00:00:24 <CakeProphet> but I mean... I guess you could just continue having a non-determined x until the program halts
00:00:32 <oerjan> oklopol: the logs are long, and you are not nickpinging me, so...
00:00:48 <CakeProphet> but you'd have to stipulate that /all/ programs halt.
00:00:53 <oerjan> whew, google seems to have removed that background annoyance again
00:01:02 <oklopol> oerjan: i redefined nondeterministic tm's and realized they have an obvious existing definition that's very different.
00:01:42 <oklopol> oerjan: what annoyance, by default there was no background annoyance was there?
00:02:42 <CakeProphet> nondetermistic branch points = parabolas. :D
00:03:13 <oerjan> oklopol: well if it wasn't default i must have triggered it, because earlier today google's front page started fading in background images (somewhat varying ones)
00:03:22 <oerjan> (i use google.no of course)
00:03:22 <CakeProphet> I guess non-deterministic functions are parabolas
00:03:35 <oklopol> non-deterministic functions? relations?
00:03:49 <CakeProphet> ...is that what a relation is technically?
00:04:24 -!- MizardX has quit (Ping timeout: 276 seconds).
00:04:35 <oerjan> "multivalued functions" are isomorphic with relations
00:04:37 <oklopol> a function from A to B is a subset S of AxB such that for each a \in A there's exactly one b \in B such that (a, b) \in S
00:04:44 <oklopol> well exactly one pair for a
00:05:09 <oklopol> in a relation you let there be more pairs for a, and don't require there to be any, so basically just any subset of AxB
00:05:41 <oerjan> although a function is single-valued by default (i.e. a multivalued function is not technically a function, at least not on the same set)
00:06:09 <oerjan> you can consider a multivalued function (and thus a relation) to be a function from A to the _power set_ of B
00:06:26 <CakeProphet> hmmm... might plan an esolang using future conditions
00:06:36 <oklopol> or you could just not assume multivalued function means function that is multivalued, but instead that it's a term.
00:07:02 <CakeProphet> I believe ?!= would have a reversed relationship of its state changes... the else-branch could be determined before halt but the true-branch cannot
00:07:17 <oerjan> oklopol: multivalued functions are important in complex analysis
00:07:40 <oklopol> perhaps you misunderstood me
00:08:02 <oklopol> i'm just saying if you say f is a multivalued X, to me that does not imply it's an X
00:08:02 -!- Oranjer has left (?).
00:08:09 <oklopol> unless that's the convention
00:08:20 <CakeProphet> hmmm... I don't think you can intermingle normal relationship operators with future-relationals
00:09:13 <oerjan> oklopol: well that's what i was saying too
00:09:27 <CakeProphet> well... it might be the same actually... but I don't know how you could implement it on tm
00:09:41 <oerjan> it's a generalization not a special case
00:10:19 <oklopol> can you think of another example where X Y isn't a Y
00:11:33 <oerjan> == division ring, i guess
00:12:23 * pikhq may have a heart attack
00:12:26 <oklopol> oerjan: so maybe we could division ring multivalued functions... how about functional relations?
00:12:41 <CakeProphet> by default the time parameter of the operation is "from current time to time of halt"
00:12:41 <pikhq> ... Did that not turn that into a CTCP ACTION?
00:12:43 <oerjan> oklopol: perhaps also non-associative ring, rings may often be associative by default
00:12:58 <oklopol> ring = associative if * is?
00:13:04 <pikhq> I was unaware such an IRC client existed.
00:13:08 <oklopol> hmm right + is always abelian and nice in every way
00:13:24 <oerjan> oklopol: it's hard to distribute over anything non-abelian
00:13:35 <pikhq> Anyways. Dresden Codak updated... A week... After his previous update.
00:13:45 <pikhq> This is the first time he has ever updated quickly. Ever.
00:14:04 <CakeProphet> what kind of abstract algebra is functions and the composition operator?
00:14:39 <oerjan> oklopol: (a+b)(c+d) = (a+b)c + (a+b)d = ac+bc+ad+bd but also = a(c+d) + b(c+d) = ac+ad+bc+bd
00:14:52 <oerjan> you see you get the middle terms switched for free
00:15:55 <CakeProphet> hmmm... I think the concept I have of future-state determinism is distinct from non-determinism
00:16:02 <oklopol> i'm not sure i see how that's a problem, but interesting point
00:16:03 <oerjan> if you let a=b=1 then that gives you c+c+d+d = c+d+c+d
00:16:13 <CakeProphet> because not all programs can be simulated via a non-halting tm
00:16:43 <CakeProphet> non-halting programs on this kind of machine can't be simulated via tm
00:16:45 <oerjan> oklopol: and if addition is a group it's then automatically abelian
00:17:15 <oklopol> should've seen that coming
00:18:17 <oklopol> i've always found the structures with two operators a bit too complicated for my taste
00:19:21 <oklopol> even though fields seem to be a locally perfect algebra (locally as in the group, ring, field, algebra over field etc family; clearly boolean algebras are the locally perfect algebra in the lattic family)
00:19:34 <CakeProphet> ha... imagine a similar construct with a while loop instead of an if
00:19:56 <oklopol> oerjan: do you agree with this very mathematical statement
00:20:21 <oerjan> oklopol: you can think of rings as essentially the endomorphisms of an abelian group, that's one reason why they tend to pop up i think
00:21:14 <CakeProphet> no you definitely cannot determinize with a while loop and the will-equal operator
00:21:54 <oklopol> so objects = endos, addition means the combined endo "take images and add them", and * composition
00:22:51 <oerjan> oklopol: i have no idea what your very mathematical statement means :D
00:23:21 <CakeProphet> perhaps we can devise a group concerning very mathematical statements
00:23:29 <CakeProphet> to develop an understanding of what oklopol means.
00:24:17 -!- biber has quit (Ping timeout: 276 seconds).
00:24:22 <oklopol> i recall you need commutativity of + to get one of the distributivities to work
00:24:22 <CakeProphet> what other turing machine ideas can I think of...
00:26:16 <oklopol> a*(b+c)x = a(b(x) + c(x)) = abx + acx simply because a is a homomorphism, (a+b)cx = acx + bcx by definition of + in the endo ring
00:27:52 <CakeProphet> hmmm... @ is actualy better if you change ?= to a ternary operator
00:30:06 <CakeProphet> first two operators are the future values to be tested for future equality and the third argument is a set of times represented in execution steps
00:31:46 <CakeProphet> hmmm... the semantics of ?= are not very elegant to simualte on a turing machine
00:31:54 <CakeProphet> they change entirely when there are two variables involved
00:32:43 <CakeProphet> well no... I guess you just "watch" those variables from then on
00:32:59 <oklopol> or maybe it was just that you needed that a is a homomorphism, it could be considered "surprising" that the proofs of left and right distributivity are different, because they look symmetric
00:34:57 <oerjan> oklopol: i think that without commutativity of +, pointwise addition of homomorphisms does not necessarily give a homomorphism
00:35:29 <oerjan> so while the homomorphisms still exist, they don't form a group
00:38:11 -!- FireFly has quit (Quit: swatted to death).
00:40:30 -!- relet has quit (Quit: Leaving.).
00:40:39 <oklopol> so (g+h)(x+y) = g(x+y)+h(x+y) = gx+gy+hx+hy = gx+hx+gy+hy = (g+h)x + (g+h)y
00:41:14 <oklopol> that didn't take me 5 minutes to prove, i was looking for a sleeping bag
00:47:33 -!- sebbu has quit (Ping timeout: 264 seconds).
00:47:41 -!- sebbu has joined.
01:01:52 -!- Oranjer has joined.
01:16:42 -!- augur has joined.
01:34:51 -!- pikhq has quit (Read error: Connection reset by peer).
01:37:29 -!- oklopol has quit.
01:59:36 -!- lament has joined.
02:02:56 <oerjan> i always thought there was something fishy about it
02:04:46 <CakeProphet> What exactly does its namesake print out as a program?
02:05:22 <CakeProphet> Hatchery Oblivion through Marshy Energy from Snowmelt Powers Rapids Insulated but Not Great
02:47:08 <AnMaster> oerjan, odd time for you to be awake during?
02:47:26 <AnMaster> (in my defence I'm doing ubuntu update)
02:47:35 <oerjan> my times to be awake are always odd
02:47:50 <AnMaster> oerjan, usually you stay on CET/CEST though
02:48:43 <oerjan> no. the only reason why you are noticing it now is presumably because _you_ for once are awake.
02:48:49 -!- coppro has quit (Ping timeout: 260 seconds).
02:49:16 <AnMaster> oerjan, my stomach. I'm hungry
02:49:29 <AnMaster> what shall I do? eat during night?
02:49:47 <AnMaster> oerjan, anyway I been awake during this time before
02:50:28 <AnMaster> oerjan, also this is just upgrade to karmic. Tomorrow (_NOT_ tonight!!) waits an upgrade to lucid
02:50:41 * AnMaster hates the ubuntu release names
02:51:23 <oerjan> Alliterophobic AnMaster
03:12:04 <Sgeo_> Attn: Everyone who loves the SCP Foundation wiki:
03:12:23 <Sgeo_> http://scp-wiki.wikidot.com/forum/t-245761/a-sad-day
03:15:30 -!- jcp has quit (Ping timeout: 276 seconds).
03:15:42 <lament> i love how Dr Gears has the audacity to call Fishmonger petty
03:17:22 <Sgeo_> You're active on the wiki?
03:22:36 <AnMaster> okay that's absurd, the bluetooth icon (16x16 or maybe 24x24) in the menu bar turned into about 4x4
03:23:32 <AnMaster> strange are the side effects of upgrade-in-progress
03:29:53 <lament> Sgeo_: no, first time i see it
03:30:37 <GreaseMonkey> although i've taken steps to de-ubuntu my laptop
03:31:49 <AnMaster> GreaseMonkey, can you get the old bluish theme and have a "username" "password prompt"
03:31:59 <AnMaster> rather than showing all user names and letting user click on them
03:32:34 <GreaseMonkey> i've managed to eliminate the stupid loading screens though
03:32:44 <AnMaster> I need that done as soon as possible
03:32:54 <AnMaster> GreaseMonkey, well it is upgraded from jaunty to karmic now
03:34:01 <AnMaster> GreaseMonkey, I want to have text thing. No splash at any point
03:34:40 <AnMaster> GreaseMonkey, I use a stock one currently. Need an initrd anyway
03:34:44 <GreaseMonkey> and then you'll have to get rid of some scripts i think
03:35:04 <AnMaster> well it seemed semi-textual for me
03:35:19 <AnMaster> GreaseMonkey, it was doing kexec to reboot
03:35:44 <AnMaster> GreaseMonkey, not during upgrade for me
03:36:02 <GreaseMonkey> wait actually this is 10.04 i'm thinking of @_@
03:37:11 <AnMaster> GreaseMonkey, right.. I'm planning that after I slept for a few hours
03:39:23 <AnMaster> GreaseMonkey, god dammit I'm going to switch from gdm to kdm or xdm just to get an usable login
03:40:07 <AnMaster> GreaseMonkey, how do I get rid of the mail icon in the notification area?
03:40:10 -!- augur has quit (Ping timeout: 245 seconds).
03:42:19 -!- coppro has joined.
03:50:46 <CakeProphet> hmmm... anyone ever try to make a Prolog quine?
03:54:38 <CakeProphet> nothing in programming is ever really too hard
03:55:40 <oerjan> hm that thing at http://www.nyx.net/~gthompso/self_prolog.txt looks more complicated than i expected
03:56:00 <CakeProphet> it would seem that the query could be used to simplify things somehow.
03:57:40 <CakeProphet> everytime I look at prolog, I have to think about it much more than I should.
