00:03:02 -!- poiuy_qwert has joined.
00:03:49 <olsner> write-only means no debugging :)
00:05:02 <tusho> Tritonio_: you don't
00:05:54 <olsner> some things really are easier to get right from the start than debug post-facto
00:06:27 <tusho> how to write a brainfuck program:
00:06:33 <tusho> 2. formalise basic idea of how it could work in head
00:06:40 <tusho> 3. figure out how you can organize it into a full program
00:06:54 <tusho> 5. if the program made by steps 4 fails, go to 3 or maybe 2
00:07:17 <olsner> basically 1. read problem, 2. think really hard, 3. write down solution
00:07:39 <tusho> olsner: so brainfuck = haskell?
00:08:06 <olsner> the difference lies in the "write down solution" step
00:08:35 <olsner> in haskell you get three lines of beauty, in brainfuck you get a steaming heap of brainfuck
00:10:51 <olsner> besides, that's not really about haskell... I think it's a feynman quote
00:11:49 <Tritonio_> well i prefer the normal way... 1 think of the problem. 2 Write down something you think it works. 3 check it for stupid mistakes. 4. put it in a debbuger and fix the rest of the stupid mistakes...
00:12:35 <olsner> tusho: http://en.wikiquote.org/wiki/Richard_Feynman#Quotations_about_Feynman
00:12:52 <Tritonio_> and for some reason, when run through wine, it still has windows decorations thus it doesn't blend with compiz nicely...
00:12:52 <tusho> olsner: yes, my joke is still valid
00:13:06 <tusho> [00:11:51] <Tritonio_> well i prefer the normal way... 1 think of the problem. 2 Write down something you think it works. 3 check it for stupid mistakes. 4. put it in a debbuger and fix the rest of the stupid mistakes...
00:13:09 <tusho> awful way to write BF programs
00:13:20 <tusho> it'll produce inferior programs & algorithms and take a lot longer to develop
00:13:59 <tusho> Tritonio_: don't you want to produce better BF programs?
00:14:04 <tusho> Besides, you can't do your current way on linux.
00:14:08 <tusho> Why not take the oppertunity to improve?
00:14:41 <Tritonio_> how would they be better? when I write the code usually it has only some spare or missing + or -
00:15:53 <tusho> Tritonio_: because you think exactly how to write the program
00:16:17 <tusho> Tritonio_: hopefully, you should have one comment per each runnable-in-your-head "unit" of code
00:16:26 <tusho> thus, if one of them has an extra/missing + or -, it'll be obvious
00:18:25 <Tritonio_> Well usually *I* miss some of the obvious mistakes... So I really need the debugger. If there is not one for linux I will make one.
00:19:32 <tusho> Tritonio_: i just explained how not to miss them
00:19:37 <Tritonio_> Normally I don't make huge mistakes, I just miss many small ones...
00:19:46 <tusho> when you write it out, you have one line being one unit of code (unit: a snippit you can run entirely in your head without problems)
00:19:47 <tusho> and comment what it's for
00:20:00 <tusho> and check that each unit does what its comment says it should
00:22:45 -!- jix has quit (Read error: 104 (Connection reset by peer)).
00:23:14 <fizzie> If your current Windows debugger is bfdebug.sf.net, it might work with very little tweaking under Linux with Lazarus.
00:23:22 -!- jix has joined.
00:23:57 <fizzie> (And I've seen at least two Javascript interpreters with at least some debugging facilities, which of course will run anywhere.)
00:24:21 <tusho> fizzie: do you agree that my method is an easier way to produce bug-free BF though?
00:25:01 <fizzie> Well, I'm not completely convinced a debugger is completely useless. Still, your method is what I've used with the few BF programs I've written.
00:25:43 <Tritonio_> i am using. brainfuck developer fizzie. it's closed source freeware...
00:25:47 <fizzie> Even with all the comments, I think there has been at least one case of a non-obvious missing "<", though.
00:26:10 <Tritonio_> especially when you work with loops with unbalanced <>
00:26:29 <fizzie> Well, there's always Wine, if you're so inclined.
00:26:30 * tusho shrugs the key is making the chunks small enough
00:26:35 <tusho> say, a chunk over 10 characters is wrong
00:27:13 <Tritonio_> that what I do now. I run it with Wine. but it doesn't come up with the rest of the windows when I use the Scale plugin of compiz or whatever it is called...
00:27:32 <fizzie> Yes, well, small chunks don't help all that much if your problem is that there's a "<" chunk missing from a multi-page [].
00:29:18 * tusho shrugs - i think they help fine because you see each one both in isolation and in the bigger picture
00:29:26 <tusho> and if the comments are sufficient you should know something's up
00:30:41 -!- calamari has joined.
00:40:34 <tusho> optbot: say hi to calamari
00:40:34 <optbot> tusho: is just added hassle.
00:40:40 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | Hum ho-hum..
00:41:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | there are few viruses that can reproduce without destroying their host, though...
