00:08:05 -!- puzzlet has quit (Remote closed the connection).
00:08:14 -!- puzzlet has joined.
00:13:03 -!- iamcal has joined.
00:33:57 -!- cal153 has quit (Read error: 113 (No route to host)).
00:37:57 -!- Rugxulo has left (?).
00:49:28 -!- BeholdMyGlory has quit (Remote closed the connection).
01:10:40 -!- coppro has quit (Read error: 110 (Connection timed out)).
01:12:40 <facsimile> any sufficiently long string on an alphabet of 22 (or 26 or any size) has Unavoidable regularities
01:13:44 -!- Gracenotes has quit (Remote closed the connection).
01:13:59 <oerjan> facsimile: http://en.wikipedia.org/wiki/Ramsey_theory
01:15:56 <oerjan> no such term in the article
01:16:19 <oerjan> not that i read it first, your comment just reminded me of this
01:17:39 <facsimile> Suppose, for example, that we know that n pigeons have been housed in m pigeonholes. How big must n be before we can be sure that at least one pigeonhole houses at least two pigeons? The answer is the pigeonhole principle: if n > m, then at least one pigeonhole will have at least two pigeons in it. Ramsey's theory generalizes this principle as explained below.
01:18:21 <oerjan> n and m are integer variables
01:19:56 <oerjan> well, natural number variables
01:20:49 -!- Gracenotes has joined.
01:21:29 <oerjan> Van der Waerden's theorem mentioned there seems like it applies to such strings on alphabets (set letters = colors)
01:30:14 -!- coppro has joined.
01:31:44 -!- Asztal has joined.
01:32:05 -!- Gracenotes_ has joined.
01:34:53 -!- Gracenotes has quit (Nick collision from services.).
01:34:55 -!- Gracenotes_ has changed nick to Gracenotes.
01:54:14 -!- facsimile has quit (Client Quit).
01:54:53 -!- facsimile has joined.
02:30:17 -!- augur has quit (Read error: 110 (Connection timed out)).
02:32:46 -!- Asztal has quit (Read error: 110 (Connection timed out)).
02:43:13 -!- facsimile has quit (Client Quit).
02:43:52 -!- facsimile has joined.
02:53:36 -!- Gracenotes has quit (Connection timed out).
02:53:40 -!- facsimile has quit (Read error: 54 (Connection reset by peer)).
02:53:52 -!- facsimile has joined.
02:55:09 -!- Gracenotes has joined.
03:01:58 -!- augur has joined.
03:32:06 -!- oerjan has quit ("Good night").
03:46:42 -!- facsimile has quit (Client Quit).
04:47:09 -!- Pthing has joined.
04:51:17 -!- Gracenotes has quit (Connection timed out).
04:52:52 -!- Gracenotes has joined.
04:54:34 -!- Gracenotes has quit (Remote closed the connection).
04:56:38 -!- Gracenotes has joined.
05:02:11 -!- Gracenotes has quit (Remote closed the connection).
05:03:51 -!- Gracenotes has joined.
05:12:53 -!- Gracenotes has quit ("Leaving").
05:29:37 -!- coppro has quit (Remote closed the connection).
05:30:13 -!- coppro has joined.
05:41:55 -!- Gracenotes has joined.
06:41:48 -!- coppro has quit ("I am leaving. You are about to explode.").
07:16:05 -!- puzzlet_ has joined.
07:18:59 -!- puzzlet has quit (Read error: 145 (Connection timed out)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:27:40 -!- immibis has quit ("ChatZilla 0.9.85 [Firefox 3.5.5/20091102152451]").
08:29:21 -!- Slereah has joined.
08:30:58 -!- Slereah_ has quit (Read error: 60 (Operation timed out)).
09:07:55 -!- MigoMipo has joined.
09:08:13 <MigoMipo> Oops, missed the / buttom again!
09:13:47 <AnMaster> MigoMipo, I think "key" is more usual for the keyboard buttons :P
09:17:19 <AnMaster> MigoMipo, are you by any chance from Sweden. considering /whois output
09:17:43 <AnMaster> if so I think I can add another Swede to the list in this channel. Hurra!
09:18:06 <MigoMipo> Yay! Hur många svenskar är det i kanalen?
09:18:56 <AnMaster> hm... Jag, du, Firefly, olsner, och kanske någon jag missade som inte är här så ofta.