03:58:42 <CakeProphet> like constructed trees by describing child-sibling-parent relationships rather than performing commands to build the structure itself.
04:00:21 -!- GreaseMonkey has quit (Quit: HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it).
04:03:33 -!- pikhq has joined.
04:06:15 -!- GreaseMonkey has joined.
04:10:46 -!- bsmntbombdood has joined.
04:11:07 -!- augur has joined.
04:12:57 -!- Mathnerd314 has joined.
04:15:33 -!- jcp has joined.
04:16:50 -!- augur has quit (Ping timeout: 245 seconds).
04:18:16 <oerjan> yay :D http://www.google.com/trends/hottrends?q=remove+google+background&date=2010-6-10&sa=X
04:18:57 <Oranjer> what is the google background
04:19:13 <CakeProphet> can anyone think of a situation in a concurrent design in which you must synchronize two threads to do something at the exact same time?
04:19:32 <CakeProphet> I wonder how you would go about doing that.
04:19:38 <oerjan> the background picture(s) that google's frontpage showed earlier today, and which i complained about
04:20:17 <oerjan> or well i guess it's yesterday
04:31:57 -!- lament has quit (Ping timeout: 260 seconds).
04:36:25 <Sgeo_> Another Radio Linden song in the wild!
04:36:32 <Sgeo_> The Hanks - Once Again
04:48:11 -!- GreaseMonkey has quit (Quit: HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it).
04:50:25 -!- GreaseMonkey has joined.
05:05:42 -!- pikhq has quit (Remote host closed the connection).
05:06:49 -!- pikhq has joined.
05:15:59 -!- Oranjer has left (?).
05:22:46 -!- oerjan has quit (Quit: Reboot).
05:32:54 -!- oerjan has joined.
05:43:22 -!- lament has joined.
06:44:22 -!- CakeProphet has quit (Read error: Operation timed out).
06:53:14 -!- coppro has quit (Ping timeout: 260 seconds).
06:58:22 -!- coppro has joined.
07:06:13 -!- tombom has joined.
07:09:23 -!- Rugxulo has joined.
07:23:07 <Rugxulo> AnMaster never heard of bologna? (pronounced "baloney")
07:24:04 <Rugxulo> "myyyyyyy bologna has a first name, it's O S C A R !!!!!!!!" ...
07:24:14 <Rugxulo> (Oscar Meyer, famous brand name)
07:24:33 <pikhq> He's not American.
07:24:56 <pikhq> And as such has not received our pro-artificial-meat-like-product propoganda.
07:25:20 <oerjan> "Bologna sausage er det nærmeste servelat man kommer i USA."
07:25:28 <Sgeo_> http://scp-wiki.wikidot.com/revised-entry
07:25:44 <oerjan> (i.e. us:bologna ~ very approximately no:servelat
07:30:20 <Rugxulo> yeah, it's just cheap, round, sliced lunch meat
07:30:42 <Rugxulo> usually a mix of several things (turkey, beef, chicken, etc... TRIPE FTW!!!!)
07:31:06 <pikhq> I'd hesitate to call it meat.
07:31:21 <Rugxulo> it's meat, just cheap and somewhat artificially mixed together
07:31:32 <Rugxulo> tastes fine to me, but some don't like it
07:31:43 -!- augur has joined.
07:33:49 -!- Geekthras has quit (Ping timeout: 265 seconds).
07:38:55 * Rugxulo still isn't quite sure he understands what SCP is (logreading)
07:39:52 <oerjan> cooperative fiction, afaict
07:40:28 <pikhq> SCP is amazing, scary-as-hell, cooperative fiction.
07:41:18 <Rugxulo> omg, u banndz me, I well telle mah lawyur awn u!!!
07:41:53 <oerjan> that threat damn well _should_ be empty, since the wiki has a CC license
07:42:19 <Rugxulo> well, the world is crazy enough, that's for sure ;-)
07:44:08 <Rugxulo> amazing how something so useless can be fought over so intensely
07:47:42 <oerjan> http://en.wikipedia.org/wiki/Sayre's_Law
07:50:38 -!- tombom has quit (Quit: Leaving).
07:53:00 -!- Geekthras has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:01:22 -!- pikhq has quit (Remote host closed the connection).
08:05:15 -!- cheater99 has quit (Ping timeout: 272 seconds).
08:12:25 -!- cheater99 has joined.
08:15:13 -!- oerjan has quit (Quit: Good night).
08:32:35 -!- Rugxulo has left (?).
09:09:24 -!- lament has quit (Ping timeout: 264 seconds).
09:27:18 -!- MizardX has joined.
10:16:56 -!- BeholdMyGlory has joined.
10:33:14 <augur> draw spaceships with me
11:09:54 -!- sebbu has quit (Ping timeout: 260 seconds).
11:09:54 -!- sebbu2 has joined.
11:09:59 -!- sebbu2 has changed nick to sebbu.
11:14:32 -!- FireFly has joined.
11:35:25 -!- GreaseMonkey has quit (Quit: HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it).
12:10:36 -!- alissed has joined.
12:11:04 <alissed> I guess that too, but why is it in the topic?
12:11:18 <alissed> BTW, not on my usual box. Remember that unholy netbook?
12:11:33 <alissed> Trying to turn it into a proper Debian install for the first time again.
12:12:00 <alissed> Which means bootstrapping an ARM debian, then making it work with its unholy bootloader. Who wants to help? :P
12:29:13 <uorygl> I looked up "furry" on Wiktionary. One of the definitions it gave was "An animal character with human characteristics; most commonly refers to such characters created by members of the furry fandom."
12:29:25 <uorygl> It had only two translations, but sure enough, one of them was Finnish.
12:33:01 <alissed> And why is this keyboard the devil
12:36:44 <AnMaster> <augur> draw spaceships with me <-- http://conwaylife.com/wiki/index.php?title=Spaceship
12:37:14 <alissed> I appear to be unable to type colons. :
12:37:28 <alissed> AnMaster: Do you want to help me? You like Linux on perverse things, surely?
12:37:48 <AnMaster> if so what is up with the nick
12:38:18 <AnMaster> alissed, as for helping you, maybe. A bit pissed off with upgrading of ubuntu atm
12:38:29 <alissed> I want to try and replace "horribly mangled semi-debian" with "debian".
12:38:44 <AnMaster> alissed, the former being ubuntu?
12:38:48 <alissed> It's an ARM box with some seemingly-custom bootloader, so this should be fun.
12:39:00 <AnMaster> alissed, okay do you have a CD drive in it?
12:39:18 <alissed> It's debian with a bunch of custom software and thinsg such as non-root users & the package manager removed. no CD, just usb: but I am going to use debootstrap
12:39:24 <alissed> So I should not need either.
12:39:37 <alissed> and nothing would install on it by stock. Insane bootloader, you see?
12:40:36 <alissed> 't a base debian isntallation smaller?
12:40:56 <AnMaster> alissed, can you boot an usb stick on it?
12:40:59 <alissed> Well, we can find out, surely.
12:41:09 <alissed> It appears to be hard-coded to boot only one thing.
12:41:18 <alissed> And the method it boots is a mystery to us, or at least it was last time.
12:41:31 <alissed> So I'm going to debootstrap /debian; that will yield a working chroot. I can go from thre...
12:41:50 <AnMaster> alissed, until we figured out how it boots I don't feel like touching the kernel image
12:42:12 <AnMaster> alissed, and this sounds like a PITA to fix
12:42:21 <AnMaster> alissed, okay I think that means it is an initrd
12:42:30 <alissed> Yeah last time i broke it just sent it back and got this shiny reflashified one
12:42:50 <AnMaster> alissed, where is the linuxrc?
12:43:10 <alissed> linuxrc starts with elf header
12:43:55 <AnMaster> alissed, ELF is a generic format
12:44:21 <AnMaster> alissed, it is supposed to be the same for all systems. linux use ELF everywhere basically
12:45:09 <alissed> So, I figure that after I have /debian, I should slowly, manually replace userspace with the Debian version. Then I can think about making a new linuxrc.
12:45:25 <AnMaster> alissed, okay linuxrc seems to be initrd, not initramfs
12:45:46 <AnMaster> alissed, not sure what it is doing in /
12:46:01 <AnMaster> alissed, and not sure how it would work there
12:46:16 <alissed> Well, it has no /boot. Or /root, even. /Desktop :P
12:46:35 <alissed> ircing as root *fuck yeah*
12:46:54 <AnMaster> alissed, what command line tools does it have? The usual set?
12:47:14 <alissed> Limited - but - yes. No file(1), for instance.
12:47:31 <AnMaster> alissed, why are you on that netbook btw?
12:48:14 <alissed> Why not? I'm too tired to lumber upstairs onto the computer,h witso might as well have some fun
12:48:19 -!- MizardX has quit (Read error: Connection reset by peer).
12:48:24 <AnMaster> alissed, where is the actual kernel image btw? linuxrc would be an userspace program
12:48:29 -!- MizardX has joined.
12:48:42 <alissed> No clue, I am searching for it now.
12:48:55 <AnMaster> alissed, maybe they put it in the firmware image or something like that
12:49:14 <alissed> That is possible. If so, I will have to endure their kernel, but nothing more, hopefully.
12:50:28 <alissed> AnMaster: It appears to not be on the FS, so.
12:51:25 <AnMaster> alissed, anyway should be near the top of dmesg
12:52:11 <AnMaster> alissed, I'm not an ARM expert
12:53:02 <alissed> Urgh I can;t paste from terminal
12:53:07 <alissed> Very odd first lines in dmesg
12:53:25 <AnMaster> [ 0.000000] Initializing cgroup subsys cpuset
12:53:26 <AnMaster> [ 0.000000] Initializing cgroup subsys cpu
12:53:26 <AnMaster> [ 0.000000] Linux version 2.6.31-22-generic (buildd@crested) (gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu8) ) #60-Ubuntu SMP Thu May 27 02:41:03 UTC 2010 (Ubuntu 2.6.31-22.60-generic)
12:53:50 <alissed> s3c2410-i2c (repeated twice): iiccon, 000...aa
12:54:24 <alissed> didnt feel like counting, sort of smal address size
12:54:49 <AnMaster> but apart from that I have no clue
12:55:14 <alissed> And that is all dmesgcontains
12:55:20 <alissed> Messgaes from that i2c thing
12:55:21 <AnMaster> alissed, anyway make sure user space works in the chroot, especially glibc being compiled to expect newer kernel might be an issue
12:55:28 <AnMaster> alissed, okay, nothing else at all?
12:55:38 <AnMaster> you pasted that "22.214.171.124-cfs-v19" line from it
12:55:39 <alissed> "Aprtfrom the opening "to DS", no.
12:55:53 <AnMaster> alissed, well, then I guess the dmesg log is filled
12:56:00 <AnMaster> that the stuff from boot is no longer in it
12:56:09 <AnMaster> alissed, dmesg is after all a cyclic buffer
12:56:36 <AnMaster> alissed, a bus for slow devices like temperature sensors and various other things
12:56:40 <alissed> I cannot middle click - or type colon in terminal -- what's the $(echo) incantation?
12:56:47 <alissed> Might be the cpu meter or battery meter then
12:57:04 <AnMaster> alissed, a bit odd that it gives so many log messages but meh
12:57:10 <alissed> 100% cpu usage and i have no idqea why
12:57:23 <alissed> $(echo -e something) or whateverto give a colon
12:57:51 <alissed> Starting top stopped the hogging. >_<
12:58:16 <alissed> Pidgin is using 20% of the shitty cpu lol
13:02:17 <alissed> fffffff i need to redownload debootstrap
13:02:47 <alissed> it wants devices.tar.gz or some thing and i think i need the arm distributtion of debootstrap
13:04:51 <alissed> it uses firefox on a like 400mhz arm with ~0 megs of ram so yeah
13:05:12 <alissed> cool busybox. so free -m does not even work
13:07:43 <alissed> AnMaster: this thing uses softfloats!
