00:06:45 -!- ihope_ has joined.
00:06:53 -!- ihope_ has changed nick to ihope.
00:10:13 <ihope> Hi, RodgerTheGreat.
00:10:21 <ihope> Everyone else is denied my hello today.
00:11:47 <ihope> I thought it somewhat appropriate because--hey, does this sketch indicate that Dr. T has a device for using his cute little tyrannosaurus arms more like human hands?
00:12:32 <RodgerTheGreat> having two claws is rather limiting for tasks involving fine manipulation
00:13:22 <ihope> Does it have a cute little AI to do with it, rather than somehow being fully controlled by only however many degrees of freedom two claws gives?
00:13:47 <RodgerTheGreat> I don't go into detail. I was thinking neural interface
00:15:39 -!- olsner has quit ("Leaving").
00:32:02 <Slereah> Why is lambda calcul named thusly?
00:32:13 <Slereah> I can see the pi <-> processus, but why lambda?
00:33:36 -!- edwardk has quit (Read error: 104 (Connection reset by peer)).
00:33:41 <ihope> Maybe when Church invented it, he decided to use a lambda to denote a function, then he named the thing after it. Or some other such, I guess.
00:35:00 <Slereah> Lambda seems like an odd choice, though
00:35:21 <oklofok> "so what did you name that construct of yours?" "lamb, duh."
00:35:22 <ihope> What would be a better choice?
00:35:22 <Slereah> He wasn't German or anything, so what could it mean in English?
00:35:31 <Sgeo> Firefox 3 was mentioned on the Cobert Report on Tuesday
00:35:59 <oklofok> i'd go with my explanation, it makes so much sense i could swallow it whole.
00:36:12 <Slereah> I dunno, since I don't know what it stands for
00:37:04 <oklofok> S:"explain X" o:"explains X..." S:"i don't know if that's right because i don't know what's right"
00:37:08 <oklofok> man, you can't question everything
00:37:15 <oklofok> some things, you just have to accept.
00:37:40 <oklofok> like peanuts, the existance of god, and the sheep fetish church had
00:38:00 <Slereah> But Church wasn't sexy, so it doesn't fit in my head
00:42:24 <oklofok> now you're just being silly
00:44:04 <Slereah> Nothing in the original article :o
00:44:16 <Slereah> Let's invoke the spirit of Alonzo Caspip.
00:59:54 -!- Corun has quit ("This computer has gone to sleep").
01:03:53 <ihope> So in pi calculus, threads send channels to each other via channels?
01:06:12 <Slereah> Well, it's a way to see it.
01:06:40 <Slereah> You can also throw the semantics away and just keep the name transfering.
01:07:53 <Slereah> Or you can use biscuit trucks.
01:08:11 <AnMaster> anyway I made an ebuild for C-INTERCAL
01:10:16 <Slereah> Problem with the names is, I don't have an easy way to mix the names and the functions.
01:12:14 <oklofok> Slereah: eagerly waiting for limp
01:12:43 <Slereah> I don't have much clue of how to deal with pi in scheme.
01:12:48 <oklofok> don't you tell me what not to be too.
01:15:44 <Slereah> I'm not too sure what language to use.
01:16:08 <Slereah> It would probably be easier in Python, since it has lambdas, function definition and is imperative.
01:27:28 <Slereah> Python is usually quite slow when I make it do huge things.
01:27:43 <Slereah> And from what I saw, it doesn't fare well in performance tests.
01:29:16 <Slereah> Any critics on the language so far?
01:47:48 <augur> how has reactance changed?
01:48:04 <oklofok> augur: i can show you logs tomorrow
01:48:14 <augur> just off the top of your head
01:48:19 <augur> what do you think has changed?
02:02:59 <Slereah> Do you think I need a particular way to differentiate strings (ie : list of numbers) from names in limp?
02:03:10 <Slereah> Or do I just try to let the context do it?
02:40:19 <augur> strings vs names??
02:40:25 <augur> whats the supposed difference?
02:40:41 <Slereah> Well, the I/O works with strings.
02:40:56 <Slereah> Actually, they are lists of numbers, for the functional part.
02:41:23 <Slereah> and actually, they're not really lists. They're consed together like a conga line (there's no null type)
02:42:00 <Slereah> On the other hand, the pi part uses names.
02:42:08 <augur> oh i see, you're doing pi calculus
02:42:13 <augur> nevermind, i have nothing to contribute
03:00:26 <ihope> Haskell doesn't have any non-deterministic bits, does it?
03:00:55 <ihope> Well, apart from any undefined behavior it may have.
03:01:22 <Slereah> Well, I'm sure it has some random number liberries
03:04:25 <ihope> Yes, but saying "flip a coin" is not the same as saying "pick one".
03:05:27 <Slereah> Well, he could pick the one on the left if it's head.
03:06:45 <ihope> Choosing "left if it's heads, right if it's tails" isn't really a choice at all, as "right if it's heads, left if it's tails" produces exactly the same result.
03:08:00 <Slereah> Then on what criterias do you chose?
03:08:08 <Slereah> And how is that non-determinism?
03:09:06 <Slereah> For some reason, I have the Meow Mix song on my MP3 player, and never knew about it.
03:20:28 <ihope> Well, if I have a programming language with a function pick(a,b) that returns either a or b, a compiler might always return a, or always b, or whichever's smaller, or whichever looks more like a giraffe in its opinion.
03:20:41 <ihope> Whichever, in the compiler's opinion, is best.
03:21:21 <Slereah> Well, wouldn't that depend on the compiler?
03:21:37 <Slereah> Then why ask about Huskoll?
03:22:55 <Slereah> Plus Hoskoll is horrible :o
03:23:30 <ihope> Why should Haskell not have nondeterministic stuff, if that's what you're asking?
03:23:52 <Slereah> I'm asking that if it depends on the compiler, why ask about the language.
03:24:10 <ihope> The language is what specifies what the compiler's allowed to do.
03:24:25 <ihope> If the compiler is given freedom, that's nondeterminism of a sort.
03:29:28 <Slereah> http://haskell.org/ghc/docs/latest/html/users_guide/bugs-and-infelicities.html
03:29:37 <Slereah> "13.1.2. GHC's interpretation of undefined behaviour in Haskell 98"
04:23:03 -!- GregorR has changed nick to GregorR[Prague].