09:19:22 <AnMaster> I still think there are more Finns than Swedes though
09:24:00 <fizzie> I'm not so sure about that; there's just me, Deewiant, Ilari and ineiros. That I know, anyway.
09:24:38 <fizzie> He's just so unnoticeable!
09:26:28 <fizzie> And maybe you can count fungot as Finnish too, though he definitely hasn't got nationality officially.
09:26:29 <fungot> fizzie: they say that the ladies were accustomed to wearing luxurious clothings and so he flew along the row stinging all the nicest things in this dungeon.
09:26:55 -!- Asztal has joined.
09:27:02 <fizzie> "They" say curious things.
09:29:07 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher ic irc jargon lovecraft nethack* pa speeches ss wp youtube
09:29:27 <AnMaster> fizzie, yeah what the hell was that a mix of
09:30:56 <fizzie> Just a moment, collecting evidence.
09:35:34 <fizzie> "They say that" from one of the rumours, "that the ladies were accustomed to" from T.H. White via the 'cornuthaum' description, "accustomed to wearing luxurious clothings and so he" from Master Kaen description (about I-Hsiu), "he flew along the row stinging all the" from 'xan' description, "all the nicest things in this" from C.S. Lewis via 'orange' and finally "this dungeon" from one of two rumours that contain it.
09:36:59 <AnMaster> fizzie, yeah the first and the last parts were not very strange
09:37:19 <fizzie> Quite a collection. I would have though at least the "ladies were accustomed to wearing luxurious clothings" part would've come from one source, since it was so coherent, but apparently not.
09:56:18 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
10:00:48 -!- clog has joined.
10:00:48 -!- clog_ has joined.
10:14:03 -!- clog has quit (Read error: 110 (Connection timed out)).
10:14:03 -!- clog_ has changed nick to clog.
10:45:31 <AnMaster> sigh, why isn't oerjan here when you "need" him,.
10:50:42 <fizzie> s#m,/m#,\./.# -- it's more cryptic-looking that way, and works as well.
10:51:35 <AnMaster> fizzie, would that be applied to the regex correcting the regex?
10:52:18 <AnMaster> fizzie, it would be even more cryptic looking if you used / in the correction of the correction :P
10:52:55 <fizzie> I guess so, but since you already started with the non-standard delimiters, I just used my own standard non-standard one.
10:53:54 <AnMaster> fizzie, just because I was lazy, I considered using / to begin with
10:54:26 <AnMaster> what if you used . for the delimiter
10:55:25 <AnMaster> and what about using \ for the delimiter
10:56:46 <AnMaster> s/./a/ seems to map to s.\..a.
10:57:02 <AnMaster> but I'm unable to get something that is s/\./a/ using . for deimilter
10:57:41 <fizzie> At least with \ as the delimiter, my sed seems to do it but then you can't use any escapes in there.
10:58:10 <AnMaster> fizzie, right. haven't tried \ as delimiter yet
10:58:20 <AnMaster> and if you have gnu sed I won't need to either
10:59:52 <fizzie> GNU sed version 4.1.5. It's a bit strange, though, that s.\..a. works like s/./a/. I would have guessed it to do s/\./a/, but it doesn't.
10:59:58 <fizzie> Don't know how to get a literally-matching-a-. when . is the delimiter.
11:00:31 <AnMaster> fizzie, probably a "don't do this then"
11:01:13 <fizzie> Er, well; you can do s.[\.].a. to get s/\.a/a/-like behaviour.
11:01:15 <AnMaster> fizzie, another funny delimiter would be s
11:02:14 <fizzie> Non-punctuation delimiters look pretty strange, especially with options. "ss\.sasg" doesn't exactly seem so regexpy at first glance.
11:02:35 <AnMaster> fizzie, does gnu sed have the ability to do negative lookbehind?
11:02:45 <AnMaster> $ echo ss\ss/sg | sed 'ss\ss/sg'
11:02:56 <AnMaster> (oh and I think irc client ate one / there)
11:03:22 <fizzie> I don't think it does much more than POSIX basic regular expressions.
11:03:39 <AnMaster> no way to match s not preceded by \ then
11:03:51 <AnMaster> anyway even with negative lookbehind you couldn't handle \\
11:03:55 <fizzie> Or POSIX extended with the "-r" flag, but those aren't *that* much extended.