13:08:14 <alissed> you think them default non-busybox userland will be ok?
13:09:48 <AnMaster> <alissed> AnMaster: this thing uses softfloats! <-- whoops
13:14:38 -!- MizardX has quit (Ping timeout: 240 seconds).
13:15:44 <alissed> The Debian architecture "arm" is different to "armel".
13:16:20 <alissed> "This is no Debian kernel and not supported by any means since Lenny." re 126.96.36.199
13:16:43 <alissed> that specific old version. not mine
13:17:46 <alissed> will have to extract the deb manually
13:18:54 -!- BeholdMyGlory has quit (Remote host closed the connection).
13:21:06 <alissed> Seems I need MAKEDEV. Anyone know where I could obtain that?
13:22:22 <AnMaster> alissed, this thing has static /dev!?
13:22:51 <alissed> Not necessarily -- but it certainly does not have MAKEDEV.
13:22:56 <alissed> PErsonally, my guess? Yes, static /dev.
13:23:17 <AnMaster> alissed, does it have the mount command
13:23:43 <fizzie> At least this Debian here has a /sbin/MAKEDEV from the "makedev" package.
13:23:45 <alissed> Yes. devpts on /dev/pts, no other dev filesystems.
13:23:55 <fizzie> (Disclaimer: I'm not following the discussion.)
13:23:57 <alissed> fizzie: Right... gotta install that too then
13:28:45 <alissed> My thing does eabi, so, I can use armel.
13:30:44 <alissed> no nano + can't type colon in terminal s so no vi. I sure hope it has ed.
13:31:16 -!- augur has quit (Remote host closed the connection).
13:31:58 <alissed> "yaffs2" root fs. No, EABI. It;s an arm thing.
13:32:40 <alissed> Will makedev work even if my kernel only does static dev?
13:32:54 <AnMaster> alissed, makedev is for static /dev
13:32:59 -!- augur has joined.
13:33:09 <AnMaster> alissed, it creates static /dev nodes
13:33:54 <alissed> Aha! It has makedevs, busybox makedev-esque yes?
13:34:24 <alissed> hmm... can't specify just some devices it seems, like you can with MAKEDEV
13:34:41 <alissed> Well, it creates the special files. but apparently not a specified list
13:35:09 <AnMaster> alissed, debbootstrap needs this on the host?
13:36:06 <alissed> MAKEDEV is not architectufre dependen6t -- a shell script? Yes, AnMaster
13:36:23 <AnMaster> alissed, and yes MAKEDEV is a shell scrip
13:37:37 <alissed> I just killed process 5, hope that wasn't too important
13:38:22 <AnMaster> root 5 0.0 0.0 0 0 ? S May29 0:20 [events/0]
13:38:36 <AnMaster> alissed, here 5 is something that couldn't be killed
13:38:56 <AnMaster> alissed, "uh oh but way worse"
13:41:01 <alissed> if [ "$RANDOM" != "$RANDOM"] <-- Fails in certain edge case (detecting a capable shell) :P
13:43:30 <alissed> colon incantation again plz? Also, makedev
13:46:04 <AnMaster> <AnMaster> alissed, $(echo -ne \\x3a)
13:46:19 <AnMaster> alissed, save it in a text file
13:47:26 <alissed> /c now contains the coloncantation :P
13:51:12 <alissed> it wants me to build pkgdetails.c from source. but i cannot find that file!
13:51:22 <alissed> Okay then, clearly fizzie must now help me.
14:09:08 -!- CakeProphet has joined.
14:15:04 <alissed> CakeProphet: YOU HELP ME THEN
14:23:31 -!- cpressey has joined.
14:26:20 <alissed> CakeProphet: I'm attempting to convert this mutant, sub-£200 Linux netbook into Debian. It's a slow ARM with a custom bootloader and a mangled system. Sound fun..?
14:26:52 -!- alissed has changed nick to somerandomentity.
14:30:17 <somerandomentity> Anyway, I'm stuck at a roadblock in this surgery...irritating.
14:31:18 <CakeProphet> I'm likw an Ubuntu desktop user that knows the file system...
14:31:53 <CakeProphet> actually you want to write code in Erlang with me.
14:33:15 <cpressey> somerandomentity: Right, like I can help. I'm running freakin' Windows too.
14:33:24 <cpressey> somerandomentity: But have you considered NetBSD?
14:34:39 <CakeProphet> for shitty old computers I like xubuntu alright.
14:35:04 <somerandomentity> cpressey: This thing is already running Linux, so it is much easier to make it into Linux.
14:35:34 <cpressey> somerandomentity: Oh. So, you're trying to install a ... bigger Linux?
14:35:38 <somerandomentity> This thing is "500mhz arm with like 3 bytes of ram and a half-eaten debian with a window manager and firefox scrawled on"
14:35:57 <CakeProphet> then yes, you actually want to prototype a rudimentary OS on it
14:36:02 <somerandomentity> cpressey: One that has users other than root... and a working man(1), say, and file(1). And... a FREAKIN' PACKAGE MANAGER.
14:36:05 <CakeProphet> and then use it as a platform for more esoteric designs.
14:36:41 <somerandomentity> Did I mention it has a bootloader that is hardcoded to boot /linuxrc, which is some ELF file? And I think the kernel is IN THE FIRMWARE so you can't replace it.
14:37:09 <somerandomentity> so I'm trying to get debootstrap working to get a debian chroot in /debian, then I'll go frolm there. I've done it before, so I should be able to do it now. Can someone unzip a file for me?
14:37:37 <cpressey> My god, I dunno. That sounds hard.
14:37:45 <CakeProphet> only if you write an Erlang program to transfer it to me
14:38:39 <somerandomentity> http://talk.maemo.org/attachment.php?s=fcb7f96a41c77dcbe428f4048c44cd88&attachmentid=926&d=1202066797
14:39:11 <cheater99> how do you want it, straight download?
14:39:29 <cheater99> http://talk.maemo.org/attachment.php?s=fcb7f96a41c77dcbe428f4048c44cd88&attachmentid=926&d=1202066797
14:39:39 <cheater99> somerandomentity, you're arousing me
14:39:51 -!- kar8nga has joined.
14:39:52 <somerandomentity> It's the attachment here: http://talk.maemo.org/showthread.php?t=16121
14:40:14 <cheater99> somerandomentity, you ask a dangerous question you get a dangerous answer
14:40:49 <somerandomentity> not rapidshare or anything though, i think that would kill the browser on the waiting screen
14:42:31 <cheater99> NOOOOOOOOOOoooooooooooooooooooooooooooooooooooooooooooooooooooOOOOOOOOOOOoooooooooooooooo
14:42:39 <CakeProphet> somerandomentity: so I have a long-term project... when I'm not working on an Android app... of making an Erlang MUD cient. Dunno if you're familiar with MUDs
14:42:42 -!- MigoMipo has joined.
14:42:56 <cheater99> i can't take it any more ^D^D^D^D^D
14:43:54 <cpressey> CakeProphet: A MUD client? I thought it was a server...
14:47:13 <cpressey> somerandomentity: 0x3a apparently
14:47:45 <cpressey> Wow, they didn't give you any way to install nothing on that POC did they.
14:47:56 <cpressey> So much for Linux being open and empowering.
14:48:29 <somerandomentity> Incidentally I can't type the colon i nthe terminal which is why this is even more hilarious
14:56:11 <somerandomentity> cpressey: i've got a working chroot before,, but then i did the surgery of moving it all into the root all in one go, then it refused to boot
14:56:27 <cpressey> I've been reading the Loper OS blog, btw. Stanislav's fundamentally wrong, of course, but it's one of the few blogs I can actually stand reading, for whatever readon.
14:57:05 <somerandomentity> Funny: I think that stanislav's basic os points are right, but his writing is unreadable vitrolic babble.
14:57:25 <cpressey> somerandomentity: Just get an interpreter of some kind on that beast so you can figure out what the ASCII value of a colon is, at LEAST.
14:57:36 <somerandomentity> He seems to be unable to agree with anyone; no camp satisfies him, no matter how similar it is to his own.
14:57:52 <somerandomentity> cpressey: How would I ask it that question if i can't type it?
14:58:05 <cpressey> somerandomentity: It is vitriolic, but at least it's not long winded and pretentious, which is what most others seem to be.
14:58:15 <somerandomentity> Anyway, it is so slow that what it really needsis a very stripped down e.g. ratpoison setup with mostly terminal apps
14:58:37 <somerandomentity> cpressey: Have you seen the recent posts? They are most certainly pretentious.
14:59:41 <cpressey> OK, point taken. His references are atrocious. But have you tried to read Unqualified Reservations?
15:00:40 <cpressey> Thankfully many other OS experimenters are smart enough to not even have blogs.
15:01:10 <cpressey> Oh, so you have. Oh, I'm not alone. I feel relieved.
15:01:37 <somerandomentity> Oh, Mencius Moldbug -- by the way Stanislav is a fan of him -- is not even the source of the Nock post on there.
15:01:47 <cpressey> I should say what I mean about Loper being "fundamentally wrong", but ... in a bit.
15:02:02 <somerandomentity> That's by C Guy Yarvin; his blog is moronlab or sometihng at blogspot. Google Moron Lab, you'll find it. Very ferdw posts, but acceptable.
15:02:56 <somerandomentity> CGY let Mencius post it first for some reason; they are friends too, though in the first Moron Lab post he calls Mencius a weirdgauy and says that frankly he wouldn't read his blog :-)
15:03:09 <cpressey> Wow, for MM not to credit him directly makes him not only insufferably pretentious, but an asshole too.
15:03:43 <cpressey> Still, MM passed it off as his own.
15:04:07 <somerandomentity> Maybe Guy didn't want to be bothered by the kind of people who read Mencius's bog.
15:04:39 <somerandomentity> I like how Mencius sometimes just decidews to dedicate a huge post to justifying pseudo-fascism.
15:04:55 <somerandomentity> Anything is moral if gifted with enough rhetorical questions and length.
15:04:59 <cpressey> Well, he said "Maxwell's eqns, I haz dem", when he could have said "This was submitted by a source who chooses to remain anonymous, and I'm hosting it for them."
15:05:22 <somerandomentity> cpressey: That whole post was written by Guy, and appears on his mMoron Lab post too.
15:05:36 <cpressey> Yeah, haven't had the stomach to look at anything political from that rot yet. And won't.
15:06:18 <somerandomentity> Do talk about how Loper is broken, but... slowly, and in short messsages. Briefl ytoo. Using this thing for real-time communication is a joke :-) Please.
15:07:40 <somerandomentity> It'd be more practical to use post! -- if not for my awful handwriting.
15:08:53 <cpressey> Wanting to program to a sane abstraction is not a bad idea. Wanting to program to the hardware is silly, because it couples you to the hardware.
15:09:27 <somerandomentity> I agree there. I do not think Stanislav wants to do that, though, perhaps I am wrong. He argues for thinner and6 fewer abstraction layers, but not none.
15:09:53 <somerandomentity> Certainly, I do not believehe itnends to write Low level lisp -- he is merely saying that a good architecture would be as close to possible to the intended highest-level language.
15:11:12 <cpressey> He does seem to accept the idea that Loper would be an abstraction layer, but he seems to consider it a necessary evil -- he has expressed that his ultimate desire is to program right on the machine, from what I remember reading.
15:11:48 <somerandomentity> Well, sometimes ultimate desire merely means "in a perfect post-singularity world with faster-than-light travel and practical quantum computing...".
15:11:59 <somerandomentity> In which case, yes, some hardware directly running high-level Lisp code would be nice.
