00:00:19 <oerjan> google throws away _very_ common words for one thing, perhaps it makes this too hard to do
00:00:44 <oerjan> um wait maybe it doesn't
00:01:18 <oerjan> maybe they did before. "the" worked fine now :D
00:01:34 <fizzie> It used to, at least in some point of data, but that changed.
00:01:56 <oerjan> hey, i tried "the or and"
00:02:06 <fizzie> Still, information retrieval is a field which has had a huge amount of published papers.
00:02:08 <oerjan> and interestingly enough, one link was this:
00:02:21 <oerjan> http://lib.colostate.edu/tutorials/boolean.html
00:03:15 <fizzie> That's about "searching efficiently" in the sense of getting good results, not in the sense of implementing the search.
00:03:40 <oerjan> (also google guide showed up)
00:03:58 <oerjan> bsmntbombdood: you cannot show it?
00:04:39 <fizzie> But a plain old-fashioned inverted index is the most basic thing used for search-engine-like things. I'm sure there are hueeg improvements to it.
00:04:49 <oerjan> bsmntbombdood: well i have norwegian language setting but i don't really see why that should make you lose it entirely
00:05:26 <oerjan> bsmntbombdood: it was the last link on the first page for me
00:05:54 <fizzie> Still, if you have 10^9 items but the tags A and B appear in 10 items each, it's obviously better to just take the intersection of A's and B's rows (which are sets of 10 elements) of the inverted index, instead of iterating through all the 10^9 elements.
00:06:15 <oerjan> i'd have expected you to get it _higher_ up, since you would have no norwegian hits before it, and it is in english...
00:06:33 <bsmntbombdood> fizzie: but in reality you have 10^9 items where tags A and B appear 10^6 times
00:07:30 <fizzie> Well, that's still only around 10^6 operations to do instead of 10^9.
00:12:07 <oerjan> lessee google often guesses _wrong_ how many hits there are until you actually look at the last hits page
00:12:42 <oerjan> this means it may actually calculate things lazily, doing only the minimal work to create the pages you view
00:12:49 <fizzie> Only thing I remember from the Google example is that their huge-ass index is divided into shards hosted by gazillion servers; so you can do those "search for this bit" and "search for that bit" operations in parallel.
00:13:14 <fizzie> I really don't know how google does their fast approximations, though, given how it needs to get highly-pageranked results first.
00:13:33 <fizzie> http://labs.google.com/papers/googlecluster-ieee.pdf is mostly about the hardware but also a bit about the software.
00:14:53 <fizzie> And of course you could easily spend an interesting week reading random papers from the few-hundred-item http://research.google.com/pubs/papers.html list.
00:16:37 <oerjan> "On average, a single query on Google reads
00:16:37 <oerjan> hundreds of megabytes of data and consumes
00:16:39 <oerjan> tens of billions of CPU cycles."
00:19:58 -!- ehird has quit ("Konversation terminated!").
00:20:01 -!- _ehird has changed nick to ehird.
00:22:59 -!- oerjan has quit ("Reboot").
00:26:08 -!- oerjan has joined.
00:41:29 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | the first version would need to be floppy based for increased pain.
00:59:25 <bsmntbombdood> BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB
01:01:33 <oerjan> ^bf ++++++++[>++++++++<-]>+++[.]
01:01:33 <fungot> CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC ...
01:04:15 <bsmntbombdood> http://www.hulu.com/watch/1404/saturday-night-live-snl-digital-short-natalie-raps
01:08:53 <oerjan> http://www.ams.org/mathscinet/collaborationDistance.html
01:10:42 <oerjan> alas there is no separate URL for the result, that i can see
01:13:03 <oerjan> alas, as i haven't been in any films
01:13:54 <bsmntbombdood> you're not the most gorgeous creature to walk the earth ither
01:14:34 <oerjan> especially not with this damn rash
01:30:09 <warrie> I want an Erdos number.
01:31:46 <warrie> I want an Erdos number of omega + 1.
01:31:50 <Slereah> I wonder if I have a Bacon number.
01:31:56 <warrie> Or, you know, a finite one.
01:32:09 <Slereah> Depends on your definition!
01:32:41 <oerjan> the question is, what definition would Erdos use?
01:33:00 <warrie> I think my Erdos number is actually the class of all ordinal numbers.
01:33:10 <Slereah> There's an easy way to have a good erdos number.
01:33:22 <Slereah> Time travel, hook up Erdos' dad with your mom.
01:33:46 <warrie> Interview someone with a low Erdos number.
01:34:12 <Slereah> Only if you both sign a piece of paper!
01:34:30 <Slereah> Ask him to help you with your math homework
01:35:39 <warrie> Honestly, it's likely that I will never need help with my math homework.
01:36:36 <Slereah> I wonder if one of my professors has an Erdos number.
01:36:45 -!- Sgeo has joined.
01:36:51 <warrie> I wonder how many of us fall under the "will never need help with math homework" category.
01:37:25 <oerjan> misinterpret your math homework in some horribly convoluted way and get a collaborator to make a paper out of it
01:37:26 <Slereah> The ones of us who have no math in their education
01:37:47 * Sgeo comes close to falling in that category
01:38:08 <warrie> Not having math homework doesn't count, Sgeo.
01:38:12 <Slereah> What do you do fo' school?
01:38:12 <Sgeo> I think I needed help once
01:40:17 <warrie> You know, I just remembered that homework in my math class is optional.
01:43:42 <warrie> Now I wonder just how smart we all are.
01:45:07 <warrie> Dumb as in "in the bottom half of the top 30 out of 600"?
01:46:25 <warrie> It wouldn't be rude of me to just tell you how smart I am, would it.
01:47:16 <Slereah> Tell us how smart you are, warrie
01:47:30 <Slereah> Does the answer involve pigs and lipstick?
01:48:54 <oerjan> that somehow seems more relevant to bacon numbers than erdos numbers
01:49:48 <Slereah> I wonder if I have a Bacon number.
01:49:51 * oerjan now wonders how many pigs with lipsticks have bacon numbers
01:49:56 <warrie> No, only being precisely tied with one other person for smartest in my class (about 600 high-schoolers), as best I can tell.
01:50:35 <warrie> It's annoying that I don't know whether I'm smarter than the other guy or vice versa.
01:51:26 <Slereah> Does it give you a boner to be smart
01:51:28 -!- GreaseMonkey has joined.
01:51:39 <Corun> I felt a bit esoteric
01:51:40 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
01:51:46 <Corun> http://rafb.net/p/4y7yOA86.html
01:52:08 <warrie> I also wonder whether it's a coincidence that he and I both have two brothers and no sisters, and I am disappointed that his brothers are both very smart and mine are not.
01:53:08 -!- Sgeo has joined.
01:53:32 <Sgeo> What did I miss, if anything?
01:53:32 <warrie> If it's not self-modifying times a constant, it... is sub-optimal.
01:53:42 <oerjan> my brothers were so smart they didn't even get born
01:53:57 <warrie> Maybe that's what happened to all my and his sisters.
01:54:10 <warrie> Sgeo: what's the last thing you saw?
01:54:24 <Slereah> They were born as energy being
01:54:30 <warrie> You missed Corun's AI and perhaps other things: http://rafb.net/p/4y7yOA86.html
01:54:32 <Sgeo> * GreaseMonkey (n=saru@unaffiliated/greasemonkey) has joined #esoteric
01:54:36 <Slereah> You're only smart because of me.
01:55:04 <GreaseMonkey> * Sgeo has quit (Read error: 104 (Connection reset by peer))
01:55:06 <Sgeo> What does it do?
01:55:24 <Corun> It's a conversational AI
01:55:30 <oerjan> Sgeo: IT'S SMART DUMBASS
01:55:57 * warrie attempts to compile it
01:56:04 <Slereah> But what does it talk about?
01:56:12 <oerjan> Corun: Whatever you do, don't tell it to make paperclips
01:56:57 * Sgeo doesn't see any wordlists or anything..