04:51:45 -!- ihope has quit (Read error: 110 (Connection timed out)).
04:56:08 -!- augur has changed nick to psygnisfive.
05:08:05 -!- GreaseMonkey has joined.
05:11:52 -!- Slereah has quit (Read error: 104 (Connection reset by peer)).
05:12:10 -!- Slereah_ has joined.
05:22:06 -!- Slereah has joined.
05:22:12 -!- Slereah_ has quit (Read error: 104 (Connection reset by peer)).
05:23:54 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
05:33:38 <Slereah> http://esolangs.org/wiki/Esme
05:35:35 -!- poiuy_qwert has joined.
05:39:47 <Slereah> Almost finished the specs of my new language.
05:40:57 <Slereah> http://esolangs.org/wiki/User:Slereah/Limp
05:49:39 <Slereah> Still need to clean up the page, though
05:53:12 -!- kwertii has joined.
06:11:06 -!- cherez1 has joined.
06:11:11 -!- cherez has quit (Read error: 110 (Connection timed out)).
06:11:43 -!- cherez1 has changed nick to cherez.
06:11:45 -!- cherez has left (?).
06:14:01 -!- kwertii has quit ("bye").
06:22:37 -!- Slereah has quit (Read error: 104 (Connection reset by peer)).
06:22:57 -!- Slereah_ has joined.
06:36:13 -!- poiuy_qwert has quit.
07:14:26 -!- olsner has joined.
07:43:07 <AnMaster> <psygnisfive> XORs are conditional inverters :) <-- yes and?
07:56:51 -!- oklopol has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:13:56 -!- oklofok has quit (Read error: 110 (Connection timed out)).
08:18:41 <psygnisfive> we need to consider two things, oklopol. ais brought them up earlier.
09:05:05 <AnMaster> anyone seen ais or ehird today?
09:05:27 <psygnisfive> whereby earlier means yesterday, but thats today for me ;)
09:05:52 <AnMaster> I mean today as in European time
09:10:59 <AnMaster> because that would be midnight
10:08:33 <oklopol> today, i'm going to DRINK MYSELF DRUNK
10:10:41 <oklopol> psygnisfive: they both used to be active
10:10:48 <oklopol> you've changed that at some point
10:10:56 <oklopol> they're both active in my implementation
10:22:53 <oklopol> i'm sure i've asked you exactly whether they x->z; y->z should keep both reactions, but cannot find that in my logs
11:05:22 -!- oklopol has quit (Read error: 110 (Connection timed out)).
11:15:43 -!- GreaseMonkey has quit ("Remote closed the previous member app").
12:05:54 -!- AnMaster has quit ("thunderstorm").
12:29:55 -!- Corun has joined.
12:41:58 -!- ais523 has joined.
12:53:40 -!- ihope_ has joined.
12:53:46 -!- ihope_ has changed nick to ihope.
12:57:41 * ais523 's exam results came out today
12:59:13 <ais523> ah, I finally figured out what I'd got wrong on my option form
12:59:21 <ais523> I was having problems counting to 120 in units of 10
12:59:28 <ais523> which should be a simple task for a programmer
13:00:34 -!- ais523 has quit ("mibbit.com: brb").
13:03:08 -!- RedDak has joined.
13:06:43 -!- Corun has quit ("This computer has gone to sleep").
13:09:09 -!- caio has joined.
13:26:34 -!- jix has joined.
13:41:54 -!- ihope has quit (Read error: 110 (Connection timed out)).
13:47:00 -!- Corun has joined.
14:12:38 -!- Judofyr has quit.
14:22:11 -!- jix has quit (Read error: 104 (Connection reset by peer)).
14:26:21 -!- RedDak has quit (Read error: 104 (Connection reset by peer)).
14:30:51 -!- jix has joined.
14:54:43 -!- Judofyr has joined.
15:01:16 -!- Judofyr has quit.
15:04:13 -!- AnMaster has joined.
15:11:18 -!- ais523 has joined.
15:11:39 <ais523> back, sorry it took so long
15:27:00 -!- Corun has quit ("This computer has gone to sleep").
15:30:13 -!- Judofyr has joined.
15:40:33 -!- tusho has joined.
15:41:19 <tusho> 're being slow, ais523
15:41:28 <tusho> ah, idle half an hour
15:41:49 <tusho> ais523: using an ajax client again...?
15:41:49 <ais523> I was looking at the wrong part of the screen
15:42:09 <ais523> so you did /version me the other time I used one
15:42:21 <ais523> either that, or it overrode my quit message
15:42:25 <tusho> ais523: colloquy VERSION's on /whois
15:42:36 <tusho> ais523: no, it's your hostname
15:42:39 <tusho> gateawy/web/ajax/mibbit.com/
15:42:44 <ais523> oh, yes that would give it away
15:42:55 <ais523> it seems to give my real hostname as my realname for some reason...
15:43:07 <tusho> ais523: yes, because all mibbit connections come from mibbit servers
15:43:09 <tusho> so for effective bans...
15:43:18 <tusho> they forward your hostname
15:43:28 <ais523> I was wondering how they did it
15:43:34 <ais523> the realname field seems a strange place to put it
15:44:34 <tusho> ais523: where do you suggest putting it?
15:44:47 <tusho> the hostname is a mibbit cloak (which is sane)
15:44:57 <tusho> the username is the mibbit identifier thing for your session
15:45:00 <tusho> so ... what's left is the real name
15:45:03 <ais523> tusho: there's a real-hostname field in the USER command IIRC
15:45:11 <ais523> but it's ignored by most servers
15:45:15 <ais523> because they have no way to trust it
15:45:24 <ais523> presumably, though, they would trust it if it came from mibbit
15:45:26 <tusho> in the latest rfc it's explicitly ignored
15:45:40 <tusho> 01:05:05 <AnMaster> anyone seen ais or ehird today?
15:45:47 <ais523> tusho: well, I just put xs there for ease of typing
15:45:48 <tusho> correct me if I'm wrong but didn't he ask that at like 5am
15:46:13 <ais523> hmm... well, I was asleep at 5am GMT (= 6am BST)
15:46:28 <tusho> er, yes, BST is what I meant
15:46:32 <tusho> god I hate summer time
15:46:41 <tusho> to me, GMT=GMT+BST-when-needed
15:47:44 <tusho> ais523: OISC is #1 on proggit
15:47:46 <tusho> though the wikipedia article..