15:12:52 <somerandomentity> My personal ideas for a perfect OS are wildly flitting around different ideas.
15:13:00 <cpressey> Well, it's got me thinking about VMs, anyway. Most of today's VMs are designed to support HLLs (JVM, .NET, etc etc). They're not designed to model hardware.
15:13:47 <somerandomentity> Something that's like Oberon, like Smalltalk, like Plan 9; like Loper, like ooc/cap, like whatever the VPRI are doing right now; like ...
15:13:49 <cpressey> A VM that actually provides a good abstraction for both the machine and for programming, is what I see as the best idea in Loper.
15:14:40 <somerandomentity> still, I seem to have some ideas or combinations of ideas that are quite uniquely mine, so that is encouraging.
15:15:20 <cpressey> Well, let 'em ferment :) And get your crap machine outfitted (sounds like that's successfully in progress) so you can experiment :)
15:15:37 <cpressey> I don't know if I have any really unique OS ideas.
15:15:56 <somerandomentity> Oh heck, experiment on this thing? The keyboard is unusable, the 7" screen almost as much, and it can barely run anything -- and I believe it could only boot Linux with a specific kernel version!
15:15:57 <cpressey> Esolangs, yes. I've got another one in the barn.
15:16:58 <somerandomentity> Aren't you well on 6yoru way to beating zzo? Have you already?
15:17:37 <cpressey> It depends on how you count. But I still think I'm third, behind zzo and Wouter.
15:18:01 <cpressey> But more of mine are implemented than zzo's... and more of my implementations are public than Wouter's.
15:18:04 <somerandomentity> I think one thing I should aim for in my OS is small size. Yes, that's irrelevant today, but you can have a complete, bootable, graphical Oberon distribution with a compiler, TCP/IP stack, web browser, and much else, that fits entirely on a 1.44 meg floppy disk.
15:18:28 <somerandomentity> It's a measure of simplicity, in a way: why do I need this code? Could I not mnake a ismpler layer? a la Forth.
15:19:02 <somerandomentity> If I set myself no constraints, I will .produce an infinitely good OS -- that has infinite system requirements, takes up an infinite amount of disk, and whose release is infinitely prolonged.
15:19:11 -!- Gregor has quit (Quit: Leaving).
15:19:23 <cpressey> I would totally aim for simplicity. That should lead to small size -- if it doesn't, it would be theoretically worrisome, but I guess not worrisome in practice.
15:19:24 <somerandomentity> cpressey: Oh, Wouter. Damn, I forgot how many languages that crazy guy made.
15:19:43 <cpressey> Well, I have more *named* languages than Wouter does :)
15:19:56 <somerandomentity> cpressey: Simplicity is totally subjective, though, so I need some metric thamt models it that is relevant to machines and not just humans. Thus, disk footprint.
15:21:10 <somerandomentity> I also want some sort of grand unifying model of STUFF, not necessarily objects. Definitely do away with disk/ram address spaceseparation.
15:21:42 <somerandomentity> Would be nice tohave some sort of stuff-persistence that can serialise any object in a format transmittable to anyone else. and all instances of stuff should be totally sandboxed from each other, so it would always be safe.
15:22:37 <somerandomentity> And of course, the itnerfce has to be so differnt and so obviously better than everything else.
15:23:00 <somerandomentity> Composable like command-line applications, but rich like graphical applications, and discoverable without using help-files, i.e. self-describing.
15:23:05 <cpressey> somerandomentity: S-Expressions are one possibility for STUFF.
15:23:20 <somerandomentity> Efficient use of space, seamlessly blending all types of stuff and doing away with applications to produce a data-basewd, not application-basewd, intreface.
15:23:50 <somerandomentity> What i am looking for is something like an object model, except possibly not even OO.
15:24:13 <somerandomentity> The whole system would just be a bunch of them, connected together.
15:24:59 <cpressey> Well, when I share something with someone else, I don't, generally, WANT it to be alive.
15:25:05 <cpressey> Or rather, when I am on the receiving end.
15:25:10 <cpressey> Please do not mail me a snake.
15:25:33 <somerandomentity> STUFF is the model for everything, it is like a Smaltlalk object environment -- smalltalk started as an OS!
15:25:43 <somerandomentity> Serialising STUFF should be possible, yes, but that is not what I am trying to find!
15:25:44 <cpressey> Smalltalk still is an OS IMO :)
15:26:10 <cpressey> Like FORTH, it's a language that brings in a whole environment with it.
15:26:19 <somerandomentity> and STUFF being alive should never be a problem: all STUFF would be secure-by-design67, some perhaps cryptogrpahic mechanism to ensure that they are perfectly sandboxed from each other
15:26:43 <somerandomentity> only interacting when the user allows it, explicitly or implicitly -- some how -- and of course I cannot articulate this because I have not found STUFF, or the interface's, true form yet
15:27:45 <cpressey> I'm working on smaller problems :)
15:28:02 <cpressey> My vision in this area is not so broad.
15:28:18 <cpressey> Hm, I wonder if I am a FORTH hypocrite.
15:28:25 <cpressey> I love the language, but I never use it.
15:28:29 <CakeProphet> silly question but I'm trying to read script(4) from erlang man pages
15:28:37 <CakeProphet> but for some reason erlang includes non-erlang manpages
15:28:43 <somerandomentity> I amn the same way ; there is something about Forth that makes ita bhorrent to existing environmentrs.
15:28:46 <CakeProphet> in its man command... and so I get some BSD tool
15:28:58 <cpressey> CakeProphet: I think erl -man just puts the erlang man pages in your manpath
15:29:12 <somerandomentity> Also, extending forth to make it high level from within itself, whioe certainly possible, involves ugly spaces in between words and the like
15:29:22 <CakeProphet> yes, that would seem to be the case... I can man ls for example
15:29:31 <CakeProphet> cpressey: but how do I get script(4) instead of script(1)
15:32:55 <cpressey> somerandomentity: OK, something more semantic than S-Exps. A computer is a piece of hardware, right? Made up of smaller pieces of hardware. So what if the OS is a piece of virtual hardware, made up of smaller pieces of virtual hardware. Then the central abstraction is whatever is common to all these pieces. That would include, at least, a rigorous definition of how they interact.
15:33:10 <somerandomentity> cpressey: With Plan 9 for instance, STuFf is files, except they don't hhave living code in there.
15:33:21 <cpressey> CakeProphet: IMO, no, not at all.
15:34:51 <cpressey> somerandomentity: Oh, just had another thought, but it's slipping away.
15:35:32 <cpressey> I want an OS/VM/Environment/whatever that makes it as easy for me to manually do the things a program does, as it is for the program to do them.
15:36:56 <cpressey> That's one of those LISP machine ideals -- visibility into what's actually going on, to support debuggability.
15:37:04 <somerandomentity> cpressey: Also, IMO, programming should become part of the user interface. The difference between users and programmers is that programmers realise that there's no difference between using and programming. Fare said that.
15:37:22 <cpressey> somerandomentity: Mostly agree, with maybe some small reservations.
15:37:23 <somerandomentity> Ideqally it should be simpler than the prorgamming of today, but then so should programming itself be.
15:37:35 <cpressey> That would be one such reservation :)
15:37:41 <somerandomentity> cpressey: Yes, I definitely think we should not be writing C every time we want to rename some files!
15:37:50 <CakeProphet> hmmm... I don't understand kernelProcess in the boot script.
15:38:36 <CakeProphet> ...so are we talking about OS design here, I take it?
15:39:03 <cpressey> I'm thinking of it as "environment design" atm, but it changes
15:39:03 <CakeProphet> is that in every instance of OS design plans I've seen
15:39:24 <CakeProphet> the progress tends to stop at design discussion
15:39:46 <somerandomentity> We're talking on the very high, theoretical level here. We are not planning to start coding tomorrow.
15:39:48 <cpressey> Ah, well implementation is a bitch, and even if you get that far, adoptation is nigh impossible :)
15:40:10 <somerandomentity> We're basically discussing higher mathematics vs doing accounting by hand.
15:40:14 <CakeProphet> but yeah, if I ever get into an OS design project
15:40:30 <CakeProphet> I will make sure I make a prototype kernel without worrying about any kind of aesthetics.
15:40:46 <somerandomentity> Then you will have made a worthless UNIX-esque clone and helped nothing in the field of OSdesign.
15:41:00 <CakeProphet> the prototype is simply to aid design later
15:41:04 <CakeProphet> because you now have something to work with
15:41:05 <somerandomentity> Hell, any realisation of my perfect OS would contain no c code whatsoever. So that would be hard to do a first-draft of.
15:41:26 <somerandomentity> Ify ou know how systems work youdon't need a kernel to find out.
15:42:13 <cpressey> Well, I for one am just blowin' ideas. I have no plans or hopes of actually implementing them.
15:42:14 <CakeProphet> I'm just saying... a quick prototype helps speed up actual implementation.
15:42:45 <CakeProphet> I suppose it matters less if you have more experience with low-level programming
15:43:11 <cpressey> I am not even thinking of it as an OS right now (I think I already said that) because I KNOW what a huge task it would be to actually build a full-blown OS, and how much of that is uninteresting, like writing device drivers.
15:43:20 <somerandomentity> My perfect os would have vvery little low level programming whatsoever :)
15:43:32 <somerandomentity> You can talk about low-level things, likehardware, in high level code.
15:43:34 <CakeProphet> I don't really think that's possible alise...
15:44:01 <somerandomentity> CakeProphet, I said very little . Not none. And that is no67t even true
15:45:52 <cpressey> I'm thinking of it as an environment, or VM I suppose, because that a) makes it feasible to implement without upsetting this ossified accretion of junk technology we call "state of the art" and b) is much more in line with language design/implementation, which I know I can do.
15:46:11 <somerandomentity> cpressey: ideal hardware is functional machine -- like reduceron, graph reduction mchine.agreed? I mean long-term-ideal
15:46:16 <CakeProphet> there's the "virtual OS"... which makes implementation easier.
15:46:34 <CakeProphet> and then you don't really need to worry about drivers in your OS/environment code.
15:46:48 <cpressey> and I should *probably* be trying to get work done, but when your vendor's service is returning 500 Server Error, and you have no answer from that, or on your side, what can you do?
15:47:12 <cpressey> somerandomentity: Um -- not entirely agreed -- but I can see that viewpoint
15:47:37 <cpressey> I just don't know if it's ideal. It's certainly nicer in many many ways
15:47:41 <somerandomentity> cpressey: what is your idqeal then? also, nothing, you can do NOTHING, except slack off
15:48:03 <CakeProphet> essentially a virtual OS needs to worry about is a) having low-level resources accessible through some kind of high-level service b) process management and scheduling c) anything else it wants to abstract.
15:48:34 <somerandomentity> A real os acheives the true goal, however: slinking off this immortal coil of shittiness
15:48:52 <cpressey> The rabble will always complain that GC and abstraction layers are too slow, of course...
15:49:07 <somerandomentity> cpressey: The worst part of making an os for me is beleieve it or not typography.
15:49:30 <somerandomentity> I'm a typography nut, and designing my own font would just be embarrasingly bad in its result
15:49:57 <cpressey> As for my ideal hardware... probably something like NVRAM FGPAs where I can build new circuits on the fly.
15:50:01 <somerandomentity> The rest is just computer stuff. I'm good at computer stuff, I can implement that...
15:50:37 <cpressey> Typography no problem for me, I would steal the Commodore 64 character set.
15:50:57 <CakeProphet> The entire Styx architecture is in a VM... I think it's a pretty good strategy for clean high-level OS designs.
15:51:45 <CakeProphet> I would just ditch everything is a file.... why would I event want to do that?