01:57:20 <Slereah> It's so smart that it can LEARN TO TALK
01:57:35 <Corun> I told it not to make paperclips
01:57:41 <Corun> And it responded with "UU$>'SV,TT'<qT)4R,'\<AG"
01:58:01 <warrie> Mmh. How do I compile it, I wonder.
01:58:13 <Corun> gcc -std=c99 tehfile.c
01:58:19 <Corun> gcc -std=c99 -lpthread tehfile.c
01:58:25 <Corun> Depends on your system really :-)
01:58:28 -!- Slereah has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | UU$>'SV,TT'<qT)4R,'\<AG.
01:58:38 * warrie goes with gcc -std=c99 -lpthread ai.c
01:58:44 <Sgeo> ..there's no "strip line numbers" link
01:58:46 <GregorR> gcc -std=c99 -pthread -vomit-frame-pointer -fuck-the-damned-loops foo.c
01:58:52 -!- z8000_ has joined.
01:59:00 <Sgeo> ..and selecting makes it work..
01:59:07 <warrie> ai.c:54: warning: incompatible implicit declaration of built-in function 'malloc'
01:59:11 -!- z8000_ has left (?).
01:59:12 <oerjan> -vomit-frame-pointer is really something
01:59:25 <GregorR> oerjan: It's -___f___omit-frame-pointer :P
01:59:57 <Corun> What do you think?
02:00:26 <oerjan> i think it's a joke, obviously
02:00:39 <warrie> It's outputting longer and longer nonsense.
02:00:58 <Sgeo> It's not responding, and my system's slowing down
02:01:01 <Corun> It's not exactly perfect
02:01:06 <Corun> I'd suggest not running it for too long
02:01:07 <Slereah> It's an AI, but it doesn't talk your language, warrie
02:01:10 <Corun> Uses up a lot of cpu
02:01:24 <Sgeo> I can't seem to make it respond
02:01:25 <oerjan> it's more the infinite number of monkey type, i take
02:01:28 <Corun> I'd suggest killing it
02:01:40 <Corun> You on a crappy OS Sgeo?
02:01:40 <warrie> I'm not running it on my system, so I can use up as much CPU time as I want.
02:01:51 <Sgeo> Corun, is Ubuntu 7.04 crappy?
02:01:57 <warrie> (Disclaimer: I rent the system, but that doesn't make it mine.)
02:02:03 <Sgeo> sgeo@ubuntu:~/c$ ./someai
02:02:16 <Corun> If it can't handle a process trying to fully utilize the cpu with 11 threads, yes :-P
02:02:29 <Sgeo> I did get a lot of errors, if that has anything to do with anything
02:02:32 <warrie> Corun, was this supposed to do something?
02:02:54 <Corun> You didn't seriously expect a 71 line of code "AI" to do something, did you?
02:03:12 * Sgeo expected it might have at least repeated what was put into it
02:03:18 <warrie> Surely you can implement a neural net in that space.
02:03:26 <Corun> It might be a neural net
02:03:35 <Corun> I'm not really sure
02:03:42 <Corun> It just kinda does stuff with numbers in an array
02:03:46 <warrie> After putting "aaaa" into it many times, it seems to be giving me especially many aaaas.
02:04:07 <Slereah> http://membres.lycos.fr/bewulf/Russell/noyoucanthaveaneuralnetwork.jpg
02:04:16 <Corun> Hah, you're right warrie
02:04:42 <Corun> It does actually do like, some stuff
02:05:07 <warrie> Outputs stuff like this: k;:9w999877jwj7hvqk76w55543jiiuk3guk3hg333g22f1gevpgf11vow00////wwwwvtrtqkwwwwwwwwvwuwuvwtqkwwwwwvvvuvwuvwuwwwwwwuwwwvvvuwuvbtuututttttswvwvwvwutwwvwwvwwvvusvtr
02:05:27 * warrie tries yes Wolfram Blitzen | ./a.out
02:05:48 <warrie> It's really stupid. :-)
02:06:11 <warrie> Not-very-representative sample: V\Le@\hWTUCflkXWTTTTTSV[WTSVV[[Bldti\ mBkXlthBekWWSSWlfj mBjlfi\leVZi[BfnnmAKaa`_l; l lheBVeVYYeVYeWUCoal_:XdmAJnAm@[aJld`_:lWoheUCcBCdeUC*eUCallmk?ZVYrelojBljr
02:06:12 <Corun> I think it's too chaotic
02:06:27 <Corun> It does just randomly mutate "brain"
02:06:30 <Corun> Every, uhh, iteration
02:06:37 <Corun> Of each of the 11 threads
02:07:18 <Corun> I challenge any of you to make a cleverer AI in less lines of code!
02:08:08 <Corun> And it can't just echo
02:08:11 <Slereah> Corun : what about a program that repeats what you said, but replaces all vowels with o?
02:08:29 <Corun> It has to atleast do something vaguelly AI like
02:08:32 <Slereah> And would simulate perfectly a /b/tard
02:08:47 <Slereah> Does your thing do something AI-like?
02:08:50 <oerjan> ah, artificial stupidity
02:09:03 <Sgeo> Corun, what, you mean like a maze solver?
02:09:19 <Sgeo> Or some other game opponent? Because I know that's not the sort of AI you mean
02:09:30 <Corun> You've lost me :-P
02:09:36 <warrie> AI is really easy to implement, and will continue to be really easy to implement until I actually try to implement it.
02:09:39 <Corun> Make a conversational AI in < 71 lines of C
02:09:43 <Slereah> Artifical stupidity is easy to do.
02:09:48 <Slereah> ffffffffffffffffffffffffffffffffffffff
02:09:49 <Sgeo> Corun, meh at "conversational"
02:10:02 <Corun> Which is _better_ than this one. But isn't just echo or some other take on echo.
02:10:25 * Sgeo doesn't know C well enough, can I write one in python and have someone translate it to C?
02:10:34 <Corun> Write on in python
02:10:38 <Corun> In less than 50 lines
02:10:43 <Corun> Cos that's. like, equivalent
02:10:58 -!- lightbringer has joined.
02:11:15 <Slereah> But... Python has indentation!
02:11:16 <Sgeo> Hm, this is worse than "a take on echo" though..
02:11:24 <Slereah> With C, you can do everything in one line
02:11:39 <Corun> Oh stop nitpickin'
02:11:46 <Corun> All I'm hearing are excuses :-P
02:12:15 * Corun goes to make his one better
02:12:44 <Slereah> It's important for a conversational AI
02:13:52 <Slereah> Maybe it should answer "I AM A CONVERSATIONAL AI BEEP BOOP" for every input
02:14:56 <Sgeo> 6 lines of an AI that's WORSE than an echo: http://rafb.net/p/XKfd7922.html
02:16:24 * Sgeo was going to make it randomly answer "Yes." or "No." but that fails with non yes/no questions
02:17:43 <Corun> I don't think 71 lines is enough
02:18:04 <Slereah> Sgeo : You could make it answer yes or no with sentences beginning with is/are/have/has
02:21:32 * Sgeo will make it do that and "do" too
02:21:46 <Sgeo> ^^intentionally interpretable dirtily
02:24:40 <Sgeo> Are you alive?
02:25:00 * Sgeo is suprised at the fact that it answered the questions in a consistant manner
02:25:03 <Slereah> It's way better than Corun!
02:25:23 <Slereah> Ask him if he wants ice cream
02:26:21 <Corun> After all, my prowess is gone
02:26:25 <Corun> I have no reason to stay
02:26:27 -!- Corun has left (?).
02:27:01 <Sgeo> http://rafb.net/p/7ZMDyW92.html
02:27:35 <Slereah> Maybe you should also make a special answer for "Who are you?"
02:27:47 <Slereah> That would involve a lot of BEEP BOOP
02:28:43 <Sgeo> really really?
02:30:49 <Sgeo> Me, or the AI?
02:31:37 <Sgeo> I am an all-powerful God-Human. You are a rather lame AI
02:32:57 <Slereah> Idea : An AI that only answers "AFK"
02:33:01 <Sgeo> http://rafb.net/p/jZ5V8k65.html
02:33:28 <Slereah> Sgeo : That doesn't seem to good
02:33:41 <Slereah> Like "Who's Julius Caesar?"