15:48:12 <ais523> personally, I like MiniMAX
15:48:16 <ais523> as being an unconventional way to do an OISC
15:48:23 <tusho> ais523: post it on the comments, then :-P
15:48:30 <tusho> it'll be someone other than me linking to esolangs.org...
15:48:40 <ais523> tusho: because you've linked it too many times already?
15:49:04 <tusho> it's a shame that my #1 voted comment is a lolcats joke
15:49:22 <tusho> http://www.reddit.com/info/61gcu/comments/c02jatt
15:49:27 <ais523> well, esolangs is linked from the Wikipedia article
15:49:40 <tusho> http://www.reddit.com/info/61gcu/comments/c02jahi
15:49:45 <tusho> if you want to actually understand my comment :-P
15:50:26 <ais523> the comment by ealf references Iota
15:50:35 <ais523> so someone else is up on the game, too
15:50:49 <tusho> ais523: people know esolangs
15:50:50 <tusho> just not esolangs.org
15:50:58 <tusho> also, iota is soooooooooooooooo a cheat
15:51:02 <tusho> sure the syntax is minimal
15:51:04 <tusho> but the semantics are huge
15:51:09 <tusho> first, you bring in the lambda calculus
15:51:09 <ais523> well, it's a reasonably complex combinator
15:51:12 <tusho> then, you bring in S and K
15:51:17 <tusho> then, you invent your own combinator on top of it
15:51:32 <ais523> tusho: no, it's a unique combinator that can be expressed in terms of S and K, or as a lambda
15:51:35 <tusho> is iota even a combinator?
15:51:40 <tusho> S and K are just shorthand for lambda expressions
15:51:41 <ais523> but still, it's a pretty complicated combinator
15:51:52 <ais523> tusho: yes, so is iota
15:51:52 <tusho> \x.x(\xyz.xz(yz))(\xy.x)
15:51:58 <tusho> that's not a combinator
15:52:32 <tusho> so iota = \x.x(\xyz.xz(yz))(\xy.x)
15:52:35 <tusho> therefore, iota is not a combinator
15:52:39 <ais523> try it with unique variable names
15:52:44 <ais523> you're shadowing there, making it harder to read
15:53:02 <tusho> \x.x(\abc.ac(bc))(\ab.a)
15:53:04 <ais523> \a.a(\b.\c.\d.bd(cd)(\e.\f.e)
15:53:20 <ais523> to make all the lambdas have unique names and separated
15:53:32 <tusho> ais523: it's still not a combinator?
15:53:43 <ais523> well, what does combinator mean, precisely?
15:53:44 <tusho> A combinator is a higher-order function which, for defining a result from its arguments, solely uses function application and earlier defined combinators.
15:53:52 <tusho> that makes combinator=lambda
15:53:59 <tusho> I'm pretty sure combinators are non-nestedlambdas
15:54:02 <ais523> tusho: no, it makes combinators=combinators
15:54:18 <tusho> but I'm probably totally wrong
15:55:22 <tusho> I attempt to learn grammar and spelling. - Quazie
15:55:29 <ais523> I like the MiniMAX single instruction partly because it's easy to express in x86 asm, and partly because it's unique compared to the other OISCs
15:55:42 <ais523> oh, and it takes no arguments, but depends on two pointers of internal state
15:55:53 <ais523> I suppose you could say that those pointers, plus memory, are its arguments
15:56:06 <ais523> the initial state of memory and the pointers is a MiniMAX program
15:57:23 <ais523> MiniMAX is harder to get one's head around than the other OISCs, though
15:57:45 <AnMaster> <tusho> 01:05:05 <AnMaster> anyone seen ais or ehird today?
15:58:08 <AnMaster> or the ebuild will fail with "trying to write outside sandbox"
15:58:28 <AnMaster> also I need to pre-create /usr/bin and /usr/lib in the DESTDIR
15:58:41 <tusho> heh, cool, ais523 used an eso-std.org userdir
15:58:49 <AnMaster> it tried to install an archive as /usr/lib64
15:59:02 <AnMaster> ais523, however I can upload it somewhere
15:59:11 <ais523> AnMaster: why did it do that?
15:59:15 <ais523> it shouldn't have done
15:59:21 <AnMaster> ais523, hm? it didn't mkdir as needed
15:59:30 <ais523> AnMaster: I have a mkdir -p in there now
15:59:38 <ais523> oh, you're using the version on intercal.freeshell.org?
16:00:00 <ais523> AnMaster: the repo's at http://eso-std.org/darcs/c-intercal
16:00:09 <ais523> I can make a tarball for you pretty easily if you like
16:00:18 <AnMaster> ais523, the ebuild itself is at http://rafb.net/p/y4VDk286.html
16:02:09 <AnMaster> http://rafb.net/p/Drn0Dy41.html
16:02:32 <AnMaster> http://rafb.net/p/Mvdg3I17.html is the other patch
16:02:37 <tusho> i am amazed that someone would put a trivial 45 line file under the gpl3
16:03:00 <AnMaster> tusho, well GPL2 would be ok too
16:03:29 <tusho> yeah, people using the gpl3 are living in the past
16:03:32 <tusho> I totally agree ;)
16:04:08 <AnMaster> http://rafb.net/p/H5ikiY96.html
16:04:23 <tusho> cool, I have power over people
16:04:23 <ais523> AnMaster: probably best to keep with your patches until the next release version of C-INTERCAL comes out, with those issues fixed
16:04:32 * tusho uses his magical powers for good and evil
16:04:44 <AnMaster> tusho, well I would do it GPL2 anyway
16:05:08 <ais523> because I kept the same licence it traditionally had
16:05:11 <ais523> well, GPL2+ to be precise
16:05:20 <AnMaster> ais523, should I submit it to bgo?
