←2010-07-19 2010-07-20 2010-07-21→ ↑2010 ↑all
00:02:53 <ais523> cpressey: is that a me reference?
00:02:59 <ais523> or are the numbers just coincidence?
00:04:45 <cpressey> ais523: Oh, an occurrence of 23 is *never* a coincidence.
00:04:55 <ais523> ah, OK
00:05:25 <cpressey> Except where it is. Verily.
00:06:03 <cpressey> ais523: Since the subject came up -- why did you pick those numbers? Born on May 23rd?
00:06:25 <ais523> cpressey: no, it was a completely random number, picked by an automatic username-allocating system
00:06:27 <cpressey> Or in March of 1952?
00:06:34 <ais523> although I am aware of the significance of the number
00:06:41 <cpressey> I see.
00:06:55 <ais523> draw whatever conclusions you like from that
00:07:21 <cpressey> I shall!
00:07:44 * cpressey starts drawing a stick figure
00:09:26 <cpressey> So I learned this about the DOM today, at least as FF sees it: If you have two nested elements, and you move the mouse into the inside one, you get two mousenter events, one for each of the elements. If however you have two elements which are *not* nested but still geometrically overlap by absolute positioning, you will only get one mouseenter.
00:09:45 <coppro> makes sense
00:09:55 <coppro> the top element hides the lower one
00:10:21 <coppro> whereas if one is a subelement of the other, entering the inner one by definition, requires you to be in the outer one
00:10:22 <cpressey> For some definition of "hide" which does not map to the user's experience and which is very inconvenient for me... perhaps.
00:10:38 <cpressey> It should go purely by the geometry imo.
00:11:19 <coppro> that would make no sense
00:12:00 <cpressey> "by definition"? You can have an inner element which displays completely outside of the element it's nested in.
00:12:11 <cpressey> I don't *like* it, but...
00:12:27 <coppro> it's not about display, it's about logical nesting
00:12:43 <coppro> if something is inside something else, to be inside it is logically to be in the outer element too
00:13:01 <coppro> that's like putting a box in a can and saying you can be in the box but not in the can
00:13:29 <Sgeo> 346126 is also a number from a user-allocation system
00:13:41 <Sgeo> Doesn't that sound much better than "It's my Active Worlds citizen number!"?
00:13:56 <coppro> yes
00:14:01 <coppro> yes it does
00:14:44 <cpressey> Were that HTML were not about display. Alas.
00:15:21 <cpressey> My world is full of boxes in cans that appear outside of them.
00:15:38 <coppro> HTML is not about display
00:17:43 <cpressey> coppro: If your HTML is not about display, then I applaud you. Recognize, however, how much of the world's HTML *is* about display, almost entirely so.
00:17:54 <coppro> CSS is
00:17:56 <coppro> HTML is not
00:18:24 <coppro> if your HTML includes display information, you're Doing It Wrong
00:18:51 <cpressey> coppro: You realize how much of the world is doing it wrong, right?
00:19:03 <Sgeo> <b>coppro</b>
00:19:16 <coppro> cpressey: unfortunately
00:19:21 <coppro> <strong>Sgeo</strong>
00:19:53 -!- SimonRC has quit (Ping timeout: 265 seconds).
00:20:04 <cpressey> coppro: I still think, since mouseenter uses geometry, and CSS uses geometry, the problem is one of display, and HTML nesting should not enter into it.
00:20:33 <pikhq> How's about we replace HTML with something better? Something without any formatting?
00:20:43 <cpressey> pikhq: GIFs.
00:20:50 <cpressey> Much better.
00:20:59 <coppro> OOh, I know!
00:21:01 <coppro> SVG
00:21:16 <coppro> it does everything HTML does
00:21:22 <coppro> except "better"
00:21:30 <pikhq> Except with more of a focus on formatting and not on content.
00:21:59 * pikhq wants a damned hypertext markup language, not a craptastic graphic design language
00:22:03 <coppro> for a focus on content, you need only use the XHTML namespace
00:22:15 <coppro> admittedly, this works both ways
00:22:19 <pikhq> Except that's a craptastic graphic design language!
00:22:22 <coppro> but importing SVG in an XHTML document is confusing
00:22:38 <coppro> let's all use TeX
00:22:45 <pikhq> Yes.
00:22:52 <pikhq> HyperTeX it is.
00:22:56 <coppro> :D
00:25:13 -!- Wamanuz2 has joined.
00:28:53 -!- cpressey has quit (Quit: Leaving.).
00:29:11 -!- Wamanuz has quit (Ping timeout: 265 seconds).
00:34:56 -!- augur has quit (Ping timeout: 240 seconds).
00:46:59 <coppro> anyone know how to get the command line of a running windows program?
00:53:39 -!- FireFly has quit (Read error: Connection reset by peer).
00:59:57 <ais523> coppro: is it even possible?
01:00:09 <coppro> ais523: it's got to be in memory somewhere
01:00:38 <pikhq> Yes, but it's not necessarily exposed to other processes.
01:10:23 -!- augur has joined.
01:12:31 -!- sshc_ has joined.
01:15:41 -!- sshc has quit (Ping timeout: 240 seconds).
01:37:43 -!- wareya_ has changed nick to wareya.
01:59:33 -!- Wamanuz2 has quit (Read error: Connection reset by peer).
01:59:35 -!- augur_ has joined.
02:00:03 -!- Wamanuz2 has joined.
02:00:15 -!- augur has quit (Ping timeout: 246 seconds).
02:07:38 -!- BeholdMyGlory has quit (Remote host closed the connection).
02:09:32 <cheater99> hello
02:21:34 -!- wareya_ has joined.
02:25:02 -!- Gregor-P has joined.
02:25:06 -!- wareya has quit (Ping timeout: 276 seconds).
02:37:27 -!- sshc_ has quit (Ping timeout: 276 seconds).
02:37:59 -!- sshc has joined.
02:38:10 -!- augur_ has changed nick to augur.
02:58:41 -!- Gregor-P has quit (Ping timeout: 260 seconds).
03:05:41 -!- Wamanuz2 has quit (Ping timeout: 240 seconds).
03:19:00 -!- Gregor-P has joined.
03:21:38 <Gregor-P> Hahaha
03:21:55 <Gregor-P> The day of the day is still there :P
03:29:53 -!- poiuy_qwert has quit (Quit: This computer has gone to sleep).
03:53:25 -!- CakeProphet has joined.
03:55:41 -!- SevenInchBread has quit (Ping timeout: 240 seconds).
03:58:27 -!- SevenInchBread has joined.
03:59:44 -!- CakeProphet has quit (Ping timeout: 240 seconds).
04:06:27 -!- poiuy_qwert has joined.
04:28:43 <AnMaster> Gregor-P, ?
04:29:29 <Gregor-P> Wiki front page
04:30:03 <AnMaster> Gregor-P, ah
05:36:32 -!- augur has quit (Ping timeout: 240 seconds).
06:00:53 -!- augur has joined.
06:05:27 -!- Sgeo has quit (Ping timeout: 276 seconds).
06:14:41 -!- GreaseMonkey has joined.
06:38:07 -!- augur has quit (Remote host closed the connection).
06:38:26 -!- augur has joined.
06:40:27 -!- coppro has quit (Quit: Reconnecting…).
06:40:56 -!- coppro has joined.
07:01:56 -!- tombom has joined.
07:31:04 -!- ais523 has quit (Remote host closed the connection).
07:32:45 -!- relet has quit (Quit: Leaving.).
07:37:20 -!- augur has quit (Ping timeout: 245 seconds).
07:48:51 -!- MizardX has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:01:25 -!- tombom has quit (Quit: Leaving).
08:02:46 -!- augur has joined.
08:04:15 -!- SimonRC has joined.
08:17:58 -!- GreaseMonkey has quit (Quit: I'm using NO SCRIPT WHATSOEVER - Download it at file:///dev/null).
08:40:44 -!- oerjan has joined.
08:45:13 -!- myndzi\ has joined.
08:49:36 -!- myndzi has quit (Ping timeout: 276 seconds).
09:07:43 -!- MigoMipo has joined.
09:33:37 -!- oerjan has quit (Quit: leaving).
09:38:42 -!- Slereah has quit (Read error: Connection reset by peer).
09:39:01 -!- Slereah has joined.
09:43:32 -!- MigoMipo has quit (Remote host closed the connection).
10:01:24 <cheater99> in my room i keep my love a tiny rubberband
10:16:33 <Deewiant> coppro: If you still need it: Process Explorer can show it
10:25:46 -!- MizardX- has joined.
10:28:03 -!- MizardX has quit (Ping timeout: 276 seconds).
10:28:04 -!- SevenInchBread has quit (Ping timeout: 264 seconds).
10:28:13 -!- MizardX- has changed nick to MizardX.
10:35:01 -!- CakeProphet has joined.
10:52:51 -!- MigoMipo has joined.
10:54:02 -!- MigoMipo_ has joined.
10:54:15 -!- MigoMipo_ has quit (Remote host closed the connection).
10:57:44 -!- MigoMipo has quit (Ping timeout: 240 seconds).
10:58:51 -!- MigoMipo has joined.
11:01:57 -!- DH____ has joined.
12:00:20 -!- BeholdMyGlory has joined.
12:01:43 -!- sebbu has quit (*.net *.split).
12:01:45 -!- fungot has quit (*.net *.split).
12:04:01 -!- sebbu has joined.
12:04:01 -!- fungot has joined.
12:07:39 -!- sebbu2 has joined.