02:34:01 <Sgeo> Actually, it would say "I don't know." because of the 's
02:34:17 <jayCampbell> post-clinton, the "'s" is also indeterminate
02:35:38 <Sgeo> http://rafb.net/p/zk6haK59.html
02:38:33 -!- oerjan has quit ("Good night").
02:39:26 -!- Corun has joined.
02:43:57 <Sgeo> Corun, want me to msg you what you missed?
02:47:04 -!- Slereah has quit (Read error: 148 (No route to host)).
02:49:31 -!- Corun has quit ("This computer has gone to sleep").
03:16:29 -!- Slereah_ has joined.
03:30:28 <Sgeo> Slereah_, what's wrong?
03:31:34 <Slereah_> But I've got to say, it looks good and didn't age.
03:34:16 -!- GreaseMonkey has quit ("rebooting into freebsd").
03:37:20 -!- GreaseMonkey has joined.
03:42:58 -!- Corun has joined.
04:44:09 -!- MizardX has joined.
04:47:56 -!- Corun has quit ("This computer has gone to sleep").
05:46:11 -!- psygnisfive has joined.
05:46:15 -!- oklopol has quit (Read error: 104 (Connection reset by peer)).
05:46:30 -!- oklopol has joined.
05:49:32 -!- oklopol has quit (Read error: 104 (Connection reset by peer)).
05:49:56 -!- oklopol has joined.
05:59:58 -!- Sgeo has quit (Remote closed the connection).
06:41:29 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | print (lambda x:x%x)("(lambda x:x%%x)(%r)").
06:45:54 -!- lightbringer has left (?).
07:17:53 -!- ab5tract has joined.
07:20:01 <psygnisfive> what languages do full propositional logic?
07:40:14 <psygnisfive> and other kinds of normal propositional logic + set theory kind of stuff
07:41:22 <psygnisfive> they're just not as easy to code as prolog because of the craziness of logic
07:53:25 -!- ab5tract has quit.
07:53:59 -!- olsner has quit ("Leaving").
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:39:23 -!- GreaseMonkey has quit ("So, how much do you love noodles?").
09:13:56 <psygnisfive> universal quantifier on x, existential on y
09:14:47 -!- sebbu has joined.
09:16:51 <fizzie> $\forall x \exists y$, in LaTeX-speak.
09:17:38 -!- sebbu2 has quit (Read error: 113 (No route to host)).
09:18:55 <AnMaster> fizzie, ok that looks the same, and doesn't explain it at all
09:19:05 <AnMaster> I guess I just lack the knowledge
09:23:13 <fizzie> The only "real logic" tool I've personally used is Otter -- http://www-unix.mcs.anl.gov/AR/otter/ -- but that's for theorem-proving, not really programming.
09:28:40 <fungot> .. !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ...
09:28:46 <fungot> AnMaster: hit the road, that works only if you ask me.
09:31:58 <fungot> fizzie: create function adder(a int(10), b int(10)) returns t
09:34:53 <AnMaster> interesting, gcc optimises the main switch block used for the core instructions in cfunge the same way even if I swap the order of the entries at -O2 and higher
09:35:06 <AnMaster> at -O1 and -O0 it generates different code
11:59:36 -!- oklopol has quit (Read error: 54 (Connection reset by peer)).
12:19:04 -!- oklopol has joined.
12:27:45 -!- oerjan has joined.
12:29:28 -!- ehird_ has joined.
12:39:50 -!- Corun has joined.
12:41:29 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | hi.
12:41:53 <optbot> oerjan: I was, obviously, thinking of something completely different.
12:51:35 -!- ehird_ has quit ("Lost terminal").
12:52:04 -!- Corun has quit ("This computer has gone to sleep").
13:20:05 -!- Corun has joined.
13:21:13 -!- Corun has quit (Client Quit).
13:24:04 -!- ehird_gobo has joined.
13:24:20 <ehird_gobo> I am using konqueror from kde4 ... in twm :S
13:35:25 -!- ehird_gobo has quit ("http://www.mibbit.com ajax IRC Client").
13:38:15 <oklopol> psygnisfive: do you mean predicate logic? you can build sets from within it, so.
13:41:49 <oklopol> maybe i should make noprob++ at some point, noprob + predicate logic.
13:41:54 <oklopol> that would be... confusing.
13:42:12 <oerjan> propositional logic is the _easy_ part. it is merely NP-complete, nothing worse. ;D
13:42:58 <oklopol> isn't first order predicate calculus too?
13:43:32 <oklopol> definitely not, at least pspace
13:43:41 <oklopol> but i mean it's not semidecidable!
13:44:17 <oerjan> and set theory has the full force of godel incompleteness
13:45:24 <oerjan> i don't quite recall but i think predicate calculus must be undecidable too
13:45:38 <oklopol> well. i don't know much about this. i just thought resolution was a complete search procedure for first order PC.
13:46:06 <oklopol> second-order is, that's for sure.
13:46:25 <oklopol> you need induction to get *numbers*... "tc"
13:46:42 <oklopol> but, as you can clearly see, going by intuition here, so :P
13:46:59 <oklopol> and induction needs 2nd order pc, naturally
13:47:03 <oerjan> can't you encode combinator calculus in predicate calculus?
13:47:14 <oerjan> it's just equations, after all
13:47:24 <oerjan> so it _must_ be undecidable
13:48:09 <oerjan> a complete search procedure may only find what's provable, after all, not everything needs to be decidable
13:48:15 <oklopol> yeah i guess you're right, can't stress enough that i don't know absolutely anything.
13:48:37 <oklopol> by completeness i meant it's complete both ways.
13:50:16 <oerjan> for all x,y,z : S x y z = x z (y z) and K x y = x
13:50:55 <oerjan> just predicate logic with functions. and functions can be recoded into relations.
13:51:44 <oklopol> yes that sounds about right
13:53:02 <oklopol> :oooooooooooooooooooooooooooooo
13:53:30 <oklopol> i don't like the way pc does functions
14:00:54 -!- oerjan has quit ("Later").
14:20:30 -!- ehird_ has joined.
14:21:47 <AnMaster> by the way, just wondering, how long have you been using gobolinux?
14:22:11 <ehird_> since yesterday. i fucked up kde because i'm a retard.
14:22:17 <ehird_> so i'm talking to you via the console about to set it compiling.
14:23:14 <AnMaster> ehird, do you still like gobolinux? :)
14:23:27 <ehird_> yes. and i still hate X.
14:23:48 <AnMaster> oh yes I can see what you mean, but doesn't xorg support auto detection these days?
14:24:23 <ehird_> Well, yeah, except it fails at getting my scrollwheel to work.
14:24:37 <AnMaster> Oh and: no I don't/won't use it, since as you pointed out I prefer the old way
14:24:38 <ehird_> Also, technically messing up KDE was my fault, but i'm blaming it on KDE.
14:24:44 <AnMaster> but it should be the way you prefer :)
14:24:57 <ehird_> Because I hate X, anyway, so KDE might as well get some of the anger.
14:25:14 <AnMaster> ehird, what did you do to poor KDE?
14:25:39 <ehird_> I installed a binary package of 4.0.something which is apparently unstable and nobody actually knows where that binary came from, I should have checked up before doing that
14:25:46 <ehird_> So I get Konqueror running in TWM.
14:26:01 <ehird_> I'm compiling it in 3..2...1
14:26:18 <AnMaster> and yes 4.0, don't you know where you got it youself?
14:26:58 <ehird_> I should have checked what KDE version was stable, what binary packages hadn't magically appeared due to faeries, etc.
14:27:15 <AnMaster> ah right. Well we all make mistakes from time to time.
14:27:22 <ehird_> So now I get to wait for it to compile the whole of KDE!
14:27:24 <AnMaster> are there no binary packages of 4.1 or whatever is the last one?