15:52:00 <somerandomentity> And interface design is human-computer interaction, information presentation (a la tufte) and typography.
15:52:46 <somerandomentity> You are also indirectly dissing Plan 9. You fail in multitudes!
15:53:16 <CakeProphet> it's just been done... and there's probably a lot of things that don't really make sense as a tree of read/write character streams.
15:53:45 <somerandomentity> CakeProphet: As long as you replace it with some OTHER unified model of STUFF.
15:53:57 <somerandomentity> also, Plan 9 seems to manage fine although its appproach DOES have flaws
15:55:19 <somerandomentity> sTUFF is exactly what it looks like: stuff. You must have a grand unified model of STUFF, be it Smalltalk objects, Plan 9 fgiles, or something of yoru own designing.
15:55:35 <somerandomentity> Not having a grand unified model of STUFF is the definifion of failure: it means you have failed to actually design your OS.
15:55:47 <somerandomentity> Culprits include... every popular OS in existence, and mostof the unpopulra ones.
15:56:15 <CakeProphet> so every kind of data and process interaction must conform to this data representation
15:56:39 <somerandomentity> pretty much -- and hopefully code too. Perhaps not process interaction, it cn be a little bit looser.
15:56:55 <somerandomentity> But you don'tw ant something ridiculous, like 3 types of IPC, both control files AND ioctl() crap, and so on.
15:57:04 <CakeProphet> I would say support for communicate more than byte streams would make sense.
15:57:15 <cpressey> "file" is like Advanced STUFF Substitute.
15:57:43 <CakeProphet> Android has a somewhat interesting IPC model.
15:58:18 <CakeProphet> I googled it and one that I found was "stuff that undermines family fun"
15:59:22 <cpressey> Simple THINGs Unified For Flexibility
15:59:40 <somerandomentity> cpressey: You can't referenceSTUFF in THING, though. It has to be mutually recursive.
15:59:57 <cpressey> THINGS, where the S stands for STUFF ?
16:01:01 <CakeProphet> you can send/receive "intents" between different "components"
16:01:38 <somerandomentity> Oh, and in my not-so-humble opinion, any realisation of STUFF that doesn't have items of STUFF being se\cured from each other by dewsign sucks.
16:01:43 <CakeProphet> the type of the intent determines where it goes. The sending process doesn't specify a specific process to receive the intent.
16:02:28 <CakeProphet> and I think you can pass along a hash table of key-values strings
16:03:10 <AnMaster> somerandomentity, how is the replacement going?
16:03:31 <AnMaster> somerandomentity, I'm still really tired, and family is making food
16:03:32 <somerandomentity> AnMaster: debootstrap is happily twidddling along configuring things\.
16:03:47 <somerandomentity> assumign the chroot works, I will gradually and slowly perform surgery to replace the system with the chroot.
16:04:24 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 188.8.131.52/20100401080539]).
16:04:34 <somerandomentity> Basically, I'm giving a retarded baby a brain transplant by moving all the neurons of a smart brain to it.
16:05:26 <AnMaster> somerandomentity, do not replace the busybox stuff on the host
16:05:51 <somerandomentity> AnMaster: i won't; nor any of its specific things like the apps, or the linuxrc
16:06:32 <AnMaster> somerandomentity, be very careful with these
16:06:41 <AnMaster> somerandomentity, first figure out what exactly linuxrc does
16:07:01 <somerandomentity> hopefully i'll have a non-glacial browser by then so i can do that properly
16:07:04 <AnMaster> somerandomentity, it might depend on calling some specific app somewhere in /usr/bin
16:07:13 <AnMaster> which must be the non-replaced version
16:07:48 <somerandomentity> since they don't seem to have been smart enough to do anything but remove shit
16:09:30 <somerandomentity> AnMaster: i may end up just making the rest of the system fit around linuxrc
16:09:55 <AnMaster> somerandomentity, flash the flash?
16:10:16 <somerandomentity> AnMaster: ok, crazy idea: if i removed almost everything and had debian in a subdirectory, then made it autorun some process as root that dismantld the system and loaded debian, like loadlin?
16:10:27 <AnMaster> somerandomentity, call company and ask for documentation
16:10:39 <AnMaster> somerandomentity, idea: logic probe during boot to figure out what it does
16:11:13 <AnMaster> somerandomentity, reverse engineer linuxrc
16:11:38 <somerandomentity> this thing cost less than two hundred pounds, and the flagship use-interwebs anywhere app uses a gprs connection to a server running ie. it sensd all link clicks and text entries, the serversends back a comrpressed screenshot with link form fiel information
16:12:43 <AnMaster> somerandomentity, what do you mean? firefox does that?
16:13:48 <somerandomentity> it has a gprs connection thta you are only allowesd to use by using their internet browser thing, that connects to one of their servers which runs IE and sends back screenshots of web pages.
16:18:45 <Deewiant> somerandomentity: Yeesh, nick length, dude
16:19:40 <Deewiant> Does the longer/different nick make him less creepy?
16:20:00 <cpressey> somerandomentity: Congrats on working chroot. Topping, what?
16:20:53 <somerandomentity> Deewiant: No, but it makes me some random entity, rather than [redacted]. My disguise, it is flawless.
16:20:55 * cpressey is half-in "Talk like a Wodehouse character" mode these days
16:21:22 <Deewiant> somerandomentity: There are shorter disguises
16:21:49 <somerandomentity> So, chroot works: now the hard part can begin. Major system-wide surgery.
16:23:16 <CakeProphet> I really do not know how to make it stop without restarting my computer
16:24:33 <Deewiant> ctrl-alt-f1; log in; pkill -u $USER
16:25:04 <cpressey> when i was using x, istr that worked to reset it
16:25:07 <Deewiant> That tends to be disabled nowadays
16:25:29 <Deewiant> A couple hours back on an NFS it took 25 seconds
16:26:20 -!- CakeProphet has quit (Remote host closed the connection).
16:27:09 <cpressey> somerandomentity: I can't imagine what would be required to fix the colon thing, and that's the most annoying.
16:27:33 <somerandomentity> Wow, aptitude is taking many seconds and full CPU to start up. Just to update the sources list.
16:27:42 <cpressey> What else is high priority to fix?
16:28:12 <cpressey> Slowness may be incurable too.
16:28:20 <somerandomentity> Basically, what I want to do right now is move over the safest stuff from the debian chroot to /.
16:28:42 <cpressey> Start with a test package, yeah.
16:28:52 <cpressey> Then I would move the package system over, if possible.
16:29:15 <somerandomentity> cpressey: Slowness will not be so incurable, if I stri]p almost all daemons and use the simplest software.
16:29:57 <cpressey> ls is pretty simple. must be a lot of daemons on that thing
16:30:03 <somerandomentity> and took several seconds first time due to slowest, disk. EVER
16:30:20 * cpressey suspects some kind of extremely cheap, slow disk
16:30:36 <cpressey> somerandomentity: any USB ports?
16:31:18 <cpressey> and links, and lose X, and that's a lot more cpu
16:33:24 <somerandomentity> Also, I am pretty sure this thing has the kernel tied into the bootloader, so yeah.
16:33:42 <cpressey> oh, you mean the chroot they supply?
16:34:40 <AnMaster> somerandomentity, devfs is dead
16:35:04 <AnMaster> somerandomentity, devfs is no longer supported by the kernel iirc
16:36:10 <AnMaster> somerandomentity, you don't put a < there
16:37:50 <AnMaster> somerandomentity, figure out device number and so on
16:38:37 <AnMaster> somerandomentity, are you on monospace font? I can't show you how to find device node info otherwise
16:39:10 <somerandomentity> I can find it with ls, I know. But will mknod persist to ddisk?
16:39:24 <AnMaster> brw-rw---- 1 root disk 8, 0 29 maj 19.54 /dev/sda
16:39:24 -!- iamcal has quit (Ping timeout: 264 seconds).
16:39:33 <AnMaster> the m in each is in the corresponding place
16:39:38 -!- lament has joined.
16:40:00 <AnMaster> somerandomentity, I told you. Switch to monospace font
16:40:05 <AnMaster> I won't help you any more than that
16:40:37 <somerandomentity> How about you just tell me which number is which? I don't care what you want me to do, I don't even know how to make Pidgin switch fonts and i won't install another client because I have no means to. So I will just ask someone else.
16:41:19 <AnMaster> cpressey, can you switch it to monospace there?
16:41:29 <cpressey> i pasted it into scite and monospaced it
16:41:47 <AnMaster> somerandomentity, you said you didn't know
16:41:49 -!- MigoMipo has quit (Quit: Quit).
16:42:01 <cpressey> unless you mean "8," is major and "0" is minor
16:42:29 <AnMaster> somerandomentity, the third number is date
16:43:06 <cpressey> which in your language has a month named "maj", just to throw in a red herring.
16:43:24 <AnMaster> cpressey, I didn't think of that
16:43:45 <AnMaster> cpressey, from when this system either was booted or when it was installed
16:43:53 <AnMaster> cpressey, since both happened in may last time
16:44:00 <AnMaster> somerandomentity, they are block devices
16:44:09 <AnMaster> somerandomentity, as indicated by brw-rw----
16:44:14 <AnMaster> somerandomentity, in my example
16:44:25 <AnMaster> somerandomentity, first letter b = block device
16:44:43 <AnMaster> then there is something for fifo and unix sockets
16:45:30 <AnMaster> somerandomentity, not surprising at all
16:45:38 <AnMaster> somerandomentity, the former will from the initramfs
16:45:56 <AnMaster> somerandomentity, which is then overlayed by the real /
16:46:46 <AnMaster> somerandomentity, but it never said squashfs there
16:47:01 <somerandomentity> HAH! Lies! Lies, from the Sea of Mount! It tells me, upon its invocation, that verily /dev/root is mounted; but lies! Lies, for it exists not!
16:47:25 <AnMaster> somerandomentity, /dev/root is special I think
16:47:46 <AnMaster> somerandomentity, like kernel makes it up for initrd
16:47:46 -!- CakeProphet has joined.
16:47:53 <AnMaster> somerandomentity, or some such
16:48:01 <somerandomentity> "Special device /dev/root does not exist," speaks chroot! Woe! WOE!
16:48:26 <AnMaster> I'll go elsewhere until you figure it out
16:48:52 <CakeProphet> use /dev/loop to make /dev/root with black magic. :)
16:48:53 <AnMaster> somerandomentity, no I think you are high
16:49:42 <somerandomentity> AHA! /dev/mtdblock2! It masquerades but it is the file system I desire!
16:49:48 <cpressey> it depends on wtf you want to do with this poc
16:49:59 <somerandomentity> What is this ludocracy, such ludicrouslousness that it contains all the files, not just the linuxrc! LIES, fstab! LIES!-----
16:50:20 <somerandomentity> but -- still -- what packages do I need for squashfs support? Could someone google? Internet on this is pain
16:50:29 <AnMaster> somerandomentity, yes I wouldn't trust fstab
16:50:31 <somerandomentity> cpressey: get it running some minimalist jazz and just sorta play with it
16:50:36 <AnMaster> somerandomentity, apt-cache search squashfs
16:50:54 <cpressey> somerandomentity: Did you ever answer: does it have a USB port?
16:51:04 <AnMaster> somerandomentity, aptitude then
16:51:06 <CakeProphet> somerandomentity: run it in background. :P
16:51:21 <AnMaster> somerandomentity, synaptic? XD
16:51:50 <cpressey> I'd plug in an external drive, flash even, and turn off the software running on it, but not try to mess with the fs it's on. too risky.
16:52:45 <AnMaster> okay that would be hilariously slow on that shit
16:53:11 <AnMaster> cpressey, yes that is what emerge is from
16:53:33 <AnMaster> CakeProphet, w3m-mode in emacs!