16:05:50 <ais523> AnMaster: may as well, if you can submit updated versions in the future
16:05:51 <AnMaster> (bgo = bugs.gentoo.org, where you submit new ebuilds as well)
16:06:42 <AnMaster> ais523, well, it should be fairly trivial to update, remove some patches as they are accepted upstream (like http://rafb.net/p/Mvdg3I17.html the current way is just plain wrong)
16:06:52 <AnMaster> NO other software try to touch the info files
16:07:09 <AnMaster> automake won't touch the dir files for example
16:07:27 <tusho> this is kind of like hobix
16:07:28 <ais523> anyway, C-INTERCAL 0.29 (under development) no longer tries that unless you're root
16:07:37 <tusho> which upgrades itself via base64 yaml transmission via http
16:07:40 <ais523> actually, unless the dir file's in your prefix
16:07:42 <AnMaster> ais523, well you could be root but in a sandbox
16:07:43 <tusho> by eval()ing some ruby from the hobix site
16:07:45 <ais523> it's mandb it doesn't do unless you're root
16:07:55 <ais523> AnMaster: maybe I should have an install-sane target
16:08:12 <AnMaster> ais523, and well gentoo simply uses DESTDIR for installing to the "stage"
16:08:30 <ais523> I mkdir -p the target dirs now if they don't exist
16:08:53 <AnMaster> + 1) sandbox.so LD_PRELOADed library that intercepts syscalls. 2) non-root access
16:09:32 <AnMaster> as the first alternative isn't as secure really, what about static binaries
16:09:55 <ais523> AnMaster: I suppose you could wrap the syscalls themselves
16:10:11 <AnMaster> gentoo can use a vanilla kernel
16:10:26 <ais523> which pretends to a program it has root priveliges even if it doesn't
16:10:42 <AnMaster> I don't know if gentoo uses something like that or not
16:11:07 * AnMaster preprares to submit the ebuild
16:27:35 <AnMaster> ais523, https://bugs.gentoo.org/show_bug.cgi?id=228563
16:30:42 <AnMaster> ais523, don't expect too much progress
16:30:50 <AnMaster> they will probably see it as a joke
16:30:58 <AnMaster> it will need some developer to be interested
16:30:58 <ais523> Debian have it as a package
16:31:24 <AnMaster> ais523, yes but debian seem to have some policy like "make everything we can a package, except when we don't like the author"
16:33:13 <tusho> anyway, c-intercal isn't a joke
16:33:22 <tusho> it's actually pretty darn popular as esolangs go
16:33:31 <tusho> and, you know, people actually compile intercal with it quite often
16:33:41 <tusho> therefore, there are likely gentoo users who have a use for it, quite a few
16:33:45 <tusho> therefore, rejecting it is silly
16:35:40 <AnMaster> but it just being forgotten == quite likely
16:39:41 <AnMaster> ais523, what was the darcs url you said?
16:40:12 <ais523> http://eso-std.org/darcs/c-intercal
16:40:26 <ais523> contains bugfixes for all the bugs you reported
16:44:01 <tusho> trac.minitage.org uses an invalid security certificate.
16:44:01 <tusho> The certificate is not trusted because the issuer certificate is unknown.
16:44:04 <tusho> FIREFOX 3 FEATURE DIE DIE DIE
16:44:38 <ais523> tusho: but all browsers report that
16:44:46 <ais523> it's an uncertified certificate
16:44:50 <ais523> anyone could have generated it
16:44:51 <tusho> ais523: but firefox displays it as an error page
16:44:55 <tusho> and you have to click 4 times to view the page
16:45:04 <ais523> tusho: epiphany's even stronger in that case
16:45:07 <tusho> (more info, Add Exception, Get Certificate, Confirm certificate)
16:45:12 <tusho> it even does it for self-signed ones
16:45:16 <ais523> you have to go to options and add an exception yourself to be able to view the page
16:45:39 <ais523> tusho: no, blindly assuming self-signed certificates are OK is the incorrect thing there
16:45:53 <tusho> so just display a notice
16:45:59 <tusho> and let you do one click to view the page
16:46:12 <ais523> tusho: people are used to just ignoring message boxes
16:46:23 <tusho> ais523: yeah, well this is the wrong way out
16:46:34 <tusho> heh, firefox's ligature support scares me
16:46:38 <tusho> 'fi' looks all weird on pages
16:46:38 <ais523> Firefox even has time limits on being able to click 'OK' on warning boxes for install, etc
16:46:53 <tusho> i think the new approach is called the UAC School of Design approach
16:48:27 <AnMaster> <ais523> tusho: no, blindly assuming self-signed certificates are OK is the incorrect thing there
16:48:35 <AnMaster> not everyone can afford verisign
16:48:55 <tusho> The error with Firefox is treating https as a security mechanism.
16:49:03 <tusho> as a _veirfying_ mechanism
16:49:04 <ais523> there should be some sort of minimum-security certificate
16:49:12 <ais523> which says "yes, this is unverified, and I know about it"
16:49:31 <ais523> which doesn't put the secure-site paraphenalia on the interface, but still encrypts the communications
16:51:39 <ais523> AnMaster: yellow address bar, lock icons everywhere
16:52:24 <ais523> also Firefox, Epiphany, IE, and Mozilla
16:52:30 <tusho> AnMaster: that's everywhere, fyi.
16:52:43 <ais523> I only normally use IE for email, though
16:52:49 <ais523> and occasionally IE-only sites
16:52:53 <ais523> I actually came across one today
16:52:54 <tusho> anyway, locks tend to make average joe think 'trustable'
16:52:55 <tusho> in a web situation
16:53:04 <tusho> which is why it's a very inappropriate metaphor
16:53:07 <ais523> which would have worked fine in Firefox except it had a bad interaction with the popup-blocker
16:53:07 <tusho> but it's really hard to find a good one
16:53:14 <tusho> since a lot of the concepts have no corresponding one in the real world...
16:53:25 <tusho> really, we just need to educate users. an icon will never explain things
16:54:16 <AnMaster> well I don't care about non-geeks
16:54:29 <tusho> I agree, let's let them all give away their credit card info
16:54:36 <ais523> AnMaster: most non-geeks do
16:54:42 <tusho> or lock off computers to people who don't know their innate workings!
16:54:46 <tusho> totally the best way to solve the problem
16:55:13 <tusho> 'stupid' != 'does not understand computers'
16:55:15 <AnMaster> helpdesk humor isn't that far from reality
16:55:32 -!- Corun has joined.
16:55:43 <ais523> I have to go for a while
16:55:46 <AnMaster> my dad is a professor, yet he needs to ask me to open a picture he got by mail
16:55:46 <ais523> but I'll be back this evening
16:56:03 -!- ais523 has quit ("mibbit.com: last seen quit message").