11:05:35 <fizzie> You can't match "s not preceded by \", true, but you can match "start of line or a non-\ character in subgroup 1, followed by s", and then use \1 backref in the substitution; that is usually enough.
11:06:03 <AnMaster> hm should try to find some existing English word that forms a valid sed expression
11:07:28 <fizzie> Here's one posix BRE that changes any unescaped s's into X's, though it fails for "\\s".
11:07:33 <fizzie> htkallas@pc112:~$ echo 's foo \s bar s and s' | sed -e 's:\(^\|[^\\]\)s:\1X:g'
11:07:33 <AnMaster> grep -E '^s([a-z]).*\1.*\1g?$' /usr/share/dict/words
11:08:16 <AnMaster> also that is one weird word in that list: "symphytically"
11:08:21 <AnMaster> I'm pretty sure it isn't spelled that way
11:09:10 <fizzie> My /usr/share/dict/words has just 39 matches, with no "symphytically" in it.
11:10:08 <AnMaster> /usr/share/dict/README claims mine is
11:10:11 <fizzie> Anyway, "sentence" I could see used somewhere; s/nt/nc/ sounds like something someone has done. Would be a nice obfuscation trick to replace that with "$foo =~ sentence;"
11:11:27 <AnMaster> fizzie, the list I get is http://sprunge.us/eUWI
11:11:38 <fizzie> Ubuntu "wamerican" package: "The Debian English word lists (wamerican*, wbritish*, wcanadian*) -- are all built from the upstream SCOWL word lists."
11:11:45 <fizzie> It's a bit small list.
11:13:21 <AnMaster> fizzie, severe also sounds probable. s/v/r/
11:14:18 <AnMaster> with corrected regex I get 253 matches (.+ instead of .*)
11:14:33 <AnMaster> hm also I think that really should be [^\1+]
11:20:41 <fizzie> Meh, I can't find type counts for our text corpuses; we have this LDC Gigaword corpus, built out of about 1756504000 whitespace-separated tokens of English, I'm sure you could make quite a large word list out of that one.
11:20:56 <AnMaster> grep -E '^s([a-z]).+\1.+\1g?$' /usr/share/dict/words | grep -Ev '^s([a-z]).*\1.*\1.*\1'
11:25:14 <fizzie> perl -ne 'print if /^s([a-z])(?:(?!\1).)+\1(?:(?!\1).)+\1g?$/;' < /usr/share/dict/words
11:25:19 <fizzie> If you want to do it with a single regex.
11:25:24 <fizzie> It's not exactly pretty that way.
11:26:51 <AnMaster> fizzie, hah. what is ?: now again... Non-capturing?
11:27:24 <fizzie> ?: is the non-capturing group, right, you could use plain () there just fine. And ?! is the negative zero-width lookahead.
11:28:09 <fizzie> So (?:(?!\1).)+ will match one of more "any character"s, as long as the contents of the first group don't start at that point.
12:01:13 -!- |MigoMipo| has joined.
12:02:00 -!- MigoMipo has quit (Nick collision from services.).
12:02:18 -!- |MigoMipo| has changed nick to migomipo.
12:05:43 -!- FireFly has joined.
12:26:31 -!- Azstal has joined.
12:30:25 -!- Asztal has quit (Read error: 60 (Operation timed out)).
12:30:26 -!- Azstal has changed nick to Asztal.
12:52:19 -!- migomipo has quit ("co'o rodo").
13:01:53 -!- FireFly has quit ("Later").
14:03:02 -!- augur has quit (Read error: 131 (Connection reset by peer)).
14:18:56 -!- BeholdMyGlory has joined.
14:24:38 -!- ineiros_ has joined.
14:32:32 -!- ineiros has quit (Read error: 113 (No route to host)).
14:36:16 -!- FireFly has joined.
14:41:07 -!- oerjan has joined.
14:44:34 <oerjan> <AnMaster> sigh, why isn't oerjan here when you "need" him,.
14:58:44 <oerjan> DOIOF doesn't often get work, but when he does he really gets overworked...
15:00:46 <AnMaster> oerjan, has anyone actually died *permanently* from him yet?