12:10:30 -!- sebbu has quit (Ping timeout: 260 seconds).
12:12:16 -!- CakeProphet has quit (Ping timeout: 260 seconds).
12:22:42 -!- Wamanuz2 has joined.
12:48:40 -!- FireFly has joined.
12:48:47 -!- MigoMipo has quit (Read error: Connection reset by peer).
13:19:16 -!- jix has joined.
13:24:16 -!- augur has quit (Remote host closed the connection).
14:01:12 -!- distant_figure has joined.
14:40:31 -!- Madk has joined.
14:46:35 <Madk> Anyone else here?
14:46:38 <Madk> Am I alone?
14:46:42 <Madk> D:
14:51:42 <cheater99> yes
14:52:48 -!- distant_figure has quit (Ping timeout: 245 seconds).
14:52:49 -!- CakeProphet has joined.
14:53:02 -!- distant_figure has joined.
14:56:16 <Madk> sweet
14:56:17 <Madk> hi
14:57:11 <Madk> I'm all fuzzy inside - my m-code language is turing complete ;D
14:57:27 <Madk> finally finished by bf interpreter in it
14:59:00 <pikhq> Ah, the simple TC proof. :D
15:02:22 -!- Madk has quit (Read error: Connection reset by peer).
15:02:35 -!- Madk has joined.
15:04:52 -!- SevenInchBread has joined.
15:05:11 -!- CakeProphet has quit (Read error: Operation timed out).
15:11:41 -!- pikhq has quit (Read error: Connection reset by peer).
15:17:16 -!- pikhq has joined.
15:18:25 -!- MigoMipo has joined.
15:24:51 -!- relet has joined.
15:38:45 -!- pikhq has quit (Read error: Connection reset by peer).
15:45:42 -!- cpressey has joined.
15:47:53 -!- pikhq has joined.
15:56:32 <Madk> So, what's everyone else doing?
15:57:12 <Madk> I'm working on changing around the syntax for non-runtime commands in m-code to free up some characters
16:02:58 <Ilari> There's way more characters than you need in ISO 10646-1. :-)
16:10:13 <Madk> I'm only using easily typable characters though :P
16:10:28 <Madk> I've got 3 lowercase letters and a handful of uppercase left
16:10:37 <Madk> oh, and 0 and @
16:11:15 <Madk> What I'm doing now will free up ] and "
16:54:47 <AnMaster> huh
16:54:52 <AnMaster> Madk, m-code?
16:55:00 <Madk> Yes?
16:55:03 <AnMaster> what is it
16:55:09 <AnMaster> don't remember hearing about it
16:55:11 <Madk> A language I made recently
16:55:18 <AnMaster> is it on the esowiki?
16:55:19 <Madk> last few days I've been polishing it
16:55:21 <Madk> yeah
16:55:30 <Madk> http://esolangs.org/wiki/M-code#.5B.5BFibonacci_sequence.5D.5D
16:56:02 <AnMaster> hm 8/16/32-bit
16:56:09 <AnMaster> Madk, does that mean memory is limited?
16:56:13 <AnMaster> if so, it won't be TC
16:56:17 <Madk> sort of
16:56:23 <AnMaster> sub-TC then
16:56:44 <Madk> Nothing can be TC, then
16:56:58 <Madk> You'll run out of bits for an address eventually
16:57:30 <AnMaster> Madk, well not in practise, but in theory sure. Any esolang run on a physical computer won't be TC in that case. But the same esolang run on a Universal Turing Machine would
16:57:43 <AnMaster> if it allows you to have infinite storage in theory (which is what matters for TC)
16:57:57 <AnMaster> that said, esolangs doesn't have to be TC and may still be quite useful
16:58:03 <AnMaster> or usable or whatever
16:58:07 <Madk> If it were possible to have limitless addresses, the ability to access the memory is there
16:58:25 <Madk> in 8-bit the maximum address is 255, of course
16:58:31 <Madk> that's the most you can do in 8 bits
16:59:01 <AnMaster> Madk, so if you want to keep it TC just allow bignum addresses
16:59:22 <AnMaster> that is, assuming it is TC in other parts
16:59:26 <Madk> What, like a string for a number?
16:59:47 <AnMaster> well, the representation is obviously up to you but see: http://en.wikipedia.org/wiki/Arbitrary-precision_arithmetic
17:00:24 <AnMaster> sigh I wish ais523 was here, he could explain this a lot better
17:00:52 <Madk> Augh, is that really necessary D;
17:01:19 <AnMaster> Madk, you could do it as a variant. Or you don't need to be TC
17:01:43 <AnMaster> besides I don't know if the rest of requirements for TCness are satisfied. I just spotted the obvious first check
17:02:16 <Madk> There's absolutely no reason why it couldn't parse itself
17:02:26 <Madk> It would be slow, given, but still tc
17:02:50 <Madk> It has all the basic arithmatic operators
17:02:57 <Madk> a ton of memory
17:03:12 -!- oerjan has joined.
17:03:28 <Madk> It's 6052 assembly with single-char mnemonics
17:03:36 <Madk> not truly, but it's not so far from it
17:03:50 <AnMaster> could you implement brainfuck with infinite data tape in it?
17:04:05 <Madk> with bignums or such, yes
17:04:16 <Madk> and, of course, a machine with infinite memory
17:04:34 <oerjan> well brainfuck with finite tape but bignums also works for proving TC
17:04:42 <AnMaster> Madk, yes it just needs to be infinite in theory. Of course in practise it won't be on physical hardware
17:05:04 <AnMaster> or that yes
17:05:15 <AnMaster> but lets not confuse poor Madk
17:05:36 <Madk> I'm not that dumb, just new to some of these terms <.<
17:05:41 <AnMaster> sorry
17:05:46 <Madk> np
17:06:10 <oerjan> Madk: i removed your empty column from esointerpreters, we sort of don't want that page to get wider than necessary
17:06:39 <oerjan> you can put it back once you've implemented M-code in brainfuck, or something ;D
17:06:40 <Madk> I noticed, yeah, I should've thought about that beforehand
17:06:47 <Madk> :)
17:06:48 <AnMaster> oerjan, which page?
17:07:00 <Madk> I'm planning on trying to make a self-interpreter eventually
17:07:01 <oerjan> http://esoteric.voxelperfect.net/wiki/EsoInterpreters
17:07:08 <AnMaster> oerjan, ah the case
17:07:10 <AnMaster> that explains it
17:07:26 <oerjan> pesky mediawiki
17:07:32 <AnMaster> oerjan, that page is like 1/3 of my screen width
17:07:38 <AnMaster> the table I mean
17:07:38 <Madk> Any obvious features or convenieves I may be missing?
17:07:55 <AnMaster> Madk, conveniences?
17:08:00 <AnMaster> in an esolang?
17:08:03 <Madk> conveniences*
17:08:11 <AnMaster> no it wasn't about the spelling
17:08:15 <AnMaster> just about the concept
17:08:20 <oerjan> it only needs a couple more columns to fill out my browser window, admittedly i don't use quite full screen width
17:08:30 <oerjan> (it's a laptop)
17:08:33 <Madk> Some, yes
17:08:48 <cpressey> <Madk> Nothing can be TC, then
17:08:57 <cpressey> You're correct that no real machine can, if the universe is finite
17:09:00 <AnMaster> oerjan, this is a 2something" wide-screen monitor
17:09:07 <oerjan> AnMaster: i've been thinking if it gets much bigger it ought to be reorganized into something with a (vertex/edge) graph instead
17:09:08 <AnMaster> 1680x1050
17:09:09 <cpressey> But languages aren't real in that sense :)
17:09:33 <Madk> AnMaster: I don't want to force people to use half the program memory for relatively simple tasks :P
17:09:37 <oerjan> although i don't really know how to do that, especially if it's supposed to be easy to edit
17:09:53 <oerjan> (well i know how to make a graph on paper)
17:09:55 <AnMaster> oerjan, a table like that is a perfectly nice way to represent a graph where there can only be one edge between a given pair of nodes in a given direction
17:10:27 <oerjan> AnMaster: well actually it isn't such a graph, you will note there are several cells with multiple entries
17:10:37 <AnMaster> oh indeed
17:10:41 <AnMaster> oerjan, what do those mean?
17:10:50 <AnMaster> oh wait I think I see
17:10:52 <oerjan> AnMaster: links to the interpreters
17:11:02 <AnMaster> oerjan, so we have labeled edges
17:11:03 <AnMaster> nice
17:11:23 <AnMaster> oerjan, why do we mark C specifically
17:11:34 <AnMaster> oerjan, oh and befunge93/98 should be separate entries
17:12:09 <oerjan> and it's not about being a nice way, it's about being big. in fact one way of shrinking things a lot without making a graph would be to separate out into a few tables: from bf/befunge, to bf/befunge, self-interpreter and others
17:12:50 -!- cheater99 has quit (Ping timeout: 240 seconds).
17:12:52 <oerjan> AnMaster: maybe, i already separated out unefunge the other day when someone added it
17:13:31 <oerjan> (because the table is very sparse except for bf/befunge)
17:13:42 <AnMaster> oerjan, well befunge-93 is quite a different beast from befunge-98
17:14:30 <AnMaster> oerjan, apart from one geocities link there seem to be only befunge-93 apart from fungot which is befunge-98
17:14:31 <fungot> AnMaster: the fnord cost fnord at least, ls -l /lib/ fnord debugging symbols found)...done.