17:09:07 <AnMaster> /tmp/ccLj01UL.o: In function `ick_og86e9b0':
17:09:08 <AnMaster> hello.c:(.text+0x1ca): undefined reference to `ick_or0'
17:10:07 <tusho> c-intercal uses hex pointer addresses
17:10:14 <tusho> so I guess something's going wrong with pointer->int
17:11:45 <tusho> AnMaster: it seems like the kind of thing openbsd would do, preventing that ...
17:11:50 <tusho> shrug. What system is it?
17:15:13 <tusho> shrug. What system is it?
17:15:21 <AnMaster> the ebuild works on one gentoo but not on another
17:15:25 <tusho> 'it breaks on one system with'
17:15:28 <tusho> what do you mean one gentoo
17:15:32 <tusho> what hardware differences
17:16:00 <tusho> AnMaster: it's the hardware, i'm sure
17:16:01 <AnMaster> tusho, both are x86_64, one got a sempron 3300+ the other got a athlon64 3300+
17:16:08 <tusho> there has to be some difference
17:16:32 <AnMaster> both got nvidia geforce 7600 cards
17:16:33 <tusho> how about the actual differences
17:16:59 <AnMaster> tusho, as far as I know there are none apart from the CPU and that one have slightly larger harddrive (500 GB vs 350 GB)
17:17:23 <tusho> AnMaster: well, there has to be something
17:17:24 <AnMaster> oh yes different mobos due to different sockets for CPUs
17:17:25 <tusho> try re-running ick?
17:17:57 <tusho> AnMaster: you could try reading ick's code
17:18:05 -!- Sgeo has joined.
17:18:10 <tusho> (there is one file that tries to be as close to idiomatic perl as possible, though)
17:18:23 <tusho> (because it was translated from a CLC-INTERCAL perl file, and ais523 kept perl idioms in there as a kind of homage)
17:18:52 <AnMaster> apart from the top line with command for compiling it
17:19:44 <tusho> AnMaster: paste the two lines
17:20:34 -!- rodrigo has joined.
17:20:59 <AnMaster> tusho, actually I was wrong, they are the same
17:21:02 <AnMaster> /* -*- mode:c; compile-command:"gcc hello.c -I/usr/include/ick-0.28 -I. -I./../include -L/usr/lib64 -L. -L./../lib -O2 -o hello -lick" -*- */
17:21:19 <tusho> AnMaster: ok, poke around in the ick sources then
17:21:27 <tusho> that can't be the problem, can it?
17:21:31 <tusho> AnMaster: ok, poke around in the generated file
17:21:39 <tusho> it can't reference a non-existant identifier on one platform and not the other
17:21:41 <tusho> that's Not Possible
17:22:01 <AnMaster> tusho, I think the issue is in /usr/lib64/libick.a
17:22:08 <AnMaster> diff says "binary file differ"
17:22:16 <tusho> AnMaster: try recompiling ick
17:22:26 <AnMaster> tusho, tried that a few times already
17:22:33 <AnMaster> also with all debuginfo left and so on
17:22:46 <tusho> poke in the ick sources for libick.a
17:23:59 <AnMaster> tusho, does this program work for you http://rafb.net/p/MWMDVo63.html
17:24:26 <AnMaster> it says "THIS PROGRAM REQUIRES CLC-INTERCAL", however it *does* compile with C-INTERCAL on one system
17:24:36 <tusho> AnMaster: ... You could have told me that.
17:24:43 <tusho> Well, remember, C-INTERCAL reports a lot of errors at runtime.
17:24:53 <AnMaster> tusho, it gives *linking errors*
17:24:59 <AnMaster> /tmp/cchXHLUO.o: In function `ick_og676800':
17:25:00 <AnMaster> hello.c:(.text+0x1ca): undefined reference to `ick_or0'
17:25:00 <AnMaster> /tmp/cchXHLUO.o: In function `ick_og675980':
17:25:00 <AnMaster> hello.c:(.text+0x24a): undefined reference to `ick_or0'
17:25:02 <tusho> the behaviour on that program is undefined
17:25:06 <tusho> it requires CLC-INTERCAL
17:25:13 <AnMaster> tusho, so why does it work on one C-INTERCAL?
17:25:28 <tusho> AnMaster: it doesn't matter
17:27:00 -!- Judofyr has quit.
17:40:13 -!- ihope_ has joined.
17:40:18 -!- ihope_ has changed nick to ihope.
17:49:14 -!- ais523 has joined.
17:50:46 <tusho> ais523: you're losing your touch my boy
17:51:40 <tusho> you didn't say hi to me again
17:52:06 <ais523> because I left with a brb-ish thing, rather than a quittish thing
17:52:14 <ais523> and you replied with wb
17:52:39 <tusho> I didn't say {hi ais523}
17:52:49 <ais523> because it's inappropriate
17:52:51 <ais523> wb is more appropriate
17:52:55 <ais523> so it's a wb vs. back race
17:54:03 <tusho> you cycle, I'll try
17:54:21 <ais523> hmm... it's a different game when we both have prior knowledge
17:54:32 <ais523> um... I haven't left yet
17:54:50 <ais523> [Error] cycle: Unknown command.
17:54:53 <ais523> since when did that happen?
17:54:57 <tusho> ais523: since you used mibbit
17:55:05 <ais523> tusho: but I'm on Konversation atm
17:55:18 -!- ais523 has left (?).
17:55:21 -!- ais523 has joined.
17:55:36 <ais523> I can't switch to #esoteric fast enough after joining
17:55:38 <tusho> ais523: i did kinda have my finger over 'enter'
17:55:38 <ais523> when you're forewarned
18:08:18 -!- Sgeo has quit (Connection timed out).
18:08:41 -!- RedDak has joined.
18:11:54 <tusho> <meta http-equiv='X-UA-Compatible' content='YOUR MOM'>
18:12:12 <tusho> ais523: X-UA-Compatible is ridiculous
18:12:23 <ais523> Slashdot's HTTP headers, which contain Futurama quotes, are more interesting
18:12:41 <tusho> ais523: well, X-UA-Compatible is meant to contain a list of browsers
18:12:47 <tusho> so the implication is slightly amusing, if nonsensical
18:12:53 <tusho> but it's the first actual x-ua-compatible i've ever seen
18:12:54 <ais523> did Microsoft invent that?