00:47:16 <tusho> calamari: (optbot is here to keep the topic fresh. he has a conversation mode because, well, because.)
00:47:16 <optbot> tusho: because ruby has to lookup the + method of Integer with every call because one can redefine the + method of Integer
01:39:17 <tusho> psygnisfive: enjoy your 2 weeks of pain
01:43:16 <psygnisfive> if you could see my fetlife profile youd know im a masochist :P
01:45:22 <psygnisfive> but hey whats life without some adventure ey?
01:47:32 -!- tusho has quit ("Leaving...").
01:48:13 -!- calamari has left (?).
02:32:47 -!- dogface has joined.
02:33:28 <dogface> I declare that the date is... oh, October -200, 1993.
02:33:47 <dogface> Actually, let's be more conservative.
02:37:26 <dogface> Tentative date that school's out...
02:39:29 <dogface> ...is June 5. Add a week and we get June 12, 2009, so if we call that October 1, 1993, today is (/me waits for his computer to unfreeze)
02:39:46 <dogface> The date June 12, 2009 is completely arbitrary, by the way.
02:41:51 <dogface> So let's say this is October -300, 1993.
02:42:04 <dogface> And that was arbitrary. Have a nice day.
02:44:27 <dogface> Add a leap day 97 times every 146097 days.
02:44:43 <dogface> Including the leap days, that is.
02:45:57 <dogface> Wait, doing evil things would actually be pretty boring.
02:46:13 <dogface> Hmm, unless I do it that way...
02:46:32 <dogface> Except there is no such way, which is unfortunate..
02:50:09 <dogface> Hmm. phi*sqrt(5) = 2+phi. This sounds like it could be exploited to yield a set of "Fibonacci numbers" that converge to sqrt(2).
02:53:49 <dogface> Cool, the generating matrix [[1,1][1,0]] has the eigenvalues phi and 1-phi. So what I want is a matrix of integers with eigenvalue sqrt(2).
03:14:27 -!- Sgeo has joined.
03:27:00 -!- jix has quit ("This computer has gone to sleep").
04:01:22 -!- CakeProphet has quit (Remote closed the connection).
04:45:07 <dogface> I am doing Evil Tuffs. Go me.
05:05:17 -!- GreaseMonkey has joined.
05:22:32 -!- dogface has quit (Read error: 110 (Connection timed out)).
05:27:38 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
05:41:38 -!- GreaseMonkey has quit ("lost connection").
05:44:00 -!- GreaseMonkey has joined.
05:51:56 -!- Corun has quit ("This computer has gone to sleep").
06:34:10 -!- Judofyr has quit (Read error: 104 (Connection reset by peer)).
06:34:31 -!- Judofyr has joined.
06:41:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | if you write a program in JavaScript does that make a web browser an OS?.
06:42:58 -!- GreaseMonkey has quit ("You're still in denial. It's easier to lie by saying that I'm lying than to face truth.").
06:49:45 -!- poiuy_qwert has quit.
07:06:02 -!- pikhq has quit ("leaving").
07:26:37 -!- Judofyr_ has joined.
07:26:37 -!- Judofyr has quit (Read error: 104 (Connection reset by peer)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:07:56 -!- kar8nga has joined.
09:01:11 -!- moozilla has quit (Read error: 104 (Connection reset by peer)).
09:03:56 -!- moozilla has joined.
09:10:35 -!- Judofyr has joined.
09:10:35 -!- Judofyr_ has quit (Read error: 104 (Connection reset by peer)).
09:11:03 -!- GreaseMonkey has joined.
09:33:13 -!- Judofyr has quit.
10:13:34 -!- Mony has joined.
10:35:01 <Mony> http://gbatemp.net/index.php?s=fa19c87f3dac3a0ff746e073da8daa1b&showtopic=99804
10:36:31 <fizzie> Well, choose another language. But don't do Befunge, otherwise I won't have anything to do.
10:38:41 -!- GreaseMonkey has quit ("You're still in denial. It's easier to lie by saying that I'm lying than to face truth.").
10:45:10 -!- kar8nga has quit ("Leaving.").
10:45:19 <fizzie> It's certainly easier to implement than to program in.
11:29:54 -!- tusho has joined.
11:30:46 <optbot> tusho: http://www.encyclopediadramatica.com/%E0%B2%A0_%E0%B2%A0
11:32:21 -!- KingOfKarlsruhe has joined.
11:37:06 <Deewiant> AnMaster: took a look now and it seems that you indeed haven't changed frontend-protocol based on my comments back on 2008-03-29
11:37:23 <AnMaster> Deewiant, what were those comments
11:37:29 <oklopol> is ds the one with a touchscreen?
11:38:01 <oklopol> . and , should be plotting and reading pixels then
11:38:04 <AnMaster> Deewiant, yes but how long ago?