17:15:31 <AnMaster> so yeah please split that
17:15:31 <AnMaster> into 93/98
17:15:31 <AnMaster> I lost my login for the wiki ages ago
17:15:31 <AnMaster> and no password set
17:15:31 -!- cheater99 has joined.
17:16:25 <oerjan> too much work, i'm lazy. just to find out which dialects they actually _are_ using, it's not like most of the link urls indicate it
17:17:08 <AnMaster> oerjan, I checked them for you
17:17:11 <AnMaster> as I said above
17:17:21 <AnMaster> oh and that one for lazyk I couldn't find anything about
17:17:21 <oerjan> what!
17:18:09 <AnMaster> oh wait found it
17:18:11 <AnMaster> 93 too
17:18:42 <AnMaster> oerjan, so from those links: for befunge: all 93 except from the fungot link which is 98 and the geocities link which is dead and thus I have no clue about
17:18:43 <fungot> AnMaster: augur: i despise the state machine. :) ( it's at http://students.depaul.edu/csweeney/ scheme.code.html if anyone cares
17:18:51 <AnMaster> oerjan, checked for you
17:19:41 <cpressey> Someone gotta implement Thue in Kipple.
17:19:46 <oerjan> AnMaster: oh and the marking for C is because we started pondering that there _are_ so few actual cycles. it's an encouragement to make more. :)
17:20:58 <oerjan> AnMaster: did you check both befunge row and befunge column?
17:21:02 <AnMaster> oerjan, yes
17:21:27 <AnMaster> oerjan, and fungot is in two places (implements bf and ul)
17:21:27 <fungot> AnMaster: there's a new ( miniscule) processor architecture, written an emulator for it,
17:22:00 <oerjan> ok and there actually no f98 in the column, iiuc
17:22:00 <AnMaster> cpressey, why that specific combo?
17:22:07 <oerjan> *there are
17:22:19 <AnMaster> iiuc ?
17:22:37 <AnMaster> oerjan, well the geocities link is dead and not in waybackmachine
17:22:40 <AnMaster> so no clue about it
17:22:48 <cpressey> AnMaster: it's feasible to me
17:23:10 <oerjan> which is not surprising, since f98 is much harder to implement but at least as easy to implement with
17:23:38 <cpressey> No one has implemented Brainfuck in Unlambda?
17:23:40 <cpressey> Surprising.
17:24:34 <AnMaster> oerjan, anyway without splitting 93/98 we could end up with false cycles
17:25:21 <AnMaster> oerjan, note also that 93 is not a proper subset to 98. String mode handling of multiple spaces differs, and that breaks a few programs
17:26:02 <cpressey> I tried building PortableFalse last night, and found both gcc and pcc choke on it.
17:27:54 <cpressey> I don't suppose we know of any False interpreter implemented in Haskell?
17:29:20 <cpressey> <AnMaster> oerjan, well the geocities link is dead and not in waybackmachine <--- try reocities?
17:29:31 <AnMaster> cpressey, never heard of that
17:29:41 <cpressey> AnMaster: try it
17:30:17 <AnMaster> how does one use it
17:30:21 <AnMaster> the front page: tl;dr
17:30:37 <cpressey> AnMaster: s/www.g/www.r/
17:30:45 <cpressey> yeah, yeah, dots in re's, whatever
17:30:54 <AnMaster> "Page not (yet!) found..."
17:30:55 <AnMaster> cpressey, ^
17:31:09 <cpressey> AnMaster: Well, it was worth a shot.
17:31:31 <cpressey> <oerjan> which is not surprising, since f98 is much harder to implement but at least as easy to implement with
17:31:54 <AnMaster> before commenting on "but at least as easy" remember he is a mathematician
17:31:57 <cpressey> Seems to me the "true" tarpits would all be equally hard to implement each other in :)
17:32:08 -!- Wamanuz3 has joined.
17:32:13 <AnMaster> cpressey, befunge is by no means a tarpit
17:32:16 <AnMaster> especially not 98
17:32:28 <AnMaster> you should know that better than anyone else
17:33:08 <AnMaster> or at least as well as anyone else ;)
17:33:08 <Madk> haaargh my changes to syntax has broken my bf interpreter
17:33:08 <cpressey> AnMaster: you misunderstood my point
17:33:08 <AnMaster> perhaps
17:33:08 <Madk> must fix
17:33:11 <AnMaster> what did you mean then
17:33:34 <AnMaster> changing syntax *does* tend to break things
17:33:40 <AnMaster> what else would you expect
17:33:47 <cpressey> AnMaster: I have better things to do than to explain it to you. Sorry.
17:33:55 <AnMaster> cpressey, mhm
17:34:06 <Madk> AnMaster I went through and changed the stuff to the proper syntax and it's still broken, that's the problem
17:34:15 <AnMaster> oops
17:34:23 <Madk> AnMaster: Hopefully I just forgot one or two things
17:34:54 <Madk> It's either jumping past the end of the loop or it's not jumping to the beginning ._.
17:35:39 -!- Warrigal has quit (Ping timeout: 240 seconds).
17:35:39 -!- Wamanuz2 has quit (Ping timeout: 260 seconds).
17:35:39 <oerjan> ok it's not in reocities (yet)
17:35:39 -!- Warrigal has joined.
17:35:49 <Madk> my fibonacci sequence is still fine, though, after I changed the syntax, so I don't really think it's the m-code interpreter itself
17:36:43 <oerjan> AnMaster: how can befunge-98 _not_ be easier to implement things with? it has a lot of extra features afaik, even if it's not strictly a superset
17:37:58 <AnMaster> oerjan, it was you who claimed it was at least as easy
17:37:59 <AnMaster> not me
17:38:06 <AnMaster> it is interesting how much harder bub is to optimise than brainfuck
17:39:55 <oerjan> AnMaster: note i said implement _with_. one of the reasons why there are so few Cs in that page is, in my opinion, that the easier a language is to implement other things with, the harder it is to implement itself.
17:39:56 <AnMaster> though if you had balanced [] you could back-translate it into a loop trivially. A lot of other optimisation would require changes though
17:40:03 <AnMaster> for the non-matching case
17:40:16 <AnMaster> oerjan, true
17:40:26 <AnMaster> oerjan, each S is a C really
17:40:37 <oerjan> also i've now edited the page
17:42:42 <oerjan> AnMaster: yes but somehow i don't think they count as much
17:42:42 <oerjan> (if you want to check the result)
17:42:42 <AnMaster> oerjan, you edited it wrong. Move the fungot entry in the bef/bf cell up
17:42:42 <fungot> AnMaster: i understood exactly what you end up with
17:42:42 <AnMaster> to 98
17:42:42 <AnMaster> or wait
17:42:42 <AnMaster> rather
17:42:42 <AnMaster> move the geocities one about
17:42:42 <AnMaster> to somewhere
17:42:43 <oerjan> what
17:42:43 <AnMaster> oerjan, okay you see the XX in b98/bf? Well one of those is a dead link. We need befunge-unknown-due-to-dead-link
17:42:43 <oerjan> i thought you said the geocities one was f98. oh wait you meant unknown.
17:42:50 <Madk> hurrah, I hauled my hello world example into the new syntax
17:42:53 <AnMaster> oerjan, maybe put dead links at a list at the end
17:42:55 -!- iamcal has joined.
17:42:58 -!- cal153 has quit (Read error: Connection reset by peer).
17:43:00 <oerjan> sheesh.
17:43:01 <AnMaster> oerjan, with a note about it
17:43:08 <AnMaster> oerjan, just a suggestion for how to handle it
17:43:23 <AnMaster> it is trivially to use an automated link checker tool for it anyway
17:43:29 <AnMaster> I believe KDE comes with one
17:43:31 <AnMaster> bbl food
17:44:58 <Madk> There's my bf interpreter
17:45:04 <Madk> forgot two "$"'s
17:46:13 <Madk> I raise my eyebrow in confusion.
17:46:29 <Madk> The comment-stripper doesn't work right anymore
17:48:16 <Madk> hup, that was a simple fix :P
17:49:17 <oerjan> AnMaster: ok i've now fixed it as much as i can be bothered
17:50:13 <Madk> oerjan: He left
17:50:35 <oerjan> just as well, there's no way he'll be satisfied with what i did anyway ;D
17:50:45 <Madk> lol
17:51:03 <Madk> What page is this? Befunge '93 and '98?
17:51:23 <oerjan> there's just one Befunge page, but the one i edited is EsoInterpreters
17:51:49 -!- tombom has joined.
17:52:22 <Madk> You forgot to add -93 to the end of Befunge in the first column
17:52:30 <oerjan> no, i removed it again
17:52:32 <Madk> unless you weren't intending to, in which case you should
17:52:44 <oerjan> because it contains one unknown interpreter
17:52:59 <oerjan> which is a dead link which we don't know which befunge dialect it is
17:53:29 <Madk> Neither link is dead for me
17:53:37 <Madk> One says a '93 interpreter in '93
17:53:48 <Madk> the other says a '93 interpreter in befunge without a year
17:53:59 <oerjan> um the brainfuck interpreter in befunge
17:54:04 <Madk> Ah
17:54:30 <Madk> If the link is broken take it down, it's of no use to anybody :/
17:54:30 <fizzie> I should perhaps point at the standalone Underload thing instead of fungot there for the Underload-in-Funge98 case.
17:54:30 <fungot> fizzie: fnord this shit doesnt achieve anything. :p)) some long, huge expression...)