18:13:08 <ais523> oh, it's the IE8 meta tag
18:13:10 <tusho> it will make it emulate different IE/Firefox versions or whatever
18:13:19 <ais523> it will make IE emulate previous IE versions
18:13:24 <ais523> and they suggested that other browser people did the same
18:13:55 <ais523> to which they effectively said "why, when the site's standard-compliant, the browser's standard-compliant, and the standards are backward- and forward-compatible"?
18:13:56 <tusho> oh, and slashdot is the only site which uses //foo/ urls
18:14:10 <ais523> tusho: Wikimedia were thinking about using them
18:14:24 <ais523> they discussed it on their mailing list
18:14:38 <ais523> but they were doing a big test of browsers first to make sure they all actually accepted them
18:14:52 <tusho> ais523: TBL has said he regretted making them
18:14:58 <tusho> iirc, he said he wanted to make urls look like this
18:15:07 <tusho> http:org.slashdot/foo
18:15:25 <ais523> I kind-of like the double-slash
18:15:51 <tusho> one stupid thing: DTD identifiers
18:15:58 <tusho> -//W3C//DTD HTML 4.01//
18:16:03 <tusho> -//W3C//DTD HTML 4.01//EN
18:16:10 <tusho> - should be + for unofficial dtds
18:16:19 <tusho> then name// then lang
18:16:29 <tusho> but putting 'DTD' at the start? even sillier
18:16:39 <tusho> and they still require you to specify a url in 90% of cases
18:16:52 <tusho> <!DOCTYPE "http://w3.org/dtd/html/4.01/strict.dtd">
18:17:24 <tusho> ais523: well, yes, but universal identifiers are useful
18:17:30 <tusho> since they can be used for automated validity checking, etc
18:17:31 <ais523> even more radical, I think
18:17:39 <ais523> the funny think is that that does identify HTML5
18:17:48 <ais523> because nothing else uses it...
18:18:06 <tusho> ais523: but the thing is that tools may want to process doc types they don't know natively.
18:18:26 <ais523> yes, and they need to find the dtd somewhere
18:18:36 <ais523> and DDOS the W3C in the process
18:18:46 <ais523> because they aren't clever enough to cache DTDs...
18:18:54 <tusho> ais523: well, yeah, that's the w3cs fault
18:19:02 <tusho> and the user agent's, of course
18:19:06 <tusho> the user agent should cache it
18:19:08 <ais523> you think it should just block broken useragents?
18:19:10 <tusho> but the w3c should optimize for that
18:19:17 <tusho> ais523: that would be nice, yes
18:19:36 <ais523> someone on Slashdot suggested that they make the DTDs deliberately slow
18:19:46 <ais523> to punish useragents that try to look them up on every parse
18:20:21 <tusho> ais523: OTOH they should just block too many requests from the same ip&useragent
18:29:54 <tusho> 'Posted 2008-06-18 23:31 with 3 comments.' ... that's far too crufty
18:30:09 <tusho> '2008-06-18 23:31 / 3 comments'
18:30:16 <ais523> tusho: compress it into one number, like 3200806182331
18:30:20 <ais523> then express it in hex for brevity
18:30:24 <tusho> i'm not crazy, ais523 :)
18:30:31 <ais523> tusho: then why are you here?
18:30:39 <tusho> shush I said it was a lie
18:34:37 -!- Judofyr has joined.
18:35:08 * tusho can't decide if he likes Helvetica Neue or not
18:40:17 -!- ihope has quit (Read error: 110 (Connection timed out)).
18:40:56 <tusho> Ooooh, Optima is nice.
18:43:26 <tusho> Thought of the moment: Long URLs suck. Especially ones including a date
18:43:51 <tusho> ais523: Singular exception. :P
18:44:07 <tusho> http://tusho.org/long-urls-suck is so much superior to http://tusho.org/blog/2008/06/long-urls-suck
18:44:16 <tusho> (Wordpress, I believe, even includes the DAY by default!)
18:44:49 <ais523> tusho: actually, blogging software putting the title of the blog entry in the URL always looked unprofessional to me
18:45:01 <ais523> I'd expect something like http://blog.tusho.org/12
18:45:08 <ais523> where the number was a sequence number
18:45:11 <ais523> blogs are inherently ordered
18:45:15 <tusho> ais523: I, too, prefer URLs which do not offer any insight into the content lying inside!
18:45:19 <ais523> so using unordered titles for them strike me as wrong
18:45:39 <ais523> and it's not as if the title's generally insightful anyway
18:45:53 <tusho> ais523: if the title isn't insightful, then it's a bad title
18:45:59 <tusho> and if it isn't unique, then you suffix it
18:46:10 <tusho> Specifically, you don't suffix the title
18:46:13 <tusho> but the slug it's available at
18:46:30 <ais523> the titles are often cut off by blogging software
18:46:40 <ais523> also they can't handle punctuation marks that aren't legal in URLs
18:46:45 <ais523> which is necessary, but also looks stupid
18:46:48 <tusho> ais523: then the blogging software is stupid
18:46:52 <tusho> good ones let you change the slug easily
18:47:21 <tusho> e.g. I'd put 'Deluxe-O-Matic 3000 v2: A Comedy of Errors' at /deluxe-o-matic-v2-review
18:47:48 <ais523> tusho: still, that gives no clue it's a blog entry, and no clue to the sequence of events
18:47:55 <ais523> blogs are inherently ordered
18:48:11 <ais523> and there's no way to guess the URL if it's title-based
18:48:18 <ais523> you have to rely on links to articles
18:48:25 <tusho> ais523: I won't be using my blog for updatey stuff, really.
18:48:26 <ais523> whereas sequence numbers are easily linked together by hand
18:48:50 <tusho> I might have /amazingsoft-v2-release for a release announcement of Amazingsoft 2, though.
18:49:00 -!- Sgeo has joined.
18:49:00 <tusho> And publishing that at /3423 is just stupid.