16:53:37 <somerandomentity> Lo, some wise traveller, please speak unto me these things ; such that by their using I could utilise my squashed file system on Debian ; these packages.
16:54:39 <EgoBot> http://google.com/search?q=Googol-faced+prophet
16:55:54 <CakeProphet> maybe Ted Stevenes was right about the internet.
16:56:38 -!- coppro has quit (Quit: I am leaving. You are about to explode.).
16:57:42 <CakeProphet> somerandomentity: test youtube: http://www.youtube.com/watch?v=13HM-bmKW2U
16:59:45 <cpressey> What did you expect, sanity and consistency? Pshaw.
17:00:14 -!- ais523 has joined.
17:02:21 <ais523> nah, obviously I wasn't paying attention
17:02:50 <somerandomentity> but anyway, its system is a horribly mangled debian, and it forces me to irc and internet as root
17:02:52 <ais523> anyway, can we have a better topic?
17:03:05 <ais523> oh, I'm starting to remember now
17:03:16 <ais523> and it explains the ~root@ in your whois data
17:03:51 <somerandomentity> Anyway, this is the state of play. Do you know which squashfs package I want? There are multiple for some reason, for different systems.
17:04:35 <ais523> and I have no idea which squashfs package is better than which other
17:05:05 <cpressey> This all makes me want my own atavistic netbook that I have to try to rehabilitate.
17:05:28 -!- tombom has joined.
17:05:30 <somerandomentity> cpressey: this thing is under 200 pounds, so you can have your enjoyment for only many weeks worth of food
17:05:54 <cpressey> It's a netbook, I damn well hope it weighs less than 200 pounds! HAHAHA
17:06:11 <ais523> cpressey: bad puns are oerjan's job!
17:06:18 <ais523> besides, he's better at them
17:06:44 <ais523> I should really get around to putting some of my esolangs up on the wiki...
17:06:55 <ais523> and implementing them, and defining syntax
17:07:06 <AnMaster> <cpressey> It's a netbook, I damn well hope it weighs less than 200 pounds! HAHAHA <-- ?
17:07:13 <cpressey> ais523: Yes, you should. You might, in fact, be 4th.
17:07:37 <cpressey> In terms of sheer number of designs.
17:07:43 <ais523> AnMaster: "pound" = "GBP/pound sterling", a unit of currency, and also an imperial unit of weight
17:07:57 <ais523> I was answering your ?
17:08:11 <AnMaster> ais523, I don't get the joke still
17:08:21 <ais523> `calc 200 pounds in kilograms
17:08:38 <ais523> `google 200 pounds in kilograms
17:08:40 <AnMaster> cpressey, I don't do imperial units
17:09:44 -!- pikhq has joined.
17:10:16 -!- cal153 has joined.
17:11:03 <ais523> hmm, a spambot pretending to be a girl that fancied me
17:11:21 <ais523> but it was obviously a spambot, due to hiding the To: line and BCCing me, and putting about fifty newlines between every line
17:13:47 <pikhq> Who is this random entity?
17:13:59 <pikhq> And no, I have never installed Linux upon an ARM.
17:14:15 <somerandomentity> Entity number too big to list. Please request more specific information.
17:14:31 <pikhq> I shall guess a one Elliot Hird.
17:14:53 <ais523> pikhq: stop accusing random people of being ehird
17:15:00 <ais523> I know it's a national sport for this channel, but still...
17:15:37 <ais523> heh, I missed the misspelling
17:15:59 <ais523> nah, I realised it was you
17:16:06 <ais523> but for some reason I thought you wanted to pretend to not be you
17:16:17 <ais523> the signs were all there, after all
17:16:26 <ais523> even the IP address starting in 9, which I use as a sanity check
17:16:58 <Deewiant> Do you remember everybody's IP address's first octet?
17:17:14 <ais523> but I whois somerandomentity so often that it would be hard not to remember
17:17:33 <AnMaster> <somerandomentity> hmm, AnMaster may be autistic <-- no
17:18:07 <AnMaster> possibly everyone in here has some traits of it though
17:19:14 <ais523> AnMaster: pretty much everyone in the universe does
17:20:59 <somerandomentity> I have Debian in a chroot, and a horribly mangled Debian outside with some linuxrc crazy bootloading thing ran by a bootloader that seems to keep the kernel in firmware.
17:21:18 <somerandomentity> pikhq: I wish to perform surgery to extract the Debian as much as possible to the outer system, and remove the outer system.
17:23:20 <CakeProphet> I don't think this black magic is going to go anyways.
17:24:10 <cpressey> Netbook, I'm going to reprogram you with a rather large axe, got it?
17:25:08 <somerandomentity> what is so strange about changing this debian for a new debian?
17:26:59 <ais523> cpressey: H2G2 misquote?
17:27:14 <pikhq> somerandomentity: chroot into your chroot, mount the old root. Remove all the outer system. debootstrap.
17:27:45 <pikhq> Yes, but not the outer system.
17:27:52 <somerandomentity> I cannot erase the outer system completely, for its linuxrc and prorgams and firmware containing kernels are much insane.
17:28:12 <somerandomentity> I can only migrate the userspace stuff to this Debian piece-by-piece.
17:28:13 <pikhq> One can retain the linuxrc and the kernel; debootstrap shan't install those, anyways.
17:28:31 <somerandomentity> Listen, last time I} replaced stuff it couldn't mount various filesystems or something and trefusewd to boot.he bootloader
17:29:12 <pikhq> rsync from inner to outer, with that option to make rsync not delete things?
17:29:32 <somerandomentity> And retain all the vicious insanity prese\nt in this system? I intend to do it quite manually.
17:30:10 <ais523> wow, SCO vs. Novell just finished altogether
17:30:17 <ais523> all the remaining points to decide were ruled in Novell's favour
17:31:00 <ais523> somerandomentity: the thing is, they already appealed and won the appeal
17:31:18 <ais523> so if they want to appeal again, they'll have to argue the opposite of what they argued last time
17:31:23 <ais523> admittedly, I think they might actually do that, but it would be hilarious
17:32:45 <AnMaster> <ais523> wow, SCO vs. Novell just finished altogether <-- are SCO going to try to get that overruled?
17:32:52 <cpressey> ais523: "misquote" sounds so ugly. I prefer "riffing on"
17:33:11 <ais523> AnMaster: they haven't said they will yet
17:33:17 <ais523> that doesn't mean they won't, ofc
17:33:46 <AnMaster> <ais523> so if they want to appeal again, they'll have to argue the opposite of what they argued last time <-- ? what why?
17:34:21 <ais523> AnMaster: basically because their last appeal, plus the current trial, adds up to "based on SCO's own arguments, SCO loses"
17:34:54 <ais523> AnMaster: there is no remotely sane way that SCO can get out of this one
17:35:01 <ais523> I'm looking forwards to whatever completely insane way they try
17:35:04 <AnMaster> ais523, but are they going to try?
17:35:10 <ais523> AnMaster: I have no idea
17:35:32 <ais523> cpressey: they're no longer a credible threat at this point, but they do provide good entertainment
17:37:25 <AnMaster> "While some of you are no doubt perfectly comfortable with solving second order differential equations in order to understand a joke in a webcomic, I'm going to assume that most of you would rather hear the good stuff. "
17:37:36 <AnMaster> from the annotation on irregular webcomic today
17:40:18 <AnMaster> somerandomentity, do not reboot while it is that way
17:40:33 <cpressey> somerandomentity: is /usr/bin built statically?
17:40:59 <cpressey> if /usr/bin loads so's from /usr/lib you could be quite fucked
17:41:01 <AnMaster> somerandomentity, because linuxrc might need it.
17:41:30 <AnMaster> somerandomentity, different versions
17:41:44 <cpressey> well, that sounds like stuff you could afford to lose. but still
17:42:03 <cpressey> live recovery disk, if you have one, would be really nice
17:42:35 <somerandomentity> I can't stay stopped forever for the fear that linuxrc won't work.
17:42:50 <somerandomentity> Like a puzzle, you know? But with tangible rewards and a realistic setting.
17:43:16 <cpressey> somerandomentity: Do you have some sort of "re-install the operating system" function on that thing?
17:44:02 <cpressey> Darn. Was hoping for an alt partition with compressed install disks or soemthing. My old laptop at home has one of those
17:44:23 <cpressey> I can only say what I would do
17:44:53 <AnMaster> somerandomentity, do not replace, add
17:45:07 <AnMaster> somerandomentity, as in, add the missing parts, but don't touch existing
17:45:24 <somerandomentity> Well, there is more present that there should not be, than is absent but should be present.
17:45:26 <cpressey> #1 would be to have a live usb thing you can boot off of, in case you fux0r sh1t up
17:45:46 <ais523> does it have a serial port?
17:45:54 <cpressey> #2 would be to get rid of X, for now, with plans to re-install it sanely later
17:45:57 <ais523> if so, you could try getting the serial console working on boot
17:46:27 <AnMaster> somerandomentity, hack the boot loader?
17:46:33 <cpressey> somerandomentity: Forgot. Then I would be very careful about touching anything that the boot sequence even might rely on it
17:46:38 <AnMaster> somerandomentity, it must be somewhere
17:47:05 <ais523> well, you're not going to be able to change the ROM, probably
17:47:15 <AnMaster> ais523, you could hack the bus
17:47:18 <ais523> even if it's EEPROM, what's the chance that there'll be a program to flash it available?
17:47:25 <AnMaster> and redirect it to something else
17:47:43 <ais523> also, do you even own a JTAG cable? it's how those things are normally flashed on systems where they don't expect the user to be able to flash it from software
17:47:55 <ais523> somerandomentity: probably old-fashioned PROM then
17:47:56 <cpressey> Finding if there is a console would be good.
17:48:07 <AnMaster> somerandomentity, check /etc/inittab
17:48:07 <ais523> somerandomentity: can you do ls /dev
17:48:11 <cpressey> It would not be uncheap to have a console, reusing existing video hw etc
17:48:16 <AnMaster> somerandomentity, and see if there are any in it
17:48:29 <AnMaster> c1:2345:respawn:/sbin/agetty -8 38400 tty1 linux
17:48:29 <AnMaster> c2:2345:respawn:/sbin/agetty -8 38400 tty2 linux
17:48:31 <ais523> see if you can find vcs1 in it
17:48:39 <ais523> AnMaster: getty clearly isn't running
17:48:48 <AnMaster> ais523, well maybe they are commented out!
17:49:21 <ais523> well, busybox getty is sane
17:49:50 <somerandomentity> it works! and they even added a welcome message from The PocketSurfer Team lol
17:50:15 <ais523> ah, I was about to suggest trying to create character device 7, 1
17:50:20 <ais523> and seeing if you could read from it
17:50:23 <somerandomentity> they call it PocketSurfer linux. "Thanks for choosing DataWind!"
17:50:29 <ais523> it's the "screenshot from tty1" device
17:51:08 <ais523> somerandomentity: I take it you're IRCing from your main laptop?
17:51:31 <ais523> how did you kill X and yet stay on IRC?
17:51:52 <somerandomentity> i cran getty and now it is trying to start things because that is whoginut in the root lhey pat t
17:52:16 <ais523> it's root's login shell that starts X?
17:52:20 <ais523> that... is pretty broken
17:53:17 <cpressey> You say broken, I say elegant, my friend.
17:53:30 <ais523> it's more mindboggling than anything
17:54:06 <ais523> <The Court> SCO argues that it is entitled to judgment as a matter of law "because the verdict cannot be squared with the overwhelming evidence and the law."11 The Court respectfully disagrees.
17:55:41 <ais523> so it's /everyone's/ login shell that starts X?
17:56:05 <cpressey> Not just elegance, but nigh god-like elegance.