15:01:37 <oerjan> that black guy in james stud, i think
15:02:15 <oerjan> although what happened to him when the universe was restarted, hm... oh wait that theme was the only one not affected
15:03:42 <AnMaster> oerjan, has there been any crossovers at all with the james stud theme and any other theme?
15:04:43 <oerjan> Me and Death are the only ones listed
15:05:10 <oerjan> oh wait also imperial rome
15:05:15 -!- augur has joined.
15:05:50 -!- MigoMipo has joined.
15:06:00 <oerjan> well that was really a dubious one
15:06:43 <oerjan> ah, except it shows DOIOF _was_ cheated out of even that one, after the fact :D
15:06:53 <AnMaster> oerjan, imperial rome/espionage crossover or what?
15:07:12 <AnMaster> I'm unable to find it in the list
15:07:12 <oerjan> http://www.irregularwebcomic.net/1178.html
15:07:38 <oerjan> AnMaster: the list is triangular to avoid duplicates, you have to look both horizontal and vertical
15:07:48 <AnMaster> oerjan, yeah I did, but I was one column off
15:09:15 <AnMaster> oerjan, anyway there are lots of possible ways to continue even after he got all possible crossovers. Like differentiating "crossover with two" and "crossover with three"
15:09:45 * AnMaster wonders how many combinations you would get that way...
15:10:01 <AnMaster> oerjan, sounds like something for you to calculate
15:10:47 <oerjan> 2^number of themes - 1
15:11:44 <AnMaster> so space/fantasy space/death fantasy/death space/fantasy/death are all separate.
15:11:58 <oerjan> 2^number of themes - number of themes - 1
15:12:02 <AnMaster> of course fantasy/space and space/fantasy would be the same
15:12:15 <oerjan> forgot to subtract singular themes
15:12:40 -!- oklofok has quit ("( www.nnscript.com :: NoNameScript 4.2 :: www.regroup-esports.com )").
15:12:59 <AnMaster> oerjan, you could have crossover with itself actually. For those that had time travel.
15:13:15 <AnMaster> space would have had that already
15:13:37 <AnMaster> there was that telephone booth thingy
15:13:54 <AnMaster> not sure what happened to it after end of universe
15:14:18 <oerjan> but if you do that there is no limit, since you can crossover any number of copies of a theme
15:15:16 <AnMaster> oerjan, maybe each theme can appear 0-2 times? How many possibilities is there then? As in space/space/fantasy/death/death and such?
15:15:35 <oerjan> 3^number of themes - number of themes - 1
15:15:38 <AnMaster> (that would be rather contrived!)
15:16:48 <AnMaster> for simple crossover (no self-crossover) there would (if I counted correctly) be 2^16-16-1 possible crossovers
15:17:13 <AnMaster> how many years would it take if he started today and managed one such every day?
15:17:59 <oerjan> just divide by 365.2425
15:18:38 <AnMaster> something like 179.384... years
15:22:22 -!- facsimile has joined.
15:50:37 -!- oerjan has quit ("leaving").
16:08:05 <AnMaster> today I was able to access a machine with SSE 4.2
16:08:19 <AnMaster> that has the new CRC32c hardware thingy
16:08:51 <AnMaster> so I tried to use that for cfunge (removing use of static area temporarily to be able to see how much was gained in the hash library part)
16:09:30 <AnMaster> for pure hash on that system mycology took and average of 0.055 seconds when using old CRC code
16:09:40 <AnMaster> with the hardware bit it was down to an average of 0.035
16:09:53 <AnMaster> with static area it runs at around 0.020 on that
16:10:02 <Deewiant> Aye, that kind of speedup can be expected
16:10:05 <AnMaster> it's a dual CPU (each quad core) system
16:10:21 <Deewiant> Which doesn't really matter since cfunge is single-threaded
16:10:23 <AnMaster> Deewiant, /proc/cpuinfo is 400 lines long on that
16:10:33 <AnMaster> Deewiant, yes but make -j16 takes like 2 seconds!
16:10:44 <AnMaster> model name : Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
16:10:54 <Deewiant> Oh, does cpuinfo list the hyperthreads as well?
16:11:22 <Deewiant> That explains that; I was wondering why it was that long
16:11:59 <AnMaster> Deewiant, with static area it is only marginally faster than my core 2 duo laptop
16:12:21 <AnMaster> (which is: model name : Intel(R) Core(TM)2 Duo CPU P8400 @ 2.26GHz)
16:12:38 <Deewiant> Yep, since funge code tends to stress memory more than CPU
16:12:52 <AnMaster> Deewiant, the xeon has larger cache though
16:13:00 <AnMaster> Deewiant, and is less loaded currently
16:13:09 <Deewiant> Hence it's marginally faster, not slower.