14:27:27 <ehird_> Woo! That certainly won't take hours!
14:27:34 <ehird_> AnMaster: Apparently not. I could make one, I guess.
14:27:51 <ehird_> AnMaster: Intel Core 2 Duo @2.17ghz (iirc), 1GB of RAM
14:27:57 <ehird_> and the rest should be wholly irrelevant to compiling :P
14:28:16 <AnMaster> compiling KDE 3.5 here on a 2 GHz Sempron takes around 5 hours in total for kdelibs, kdebase, kdesdk and parts of the other packages
14:28:25 <AnMaster> so it should be faster for you
14:28:41 <AnMaster> ehird, well this is a single core sempron
14:28:58 <AnMaster> I bet your cpu is quite a bit better
14:29:03 <ehird_> Eh, I have irssi and w3m. I'll be fine.
14:29:08 <ehird_> I could do with an IM client, though.
14:29:15 <AnMaster> there is w3m-mode for emacs btw
14:29:27 <ehird_> There's also a window manager for xemacs.
14:29:43 <ehird_> Grr, w3m needs to let me set an artificial width for page rendering.
14:29:45 <AnMaster> I'm probably moving to awesomwm rather than KDE4
14:29:59 <ehird_> Tiny fonts + 21" widescreen = awful reading experience.
14:30:13 <ehird_> Also, the GoboLinux console font is a bit ugly.
14:30:32 <ehird_> The console is preferable to twm, AnMaster.
14:31:15 <AnMaster> if yes links -g could work okish
14:31:35 <ehird_> Framebuffer, I assume, since w3m is rendering images.
14:31:40 <ehird_> And I don't have links :p
14:31:56 <ehird_> I'm not installing it, I don't know if Compile would barf if I run two instances at once.
14:32:22 <AnMaster> Gentoo Portage handles multiple instances fine
14:32:40 <ehird_> Compile probably does too, but I'm allowed to be careful when I'm donating hours and hours to compile KDE>
14:32:44 <AnMaster> (it uses lock files for when it is merging from the fake install root to the actual real file system)
14:33:37 <AnMaster> and yes I would be careful too. I certainly was until I found out gentoo handles it just fine
14:33:45 <ehird_> I wonder if the Linux kernel devs would accept my hypothetical patch to allow proportional fonts on the console. :P
14:34:09 <ehird_> Although if they would they might as well accept my other patch, "rm -rf console; ln -s X11 console"
14:34:36 <AnMaster> most likely not. IIRC the "non-graphic" OS X mode (use odd key combo I forgot) is monospace iirc
14:35:00 <AnMaster> also ever looked at ls output in a non-monospace font?
14:35:19 <AnMaster> Very often the output is very hard to read
14:35:25 <ehird_> Well, that's why it'd be optional.
14:35:29 <AnMaster> since almost all programs are coded for monospace
14:35:34 <ehird_> proportional; w3m; proportional off
14:35:45 <ehird_> Better idea: rm -rf console; ln -s X11 console
14:35:57 <AnMaster> ehird_, still, RTC: Rich Text Console
14:35:57 <ehird_> rm -rf X11; ln -s /OSX/Quartz X11
14:36:25 <AnMaster> since everyone would need to buy their stuff to get that
14:37:24 <ehird_> Ok, Qt doesn't want to compile.
14:37:38 <ehird_> How come all my linux usage ends up being stereotypical to the max? :-P
14:37:47 <ehird_> AnMaster: It thinks the compiler isn't supported.
14:37:50 <AnMaster> ehird, um I don't have such problems on gentoo
14:37:54 <ehird_> //Long/Path/Here/linux-g++
14:38:01 <ehird_> AnMaster: I'm just unlucky
14:38:12 <AnMaster> "checking if compiler is GCC"?
14:38:13 <ehird_> "Compiler not supported, read the qt readme you moron"
14:38:32 <AnMaster> report a bug to gobolinux I guess
14:38:38 <ehird_> Well, I've asked in #gobolinux.
14:38:42 <ehird_> I could use a Qt binary pkg.
14:39:07 <AnMaster> and if you have several installed, is QT using the right one?
14:39:10 <ehird_> 4.1.2. I'm going to mass-upgrade the distribution overnight; so it's a bit old.
14:40:15 <ehird_> Trying to install 4.4.x, the binary package is 4.3.x.
14:40:25 <ehird_> Qt takes ages to compile, I know this... so I'm leaning to the binary package
14:41:14 * ehird_ wonders how the UK's national id card scheme is getting along. Freeeeeeedommmmmmmm
14:41:23 <AnMaster> well yes qt 4 is really slow to compile
14:42:32 <ehird_> The compiled Qt is SEVENTY SEVEN MEGABYTES
14:42:48 -!- Slereah has joined.
14:44:36 <ehird_> Downloading @ 800KB/sec, though.
14:44:39 <AnMaster> ehird, which is why I'm going to leave KDE after my distro declares 3.5.x as end of life
14:45:44 <ehird_> AnMaster: I know about awesomewm.
14:45:49 <ehird_> But KDE4 looks really nice.
14:46:02 <ehird_> It seems to fix everything I didn't like about KDE, although that's probably everything you liked.
14:46:15 <AnMaster> ehird, well I expect you to complain at least twice about how it is just a bad copy of the special effects in OS X ;P
14:46:27 <ehird_> No, that's Beryl/Compiz/whatever it's called now.
14:46:56 <ehird_> Ah. The binary Qt is too old.
14:47:38 * ehird_ considers patch to w3m to run flash.
14:47:43 <ehird_> YOUTUBE ON THE CONSOLE
14:49:05 <AnMaster> ehird, how well would you say gobolinux handles dependencies between packages?
14:49:21 <ehird_> Perfectly; I haven't seen any problems so far.
14:49:28 <ehird_> It just has a list of dependencies/versions and recursively calls itself for them.
14:50:03 <AnMaster> I have bad experience with for example debian when it comes to that, issues figuring out exactly what version is needed and so on
14:50:29 <AnMaster> (oh and portage handles it perfectly too)
14:50:47 -!- oklopol has quit (Read error: 104 (Connection reset by peer)).
14:50:53 <ehird_> Debian do their best shot; it's just a problem because they have such a comprehensive package base
14:51:11 -!- oklopol has joined.
14:51:20 <AnMaster> and yum sucked when I last tried, it was a few years ago, no idea how it is these days
14:51:36 <ehird_> But Debian is too steeped in the past, anyway; RPM/APT are pretty much the worst packaging systems that anyone actually uses today :-P
14:51:40 <ehird_> Apart from Cygwin's package manager. That is awful.
14:52:02 <AnMaster> if that is what you want really then Arch Linux should be quite interesting for you. For example on Arch Linux GCC 4.3.2 hit stable about 4 days after upstream released it
14:52:05 <ehird_> But if a program's been out for, say, 5 months, I expect it to be in.
14:52:12 <ehird_> So: Up-to-date, but not bleeding-edge alphas.
14:52:18 <AnMaster> usually kernel takes a week or two
14:52:58 <AnMaster> oh and not alphas, but I mean fast getting packages into the stable version
14:52:58 <ehird_> Yeah, that's a bit too bleeding-edge for my tastes.
14:53:13 <ehird_> However, a system which is constantly updated to the very bleeding edge would be ... interesting ...
14:53:40 <AnMaster> ehird, ah then you would like the other end of arch linux. libcap is on version 1.10 or so iirc. But all other distros are around 2.11
14:53:47 <AnMaster> I guess a lazy package maintainer
14:53:58 <AnMaster> oh and libcap 1.x is for 2.4 kernels
14:54:04 <ehird_> macports is pretty good in my experience
14:54:14 <ehird_> their maintainers keep spontaneously disappearing, though.
14:54:19 <AnMaster> hm gobolinux is rolling release right=
14:54:32 <ehird_> You mean, constantly updated
14:54:37 <ehird_> with a snapshot as a release?