17:55:42 <oerjan> Madk: the point is reocities is adding old geocities links slowly, so it _might_ become found again and i don't like to delete dead links completely without replacement
17:56:03 <oerjan> if we delete them then we'll _never_ remember to check if they've reappeared
17:56:07 <Madk> Anyhow, I'm off to write the collatz and the seive of eratosthenes in m-code :D
17:56:13 <Madk> oerjan: Oh, ok
17:56:26 <Madk> oerjan: What about a separate wiki page for dead links
17:56:44 <Madk> Go check them every now and then and if one works, put it wher it belongs
17:56:52 <oerjan> also the wayback machine, which i read somewhere hasn't added anything to the publicly available database for years
17:57:10 <fizzie> The EsoInterps table could be wider; I should perhaps implement something that doesn't have an esointerp yet. (Then I could transmogrify it into fungot, perhaps.)
17:57:11 <fungot> fizzie: definitely worth .25, clearly wrong, since the last time you took a turning test? to see why:
17:58:19 <cpressey> I adore fungot.
17:58:20 <fungot> cpressey: the price had went down from humongous to merely ginormous. the latter one doesn't look very probable though.) i think i actually need four quite small tunes
18:08:46 -!- oerjan has quit (Quit: Reboot).
18:10:13 -!- augur has joined.
18:10:49 <fizzie> AnMaster: Since we've usually been swapping these, here's a freehand (to excuse the very wobbly horizon) 360-degree huginization from a nearby cliff-thing: http://zem.fi/~fis/P1070725-739.jpg
18:11:56 -!- oerjan has joined.
18:16:12 <AnMaster> fizzie, hm nice
18:16:38 <AnMaster> oerjan, well your way is kind of okayish
18:16:43 <AnMaster> but not very probable
18:16:53 <AnMaster> since bf in 93 would mean very short tape
18:17:17 <AnMaster> fizzie, it is hard to tell that it is wobbly. It looks like a hill at first glance
18:17:45 <AnMaster> oh it is a hill
18:17:46 <AnMaster> XD
18:18:01 <oerjan> hillarious
18:18:12 <AnMaster> fizzie, yeah landscape too hilly in the near area to be able to tell if the horizon is wobbly or not
18:18:44 <AnMaster> fizzie, also very little parallax I see. Nice
18:18:53 <AnMaster> I don't see any seam
18:18:55 <fizzie> Yes, it is a hill. Well, the freehandiness also meant I had to leave quite a lot of black in there. (Or crop a lot out.)
18:19:07 <AnMaster> well yes
18:20:02 <fizzie> There's one visible seam near the left edge, in the roof/top-of-balcony of the bricky building visible through a gap in the trees.
18:20:33 <AnMaster> oh indeed, I thought it was strange architecture
18:20:47 <AnMaster> like another house behind it
18:21:43 <AnMaster> since it neatly lines up with the balcony
18:22:14 <fizzie> There's also source images for at least one (if not two) full-circle panoramas (plus a few wide-angle shots) more from different spots on the hill, but I haven't had time to combinate them yet.
18:22:27 <AnMaster> fizzie, a way to solve having to crop as much would be taking pictures in a lower line as well
18:22:35 <AnMaster> so you get a larger hfov
18:22:37 <AnMaster> err
18:22:39 <AnMaster> vfoc
18:22:42 <AnMaster> vfov*
18:23:15 <AnMaster> fizzie, where are those hills?
18:24:44 <fizzie> It's called Laturinkallio, and it's approximately here: http://maps.google.com/?ll=60.223788,24.816624&z=17
18:25:57 <fizzie> The houses with the seam are in the "triangle" between Laturinkatu / Porarinkatu / Mestarintie/Kehä I in that map; Google's satellite image is too old to show them. (They've been built mostly during this year.)
18:26:06 <AnMaster> how strange it is to see both Swedish and Finnish names on things
18:26:30 <AnMaster> Though "Laddargränden" is a strange Swedish name for an alley
18:26:46 <AnMaster> Laddar is not a Swedish word that I know
18:27:06 <fizzie> That particular bit of road is "Laturinkuja" (where fi:kuja == en:alley) in Finnish, too.
18:27:48 <AnMaster> No it says "Nikkaripolku" on that one?
18:28:01 <fizzie> It changes name there.
18:28:02 <AnMaster> or does the road rename is the middle?
18:28:03 <AnMaster> ah
18:28:33 <fizzie> It doesn't continue so smoothly as the map makes it look like; there's a series of steps there in the middle, you can't drive a car through.
18:28:53 <AnMaster> hah
18:29:11 <AnMaster> fizzie, besides, I doubt it goes through those houses
18:29:34 <fizzie> Right. Openstreetmap represents it a lot better: http://osm.org/go/0xPMNrFE?layers=M
18:30:12 <fizzie> Lacks the Swedish names, though. (Or doesn't show them, anyway; I would think they are in the metadata still.)
18:30:20 <AnMaster> it still goes through the houses
18:30:29 <AnMaster> that red line I mean
18:31:26 <fizzie> Well, it goes right next to the houses. There's no satellite-map overlay in osm, so I'm not sure how well it is actually placed.
18:31:43 <fizzie> OSM also shows the current and future routes of Kehä I; there's a large road project going on there.
18:33:02 <AnMaster> ah
18:33:37 <AnMaster> fizzie, on the north or south side of them?
18:33:49 <fizzie> South.
18:34:18 <AnMaster> fizzie, also you could drive a car there. With sufficiently large wheels
18:35:02 <fizzie> It's not a very wide path. But yes, for some values of "car" you could drive one through.
18:35:20 <fizzie> http://kartat.eniro.fi/m/pKpro shows both the houses *and* the road, if you're curious.
18:35:32 <cpressey> oerjan: Shelta has been implemented in itself -- is that worth adding to EsoInterpreters?
18:35:50 <AnMaster> fizzie, hey I thought eniro was Swedish!?
18:36:15 <fizzie> Also, here's the beginning of the steps: http://kartat.eniro.fi/m/pKpmQ -- it should go directly to eniro's streetview-alike.
18:36:39 <oerjan> cpressey: sure
18:36:41 <fizzie> Admittedly there are those ramps for wheelchairs, those might help a bit in driving up there.
18:36:49 <fizzie> (Down would be easier, I guess.)
18:37:03 <AnMaster> fizzie, link doesn't work
18:37:17 <AnMaster> as in, I get a blank photo and no "plugin missing" bar at the top
18:37:32 <fizzie> Curious; it works-for-me(tm).
18:37:56 <AnMaster> fizzie, well with _sufficiently_ large wheels it shouldn't be any issue anyway. Heck with *sufficiently* large wheels that hill wouldn't pose a problem at all, it would just like a bump in the "road"
18:38:07 <AnMaster> fizzie, enabled javascript too
18:38:13 <AnMaster> still doesn't work
18:38:24 <fizzie> For both eniro.fi and eniro.com?-)
18:38:30 <AnMaster> yes
18:38:55 <fizzie> Hm, well. You can click "Katunäkymä" in the earlier map and try to navigate there manually.
18:38:59 <AnMaster> oh wait, if I enlarge the window I see far out on the right: "Gatuvy behöver Adobe Flash."
18:39:06 <fizzie> Heh.
18:39:09 <AnMaster> what a silly place to put it
18:39:33 <AnMaster> strange it was translated to Swedish
18:39:44 <AnMaster> since browser is set to English
18:39:52 <fizzie> Google hasn't bothered to run their streetview car into Nikkarinkuja. But they have driven to the end of Laturinkuja, you just can't see the steps so well from there.
18:40:09 <AnMaster> fizzie, okay, localising google earth binary
18:40:13 <AnMaster> it should be *somewhere*
18:40:24 <AnMaster> maybe on a previous install
18:40:32 <fizzie> http://maps.google.com/?ie=UTF8&ll=60.223373,24.81612&spn=0.005014,0.013036&z=17&layer=c&cbll=60.223491,24.816834&panoid=DOmQFc4jsxgXEgAuw9xc2w&cbp=12,202.36,,0,-3.3 points at Google's streetview hopefully.
18:40:36 <fizzie> (Horrible link.)
18:40:58 <AnMaster> fizzie, can't do in browser due to previously mentioned lack of flash
18:41:05 <AnMaster> should be able to do it in google earth
18:41:24 <AnMaster> fizzie, or you could just upload a screenshot of it for the .fi one
18:41:34 <fizzie> I was just about to suggest that. A moment.
18:41:44 <AnMaster> thanks a lot
18:41:57 <AnMaster> hm where *is* the google earth steuff
18:41:58 <AnMaster> stuff*
18:42:03 <AnMaster> I know I had it installed
18:42:18 <cpressey> oerjan: Er... it's actually a Shelta->8086 compiler written in Shelta. Not technically a self-interpreter.
18:42:38 <cpressey> oerjan: So... ignore me :)
18:42:45 <fizzie> AnMaster: http://zem.fi/~fis/steps.jpg
18:42:52 <cpressey> The whole page is about *interpreters*
18:43:02 <AnMaster> fizzie, btw I wonder how google solves the parallax in their pictures. Since their cameras can't all be in the same place
18:43:17 <AnMaster> fizzie, ah yeah indeed some problems for your average car
18:45:38 <fizzie> The streetview car (well, one of them) looks funky: http://sfcitizen.com/blog/wp-content/uploads/2008/06/2584018127_c2701eaef8_o.jpg
18:45:58 <AnMaster> fizzie, sure that ramp is for wheelchairs and not baby wagons (is that the English term for it?