17:56:49 <ais523> <The Court> It is true that SCO presented more witnesses who testified that it was the intent of the parties to transfer the copyrights as part of the deal but, as the jury was instructed, the number of witnesses is not determinative.
17:56:59 <ais523> you have to love some of SCO's arguments
17:57:48 <somerandomentity> ais523: So, uh... What I am going to do is, replace the X-starting things with gettys, and install irssi in the debian chroot.
17:58:08 <ais523> putting getty in everyone's login shell is surely just as wrong
17:58:55 <ais523> wouldn't it create infinitely recursive login terminals?
17:59:12 <ais523> hmm, is /tmp wiped on every boot?
17:59:21 <ais523> you could create a file in /tmp to say that you'd already created the gettys
17:59:26 <ais523> and thus had no need to create them again
18:00:26 <ais523> does GNASH actually work yet?
18:01:27 <ais523> btw, someone reimplemented Flash in JS
18:01:46 <ais523> presumably rather slowly
18:03:12 <somerandomentity> would you guys be able to help more with an ssh connection? :P
18:03:58 <somerandomentity> I'd feel better if there were little notes of horror scattered by the devs around the system
18:04:16 <ais523> somerandomentity: is there any way you can post it to the daily WTF?
18:04:17 <somerandomentity> "boss says it has to be this way" "oh god why" "This is insane, but then so is everything else."
18:04:19 <ais523> or is it not /that/ bad
18:04:38 <ais523> somerandomentity: but then you'd no longer own it
18:05:59 <somerandomentity> ais523: want to poke around the ssh of this thing? It'd be amusing, if nothing else.
18:06:03 <ais523> haha, SCO tried more than once to argue that they had the better case because they had numerically more witnesses
18:06:20 <ais523> somerandomentity: you trust me enough to do that?
18:06:36 <somerandomentity> ais523: I think you're probably one of the most trustworthy people alive outside of nomic.
18:06:38 <Deewiant> You're not exactly known to be malicious
18:06:57 <ais523> somerandomentity: I'm trustworthy in nomic too, just with a different and very pedantic definition
18:07:36 <somerandomentity> You can just be root, I don't feel like battlin]g with this to add a new user
18:09:17 <ais523> ooh, it seems possible looking at the docs
18:09:36 <ais523> what's the specific version of the shell over there
18:09:51 <ais523> as in, busybox ash? debian ash?
18:10:53 <ais523> failing that, try which sh
18:10:58 <ais523> and seeing if the result is a symlink or not
18:11:31 <ais523> not even in man busybox over here
18:11:34 <ais523> most of the commands have docs
18:11:45 <ais523> but ash's documentation is "ash ash #define ash_full_usage"
18:12:04 <ais523> can you try running ash -i, followed by ps?
18:12:15 <ais523> and verifying that the resulting shell is called "busybox" or "ash" in the ps listing?
18:13:30 <ais523> yep, instead of telling it to start a shell
18:13:38 <ais523> you tell it to run a particular command that happens to be a shell
18:13:43 <ais523> and get a non-login shell that way
18:14:38 <cpressey> Well, good luck, you two. I gotta be off.
18:14:54 <ais523> I think I know why that thing isn't running sshd by default
18:14:56 -!- cpressey has left (?).
18:15:17 <ais523> well, there's no obvious way to authenticate a login
18:15:38 <ais523> a crazy method would involve writing your own PAM plugin that requires a password only remotely an has nothing to do with the user
18:15:52 <ais523> alternatively, you could accept connections only from my public IP, that would likely work
18:15:59 <ais523> assuming you have a clever enough firewall
18:17:37 -!- cal153 has quit (Ping timeout: 265 seconds).
18:17:59 -!- relet has joined.
18:18:40 <ais523> it seems this laptop doesn't actually have sshd installed
18:23:01 -!- cal153 has joined.
18:29:38 -!- augur has quit (Remote host closed the connection).
18:31:23 <ais523> and why not change it to a different DNS server?
18:31:33 <ais523> or is it the DNS client at your end that's broken?
18:32:12 <ais523> or you could try google's at 184.108.40.206
18:32:21 <ais523> but I trust a random company I've never heard of more than google
18:32:31 <ais523> there must be something wrong with me
18:34:16 <ais523> it's a really big telecom company, Level3
18:34:51 <somerandomentity> heh this thing is completely solid state so you only know it's struggling becaues of the little cpu indicator
18:35:19 <ais523> I've only heard of Level3 because of the DNS
18:36:18 <somerandomentity> well, you deend on them for access to the internet. At some point, most likely
18:36:37 <somerandomentity> iirc they maintain a few root dns servers, and many many many things utilies their network
18:37:15 <Deewiant> Nothing unless you're running some kind of DNS cache
18:39:50 -!- relet has left (?).
18:43:30 -!- somerandomentity has quit (Read error: Connection reset by peer).
18:45:09 -!- alissed has joined.
18:45:23 <alissed> ais523: you know the time, you know the place, something that rhymes, to save the human race.
18:45:31 <alissed> Less obscurely, you can connect now. I think.
18:45:44 <ais523> I'm missing several relevant pieces of info
18:45:48 <ais523> should I use the IP in your whois?
18:45:55 <ais523> actually, that's the main one
18:46:09 <alissed> What are the other pieces of info?
18:46:18 <ais523> I was going to say port, but we agreed that alreayd
18:46:58 <ais523> there's no prompt, though
18:46:58 <ais523> although I get responses to commands
18:47:43 <alissed> wow, I can't type capital P in the terminal either
18:47:53 <ais523> also, ls isn't outputting in columns, making it rather hard to use
18:48:50 <alissed> what options does busybox sh have, could you look it up?
18:49:38 <ais523> still no prompt, otherwise working
18:49:49 <ais523> I can do without a prompt, I'm just wondering wtf is going on
18:50:13 <ais523> are the prompts being shown on your screen?
18:51:10 <ais523> (presumably, anything going to stderr goes to your screen, anytthing going to stdout goes to mine)
18:51:31 <alissed> no, stderr is not being seen
18:53:08 <ais523> if I send to stderr, i get the output
18:53:17 <ais523> setting PS1 did nothing
18:53:34 <alissed> you know what, I'll connect myself
18:53:55 <ais523> busybox claims it has telnetd compiled in
18:54:12 <ais523> you'd have to run it as "busybox telnetd" if there isn't a symlink, though
18:54:29 <fizzie> If there's no prompt, the shell might be in the non-interactive mode.
18:55:04 <alissed> interactive mode doesnt work though
18:55:23 <ais523> what if I try to start an interactive shell from the noninteractive shell?
18:56:31 <ais523> yep, thought it might be
18:56:35 <alissed> hahaha fuck you windows EVERYWHERE :D
18:56:46 <ais523> I got a fatal X error, saying it already ran
18:56:49 <alissed> two battery indicators fuck yeah
18:57:04 <ais523> which exited because it was already running
18:57:19 <ais523> OTOH, I now have a shell that works
18:57:47 <ais523> connection closed by foreign host
19:00:08 <alissed> failing that we'll go back to nc + starting interactive
19:01:24 <ais523> and it's taking a noticeable amount of time, maybe around 200ms, to reply to trivial commands like echo
19:01:57 <alissed> ais523: it's possible this thing is just slow
19:02:01 <ais523> also, less and more aren't waiting for paging, they're just outputting the whole thing
19:02:09 <ais523> not sure what to use it /for/, though
19:02:19 <alissed> Well, I guess, just try and figure it out.
19:02:49 <alissed> /debian is my chroot. /etc/init.d has stuff, /etc/inittab too. /etc/profile is the devil.
19:03:02 <ais523> hmm, at least reading /etc/profile explains the insanity
19:03:08 <ais523> it's echoing PATH after it sets it
19:03:13 <alissed> So, the question is: How does this system really work? And how can I surgeryify /debian into it?
19:03:17 <ais523> which explains why telnet blurted out the path for no apparent reason
19:03:17 <alissed> also, what the hell is linuxrc?
19:04:05 <ais523> is /linuxrc text or binary?
19:04:13 <ais523> I'm not sure how to tell with the commands on there, other than trying to look at it
19:04:34 <alissed> ais523: you may want to chroot into /debian
19:04:40 <ais523> anyway, linuxrc seems to be a SuSE thing
19:04:43 <alissed> it has modern utilities, like nano and vi and a good shell
19:04:53 <alissed> ais523: no, its definitely debian
19:05:02 <alissed> this machine is most certainly debian
19:05:17 <ais523> aha, it's the bootloader for the SUSE installer
19:05:31 <ais523> the docs say you can use it as a bootloader for an installed system too if you really want to
19:05:32 <alissed> it seems to be initrd or something
19:05:36 <alissed> common in embedded devices i guess
19:05:47 <ais523> http://en.opensuse.org/Linuxrc
19:05:56 <alissed> ais523: anyway, use the /debian chroot; it is far more pleasant
19:06:08 <ais523> I thought I was supposed to be helping you figure out how the system was so insane?
19:06:18 <ais523> if I chroot out of the insane bit, I just get a standard debian
19:06:19 <alissed> but you can do that with sane tools
19:06:31 <ais523> hmm, you mean I should chroot to debian, then break the chroot?
19:06:33 <alissed> Mount Poop, the most dreaded of mountains
19:06:44 <alissed> look in /mnt/poop, inspect with high-tech tools
19:06:49 <ais523> wow, "cd /" took almost half a second
19:06:57 <alissed> archaeologists don't use primitive tools to examine their samples
19:07:39 <ais523> whoops, exited netcat by mistake
19:07:42 -!- alissed has left (?).
19:07:50 <ais523> hmm, that's a bad sign
19:08:19 -!- alissed has joined.
19:08:21 <ais523> "linuxrc parameters are case-insensitive and you can add as many hyphens, underscores, or dots as you want."
19:09:11 <ais523> it's working fine, except /mnt/poop seems to be empty
19:09:23 <ais523> "/dev/mtdblock2 on /mnt/poop type yaffs2 (rw)"
19:10:16 <alissed> umount it and remoutn it, then
19:11:01 <ais523> what's the command to mount?
19:11:06 <ais523> it needs a few settings I can't remember offhand
19:11:16 <alissed> mount -t yaffs2 /dev/mtdblock2 /mnt/poop
19:11:19 <ais523> "mount: you must specify the filesystem type"
19:11:21 <ais523> and I can't remember how
19:12:29 <ais523> hmm, /etc/install.inf, linuxrc's config file, seems not to exist
19:12:46 <alissed> its not that kind of linuxrc afaik
19:12:58 <ais523> it seems plausible that it would be
19:13:05 <alissed> ais523: hey can you start icewm --replace in the background plz?
19:13:21 <ais523> icewm: command not found
19:13:25 <ais523> presumably because I'm inside the chroot
19:13:41 <alissed> well, exit temporarily then plz :P
19:13:46 <alissed> no disown btw so youd need nohup
19:16:23 <ais523> whoops, exited by mistake again
19:16:33 <ais523> the real problem with nc is that it exists on things like ^C and ^D
19:16:39 <ais523> rather than sending them into the inside shell
19:17:48 <alissed> ais523: well, feel free to try and get telnetd working inside :P
19:18:01 <alissed> anyway, I'm just unsure which part I should migrate from /debian first.
19:18:09 <alissed> Or even what init it is using.
19:18:14 <alissed> linuxrc holds the key, but how to inspect it?