16:13:48 * AnMaster waits for a CPU where the entire mycology can fit inside L2 (or L3)
16:14:00 <Deewiant> Funge access patterns aren't the linear kind that CPU caches like, so it doesn't help that much unless it fits everything
16:14:40 <AnMaster> Deewiant, well, in the static area you will get linear when you go < or > but not other directions
16:14:47 <AnMaster> (as it is stored currently I mean)
16:15:00 <AnMaster> of course that won't help for reading data from elsewhere
16:15:09 <Deewiant> Yes, but that typically lasts for at most a couple dozen accesses, not hundreds.
16:15:17 <Deewiant> And then you're jumping randomly again.
16:15:30 <Deewiant> And yes, good point, p and g and their cousins will mess things up as well.
16:15:50 <AnMaster> and hash libraries are very unfriendly anyway
16:16:30 <Deewiant> Yes, I was thinking only of array storage.
16:16:56 <AnMaster> Deewiant, btw you can't use valgrind on the hardware CRC32 one
16:17:10 <AnMaster> vex amd64->IR: unhandled instruction bytes: 0xF2 0x48 0xF 0x38 0xF1 0x7
16:17:33 <AnMaster> (currently, I assume this will be fixed soon)
16:17:46 <AnMaster> (me should try out trunk on that system and then possibly report a bug
16:17:57 <AnMaster> that was supposed to be / not (
16:20:10 <AnMaster> Deewiant, anyway, as far as I can tell it works but I don't trust it to be correct. Like: I have been unable to find if the initial value should be something specific or such
16:20:28 <AnMaster> (like it should for normal crc
16:23:52 <AnMaster> heh I think I can reduce it even more by making it inline it instead...
16:25:44 -!- Pthing has changed nick to |prettyboy.
16:26:01 -!- |prettyboy has changed nick to Pthing.
16:35:01 -!- FireFly has quit ("Later").
17:12:50 -!- kar8nga has joined.
17:34:40 -!- puzzlet_ has quit (Remote closed the connection).
17:34:48 -!- puzzlet has joined.
17:44:30 -!- Pthing has quit (Remote closed the connection).
18:17:30 -!- ais523 has joined.
20:25:51 -!- clog has joined.
20:25:51 -!- clog has joined.
21:01:29 -!- Sgeo has joined.
21:36:27 <AnMaster> I managed to find the reference/manual/extras CD for CodeWarrior on classic mac
21:36:36 <AnMaster> this seems to include a Perl for mac for example
21:36:54 <AnMaster> and a CodeWarrior<->MPW bridge of some sort
21:37:24 <AnMaster> oh and the experimental parts are stored in a funny folder name
21:39:13 -!- bsmntbombdood has quit (Read error: 113 (No route to host)).
21:45:38 <AnMaster> ais523, hey there is even an Eiffel compiler here
21:47:17 <ais523> AnMaster: it means that I've heard what you've said, and I think it's interesting, but I don't think any reply but ` is particularly necessary
21:47:46 <AnMaster> ais523, so the definition of it is recursive?
21:48:22 <ais523> AnMaster: not really, its definition mentions itself, not uses itself
21:48:44 <AnMaster> oh great an actual hex editor, that is useful, at one point I was seriously considering writing my own for MacOS
21:48:50 <ais523> sort of like a function void(*f)()(void) { return (void(*)())f; } in C
21:48:54 <ais523> which returns a pointer to itself
21:49:37 <AnMaster> ais523, there is a reason I use typedefs when dealing with function pointers in C
21:49:48 <AnMaster> and that is being able to understand what the heck is going on :P
21:49:52 <ais523> meh, they're not too hard once you get used to them
21:50:12 <ais523> although I might have missed a pair of parens in there somewhere
21:50:35 <AnMaster> it looks not like a function returning a function pointer
21:50:54 <Deewiant> Would you prefer void(*f)(void)(void)
21:51:10 <ais523> I'm using void(*)() as a generic function pointer type there
21:51:15 <Deewiant> Or might as well bracket it up a bit more and (void)(*f)(void)(void)
21:51:16 <ais523> as you aren't allowed to cast function pointers to non-function pointers
21:51:27 <AnMaster> Deewiant, a bit more readable. But I would actually use a typedef
21:51:40 <AnMaster> so I can just worry about that in one place
21:52:10 <AnMaster> ais523, also is that cast of f strictly necessary in the return? As far as I know it shouldn't be
21:52:22 <ais523> AnMaster: it is, can you work out why?