18:45:59 <AnMaster> )
18:46:13 <AnMaster> fizzie, it seems quite horrible to drive an electric wheelchair up that
18:46:33 <AnMaster> or a manual for that
18:46:50 <fizzie> Yes, I guess it would be mostly used by baby carriages (I think that's the term).
18:46:52 <AnMaster> fizzie, is that a policeman behind it?
18:47:04 <fizzie> It looks like one; I'm not sure what's happening in the picture.
18:47:21 <AnMaster> fizzie, anyway those cameras look really strange
18:47:39 <fizzie> The four beige blocks are some sort of laser range-finders.
18:47:43 <AnMaster> err, if the cameras are the black ones at top
18:47:49 <AnMaster> if they are the white ones then even stranger
18:48:01 <AnMaster> fizzie, you mean white
18:48:38 <fizzie> I guess. It said "tan" in the page describing the photo, I just picked yet another light colour.
18:48:58 <AnMaster> fizzie, link to that page?
18:49:28 <AnMaster> fizzie, still it doesn't explain how to 1) solve parallax 2) solve nadir image
18:49:40 <AnMaster> the nadir one tends to have parallax at google though
18:49:59 <AnMaster> I suspect they extract that one from the next/previous one
18:50:45 <fizzie> http://www.byetman.com/2008/06/17/google-street-view-car-busted-in-more-ways-than-one/ claims that they're range-finders.
18:51:31 <oerjan> cpressey: ok
18:51:41 <fizzie> AnMaster: Ooh, here's a better look at the camera: http://ekstreme.com/images/google-streetview-camera-1.jpg
18:54:32 <AnMaster> okay the camera thing is definitely the one at the top
18:54:40 <AnMaster> still how the fuck do they solve parallax
18:55:32 <fizzie> Well, there's not *that* much of it. And they do know the (relative) positions of their cameras and all.
18:55:55 <AnMaster> fizzie, oh and a fun thing I noticed at E20 of google street view: on the north bound side it showed a road covered with rain water, on the south bound side it was sunny
18:56:33 <oerjan> solving parallax is such a paradox
18:56:33 <fizzie> Yeah, "big" roads like that have some discontinuities.
18:56:58 <AnMaster> fizzie, plus since there were two lanes in each direction on that section it showed that there was cloudy in the outer lane in the south bound direction :D
18:57:22 <AnMaster> cloudy but not raining that is
18:57:31 <fizzie> Ha, and as for "solving" parallax, I just rotated the "Laturinkuja" Google streetview image around, and there's a very nice ghostly duplicate of one tree here. I'll screenshotify it.
18:57:47 <AnMaster> fizzie, they use cars to do it too?
18:58:16 <AnMaster> <oerjan> solving parallax is such a paradox <-- that one was so bad
18:59:12 <oerjan> virtually without parallel
18:59:17 <fizzie> http://zem.fi/~fis/ghost-tree.jpg <-- I wouldn't call that a "solved" merging of their panorama images.
18:59:38 <fizzie> (There's a bit of a seam in the sidewalk too.)
19:02:25 <AnMaster> fizzie, indeed, even hugin does better
19:02:49 <AnMaster> fizzie, hugin would give a cut off tree instead
19:03:20 <AnMaster> unless you use enfuse
19:03:27 <fizzie> Yes, their image-blender seems more enfusey than enblendy.
19:03:42 <AnMaster> fizzie, but enfuse is so much slower than enblend that I doubt it would be feasible to run that on so much data
19:04:11 -!- ais523 has joined.
19:06:11 <fizzie> AnMaster: Well, while we're at it, here's one of the wide-angle huginizations: http://zem.fi/~fis/P1070756-760.jpg -- the road that goes parallel to the picture there is Laturinkuja/Laddargränden.
19:14:40 <cpressey> oerjan: At any rate, that table has got me thinking about the feasibility of implementing esolangs in other esolangs. Like, why implementing Thue in Kipple, or Kipple in Thue, would both be hard, for different reasons.
19:15:24 <oerjan> thue has that pesky no good IO problem
19:15:51 <cpressey> oerjan: I was just about to say that -- it extends to initial data being pesky as well, doesn't it?
19:16:39 <cpressey> e.g. The initial data would be a Kipple program -- you'd first need to transform that into something you could interpret -- and that would be ugly. Well, I guess not impossible though.
19:16:53 <oerjan> well initial data is no more allowed to be free format than input is
19:17:25 <cpressey> For some languages, for some small character sets, it might be impossible
19:17:30 <oerjan> the thing is _everything_ in free format can clash with the thue program source
19:17:30 -!- BeholdMyGlory has quit (Remote host closed the connection).
19:17:40 <cpressey> Exactly
19:18:00 <cpressey> You need disjoint character subsets for the initial data and the "intermediate representation"
19:18:23 <cpressey> Or... boom (potentially) :)
19:19:03 -!- Madk has quit (Read error: No route to host).
19:19:11 <oerjan> Itflabtijtswi, which is also substitution based, solves this by reading only single characters at a time
19:19:30 <oerjan> *Itflabtijtslwi
19:20:56 <ais523> cpressey: do you allow compiling esolangs into esolangs?
19:21:18 <ais523> the big problem with Thue, really, is a sort of wire-crossing problem
19:21:36 <ais523> it's a pain to move data from one end of the program to the other
19:21:46 <cpressey> ais523: I just now wanted to add the Shelta->8086 compiler written in Shelta to the page, but decided against it because it's excpliticly about EsoInterpreters.
19:21:47 <oerjan> if someone starts compiling esolangs into esolangs with esolangs, we will certainly need to put that on the wiki somewhere
19:22:02 <ais523> cpressey: try adding it to the Shelta page
19:22:06 <ais523> if it isn't there already
19:22:07 <cpressey> ais523: I should.
19:22:25 <ais523> oerjan: does Perl count?
19:22:30 <oerjan> heh :D
19:23:05 <ais523> compiling esolangs into esolangs with esolangs is something I've been considering, incidentally
19:23:11 <ais523> I really need to finish Cyclexa sometime
19:23:15 <ais523> or, well, properly start it
19:26:57 <oerjan> i thought that was most of the point with your underlambda thing
19:28:41 <ais523> yes, agreed
19:28:50 <ais523> underlambda's a language to compile via, or will be when I finish it
19:28:56 <ais523> hmm, RL gets in the way of esolanging so much
19:34:20 -!- augur has quit (Remote host closed the connection).
19:37:20 -!- augur has joined.
19:40:30 <cpressey> I should learn Underload sometime.
19:41:13 <cpressey> Has anyone designed a language with the goal of making quines hard to write?
19:42:38 <ais523> someone tried to make a language where they were impossible to write
19:42:46 <oerjan> http://esoteric.voxelperfect.net/wiki/User_talk:Smjg
19:42:47 <ais523> on the basis that any program that was a quine was rejected, anything else ran fine
19:43:00 <cpressey> Well yes, they argued with me that such a language exists *in principle*, which is true ;)
19:43:17 <oerjan> cpressey: not only did they do so, but you invented narcissists because of it
19:44:11 <cpressey> oerjan: Indeed
19:44:27 <oerjan> but their idea iirc is a trivial modification that can be applied to any language
19:44:30 <cpressey> But right now I'm just thinking of a language that would make quine-writing ugly. For lack of a better way to put it.
19:45:26 <oerjan> well the more awkward computing and printing strings is...
19:46:07 <cpressey> oerjan: It was Underload's "in output, parens must be balanced" that got me wondering about it.
19:48:18 <oerjan> well then, anything that makes it _impossible_ to print a program source in the language would pretty much ruin things. see intercal-72.
19:48:31 <ais523> well, yes
19:48:34 <ais523> you could still encode it somehow, though
19:48:44 <cpressey> Yeah, yeah. Not what I'm thinking of :)
19:48:58 <oerjan> encoding output is entirely contrary to the spirit of quines
19:49:06 <cpressey> btw, why are []<> reserved in Underload anyway?
19:49:20 <cpressey> Seems a bit arbitrary, reading through the spec
19:49:23 <ais523> in a language like Unlambda, quines are hard because of the difficulty of copying data
19:49:26 <ais523> and yes, it is rather arbitrary
19:49:29 <oerjan> future expansion?
19:49:32 <ais523> it's for backwards compatibility with something that doesn't exist
19:49:53 <cpressey> I see. I notice <>'s appear in the stack when you single-step trhough the JS interpreter
19:50:02 <ais523> they're the stack separator
19:50:08 <cpressey> Indeed.
19:50:09 <oerjan> i'd say with unlambda it's more about the awkwardness of string representation
19:50:27 <ais523> I use that because it's illegal in the input program, or in the working
19:50:58 <ais523> a "neater" representation, though, is simply to parenthesise the stack elements
19:51:06 <oerjan> the shortest unlambda quine is a rather evil continuation hack
19:51:33 <ais523> ooh, ingenious
19:51:36 -!- ais523 has changed nick to ais523_.
19:51:45 <oerjan> (btw the one in the distribution can be shortened somewhat, because it misses the optimization `kv = k or something like that
19:51:46 <ais523_> ooh, ingenious
19:51:48 <oerjan> )
19:51:51 <ais523_> err, what?
19:51:53 -!- ais523_ has changed nick to ais523.
19:52:07 <oerjan> let me look it up
19:52:14 <ais523> and you mean `kv=v, surely?