19:18:31 <ais523> take its SHA1 hash, then google it
19:18:42 <alissed> nobody else uses this thing
19:19:19 <ais523> pity, I thought that was actually a decent idea for finding executables
19:20:03 <ais523> file isn't installed, either on the inside or the outside chroot
19:20:27 <alissed> aptitude takes minutes to start
19:20:53 * ais523 sets PS1 to a sensible value
19:21:18 <ais523> hmm, apt-get is also far from instant
19:21:23 <ais523> the thing seems to be rather slow
19:21:36 <alissed> well, its a few hundred mhz arm with 128 megs of ram or so
19:21:40 <alissed> with x11 and pidgin running
19:22:01 <alissed> would be nice to be able to communicate in-shell
19:22:01 <ais523> ptys seem not to work altogether
19:22:23 <ais523> because /dev/pts doesn't exist inside the chroot
19:22:25 <ais523> does it exist outside?
19:23:05 <ais523> oops, I've started wall and now have no way to exit it
19:23:09 <ais523> could you kill my wall process?
19:23:24 <ais523> (^C and ^D both kill the outside netcat)
19:23:34 <alissed> no kiillall so i have to mount proc to use ps to find the id
19:24:04 <ais523> "sh: [4080: 1] tcsetattr: Invalid argument"
19:24:19 <alissed> ais523: im echoing to ptses
19:24:27 <ais523> as well as not pressing control-anything, I now also have to remember to use heredocs
19:24:58 <ais523> hmm, /dev/pts is quite a bit smaller than /mnt/poop/dev/pts
19:25:03 -!- augur has joined.
19:25:28 <alissed> on a scale of one to ten, how weird is it to be talking technically about a directory called poop when i was just test-mounting it?
19:25:38 <ais523> only about 3, i know you
19:25:49 <ais523> try 5 if you're aiming for me
19:26:10 <ais523> it's likely to be one of the high-numbered ones
19:26:13 <alissed> all except 0 and 1 which is me
19:26:40 <ais523> I'm not in a pty at all
19:27:00 <alissed> and not a named one either
19:27:13 <ais523> I tried running script to create a pty
19:27:18 <ais523> and got "openpty failed"
19:27:33 <alissed> chroots tend to not like doing such things
19:27:56 <ais523> I wrote "exit" and it exited every level at once
19:28:02 <ais523> and dropped me back to my own shell
19:28:03 <alissed> ais523: incidentally, bit of history: both devs are static
19:28:15 <ais523> no idea why it did that
19:28:19 <alissed> oh i could install ssh in the chroot
19:28:23 <alissed> you could escape it anyjway
19:28:51 <ais523> yes, but that's a pain
19:29:03 <ais523> "chroot /mnt/poop" would have been enough
19:29:06 <ais523> but then I'd be inside two nested chroots
19:29:10 <ais523> in opposite directions
19:29:25 <alissed> so is everything else about this machine
19:29:58 <alissed> ais523: is this the only machine to use both busybox and firefox?
19:30:16 <ais523> my laptop has busybox installed, after all
19:30:35 <alissed> but i mean using busybox primarily
19:31:38 <alissed> ais523: what do yout hink would be the safest thingt o move out of the chroot first?
19:31:53 <fizzie> My phone uses busybox primarily, and does have Firefox (well, Fennec... but it's related).
19:31:54 <ais523> I say, just change the init script to boot into the chroot
19:32:10 <ais523> and ignore the surrounding level of insanity
19:32:26 <alissed> ais523: oh, but that's just a chroot, not a real boy!^Winstall!
19:33:08 <fizzie> (Or is it even "Firefox Mobile" officially? I think it is.)
19:34:22 <ais523> and ignore the surrounding level of insanity
19:34:43 <ais523> alissed: can I write a shellscript into /root inside the chroot?
19:34:53 <ais523> that sets PS1 to a sane value?
19:35:03 <ais523> it'd save having to copy it over all the time
19:36:13 <ais523> argh, /no/ control code works
19:36:17 <ais523> not even control-X to exit nano
19:36:33 <ais523> could I have a process kill again? I'll use ed, that should work
19:36:37 <ais523> assuming it's installed
19:38:23 <alissed> can chroots start network connections in the host
19:39:02 <ais523> although nc is currently using the same port
19:39:26 <ais523> it wants a password, and won't accept the null string
19:39:41 <ais523> that could be a bad idea
19:39:47 <ais523> because then you might not be able to boot
19:39:57 <ais523> given the insanity of the boot process
19:40:13 <alissed> password is the most common algebraic placeholder variable
19:40:43 <alissed> for some reason, /root/.bashrc is not running
19:40:49 <alissed> just run that instead of your ps1-setter
19:41:34 <ais523> "PTY allocation request failed on channel 0"
19:41:42 <ais523> nothing inside the chroot can create ptys, it seems
19:41:52 -!- b3n4dd1 has joined.
19:41:52 <ais523> probably you're missing something in /dev
19:42:01 <ais523> IIRC, that's how you create ptys, you ask a pty-creation device for one
19:42:14 <alissed> i linked the chroot pts to the outside one
19:43:22 <ais523> what sort of link, hardlink?
19:43:22 <alissed> can you hardlink a directory?
19:43:30 <alissed> symlink, which is why i suspect it fails
19:43:32 <ais523> and it depends on the filesystem
19:43:43 <alissed> also, im paranoid and always remove symlinks with unlink, not rm; am i weird?
19:43:47 <alissed> ais523: some crazy embedded one
19:43:57 <ais523> even on the ones that let you, you need to be root and give a special arg to ln to say "yes I really mean this"
19:45:36 <alissed> i have a feeling it would not work on the outer system
19:46:29 <alissed> ais523: ill just go for telnet then
19:46:43 <ais523> I don't think either will work
19:47:05 <ais523> hmm, rsh probably doesn't need ptys, come to think of it
19:47:29 <alissed> this is fun, in a sort of really demented way, isn't it?:P
19:47:48 <alissed> wow, removing packages is slow
19:49:03 <ais523> I'm not entirely sure what we're trying to achieve
19:49:17 <ais523> other than making your system saner, which everyone here but you seems to think is a bad idea to even try to accomplish
19:49:26 <alissed> Well, I'm trying to abolish all insanity in / and insert sanity in the form of Debian.
19:49:38 <alissed> ais523: but why? Because it might break it? It is useless as it is anyway.
19:49:55 <ais523> it wouldn't be useless if you just made it boot into the chroot
19:50:11 <ais523> it'd be as useful as a sane version would be
19:52:27 <alissed> ais523: try now, standard rsh port
19:52:35 <alissed> oh i have not forwarded it
19:53:22 <ais523> I do "man rsh" and get the man page for ssh
19:53:46 <alissed> shall i just put the nc back up?
19:55:11 <alissed> ais523: ok, how about we get it booting to the chroot
19:55:16 <alissed> as, at least, a first step
19:55:20 <alissed> now, we need it to do getty stuff
19:56:10 <alissed> ** consider /etc/profile locked for editing
19:57:19 <alissed> ais523: should it run the chroot's init?
19:57:39 <ais523> well, is the outside init working?
19:57:46 <ais523> if so, you don't need an inside init
19:57:49 <ais523> just something to start services
19:59:20 <ais523> running two copies of init is simple?
19:59:24 <alissed> im going to try running the chroots /sbin/init to see what happens
19:59:27 <alissed> ais523: simpler from a debian point of view
19:59:45 * ais523 reads Slashdot story about company putting a content warning on the US constitution
19:59:56 <alissed> whats the default init level?
20:00:44 <pikhq> I'm amazed at Puppy Linux.
20:00:52 <pikhq> It has managed to make GTK not seem slow.
20:01:36 * alissed replaces /dev/initctl with mnt poop version
20:02:33 <alissed> ais523: ok, how can i start just the init.d stuff?
20:02:47 <ais523> it used to be done by shellscript
20:02:56 <ais523> presumably that method still works
20:03:50 <alissed> like... manually running them?
20:04:03 <ais523> a shell script ran everything in the right order
20:04:07 <ais523> and was the only thing init ran
20:04:18 <alissed> well, yes, but theres already a tangle of init.d stuff id like to use
20:04:33 <alissed> so how can i just use the sysv part of sysvinit?
20:05:10 <alissed> its /etc/init.d/rc i tihnk
20:05:27 -!- ais523 has left (?).
20:06:44 -!- alissed has quit (Quit: Leaving.).
20:06:49 -!- bsmntbombdood_ has joined.
20:09:09 -!- bsmntbombdood has quit (Ping timeout: 264 seconds).
20:20:41 -!- alissed has joined.
20:23:17 <fizzie> Quit with a fungot quote in the quit message; that is messed-up.
20:23:18 <fungot> fizzie: mother told me some of the art in solving the eopl problems specific to their business.
20:23:36 <alissed> Whenever he doesn't set a quit message, which is rare.
20:26:49 -!- MigoMipo has joined.
20:33:36 -!- Leonidas_ has joined.
20:34:58 -!- alissed has quit (Read error: Operation timed out).
20:37:31 -!- Leonidas_ has quit (Client Quit).
20:37:38 -!- Leonidas_ has joined.
20:40:03 -!- Leonidas_ has quit (Client Quit).
20:40:13 -!- Leonidas_ has joined.
20:40:24 -!- olsner has joined.
20:41:33 -!- Leonidas_ has quit (Client Quit).
20:42:21 -!- Leonidas_ has joined.
20:43:51 -!- MizardX has joined.
20:44:37 -!- b3n4dd1 has quit (Ping timeout: 260 seconds).
20:44:50 -!- Leonidas_ has quit (Client Quit).
20:46:08 -!- olsner has quit (Read error: Connection reset by peer).
20:48:28 -!- olsner has joined.
20:48:29 -!- olsner has quit (Read error: Connection reset by peer).
20:49:18 -!- olsner has joined.
20:55:28 -!- Leonidas_ has joined.
20:57:27 -!- cheater99 has quit (Ping timeout: 260 seconds).
20:58:41 -!- cheater99 has joined.
21:10:34 -!- Leonidas_ has changed nick to Leonidas__.
21:12:05 -!- Leonidas has changed nick to Leonidas_.
21:12:20 -!- Leonidas__ has changed nick to Leonidas.
21:13:26 -!- Leonidas_ has left (?).
21:17:12 -!- olsner_ has joined.
21:19:57 -!- olsner has quit (Ping timeout: 260 seconds).
21:19:57 -!- olsner_ has changed nick to olsner.
21:39:49 <pikhq> Hello, cheater #99.
21:59:28 <pikhq> I faireth well today.
21:59:44 <pikhq> I fare well today.
21:59:49 <pikhq> However, my English does not.
22:05:43 <olsner> pik is a kind of fish in swedish, I think
22:06:17 <olsner> incidentally, pike in english is a different fish and also a sharp pole (which pik in swedish is also)
22:06:45 <olsner> that last paren could actually use a second 'also'
22:09:01 <olsner> nope, I'm just making things up
22:13:07 -!- olsner has quit (Quit: olsner).
22:25:59 -!- tombom_ has joined.
22:28:29 -!- tombom has quit (Ping timeout: 272 seconds).
22:29:34 <CakeProphet> I finally found a store that supplied vermiculite
22:32:19 -!- charlls has joined.
23:21:57 <AnMaster> okay so my sleep pattern is extremely fucked up atm
23:27:03 <pikhq> That's pretty fucked up.
23:27:35 <AnMaster> pikhq, at least I get to talk to new people
23:36:49 <Leonidas> hmm, waking up at 0:05 sounds like fun to me. I like being awake during the night.
23:37:02 -!- CakeProphet has quit (Ping timeout: 260 seconds).
23:38:55 -!- CakeProphet has joined.
23:41:39 -!- charlls has quit (Quit: Saliendo).
23:41:53 -!- oerjan has joined.
23:55:20 -!- jcp has quit (Quit: Later).
23:58:05 -!- jcp has joined.
23:59:02 -!- tombom_ has quit (Quit: Leaving).
23:59:48 -!- MigoMipo has quit (Remote host closed the connection).