14:54:41 <AnMaster> as in you don't need to wait for a new release to get anything but bugfixes
14:54:50 <AnMaster> and releases are just snapshots for easy install yeah
14:55:00 <AnMaster> right that is a MAJOR requirement for me
14:55:02 <ehird_> Any other kind of distro is just paranoia
14:55:11 <ehird_> "but what if it breaks?" Test it, stupid.
14:55:12 <AnMaster> rolling release are the only maintainable ones
14:55:22 <AnMaster> because on them upgrading is easy
14:55:27 <ehird_> I run Ubuntu on eso-std.org, though.
14:55:48 <AnMaster> while for example ubuntu and such in my experience tends to be a bit of a pain to upgrade to a new release
14:55:53 <ehird_> Because I'd rather have it certain to be stable and reliably-the-same than be on the cutting edge with it
14:55:57 <AnMaster> at least on a running remote system
14:56:12 <AnMaster> though i have done remote freebsd upgrades a few times
14:56:24 <AnMaster> but yes rolling release makes it all much simpler
14:56:27 <ehird_> You can actually run any distro on Slicehost as long as it works with the kernel they have (They give you a list of distros, due to their Xen setup, so they can auto setup them)
14:56:40 <ehird_> You have to install it into a chroot, then methodically rip out the current system and put the stuff from the chroot in.
14:56:43 <ehird_> Then reboot and hope it still runs.
14:56:47 <AnMaster> arch linux, gentoo linux, and (as you said) gobolinux are all rolling release
14:56:57 <ehird_> i'm not crazy enough to try, though
14:57:39 <AnMaster> I heard of a nice way to do that for hosting only installing certain distros on dedis
14:57:50 <AnMaster> you could ask them for specific partitioning scheme
14:58:10 <ehird_> the cool thing is, slicehost actually unofficially endorse the crazy method
14:58:13 <AnMaster> so the person who told me about this asked them to put an 1 GB partition at the end of the disk with debian, and have the rest as home
14:58:22 <AnMaster> then he reversed it installing gentoo on the big one
14:58:26 <ehird_> i'd have expected them to put huge THIS WILL VOID YOUR WARRANTY or sth on it
14:58:31 <AnMaster> and converting the last GB to swap
14:58:45 <AnMaster> and no I wouldn't use gentoo on a server
14:59:27 <ehird_> Xen doesn't work very well with BSD right now, unfortunately
14:59:43 <ehird_> Because it has some neato stuff.
14:59:57 <ehird_> Yeah. Freezing processes to disk, then restarting them from that
15:00:12 <ehird_> A replacement to "chroot-as-secure-hole", jails
15:00:16 <AnMaster> hm useful for debugging and for a laptop user
15:00:28 <AnMaster> it was freebsd that invented them
15:00:35 <ehird_> AnMaster: I believe they stole that from dragonfly, actually
15:00:41 <ehird_> or at least, dragonfly has a variation on them
15:00:50 <ehird_> but i know that dragonfly's is different :-P
15:01:10 <ehird_> Out of the "conventional" BSDs, i'd use netbsd.
15:01:23 <AnMaster> "DragonFly BSD is a free Unix-like operating system created as a fork of FreeBSD 4.8"
15:01:28 -!- Slereah_ has quit (No route to host).
15:01:34 <ehird_> One of the lead devs of FreeBSD forked i t.
15:02:09 <ehird_> glibc = slow ass-compile.
15:02:10 <AnMaster> in fact I think freebsd based it on an idea from solaris
15:02:22 * AnMaster tries to find where he read that
15:02:30 <ehird_> solaris seems to have all the neat stuff
15:02:34 <ehird_> that trace thing, that xfs thing, etc
15:02:44 <AnMaster> "The jail(8) utility and jail(2) system call first appeared in FreeBSD 4.0."
15:02:50 <AnMaster> well that predates 4.8 at least
15:05:21 <AnMaster> 4.0-RELEASE appeared in March 2000 and the last 4-STABLE branch release was 4.11 in January 2005. FreeBSD 4 was a favorite operating system for ISPs and web provider during the first .com bubble, and is widely regarded as one of the most stable and high performance operating systems of the whole Unix lineage.
15:05:34 <AnMaster> wikipedia claims solaris containers showed up in 2005
15:08:27 <ehird_> I saw a Linux distro a while ago written by a guy who uses it as his main OS; it was the Linux kernel but with a BSD userspace instead of GNU
15:08:39 <ehird_> It looked very neat, it was called heretix iirc
15:08:52 <AnMaster> hm I heard of a few similar things before
15:09:04 <AnMaster> some just for seeing "is it possible" and some serious
15:09:27 <ehird_> It had an incredibly spaced apart release, iirc from 2000-2008 fit in about 7 lines, one per release.
15:09:41 <ehird_> So less than 1 release a year, he seemed to just develop it when he ran into issues with it personally.
15:10:10 <ehird_> There also seemed to be a gap 2003-2005, iirc, with no releases.
15:10:44 <AnMaster> http://distrowatch.com/table.php?distribution=heretix <--?
15:12:54 <ehird_> AnMaster: mastodon.biz
15:13:11 <ehird_> heh, the newest release is 7 years old
15:13:15 <ehird_> guess i underestimated
15:13:35 <AnMaster> "roll to one of the super-bloated newer Linux kernels or write my own USB stack plus SATA and UDMA drivers for 2.0.28"?
15:13:59 <ehird_> I like how he doesn't use ELF or glibc, it's a kind of radical because-I-canism that often leads to good things.
15:14:00 <AnMaster> ehird, well clearly he is way worse than me when it comes to disliking new technology
15:14:26 <ehird_> AnMaster: Not really, I've used some of his other software projects...
15:14:30 <ehird_> He just seems very opinionated.
15:14:43 <AnMaster> where does it say he doesn't use ELF?
15:14:50 <ehird_> AnMaster: the part where it says it's completely a.out
15:15:11 <ehird_> just the systems programs, apparently
15:15:12 <AnMaster> oh well he got issues then, the Linux kernel is planning to drop a.out support soon iirc
15:15:15 <ehird_> So I guess you can _build_ programs as ELF
15:15:36 <ehird_> AnMaster: iirc, he maintained a set of patches to update the old unix libc so it still worked on modern systems
15:15:43 <ehird_> I doubt he'd have problems with porting it across all the time :-P
15:15:46 <AnMaster> clearly not an issue for him then yeah
15:16:06 <ehird_> I should try Linux From Scratch sometime.
15:16:40 <ehird_> wtf, apparently I need a kernel update for gcc.
15:16:56 <ehird_> AnMaster: how long does it take to compile the kernel
15:17:37 <AnMaster> ehird, hm well I do a manual kernel, a standard kernel would have to include more modules to support different hardware and such
15:17:49 <AnMaster> but for me, about 15-20 minutes
15:18:06 <ehird_> The best thing is, Gobolinux lets you just skip a dependency.
15:18:11 <ehird_> So I'm gonan do that and hope for the best.
15:18:18 <AnMaster> ehird, what kernel version do you have now?
15:19:01 * AnMaster tries to think what on earth got added after that which would be needed for gcc
15:19:10 <ehird_> Perhaps i should learn some nice assembly, like ppc assembly, and write something like colorforth in 100 lines.
15:19:18 <ehird_> Then write a TCP stack in 50 lines of the crazy language I make
15:19:19 <AnMaster> is it a direct dependency of gcc?
15:19:22 <ehird_> Then a web browser in 300 lines
15:19:31 <AnMaster> or a dependency of for example linux-headers
15:19:36 <AnMaster> which would be needed for glibc
15:20:00 <AnMaster> because if you got newer kernel headers in /usr/include that you have kernel you could potentially run into issues
15:20:11 <AnMaster> if it tries to use a newer feature than exists on the current kernel
15:20:38 <AnMaster> so if glibc is going to be compiled I would definitely update kernel first
15:20:56 <AnMaster> and compare the linux headers version with the kernel version
15:21:18 <AnMaster> but other than that it should work fine to wait with kernel upgrade I guess
15:21:34 <AnMaster> (I Am Not A Toolchain Maintainer)
15:21:52 <oklopol> i don't know about you but my nick is oklopol
15:21:55 <AnMaster> so insert clause of no liability and so on
15:23:09 <AnMaster> well yes it will bootstrap itself
15:23:12 <ehird_> FREE HOT GCC SELF-PLEASURE ACTION
15:26:15 <AnMaster> ehird, also doesn't gobolinux provide binary packages for last gcc and so on?