19:52:22 <oerjan> er yes
19:52:33 <ais523> you can think of v as a fixpoint for k
19:52:52 <oerjan> quine06
19:53:09 <oerjan> yep, it has a `kv in it
19:53:37 <ais523> incidentally, I think I have a continuations impl for Underlambda written in Underlambda preprocessor somewhere
19:53:43 <ais523> and it would be easy enough to adapt to Underload
19:54:07 <ais523> as all Underload commands but S translate directly to Underlambda
19:54:27 <oerjan> and when you simplify that, you shave off ``sc.```s``sc.k``s as well
19:54:46 <ais523> you'd need to modify the string representation too, wouldn't you?
19:54:52 <oerjan> (since that's the part of the function representing the `k)
19:55:00 <ais523> ah
19:55:02 <oerjan> that's what i'm saying
19:55:28 <ais523> more languages need continuations
19:55:53 <ais523> I'm still really proud of continuation.i
19:56:30 <oerjan> hm?
19:57:04 <ais523> I implemented continuations in INTERCAL
19:57:15 <oerjan> ah
19:57:18 <ais523> pure INTERCAL, that is, it doesn't even use an external C library
19:57:30 <ais523> (pure C-INTERCAL; doing so in INTERCAL-72 would be insane)
19:57:38 <cpressey> Jeez. That suggests CONTINUE FROM to me.
19:57:45 <oerjan> hey insanity is an advantage here
19:58:04 <ais523> cpressey: that's a three-liner; NEXT FROM, make a continuation, RESUME
19:58:08 <cpressey> My mind just drove into a tree.
19:59:23 <cpressey> Ah, but it reminded me of the idea I had earlier this morning, that I forgot. Thank you!
20:01:42 <ais523> I actually really like C-INTERCAL's control flow, more languages should use it
20:01:55 <ais523> it's a lot more natural, in a way, then the kludges many other languages use to do much the same thing
20:02:19 <oerjan> ais523: you need a *MWAHAHAHAHA* after saying that
20:02:30 <ais523> no, not really
20:02:51 <oerjan> tsk, such denial
20:08:25 <AnMaster> <fizzie> AnMaster: Well, while we're at it, here's one of the wide-angle huginizations: http://zem.fi/~fis/P1070756-760.jpg -- the road that goes parallel to the picture there is Laturinkuja/Laddargränden. <-- nice
20:09:15 <AnMaster> fizzie, you should like totally take multiple 360° panos and link them together like a virtual fizzie off-street view :D
20:18:35 -!- Madk has joined.
20:25:13 -!- Sgeo has joined.
20:31:33 -!- BeholdMyGlory has joined.
20:37:03 <AnMaster> ais523, hi there
20:37:49 <AnMaster> ais523, I have a question about nwn toolset. How on earth do I create new items of cloth type armour. I seem unable to manage anything but light armor.
20:38:10 <AnMaster> ais523, editing cloth type also turns it into light armor
20:38:15 <AnMaster> ais523, I suspect a bug
20:38:34 <AnMaster> ais523, I think I have 1.69 or whatever the last one was called. Diamond edition iirc
20:39:04 -!- BeholdMyGlory has quit (Remote host closed the connection).
20:39:46 -!- BeholdMyGlory has joined.
21:04:48 <Sgeo> "A regular expression to check for prime numbers"
21:04:54 <Sgeo> I'm assuming that that's a joke
21:05:14 <Sgeo> Oh, it's supposedly for un.. wait, still
21:07:15 <ais523> in unary, with Perl 5.10 regexen, ^(aa+){2,}$(*PRUNE)(*FAIL)|
21:07:44 <ais523> actually, Perl's been stealing a whole bunch of features from Cyclexa, just with more unwieldy syntax
21:07:45 <coppro> unary?
21:07:57 <ais523> coppro: representing, say, 9 as aaaaaaaa
21:08:06 <coppro> ah
21:08:12 <ais523> length of a string used to represent the number
21:08:20 <coppro> I thought you meant a language or something
21:08:22 <ais523> pattern-matching esolangs tend to use it, as it's easiest
21:08:37 <ais523> and there is a lang called Unary, but it's one of those concept things like HQ9+
21:08:57 <coppro> what do *PRUNE and *FAIL do?
21:09:04 <ais523> cut and fail from Prolog
21:09:16 <coppro> remind me
21:09:17 <AnMaster> ais523, so no idea?
21:09:21 <Sgeo> Wait, that _works_?
21:09:30 <Sgeo> I assume it must be inefficient for large numbers?
21:09:39 <ais523> AnMaster: change the torso model, the material and AC change to match
21:09:45 <coppro> oh wait, I get it
21:09:47 <ais523> Sgeo: yes, very
21:09:54 <AnMaster> ais523, huh, material is where?
21:10:08 <coppro> heh, that's awesome
21:10:15 <ais523> AnMaster: not visible directly, but, say, if you recolour leather, looking at what changes colour lets you see what's made of leather
21:10:28 <coppro> is there proof that Perl 5.10 regexes are TC yet?
21:10:28 <AnMaster> ais523, hm...
21:10:42 <ais523> and (*PRUNE) = discard backtracking points before here, (*FAIL) = try to backtrack
21:10:45 <ais523> the combo forces the regex to fail
21:10:53 <coppro> right
21:11:45 <ais523> actually, that regex needs special handling for 0 and 1
21:11:50 <ais523> or it marks them as prime
21:11:52 <ais523> but you get the idea
21:11:55 <coppro> true
21:12:02 <coppro> that's easy enough
21:12:08 <AnMaster> ais523, link to this Cyclexa
21:12:11 <AnMaster> I can't find it
21:12:16 <ais523> there isn't one
21:12:20 <ais523> as I said, it's unfinished
21:12:26 <AnMaster> ais523, then how can perl steal from it
21:12:31 <ais523> unknowingly!
21:12:32 <AnMaster> ais523, is it your language?
21:12:32 <coppro> ^($|a$|(aa+){2,})$(*PRUNE)(*FAIL)|
21:12:40 <coppro> err
21:12:43 <coppro> remove the two $s
21:12:57 <cpressey> that
21:12:58 <cpressey> is
21:12:59 <cpressey> unholy.
21:13:02 <AnMaster> ais523, if so, a description and dare share your work earlier ;P
21:13:16 -!- Behold has joined.
21:13:30 -!- BeholdMyGlory has quit (Disconnected by services).
21:13:42 <ais523> AnMaster: a cross between regexen and Prolog
21:14:09 <ais523> with a built-in primitive for recursive replacements on parse trees
21:14:27 <ais523> (most langs don't have a primitive /that/ specialised; Cyclexa was intended to write compilers)
21:15:34 <cpressey> I have to give Cyclexa points for sounding like the marketing name for a drug.
21:15:38 <cpressey> Sorry, *medication*.
21:15:44 <AnMaster> ais523, hm
21:16:03 <AnMaster> cpressey, you are right
21:16:04 <AnMaster> it does
21:16:05 <cpressey> Talk to your doctor about Cyclexa!
21:16:05 <ais523> also, it has antitext
21:16:18 <AnMaster> huh?
21:16:45 <ais523> if you match a string against normal text, that text is removed from the string
21:16:52 <ais523> if you match a string against antitext, it's added to the string
21:16:54 <AnMaster> hm the crickets are *really* loud outside today
21:17:06 <ais523> thus you end up with a TC language, rather than an LBA
21:17:39 <AnMaster> ais523, weird
21:17:54 <AnMaster> ais523, is it always added? Or only in case of a non-match?
21:17:58 <AnMaster> or in case of a match?
21:18:12 <ais523> how can it fail to match?
21:18:15 <ais523> so yes, always added
21:18:22 <ais523> but ofc that bit of the regex might not run at all
21:19:04 <AnMaster> ah
21:19:27 <AnMaster> ais523, well fail to match = only added if not there already
21:19:36 <Madk> nuuu
21:19:40 <ais523> err, adding something != removing it
21:19:41 <Madk> esolangs.org is down
21:19:45 <ais523> I don't see why you'd consider those the same
21:19:49 <Madk> whup
21:19:51 <Madk> that was short
21:19:57 <AnMaster> ais523, I meant for antitext?
21:19:58 <ais523> Madk: working fine for me
21:20:21 <cpressey> Howabout, nondeterministic regexp replacements: so s/a/xy/ on "baza" yields {baza, bxyza, bazxy, bxyzxy}
21:20:32 <AnMaster> cpressey, nice idea
21:20:45 <cpressey> with n modifier: s/a/xy/n
21:21:06 <AnMaster> cpressey, n meaning?
21:21:09 <cpressey> Too bad Perl insists on being deterministic. I mean in general
21:21:21 <AnMaster> cpressey, doesn't it have an rng?
21:21:23 <AnMaster> err
21:21:25 <AnMaster> prng*
21:21:30 <ais523> I'm reasonably certain you can do something like that in Cyclexa
21:21:46 <cpressey> AnMaster: I mean the theory-of-computation meaning of nondeterminstic
21:21:50 <AnMaster> ais523, show us the spec
21:22:01 <ais523> not finished!
21:22:04 <AnMaster> cpressey, right
21:22:11 <AnMaster> ais523, release draft then!
21:22:15 <ais523> but backtracking does nondeterminism, in a rather inefficient way
21:22:29 <cpressey> ais523: That's why I added "in general" :>
21:22:35 <ais523> I imagine something like a=(xy|) would be enough to do what cpressey suggested
21:22:40 <ais523> (in Cyclexa, that is)
21:23:03 <AnMaster> ais523, explain that syntax or post daft spec
21:23:19 <ais523> daft specs would seem a rather appropriate thing for this channel
21:23:21 <augur> ruby! \o/
21:23:21 <myndzi\> |
21:23:21 <myndzi\> /<
21:23:37 <AnMaster> ais523, err? isn't that a good thing?