18:49:16 <ais523> well, it's easy enough to look at 3422 to see the previous entry
18:49:23 <ais523> whereas with your scheme I can't even guess what the URL is
18:49:34 <ais523> I prefer URLs to be meaningful so that they can be adjusted by humans
18:49:41 <ais523> not to contain content that should be in the page
18:49:48 <ais523> i.e. contain keys used to fetch the data
18:49:52 <ais523> not the result of fetching that data
18:50:06 <tusho> ais523: This is some definition of 'meaningful' with which I was not previously aware.
18:50:06 * Sgeo is missing his interview so he can chat right now.
18:50:09 <ais523> don't tell me blog software seriously keys the blog entries by title
18:50:13 <tusho> and no, they don't
18:50:19 <tusho> but I don't think you understand what meaningful URLs should be
18:50:32 <ais523> tusho: URLs should be meaningful to the server that receives them
18:50:36 <ais523> I feel that, even as a user of the URL
18:50:43 <tusho> Meaningful URLs mean meaningful to a human.
18:50:52 <tusho> And, for what it's worth, looking up by the 'slug' key is not particularly difficult.
18:51:04 <ais523> on Wikipedia, I can append things like ?action=parse and so on to URLs without trouble
18:51:05 <tusho> The slug is just normally made from the title, unless you change it. Which you should, a lot of the time.
18:51:08 <ais523> and it actually happens
18:51:12 <Sgeo> So what happens if two entries would happen to have the same slug?
18:51:13 <ais523> that's what I mean by meaningful URLs
18:51:35 <tusho> Sgeo: They can't. A slug is, by definition, unique.
18:51:48 <tusho> It is a unique, meaningful, URL-safe key.
18:52:16 <Sgeo> So what if there are two entries with the same title? What's changed so that the slugs can be unique?
18:52:31 <ais523> tusho: a slug is not inherently meaningful
18:52:35 <ais523> no matter how much you might want it to be
18:52:37 <tusho> ais523: yes it is, because you make it so
18:52:48 <ais523> tusho: that would prevent it being /inherently/ meaningful
18:53:07 <tusho> ais523: if you use a slug called 'x3454', then you've missed the point of slugs.
18:53:15 <tusho> Unless it's about the X3454 dishwasher..
18:53:26 <ais523> but I'm saying the point of slugs is not useful
18:53:35 <ais523> seriously, meaningful names are too hard to guess
18:53:40 <ais523> and this is a real problem
18:54:16 <ais523> imagine if all the Agora CFJs had 'meaningful' URLs with your meaning, then I'd have to use a search engine to find individual CFJs rather than just typing in the URL
18:54:18 <Sgeo> Easier to guess than meaningless names, I'd think </random-tired-contribution>
18:54:46 <Sgeo> ais523, don't the urls have the CFJ number?
18:55:04 <Sgeo> Which isn't the software, it's part of Agora
18:55:06 <ais523> Sgeo: yes, they do right now
18:55:17 <ais523> and the ID number is a useful key to use
18:55:27 <ais523> it increases in a regular manner
18:55:44 <ais523> if the subject of a CFJ was used as a slug, it would look nicer to the reader but be harder to guess and less useful
18:56:12 <tusho> ais523: yes, but the cfj numbers are exposed to players of agora regularly
18:56:20 -!- Corun has quit ("This computer has gone to sleep").
18:56:22 <tusho> the number of a blog entry isn't on the interface
18:56:26 <tusho> and shouldn't be; nobody cares about it
18:56:28 <tusho> they care about the content
18:56:34 <ais523> tusho: think about xkcd
18:56:40 <ais523> do you agree with their URL scheme?
18:56:45 <ais523> it has numbered versions
18:56:49 <ais523> also text versions IIRC
18:56:49 <tusho> yes, because the comic number is exposed via the interface
18:56:52 <tusho> because it makes sense
18:56:55 <ais523> but the numbered versions are a lot more convenient
18:57:01 <ais523> tusho: yes, why do the comic numbers make sense?
18:57:09 <tusho> because they chose to expose them
18:57:13 <ais523> tusho: so why don't blogs
18:57:20 <ais523> the real reason is, because the comics come in an order
18:57:26 -!- cherez has joined.
18:57:27 <ais523> and the numbers allow people to deduce the order
18:57:40 <ais523> and maybe they should expose numbers too
18:57:43 <tusho> I explicitly opted out of this discussion, I merely offered one sentence.
18:57:45 <ais523> reading a blog in order is not too much to ask
18:57:56 <tusho> If you'll violate that opt-out, then I'll opt out of the channel.
18:58:16 <ais523> this is an interesting new concept of opt-out
18:58:34 <tusho> do I smell sarcasm?
18:58:44 <ais523> s/$/.~/ if you really care
19:01:05 <ais523> psygnisfive: I'm impressed, you had a chilling effect on a conversation and actually meant to have a chilling effect on the conversation, normally they don't go together in IRC
19:01:43 <psygnisfive> so i dont know what oklopol is smoking but with two reaction equations, they werent both supposed to be active
19:01:57 <psygnisfive> but he never really paid attention to anything i said anyway
19:02:19 <ais523> psygnisfive: ah, you're augur
19:02:28 -!- psygnisfive has changed nick to augur.
19:02:36 <ais523> I think I knew you by the psygnisfive nick first
19:02:48 <ais523> and then spent a while wondering who augur was
19:03:13 <ais523> anyway, having both active is an interesting idea, and I think there should be some way to do it
19:03:27 <ais523> but quite probably not with simple syntax
19:03:45 <augur> but there should also be a way to invalidate reactions
19:04:01 <augur> i think simultaneous activity should be special syntax, defined as a single reaction
19:04:17 <augur> something like, maybe, x $ y -> z
19:04:23 <augur> or something like that
19:04:23 <ais523> or alternate activity?
19:04:30 <ais523> i.e. whenever x or y reacts, send its value to z
19:04:36 <augur> alternate activity?
19:04:52 <ais523> well, "when x reacts, x -> z, and when y reacts, y -> z"
19:04:54 <augur> thats simultaneously active reactions...
19:04:54 <ais523> that isn't simultaneous
19:05:03 <ais523> ah, I see what you mean
19:05:08 <augur> no, but both reactions _exist_ simultaneously
19:05:11 <ais523> simultaneously active is different from simultaneous
19:05:17 <augur> whatd you think i meant before?