15:26:37 <ehird_> So, um, not exactly ancient, but Qt sucks ass, apparently.
15:26:40 <AnMaster> err 4.3 is last, well 4.3.2 is
15:27:02 <AnMaster> ehird, or the actual issue is something else
15:27:49 <AnMaster> however one good thing: Once you got all the stuff set up, you can usually run the upgrades in one terminal and do something else while it compiles
15:28:15 <AnMaster> since your is dual core (right?) it would hardly even be noticeable on responsiveness
15:28:22 <AnMaster> heck I don't notice it on a single core system
15:29:11 <ehird_> I'm going to run the mass-upgrade overnight.
15:39:08 <AnMaster> ehird, likely the cause is something else
15:39:29 <AnMaster> if it uses autotools there would be config.log iirc
15:39:34 <AnMaster> My implementation of John Gruber’s Markdown markup language. This implementation is written in C, so I can use it without having to install a modern vanity language."
15:40:01 <ehird_> Yes, he seems to dislike most non-C, non-shell languges.
15:40:04 <AnMaster> "Markdown requires Perl 5.6.0 or later. Welcome to the 21st Century. Markdown also requires the standard Perl library module Digest::MD5, which is probably already installed on your server."
15:40:15 <AnMaster> perl is a "modern vanity language"
15:40:26 <ehird_> his weblog software is written in C
15:40:30 <AnMaster> I would never think if it like that
15:40:32 <ehird_> Why anyone would write a CGI in C I don't know, but :D
15:40:47 <ehird_> AnMaster: Well, he said that his latest distro release was "not that recent" = 7 years
15:40:58 <ehird_> Sure, that's a bit of sarcasm, but I imagine he's been using these computery things before Perl was around
15:41:15 <AnMaster> ehird, about cgi in C: maybe to implement a scripting language. but that doesn't apply to his case
15:41:47 <ehird_> Python can do CGIs, but it isn't a CGI written in C. :-P
15:42:10 <AnMaster> which is why I said "but that doesn't apply to his case"
15:42:16 <ehird_> Considering CGIs are just a couple of environment vars that you get, and you just echo out an http response.
15:42:18 <AnMaster> he isn't writing a scripting interpreter
15:42:35 <AnMaster> ehird_, yes except no one uses that any more because it is so inefficient
15:42:47 <AnMaster> these days there are things like fastcgi, or mod_foo and so on
15:42:50 <ehird_> Well, I use it for the occasional trivial hack
15:43:13 <ehird_> But mostly I use the language-specific solutions because they tend to be better at it than fastcgi; e.g. mod_passenger for Ruby, mod_wsgi for Python
15:43:25 <ehird_> (Not mod_ruby or mod_python, which are both terrible - go figure)
15:44:22 <AnMaster> Cwatch is a log watcher much like the well-known swatch utility. Unlike swatch, cwatch is written in lex, yacc, and C, so it can run on a system that doesn’t have perl (or, as in my case, on a system where I don’t have the dynamic linking capacity that modern versons of perl require.)"
15:44:29 <AnMaster> wtf is wrong with dynamic linking?
15:44:42 <ehird_> Oh, dynamic linking has caused me tons of problems.
15:44:44 <AnMaster> most of those security hacks, like randomization of addresses require dynamic linking even
15:44:47 <ehird_> Plan 9 has no dynamic linking, either, it's all static
15:44:56 <AnMaster> ehird, well ok, but everyone use it these days
15:45:11 <ehird_> AnMaster: everyone also uses ELF and GNU and Perl
15:45:29 <AnMaster> and to be frank: I don't see the issue with ELF of Perl
15:45:42 <ehird_> Plan9's solution is unixy; you static-link the important stuff and the rest is in commands.
15:45:52 <ehird_> AnMaster: ELF, iirc, is bloated compared to a.out
15:45:59 <ehird_> I'm not sure of his exact gripe, though.
15:46:11 <ehird_> Perl presumably because as he said he can't install Perl, due to no dynamic linkin
15:46:12 <AnMaster> well yes it got a certain amount of metadata overhead
15:46:33 <AnMaster> but considering modern binaries... it is an insignificant amount
15:47:05 <AnMaster> like much less than 1 KB compared to a binary of often over 30 KB at least
15:47:34 <ehird_> I don't know, ask him. :p
15:47:46 <ehird_> Perhaps a.out files are faster to run/load.
15:48:04 <ehird_> Perhaps he just sees the features of ELF compared to a.out as not needed.
15:48:06 <AnMaster> also I never thought I would end up arguing on the same side as ehird in that question
15:48:24 <AnMaster> because I didn't think anyone could be much more prefer the old way than me heh
15:48:39 <ehird_> I thought you were a liberal, anyway, isn't that a bit paradoxical? :p
15:49:22 <AnMaster> I'd rather not discuss that. Suffice to say that liberal may be too much to the right for my taste
15:49:41 <ehird_> I don't think liberal actually has any fine pin-pointing, but whatever.
15:49:52 <ehird_> I know you've said you like socialism in the past, so I guess far-left. I'm pretty much the same.
15:50:02 <oklopol> a birdie told me AnMaster is a socialist
15:50:17 <ehird_> oklopol: a birdie known as clog?
15:50:33 <oklopol> no i think it was AnMaster.
15:50:58 <AnMaster> ehird, I guess it depends on what issues it is
15:51:16 * ehird_ supports ABORTING EVERY NEWBORN BABY & KITTEN
15:51:34 <oklopol> socialism doesn't really work for me, i don't really give a shit about people i don't know, and i don't even like the idea of a society where people are nice to strangers.
15:51:38 <ehird_> yes, aborting them POST-BIRTH
15:51:47 <AnMaster> freedom of speech, freedom of the press, freedom of opinion. Liberal I guess when it comes to that.
15:51:55 <oklopol> yeah that's the humane thing to do, because you know whether they deserve it
15:52:17 <oklopol> if they aren't born yet, who knows, they could be really cool guys.
15:54:04 <oklopol> i was doing something, i'm sure
15:54:12 <oklopol> and now i'm staring at this window and have no clue.
15:54:21 <oklopol> so i guess i'll socialize a bit
15:55:09 <oklopol> or like gimme-some-hot-tea-or-i'll-most-surely-die cool?
15:55:30 <oklopol> well heck, what am i doing here then :\
16:20:15 -!- ehird_ has quit ("Lost terminal").
16:40:08 -!- comex has quit (kornbluth.freenode.net irc.freenode.net).
16:40:08 -!- optbot has quit (kornbluth.freenode.net irc.freenode.net).
16:42:38 -!- comex has joined.
16:42:38 -!- optbot has joined.
16:44:03 -!- Hiato has joined.
16:44:40 <Hiato> Guten[nag/dag] world
16:47:11 <Deewiant> in which language is it "nag"? O_o
16:47:52 <Slereah> I think he means "nacht" or something
16:48:03 <Deewiant> just wondering if anybody would actually say or write that
16:48:25 <Hiato> Anyone got either of the following: A good way to validate a large prime (found with probably miller-rabin), an arbitrary precision calculator so I can see the number in all its glory?
16:48:26 <Deewiant> I thought it was obvious what it meant :-P
16:48:51 <Deewiant> Hiato: if you answer my question I might be able to run something through Mathematica
16:49:53 <Deewiant> but evidently "guten" doesn't fit in
16:50:51 <Deewiant> Hiato: anyhoo, I've got Mathematica here... so what are you looking for
16:51:13 -!- Slereah2 has joined.
16:51:40 <Hiato> yip, nicely done :P
16:52:15 <Hiato> Goeienag, and "snags" means funny or in the evening
16:53:26 <Hiato> (sorry, IRC client seems to be having some bad delays)
16:53:54 -!- AquaLoqua has joined.