21:52:47 <ais523> AnMaster: it's to do with data types
21:52:59 <ais523> a function can't return a pointer to itself without casting it to an incompatible type
21:53:12 <ais523> as otherwise you'd need a type function returning (function returning (function returning (function returning...
21:53:22 <AnMaster> I always wondered why the name of the thing you were defining with function pointers were written in such a weird place
21:53:29 <ais523> and you need a cast to convert a pointer to a different incompatible one
21:53:34 <ais523> AnMaster: it's because the declaration reflects the use
21:53:50 <ais523> if you wanted to call the function returned by f, you'd write (*f)()();
21:54:21 <ais523> basically, the way you write the declaration is exactly the same as the code to get a basic data type out of it as an expression, possibly with extra voids added
21:56:33 <AnMaster> ais523, btw I'm using netscape in sheepshaver, because MSIE crashes it. (And there is only one other alternative and that is icab, which is shareware)
21:56:36 <Deewiant> ais523: Actually, the syntax you used to define the function /was/ wrong
21:57:07 <ais523> ah, thought I was missing a pair of parens somewhere
21:57:14 <ais523> but I'd accidentally put them in the wrong place
21:57:43 <fizzie> cdecl> explain void(*f)()(void);
21:57:44 <fizzie> declare f as pointer to function returning function (void) returning void
21:57:44 <fizzie> cdecl> explain void(*f())(void);
21:57:44 <fizzie> declare f as function returning pointer to function (void) returning void
21:58:34 <fizzie> I like cdecl's help texts: explain <gibberish>
21:59:14 <Deewiant> Hmmh, my cdecl binary at school no longer works due to missing libreadline version 5, and nobody's bothered to package it for Arch so I haven't bothered to install it at home
21:59:24 <fizzie> Later they define the nonterminal: "gibberish: a C declaration, like 'int *x', or cast, like '(int *)x'". The examples might not be very gibberishy, but I still like the word.
22:02:16 <fizzie> c++decl> declare x as array 4 of const pointer to member of class X function (pointer to function (int) returning int) returning void
22:02:16 <fizzie> void (X::* const x[4])(int (*)(int ))
22:02:23 <fizzie> Member-pointers are so wonky too.
22:03:42 <Deewiant> Not really, barely wonkier than the standard function pointers
22:04:11 <fizzie> I think it's pretty wonky how their sizes vary so very very much.
22:04:41 <Deewiant> Blame wonky compilers for that
22:05:02 <fizzie> I remember seeing a page about the various bits in there, but can't find it right now.
22:05:26 <Deewiant> http://www.codeproject.com/KB/cpp/FastDelegate.aspx is a good write-up.
22:05:33 <fizzie> Yes, I just got that opened.
22:05:47 <fizzie> It was probably exactly what I was thinking of.
22:06:06 <Deewiant> The writer has been patching the D compiler quite heavily lately.
22:08:10 -!- puzzlet has quit (Read error: 60 (Operation timed out)).
22:09:50 -!- puzzlet has joined.
22:14:39 -!- oerjan has joined.
22:26:03 <ais523> wow, I think today's Dudley's Dungeon comic was written by someone with the writing style of zzo38
22:26:20 <ais523> "Your wish cannot be satisfacted because the computer is not yet invented, you playing NetHack would generate a paradox, it would be impossible for you to use a computer as there is no electricity here and finally that would be too expensive even for us."
22:26:34 <ais523> http://alt.org/nethack/dudley/?f=2009.11.20
22:44:16 -!- augur has quit (Read error: 110 (Connection timed out)).
22:48:27 -!- kar8nga has joined.
22:49:47 -!- MigoMipo has quit.
23:02:29 -!- augur has joined.
23:06:21 -!- ais523 has quit (Remote closed the connection).
23:38:19 -!- kar8nga has quit (Remote closed the connection).