19:05:17 <ais523> so it's just a language thing
19:05:27 <ais523> augur: two things reacting at the same time
19:05:34 <ais523> i.e. act only when x and y both react at the same instant
19:05:42 <ais523> say because they were both triggered by the same thing
19:08:32 <augur> well, one thing i thing we need to be able to do is have just something that says when a variable changes
19:09:32 <augur> or something, where the value pushed into y is short lived, perhaps almost instantaneously gone
19:09:39 <augur> just a flag saying, x changed
19:09:50 <augur> this might be doable instead with just functions, i dont know.
19:10:08 <augur> x -> { something about y }
19:12:32 -!- atsampso1 has quit (Read error: 54 (Connection reset by peer)).
19:13:29 <tusho> oh look, another programming distraction
19:17:11 -!- RedDak has quit (Read error: 104 (Connection reset by peer)).
19:19:12 <augur> i got new glasses yesterday :)
19:20:21 -!- Corun has joined.
19:22:43 -!- atsampson has joined.
19:33:58 -!- ais523 has quit (Remote closed the connection).
19:34:15 -!- ais523 has joined.
19:38:45 -!- kar8nga has joined.
19:40:35 -!- ais523 has quit (Remote closed the connection).
19:40:49 -!- ais523 has joined.
19:41:09 -!- kar8nga has left (?).
19:42:48 -!- atsampson has quit (Read error: 104 (Connection reset by peer)).
19:44:00 -!- atsampson has joined.
19:44:42 <augur> i wish there was a way to speed up my cognition, my perception of time, so the world seemed slower, without affecting my thought
19:45:41 -!- augur has quit ("Leaving...").
19:45:52 -!- tusho has changed nick to ehird.
19:46:02 -!- ehird has changed nick to tusho.
19:46:28 <tusho> Darn. If I get tusho.org I need usho.org
19:46:43 <tusho> And usho.org IS TAKEN!!!!
19:46:43 <ais523> tusho: that so isn't worth it
19:46:47 <Slereah_> Can't you give that money to orphans instead?
19:47:03 <tusho> x@tusho.org is okay I guess
19:47:17 <tusho> or id@tusho.org, but that's just geeky
19:47:24 <tusho> (which leads to ego@tusho.org, heh)
19:53:12 <tusho> I like ω@tusho.org better, though
19:53:21 <ais523> tusho: just have @tusho.org
19:53:27 <ais523> that confuses spambots like mad
19:53:29 <tusho> ais523: is that a zero-width joiner?
19:53:37 <ais523> no, it's a null string
19:53:41 <ais523> that's legal, apparently
19:53:47 <tusho> how many things accept it?
19:59:11 -!- augur has joined.
20:00:25 -!- jix has quit ("CommandQ").
20:06:17 <augur> what would you suggest for the representation of both reactions being active?
20:09:01 -!- jix has joined.
20:21:55 <augur> thats bitwise logic :p
20:23:00 <augur> we need a neurohacking channel
20:23:57 <augur> ::hacks ur nurons::
20:34:07 <fizzie> My reading of RFC2822 would say that a simple @example.org address is not valid. The syntax says it's local-part "@" domain where local-part is either dot-atom, quoted-string or obs-local-part; few definitions deeper in they mostly start with 1*atext, meaning there must be at least one character. Although a zero-length quoted string should be valid, but ""@example.org is not as pretty.
20:34:47 <ais523> someone said @example.org worked, but didn't give evidence
20:34:53 <ais523> I wonder if most email transport agents support it?
20:35:51 <fizzie> Wwouldn't be surprised if it were so; but I'm sure there'd be at least some systems that would be confused.
20:36:00 <tusho> fizzie: I dislike you for your usage of facts and evidence.
20:36:31 <tusho> t@usho.org would be awesome, still
20:36:32 <ais523> wow, I've just found another thing to disagree with tusho on
20:36:40 <tusho> though if I was tasho, t@sho.org would be funner
20:36:44 <tusho> but sho.org will be even more taken
20:36:47 <tusho> ais523: oh come on, it was a joke
20:39:02 <ais523> oh, for all the IE-bashers out there: http://mosspod.com/ie7_and_gmail.html
20:39:26 <ais523> that's almost as bad as MSN Messenger blocking links to YouTube, or Hotmail blocking messages from Yahoo mail
20:47:46 <tusho> hmm, i should start using #pfft more
20:47:52 <tusho> I don't like clogging ##nomic and #esoteric with random crap
20:48:36 -!- jix has quit ("CommandQ").
20:53:11 -!- kar8nga has joined.
20:53:47 -!- jix has joined.
21:02:49 -!- ais523 has quit ("(1) DO COME FROM ".2~.2"~#1 WHILE :1 <- "'?.1$.2'~'"':1/.1$.2'~#0"$#65535'"$"'"'&.1$.2'~'#0$#65535'"$#0'~#32767$#1"").
21:09:48 -!- mckiko has joined.
21:13:54 -!- Slereah has joined.
21:14:05 -!- Slereah_ has quit (Read error: 104 (Connection reset by peer)).
21:46:17 -!- mckiko has quit.
22:04:37 -!- caio has quit ("Leaving").
22:10:11 -!- jix has quit (Read error: 113 (No route to host)).
22:12:13 -!- jix has joined.
22:15:38 -!- Corun has quit ("This computer has gone to sleep").
22:35:07 <Slereah> http://esolangs.org/wiki/PlayGround
22:35:15 <Slereah> What, there's JumpRope but no tag?
22:39:07 <tusho> before becky there is tom
22:40:25 -!- Corun has joined.
22:52:56 -!- jix has quit ("This computer has gone to sleep").
22:55:30 -!- ihope__ has joined.
22:55:35 -!- ihope__ has changed nick to ihope.
22:55:45 <tusho> IHOPE IHOPE IS HERE
22:56:50 <tusho> I tend to dislike things that don't exist, though :(
23:03:26 -!- kar8nga has left (?).
23:05:08 <tusho> http://esolangs.org/wiki/Esme
23:05:15 <tusho> most uesless page imaginable
23:05:43 <Slereah> I'm sure we'll soon discover that it actually has BF's instruction set.
23:38:52 -!- rodrigo has quit ("rfcregnb ip, urva?").
23:57:41 <Slereah> Here's the sourcecode then :
23:58:26 * tusho is waiting for commands
23:59:04 <tusho> [output] Print this sentence.