16:54:13 <AquaLoqua> sorry about that Deewiant, Pidign's IRC is dying
16:54:41 <AquaLoqua> if not dead. What was this quesiton that I should answer?
16:54:54 <Deewiant> you already answered it with "afrikaans" :-)
16:55:34 <AquaLoqua> Oh, I see. Then perhaps you might be able to run 3^2521 -2 through Mathematica for me?
16:55:55 -!- Hiato has quit (Read error: 54 (Connection reset by peer)).
16:55:57 <Deewiant> bah, that's a tiny number, all too easy :-P
16:56:10 <Deewiant> 66478796450889882937915975072813026539507576193402750526873297432119962224642117697219503382301330355047976113745824270903602329332161174576135552602173854297305538263330207230078187030679151177348747502892260949055429524724529301567307108189680225162226220405876105939921908744539875799782857201748186068248330693464490504074543525642597576445339082810885745581185402915395442133718406239632547971222498562638212574353711158105923192118705598166683582404
16:56:16 <Deewiant> 44999945951305275385112133493371508826394490679609109227534237764060131571928075430416911840882836753074732482507345880974398805950602433590637696036075148851407585332923838239726403418149714502307108083034929437313801039719746183968015332703268731287381514837788933821418626391959087957344640857464802443995702465243624042634594729865724564322979967970112323691163488583513564666359520981358727859643686500018828569529703568272721740614434763305472853925
16:56:22 <Deewiant> 51921038453893239515977131635925752931423550512603272087608800954436545497189268218875057633342317563395219199506400913166248245405520867428314613778858986060570975429236265467079808998631080351307401919284358880660711375210653975341720517847689216821848101278129662281995532069183363384503201
16:56:28 <Deewiant> the result came instantaneously
16:56:40 <Slereah2> Remember, guys, we're in the future
16:56:57 <Deewiant> AquaLoqua: and you wanted to know if it's prime?
16:57:00 <AquaLoqua> wow, damn, oh well, my programme is still chugging away, hopefully it will come up with something large soon :P
16:57:31 <AquaLoqua> Hah, that was fast, what is this programme?
16:57:36 <ehird> that will be a bit slow
16:57:39 <ehird> for such a huge number.
16:57:46 <AquaLoqua> Ah, what language might that be? (Lisp?)
16:57:52 <ehird> AquaLoqua: That's scheme.
16:57:58 <Slereah2> Although I guess it works in lisp
16:57:59 <ehird> Deewiant's system is Mathematica
16:58:02 <Deewiant> AquaLoqua: http://en.wikipedia.org/wiki/Mathematica
16:58:04 <ehird> Slereah: it doesn't
16:58:25 <AquaLoqua> Blah, the only functional language I know (and by now i mean have met) is Haskell. Thanks Deewiant, will take a look now
16:58:37 <ehird> Slereah: Everything.
16:58:49 <Deewiant> AquaLoqua: well, this number is small enough that this should be fairly doable in Haskell as well ;-)
16:59:31 <AquaLoqua> Hah, not gonna try - not with GHC at least. All the error codes are encrypted using AES, can't make heads or tails of them
17:00:55 <Slereah2> Well, it's been a minute and it's still checking for primolarity :o
17:02:10 <AquaLoqua> Heh, what have we got you into Slereah? :P
17:02:15 <Deewiant> Slereah2: you do realize that you're going to have to loop through N iterations where N is a number with over a thousand digits
17:02:23 <Deewiant> because it's been established that it /is/ prime
17:02:35 <AquaLoqua> why don't you just loop to the approx Sqrt?
17:02:41 <Deewiant> Slereah2: if you want to take a head start, start from the square root
17:02:56 <Deewiant> if you start from the square root now
17:03:01 <ehird> Your system resources.
17:03:02 <Deewiant> you'll be a LOT further than you are now :-P
17:03:04 <ehird> Call me back in 500 years
17:03:19 <Deewiant> Even if you take the square root
17:03:22 <Slereah2> I once ran a prime checker for Mesrene primes.
17:03:43 <ehird> ah, just let Slereah burn his system
17:03:47 <Slereah2> The program was called "Prime something something"
17:04:00 <ehird> PRIME SOMETHING SOMETHING
17:04:10 <Deewiant> it doesn't use that algorithm, I assure you :-P
17:04:32 <Deewiant> Because, again, they prefer 'few months' to '500 years' :-P
17:04:49 <Slereah2> www.mersenne.org/gimps/t/status.txt
17:05:23 <Slereah2> Of course, I just checked one number, and it wasn't even prime
17:05:50 <AquaLoqua> Well, considering they're getting the 50 000 smackers for finding the 46th mersenne prime, what's a few CPU cycles?
17:06:21 <Deewiant> I think I checked 2 or so numbers with prime95
17:06:22 <Slereah2> Well, they are, not the users :o
17:06:44 -!- Slereah has quit (Connection timed out).
17:06:59 <AquaLoqua> Deewiant: Shame, Slereah2: but you get your name in lights
17:07:23 <AquaLoqua> What are the odds anyway, of finding a mersenne prime if you randomly strike odd powers of two?
17:08:04 <Slereah2> AquaLoqua : To know the odds, you need to know the distribution of primes.
17:08:13 <Slereah2> And I'm not sure you can solve that problem!
17:08:15 <Deewiant> OR we can just look at the history.
17:08:29 <AquaLoqua> well, isn't the minimum distribution of primes in an interval given by a formula?
17:08:36 <Deewiant> How many numbers have been tested, how many were prime.
17:08:54 <Slereah2> Deewiant : Well, for that to work, you have to assume a somehow uniform distribution
17:09:28 <Deewiant> Or depends on what you mean by 'work'. We'll get a value, it's probably not correct but at least it's something :-P
17:09:48 <Deewiant> Hmm, looks like they don't go in order
17:10:10 <AquaLoqua> http://en.wikipedia.org/wiki/Prime_number#Distribution
17:11:32 <Deewiant> I guess GIMPS doesn't publish how many numbers they've tested :-/
17:12:03 <AquaLoqua> On a side not, I can only get my hands on Maxima (http://maxima.sourceforge.net/) in place of Mathematic, unless I be a pirrrate. Is it as good (should anyone know)?
17:12:26 <Deewiant> Mathematica is the best program out there for some things
17:12:49 <Deewiant> Mathematica and Matlab are the two top programs for basically 99% of things that can be done that you might want to do :-P
17:13:07 <AquaLoqua> Blarg, oh well, I gues I just most certainly will not obtain it illegally
17:13:23 <Deewiant> AquaLoqua: Scilab is a free program similar to Matlab.
17:13:44 <Deewiant> oklopol: No, I haven't proved it or anything. :-)
17:14:11 <Deewiant> My school's computers all have Mathematica 6 installed :-)
17:14:13 <oklopol> i don't know those languages, but that sounds like a lie
17:14:16 <AquaLoqua> Well, I'll give it a bash though, thanks
17:14:51 <Deewiant> Ah, and Maple is another popular one.
17:14:53 <AquaLoqua> aha, there's were you luck Deewiant, my schools PC's don't even have calculator, well at least you acn't get to it (well, you're not supposed to be able to anyway)
17:15:16 <Deewiant> Heh, you're stuck to Windows? I feel your pain :-)
17:15:55 <Deewiant> Mathematica is reputedly the best for symbolic computation, Matlab is faster for numerical stuff
17:16:11 <Deewiant> R is what statisticians prefer
17:16:39 <Deewiant> AquaLoqua: oh, and see also Octave, another Matlab clone
17:16:44 <Slereah2> Them Mathematica is hard to use though
17:16:57 <Slereah2> It's nifty and all, but it's hard to find errors
17:17:00 <AquaLoqua> Like Primes? Perfect. You see: 2008/11/13 @ 06:28:31 PM : 3^2521-2 was the last Prime it came up with, it now being 7:19
17:17:26 <Deewiant> evidently Maple is more like Mathematica
17:17:29 <AquaLoqua> Roger that, will check it out (something tell's me I've used Sciab before)
17:18:18 <Deewiant> I saw some good benchmarks many months ago, am trying to dig up
17:20:01 <Deewiant> School computers appear to have matlab as well
17:20:15 <Deewiant> Maybe I'll just test that one number and see what kind of speeds come out :-P
17:20:48 <Deewiant> now if I knew how to use this thing
17:20:53 <AquaLoqua> (Are you connecting to a virtual school network from home?)