21:23:37 <ais523> AnMaster: "replace a with xy or nothing"
21:23:46 <AnMaster> augur, yes ruby is horrible from what I seen
21:23:52 <ais523> I think you'd need a modifier to do a global replace, but I forget what it is offhand
21:24:01 <augur> AnMaster: i love ruby
21:24:06 <augur> i just discovered something wonderful about ruby :D
21:24:12 <AnMaster> augur, isn't haskell better?
21:24:33 <augur> yes, but not for scripting and the like
21:24:48 <augur> but i discovered that ruby has minor pattern matching :D
21:25:11 <AnMaster> augur, erlang has pattern matching on par with haskell. better for some things, slightly worse for some
21:25:15 <AnMaster> and there is escript
21:25:20 <AnMaster> which is erlang as script
21:25:31 <augur> perhaps. but i like being able to dick around in TextMate and test things like and so forth
21:25:41 <AnMaster> augur, I have no idea about that editor
21:25:47 <ais523> Prolog's pattern matching beats Haskell's
21:25:49 <augur> its a mac editor
21:26:04 <augur> but oh man
21:26:07 <augur> check this out
21:26:08 <AnMaster> ais523, can't it do erlang?
21:26:16 <cpressey> ais523: I think I saw someone that tried to add regexps to Haskell's pattern matching !
21:26:23 <ais523> AnMaster: that's rather out of context...
21:26:24 <ais523> cpressey: ouch!
21:26:35 <AnMaster> err
21:26:36 <augur> [[1,2], [3,4]].map { |(a,b)| a+b } == [3,7]
21:26:38 <AnMaster> ais523, augur*
21:26:41 <AnMaster> tab failure
21:26:47 <augur> wonderful!
21:26:51 <cpressey> ais523: It's not a bad idea, I don't think, but the potential for mess... yeah.
21:27:13 <augur> its like, i can do map (\[a,b] -> a+b) [[1,2], [3,4]]
21:27:16 <augur> :D
21:27:33 <ais523> surely the clean method would just be 'myFunction a | a `matchesRegex` "regex" = ...'
21:27:45 <AnMaster> augur, and you can't in haskell?
21:27:51 <augur> no no, you CAN in haskell
21:27:58 <AnMaster> see
21:28:01 <AnMaster> haskell >> ruby
21:28:01 <augur> but i discoverd you can do the thing above that in ruby!
21:28:07 <augur> haskell >> ruby?
21:28:12 <augur> do haskell, then do ruby/
21:28:16 <augur> well if i must!
21:28:17 <AnMaster> augur, no no
21:28:21 <augur> but i'd prefer >>= to >> honestly
21:28:24 <AnMaster> augur, "much better than"
21:28:25 <augur> i mean lets be serious here
21:28:27 <AnMaster> in this context
21:28:33 <augur> i know what you meant, spergface
21:28:34 <augur> :|
21:28:43 <AnMaster> spergface? I never heard that insult before
21:28:53 <pikhq> [a+b|a<-[1,2],b<-[3,4]] -- Better.
21:29:15 <AnMaster> pikhq, that hurts. Interestingly I think it also is valid erlang, or nearly so
21:29:19 <augur> pikhq, yes, but not as illustrative
21:29:35 <pikhq> augur: Mmm, true.
21:29:45 <AnMaster> hm it isn't strangely enough
21:29:48 <AnMaster> even with ||
21:29:56 <AnMaster> I wonder why it returns an empty list
21:30:00 <AnMaster> that makes no sense
21:30:07 <pikhq> (\[a,b]->a+b)<$>[[1,2],[3,4]]
21:30:09 <pikhq> Thar; done.
21:30:38 <AnMaster> pikhq, couldn't you do [a+b|[a,b]<-[[1,2],[3,4]]]
21:30:39 <cpressey> Haskell has some srsly gnarly operators for being an academic language.
21:30:50 <pikhq> AnMaster: Yes.
21:30:59 <AnMaster> pikhq, would it do the same though
21:31:34 <pikhq> cpressey: Yeah, well. Hooray, functions.
21:31:56 <pikhq> <$>, BTW, is not in Prelude.
21:32:00 <pikhq> Data.Applicative has it.
21:32:11 <AnMaster> pikhq, what does it do?
21:32:15 <AnMaster> I don't think I got that far yet
21:32:22 <pikhq> AnMaster: It's infix fmap.
21:32:26 <augur> applicative functors! :D
21:32:34 <augur> i just read that in Learn You a Haskell
21:32:50 <pikhq> It goes along with <*>, which is the applicative functor operator...
21:33:06 <augur> <$> :: (a -> b) -> f a -> f b
21:33:17 <augur> <*> :: f (a -> b) -> f a -> f b
21:33:25 <pikhq> *Basically* it lets you apply the function in a functor to another functor. <*> :: f (a -> b) -> f a -> f b
21:34:12 <augur> <$> is sort of like liftA1 isnt it there
21:34:14 <AnMaster> pikhq, ouch that hurts to think about
21:34:47 <pikhq> AnMaster: Most obviously useful for applying a list of functions to a list. ;)
21:35:01 <AnMaster> pikhq, yes quite. Still somewhat mindbending
21:36:12 <AnMaster> bbl
21:36:17 -!- augur has quit (Read error: Connection reset by peer).
21:36:26 -!- augur has joined.
21:38:15 <pikhq> Also, curried functions.
21:38:26 <pikhq> (+)<$>[1..]<*>[2..]
21:39:22 <oerjan> i don't think that ever gets past the first 1
21:39:30 <coppro> yeah, that shouldn't work AFAICT
21:40:02 <pikhq> Baaah
21:40:05 <oerjan> the applicative instance for lists is cartesian product based, not zipping
21:40:05 <cpressey> One degree of gnarliness too far.
21:40:12 <oerjan> i.e. same as the monad
21:40:19 <pikhq> Right, right.
21:40:49 <pikhq> The zipping instance doesn't get you a monad, so that's not the default.
21:41:01 <oerjan> you could use ZipLists but that's awkward
21:41:18 <oerjan> actually it _does_ give a monad, it's just a different one
21:41:44 <pikhq> I thought it violated monad laws if you allowed for finite ziplists?
21:41:47 <oerjan> and it's not defined in the source
21:42:04 <pikhq> Regardless, it's not a ziplist because the monad instance isn't.
21:42:21 <pikhq> And inconsistent instances make me sad.
21:43:10 <oerjan> i believed i checked it once, and you can define it. join is essentially taking the diagonal, with some care (all the previous lists must be at least that length or you do get a law failure)
21:43:34 <oerjan> *believe
21:45:00 <pikhq> Okay, so you get a valid instance, with some restrictions.
21:45:17 <pikhq> Whereas the actually used instance has no such restrictions.
21:45:23 <oerjan> um i didn't mean restrictions
21:45:49 <pikhq> So I misunderstood.
21:45:59 <Madk> what other than the collatz and fibonacci sequences should I make examples for in my language?
21:46:06 <oerjan> i meant you need to take care when defining join that you cannot use a diagonal element unless all the "square" above and to the left are also defined
21:46:12 <pikhq> Ah.
21:46:21 <Madk> I'm thinking juggler, but then I have to figure out how to get a square root
21:46:28 <cpressey> Madk: Kolakoski sequence
21:46:38 <cpressey> It's cool
21:46:40 <cpressey> it's my fave
21:46:53 <oerjan> Madk: thue-morse
21:50:11 <Madk> preferably not a gigantic one :|
21:50:33 <Madk> also, wikipedia isn't clear on how the kolakoski sequence works
21:51:33 <oerjan> the nth number is the length of the nth block of equal digits. blocks alternate with 1s and 2s, and 1 starts the sequence
21:51:38 <coppro> 1, 11, 21, 1211, 111221, 312211, etc.
21:51:43 <oerjan> *the nth digit
21:52:03 <oerjan> coppro: um that's look-and-say
21:52:09 <coppro> yes, exactly
21:52:21 <oerjan> that's not kolakoski
21:52:32 <coppro> I didn't say it was
21:52:49 <oerjan> Madk: oh also factorial
21:56:40 <augur> google
21:56:48 <augur> 's new image search design is nice
22:00:46 -!- Madk has quit (Quit: Leaving).
22:03:12 -!- distant_figure has quit (Quit: underflow).
22:09:03 <cpressey> There are two things that seem "obvious" to me but for which no proof has been found. 1. P < NP 2. There are an equal number of 1's and 2's in the Kolakoski sequence.
22:10:13 <augur> cpressey: the wonderful thing about proofs is that seeming is irrelevant
22:11:01 <cpressey> augur: Indeed their power to dispel illusion is unmatched.
22:11:30 <augur> i await the day when we have a proof that P = NP
22:11:35 <augur> just so i can shove it in your face. :D
22:11:39 -!- augur has quit (Read error: Connection reset by peer).
22:11:49 -!- augur has joined.
22:12:06 <augur> just so i can shove it in your face
22:12:08 <augur> and be like
22:12:11 <augur> EXPEL THIS
22:12:11 -!- coppro has quit (Ping timeout: 240 seconds).
22:12:24 <cpressey> ... "expel"?
22:13:08 -!- DH____ has quit (Quit: Trillian (http://www.ceruleanstudios.com).
22:13:18 <augur> or whatever
22:14:43 <augur> eh
22:15:03 <augur> dispel sorry. i thought you had said expel (for god knows what reason) and my client died before i could get a good look
22:15:04 <augur> ahahaha
22:15:23 <augur> pretend everything i said made sense!
