott> pikhq: Of course, if you're too WIMPY, you can use sv(8) whose interface I basically stole from runit to do the same thing.
23:31:06 <elliott> pikhq: HOWEVER I do need an opinion on a matter of most importance. Is stopping a service unlinking (whatever the path is) /sv/foo/started, or is it, I don't know, making /sv/foo/run -x?
23:35:08 -!- Mathnerd314 has quit (Ping timeout: 245 seconds).
23:38:20 -!- FireFly has quit (Quit: swatted to death).
23:41:29 -!- oklopol has quit (*.net *.split).
23:41:29 -!- Quadlex has quit (*.net *.split).
23:41:41 -!- oklopol has joined.
23:42:03 <elliott> oklopol klos the pols in o
23:42:09 <elliott> that's why he's the o klo-pol
23:43:11 <elliott> pikhq: if you don't want me to hardcode a maximum path limit NOW'S THE TIME TO STOP MY IRRESPONSIBILITY
23:44:01 -!- sftp has quit (Remote host closed the connection).
23:44:19 <Sgeo> Am I capable of stopping your irresponsibility?
23:44:37 <Sgeo> Also, where, exactly, have I heard the phrase "None, one, or many"?
23:44:45 * Sgeo can't figure it out
23:45:14 -!- ais523 has quit (Remote host closed the connection).
23:46:40 <olsner> I've heard that said as "two is an impossible number"
23:50:26 <Sgeo> http://c2.com/cgi/wiki?ZeroOneInfinityRule
23:50:34 <Sgeo> http://c2.com/cgi/wiki?TwoIsAnImpossibleNumber
23:51:23 -!- Quadlex has joined.
23:52:18 <elliott> hmm, why is there no strmulticat function in C?
23:52:25 <elliott> takes a string and an array of strings
23:53:47 <elliott> it would be ok if strcat at least returned dest + strlen(dest)
23:54:48 <olsner> you need: memcpy, strlen, addition
23:55:58 <Sgeo> "Let's say you're designing an improvement over WIMP"
23:56:09 <Sgeo> This sounds like the Richard guy, although it's unsigned
23:57:34 <Sgeo> Um, some UI concept, I think. Basically the norm for general purpose computers, unless you're using a tiled WM?
23:57:45 <Sgeo> 'In humancomputer interaction, WIMP stands for "window, icon, menu, pointing device", '