17:21:15 <Deewiant> It's where I'm running my IRC client as well :-)
17:21:33 <AquaLoqua> NICE! Damn, would I kill for that (hell, we don't even have a functioning website let alone telnet and co)
17:24:52 <Deewiant> hmm, I think I'd need to run matlab in X to actually see the results :-/
17:25:07 <Deewiant> the profiler doesn't seem to work on the console
17:25:33 <AquaLoqua> .. Wouldn't know :P What distro do you use?
17:26:32 <Deewiant> ??? Error using ==> isprime at 22
17:26:32 <Deewiant> The maximum value of X allowed is 2^32.
17:26:39 <Deewiant> so yeah, that's why matlab is faster :-)
17:27:00 <Deewiant> now how do I pull arbitrary precision out of this thing
17:27:04 <AquaLoqua> Ah, so you're on Windoze at home? I suppose if you're a gamer then it's alright. HAHA, yay, my puny number broke something :P Well, I know now, thanks :P
17:27:50 <Deewiant> Yeah, I use windows at home, although my laptop is mainly gentoo (dual boot there)
17:28:42 <Deewiant> well hmm, looks like matlab can't do this at all
17:29:06 <AquaLoqua> Cannot be, but I guess it's much like GHC in it's cryptic-ness
17:29:36 <AquaLoqua> We know it's prime, and that's all that matters, so 7 interations of Miller-Rabin is good enough for at least about 1000 odd digits
17:30:11 <AquaLoqua> Well, after reading YAHT (about half a dozen times) I still don't know what it's on about, and HUGS just plain sucks :P
17:31:14 <Deewiant> Two newish ones that I hear are good are http://www.learnyouahaskell.com/ and http://book.realworldhaskell.org/read/
17:31:15 <Asztal> I always thought I just couldn't find the option for loading multiple modules at once in Hugs... but apparently there isn't one
17:31:52 <AquaLoqua> Heh, another hole in the wall I suppose
17:32:03 <AquaLoqua> Deewiant: will check them out, thanks
17:32:49 <AquaLoqua> Kind of reminds me of Why's Poignant Guide
17:34:09 <Deewiant> yeah, the first one is in that style
17:34:18 <Deewiant> the second one is a complete book
17:41:05 <AquaLoqua> "On Windows it's just a matter of downloading the installer and clicking "Next" a couple of times and then rebooting your computer. On Debian based Linux distributions you can just do apt-get install ghc6 libghc6-mtl-dev and you're laughing." How sad but true
17:43:40 <Deewiant> Run before the lambdas get you
17:43:52 <Deewiant> I'm already caught, I can't escape
17:43:53 <Slereah2> It's the monads you have to watch out for
17:43:57 <Deewiant> All other languages seem to suck now
17:44:30 <AquaLoqua> Slereah2: Am I sensing some kind of tension ehre? Why don't you like Haskell? Either way, do you think there's a better functional language than Haskell for a newb at functional?
17:45:05 <Deewiant> there was somebody on #haskell just yesterday for whom it was his first programming language ever
17:45:30 <AquaLoqua> Heh, wow, talk about brave (or stupid)
17:45:35 <Deewiant> he was reading the intro on LYAH and said "aren't all languages like this? if not, what are those other muppets doing" or something to that effect :-D
17:45:53 <Deewiant> I think Haskell is perfectly fine as a first language
17:46:10 <Deewiant> I suspect it's much harder as a first functional language after imperative languages
17:46:24 <Deewiant> But I don't know, I only went the latter way and there was much head-bashing involved :-P
17:46:32 <AquaLoqua> Well, I can only talk from experience when I say, as approx a sixth langauge, I still don't get it. Then again, I'm not the sharpest tool
17:47:21 <AquaLoqua> The imperative trap holds one fast. Though, I wonder if it isn't a more natural (if somewhat less tangible system)
17:49:00 <Deewiant> I don't know. It may or may not be.
17:49:17 <Deewiant> What I will say is that prolog is definitely unnatural. ;-)
17:49:42 <AquaLoqua> Heh, apples and .. well, llamas my friend - beyond comaprison :P
17:53:06 -!- Mony has joined.
18:05:15 <Slereah2> I mostly went to scheme because it's the first functional language I saw
18:05:27 <Slereah2> (It's what Lazy K is writen in, and I saw that)
18:09:03 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | quite, but is it t or f.
18:09:05 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | $ ls tests/.
18:09:07 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | yes.
18:09:10 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | as in your example, double each integer is equivalent to like.
18:09:33 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | what is agora?.
18:09:36 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | !hangman s.
18:09:38 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | fizzie: I think it may be quite possible to do hot code change of parts of efunge without stopping :D.
18:09:40 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | no.
18:09:47 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | did it have static scoping?.
18:09:55 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | <image x="50" y="50"><pixel x="1" y="1"><red value="255"/><blue value="0"/><green value="0"/></pixel>....
18:10:05 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | ....
18:20:34 -!- Doitle2 has quit (Read error: 104 (Connection reset by peer)).
18:37:54 -!- kar8nga has joined.
18:41:24 -!- jix has joined.
18:41:29 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | that's a very old version.
18:48:35 -!- Sgeo has joined.
18:51:45 -!- olsner has joined.
18:54:47 -!- AquaLoqua has quit ("Dana").
19:09:28 -!- ab5tract has joined.
19:11:19 -!- Doitle2 has joined.
19:27:25 -!- jix has quit ("Computer has gone to sleep").
19:29:44 -!- jix has joined.
19:37:56 -!- ab5tract has quit.
19:54:24 -!- jix has quit ("Computer has gone to sleep").
20:08:40 -!- Judofyr has joined.
20:12:37 -!- Mony has quit ("Join the Damnation now !").
20:40:30 -!- jix has joined.
20:51:11 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=41
20:53:41 -!- oerjan has joined.
20:57:02 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=94 muahaha
20:57:12 <oerjan> <ehird_> you are not oerjan
20:57:20 <oerjan> i am glad we finally agree on that
21:04:26 <Sgeo> Bonobo Conspiracy
21:06:21 <Sgeo> Geeky stuff, apparently. The bald guy is the TA of some course
21:06:32 <Slereah2> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=3
21:06:40 <Slereah2> Is this comic just jokes about computation theory?
21:07:24 <Sgeo> No, there are other types of jokes in there
21:11:10 <Slereah2> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=10
21:11:16 <Slereah2> Author is too lazy to change any sprite
21:11:51 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=66
21:13:21 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=69
21:37:55 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=129
21:39:39 -!- kar8nga has quit (Read error: 110 (Connection timed out)).
21:47:50 <Sgeo> http://ansuz.sooke.bc.ca/bonobo-conspiracy/?i=148
22:16:28 -!- oklokok has joined.
22:25:55 -!- Corun has joined.
22:33:21 -!- oklopol has quit (Read error: 110 (Connection timed out)).
22:35:50 -!- jix has quit ("...").
22:52:23 -!- oklopol has joined.
22:52:34 -!- oklokok has quit (Read error: 54 (Connection reset by peer)).
23:01:06 -!- olsner has quit ("Leaving").
23:18:25 -!- sebbu2 has joined.
23:27:57 -!- sebbu has quit (No route to host).
23:27:57 -!- sebbu2 has changed nick to sebbu.
23:28:12 -!- AnMaster_ has joined.
23:30:51 -!- AnMaster has quit ("here we go, changing to znc").
23:31:37 -!- AnMaster_ has changed nick to AnMaster.
23:31:39 * oerjan gets a znc-ing feeling