22:15:42 <cpressey> OK!
22:15:48 <augur> :D
22:15:56 <cpressey> I have a lot of practice doing that all the time at work anyway.
22:25:22 -!- coppro has joined.
22:36:41 -!- GreaseMonkey has joined.
22:37:15 <augur> oooh
22:37:18 <augur> brainpipe looks good
22:38:40 -!- sftp has quit (Remote host closed the connection).
22:41:23 * Sgeo needs a set of earphones that aren't so fragile that I will end up breaking the wire and being unable to listen with both ears
22:42:23 <pikhq> Sgeo: Get. Proper. Headphones.
22:42:38 <Sgeo> pikhq, I _had_ headphones. They broke.
22:42:54 <pikhq> That sounds improper.
22:42:57 <Sgeo> I think one of the wires broke or something
22:42:59 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~())~*^(a(:^)*~a(*()~)~*^~^):^
22:42:59 <fungot> 122 ...out of stack!
22:43:04 <oerjan> dammit
22:43:19 <Sgeo> Because now audio only comes out of one.. ear thing, unless I hold the thing in position just right
22:43:32 <pikhq> Pay money for ones with replacable wires. Or at *least* one with not-shitty wires.
22:44:10 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~())~*(b1)S^(b2)S(a(:^)*~a(*()~)~*^~^):^
22:44:10 <fungot> 12b12 ...out of stack!
22:44:46 <Sgeo> Something similar happened to my earbuds just today
22:44:50 <Sgeo> So I think it's me
22:45:06 <Sgeo> erm, yesterday
22:45:16 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~())~*aSaSaS
22:45:17 <fungot> 12(~()(2)S:**a(~:)~*^~)((2)S:**a(~:)~*^~)((1)S*a(~:)~*^~)
22:45:22 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~())~*aSaSaSaS
22:45:22 <fungot> 12(~()(2)S:**a(~:)~*^~)((2)S:**a(~:)~*^~)((1)S*a(~:)~*^~) ...out of stack!
22:46:13 <pikhq> Sgeo: About how thick would you say those wires are?
22:46:29 <oerjan> oh
22:46:29 <Sgeo> Which ones, the headphone ones or the earbud ones?
22:46:41 <pikhq> Both, I guess.
22:47:01 <Sgeo> The headphone wires look thick
22:47:07 <Sgeo> Earbuds, not so much
22:47:18 <Sgeo> Let me take a pic
22:47:22 <pikhq> Mmkay. Get ones with replacable cables.
22:47:51 <pikhq> (the attached headphones, BTW, will almost certainly be absolutely wonderful to listen to)
22:47:52 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~:()~)~*^(a(:^)*~a(*()~)~*^~^):^
22:47:54 <fungot> 122112122122112112212112122112112122122112122121121122122112122122112112122121122122112122122112112212112122122112112122112112212112112212211212212112212212112112212211212212112112212112122112112122121122122112122122112112122112112212212112122112112212112112212212112122112112122122112122121121122122121122122112122122112112 ...too much output!
22:48:01 <cpressey> Yessssss
22:48:08 <ais523> oerjan: what sequence is that?
22:48:18 <oerjan> kolakoski, assuming it's correct
22:48:46 <Sgeo> http://i.imgur.com/DVIRu.jpg
22:49:19 <pikhq> Sgeo: "image not found"
22:49:40 <Sgeo> pikhq, hm, I see
22:50:08 <Sgeo> I have no clue why, though
22:50:09 * oerjan notes it seems to match what's on wikipedia
22:50:12 <Sgeo> Weird font on the phone?
22:50:43 <Sgeo> http://i.imgur.com/DVlRu.jpg
22:51:02 <pikhq> There we go.
22:51:32 <pikhq> Definitely need replacable wiring.
22:51:51 <pikhq> Sad, though; noise cancelling headphones that can have a broken wire.
22:52:03 <Sgeo> I don't use the noise cancellation
22:52:05 * oerjan adds to wiki
22:52:06 <Sgeo> These were my dad's
22:52:35 <Sgeo> If I hold the wire just right, it works
22:52:45 <pikhq> Get you some Sennheiser phones.
22:53:27 * cpressey translates the poem from Greek into an archaic form of English so that it *still* doesn't rhyme.
22:54:09 <oerjan> cpressey: now you have even more reason to learn underload *evil cackle*
22:54:24 <cpressey> oerjan: Sigh, yeah
22:54:47 -!- bsmntbombdood_ has joined.
22:54:58 <Sgeo> Are wireless headphones any good?
22:55:11 -!- bsmntbombdood has quit (Ping timeout: 240 seconds).
22:55:27 <Sgeo> WTF at $1000 headphones
22:55:34 <oerjan> ais523: you're right, using a(...)~*^ is convenient :)
22:56:35 <pikhq> Sgeo: And you'll never have to buy them again.
22:56:35 <pikhq> :P
22:56:51 <Sgeo> I am not paying $1000 for headphones
22:57:32 <pikhq> Darnit, you're not made of money.
22:59:25 <Sgeo> http://www.sennheiserusa.com/private_headphones_audiophile-headphones_005341 is that decent?
23:00:04 <pikhq> If by "decent" you mean "as good as you can get without selling organs", then yes.
23:02:40 <Sgeo> My dad says he has headphones in the house
23:02:48 <Sgeo> And that he'll try something called a "relief fold"
23:10:23 <Gregor-P> Somebody in here mentioned a very long time ago a company that prints custom cards for card games ... any recollection?
23:11:54 * oerjan recalls zzo38 being involved in such a discussion
23:12:34 <oerjan> and it wasn't _that_ long ago, i think it was this year
23:12:58 <Gregor-P> For me, that's a lifetime ago.
23:13:16 <oerjan> for some reason i have a feeling pikhq was involved, too
23:13:29 <pikhq> No, but I'm now curious.
23:15:09 <Gregor-P> My mother-in-law is convinced that I need to adjust Hydra to involve some cusom cards, then produce it and try to pitch it to e.g. Mattel or Hasbro. Which is hilariously absurd, but I like step 1 :P
23:15:10 <oerjan> well that's as much as my vague memory contains. well that and that it was expensive to do.
23:16:11 <oerjan> hydra? is this anything to do with that goldstein sequence thing?
23:16:19 <cpressey> ha
23:16:29 <pikhq> Gregor-P: If by "Hasbro" she means "Wizards of the Coast", it might almost work.
23:16:32 <pikhq> :P
23:16:38 <cpressey> Yes, it's a card game that relies on transfinite induction.
23:17:01 <cpressey> Actually, I hope it is.
23:17:12 <oerjan> er, *goodstein
23:17:51 <Gregor-P> 'fraid not, folks
23:18:28 <oerjan> darn
23:19:15 -!- Gregor-W has joined.
23:19:24 <Gregor-W> Tired of typing on my phone :P
23:19:35 <Gregor-W> http://codu.org/wiki/Hydra <-- Hydra is this
23:21:41 <Gregor-W> (People seem to enjoy it because the rules are quite simple but, as the number of heads increases, it has a surprising amount of strategy involved)
23:22:31 <Sgeo> These aren't circumaural
23:23:08 -!- GreaseMonkey has quit (Quit: I'm using NO SCRIPT WHATSOEVER - Download it at file:///dev/null).
23:24:04 <oerjan> transaural, for airheads
23:26:02 -!- SimonRC has quit (Ping timeout: 260 seconds).
23:26:36 <Sgeo> http://i.imgur.com/Qb7ns.jpg
23:29:00 * Sgeo pokes pikhq
23:29:26 -!- tombom has quit (Quit: Leaving).
23:29:36 <pikhq> Sgeo: Insufficiently Sennheiser!
23:30:18 <Gregor-W> I see nobody has opinions on my card game :P
23:30:42 <Sgeo> Will these at least have decent sound quality?
23:31:38 -!- MigoMipo has quit (Quit: Quit).
23:33:29 -!- Madk has joined.
23:33:32 -!- _Madk has joined.
23:34:54 <coppro> Gregor-W: Seems
23:35:04 -!- SimonRC has joined.
23:39:45 <Sgeo> I think these are minimum-volume headphones
23:40:05 <oerjan> you need a very small head to use them
23:40:57 -!- relet has quit (Quit: Leaving.).
23:53:21 -!- SevenInchBread has quit (Ping timeout: 265 seconds).
23:55:55 <ais523> oerjan: heh, why did you put the a there?
23:56:02 <ais523> I suppose it works the same way there, and after the ~
23:56:06 <ais523> Underload is a rather redundant language
23:57:48 <oerjan> wait where?
23:58:09 <oerjan> ^ul (12)S(*a(~:)~*^~):((1)S)~*~((2)S:*)~*:(~:()~)~*^(a(:^)*~a(*()~)~*^~^):^
23:58:10 <fungot> 122112122122112112212112122112112122122112122121121122122112122122112112122121122122112122122112112212112122122112112122112112212112112212211212212112212212112112212211212212112112212112122112112122121122122112122122112112122112112212212112122112112212112112212212112122112112122122112122121121122122121122122112122122112112 ...too much output!
23:58:11 -!- wareya_ has quit (Read error: Connection reset by peer).
23:58:34 <oerjan> oh right
23:59:12 -!- wareya has joined.
23:59:20 <oerjan> i guess i thought of ~* as more of a unit
←2010-07-19 2010-07-20 2010-07-21→ ↑2010 ↑all