11:38:05 <Deewiant> AnMaster: hmm, it seems that you had a second draft but it's not published
11:38:06 <Deewiant> 2008-03-31 17:33:12 * AnMaster looks for the first (now outdated) draft
11:38:06 <Deewiant> 2008-03-31 17:33:16( AnMaster) the second isn't yet redy
11:38:13 <AnMaster> Deewiant, I archive old on cds
11:38:13 <Deewiant> AnMaster: 2008-03-29, like I said
11:38:49 <Deewiant> AnMaster: but if you've got that second draft somewhere it might have the stuff in it
11:38:52 <tusho> use the tunes.org logs
11:38:57 <tusho> they're in the bloody topic
11:39:04 <tusho> http://tunes.org/~nef/logs/esoteric/08.03.29
11:39:24 <tusho> Deewiant: maybe he's afraid they've been tampered with
11:39:28 <tusho> how can he trust them?!
11:39:36 <optbot> Deewiant: I could say "the process providing the 'fs' service" or "the process that's an instance of /foo/bar" or "the process that's using all the CPU time".
11:39:49 <tusho> Deewiant: he's saying what the editor might make him say
11:39:51 <Deewiant> I think he's been tampered with too
11:39:51 <tusho> in the edited version
11:40:13 <fizzie> oklopol; One could also do the "bfvga" thing and map the brainfuck memory array to the 256x192-pixel non-touchscreen display.
11:40:52 -!- lament has quit (Read error: 110 (Connection timed out)).
11:40:55 <AnMaster> <Deewiant> 2008-03-31 17:33:16( AnMaster) the second isn't yet redy <-- not in said log file that tusho posted
11:41:16 <Deewiant> AnMaster: you do realize that the thing at the beginning of the line is a date/timestamp
11:41:17 <tusho> ur gud @ reading comprehension
11:41:32 <tusho> here let me babysit you. http://tunes.org/~nef/logs/esoteric/08.03.31
11:41:37 <Deewiant> and like tusho said, 29 and 31 aren't the same day. ;-)
11:42:14 <AnMaster> Deewiant, I'm still kind of sleepy, yes I know it isn't morning any more but that still doesn't help :P
11:42:40 <tusho> why are you writing a language spec
11:42:46 <tusho> if you can't make out the difference between "29" and "31"
11:42:59 <tusho> i wasn't being funny
11:43:03 <tusho> generally a clear mind would be reccomended...
11:43:07 <Deewiant> tusho: I gather he's not writing it right now
11:43:17 <tusho> Deewiant: but he's trying to look up stuff for it
11:43:41 <Deewiant> well, it's one way to clear one's mind :-P
11:44:42 <AnMaster> Deewiant, oh the lyx file is more uptodate it seems
11:45:22 <AnMaster> Deewiant, I don't see your comments on what the issues where in that log
11:45:26 <Deewiant> if it hasn't changed since 2008-03-31 I guess it's still a work in progress
11:45:44 <AnMaster> Deewiant, it has, the local one differs in page count at least
11:45:45 <Deewiant> AnMaster: starting around 07:04:51
11:46:10 <AnMaster> what timezone are the tunes.org logs in?
11:47:21 <tusho> AnMaster: just search for deewiant
11:47:32 <AnMaster> tusho, there are lots of other funge stuff there too
11:47:41 <AnMaster> it seems I were implementing FPSP at that time
11:47:46 <Deewiant> tusho: EST can refer to 5 different time zones...
11:47:55 <tusho> Deewiant: yes. one of them
11:47:56 <Deewiant> but I guess you meant the american one?
11:48:06 <AnMaster> Deewiant, can you give me some comment to search for please
11:48:40 <AnMaster> Deewiant, there is nothing around then in the http://tunes.org/~nef/logs/esoteric/08.03.31, at that time I haven't even posted the link to that draft
11:48:57 <Deewiant> AnMaster: 31 is just you saying that there exists a second draft
11:49:17 <Deewiant> 29 is when you posted the first
11:52:58 <AnMaster> Deewiant, it seems the local one is more uptodate
11:54:43 -!- jix has joined.
11:55:36 <tusho> and I want to use the "M" from each one in a bit of code
11:56:06 <AnMaster> tusho, you could use FING to remap one of the M to something else, that you don't need
11:56:09 <Deewiant> tusho: you use the FING or FNGR or IMAP fingerprints
11:56:22 <tusho> Deewiant: what if I need A-Z for both?
11:56:24 <AnMaster> tusho, or you could just load and unload them
11:56:31 <tusho> just do "2GNF"(M) each time I want it?
11:56:33 <AnMaster> tusho, then you need to switch between them a lot
11:56:48 <tusho> "1GNF"(M"2GNF"(M)M)
11:56:58 <oklopol> fingerprints should be implemented as befunge modules
11:57:02 <AnMaster> tusho, depends on what M does to the stack
11:57:12 <Deewiant> oklopol: they can be, it's called mini-funge or fungelib
11:57:47 <tusho> there should be a FGPT fingerprint
11:57:55 <tusho> that lets you make a fingerprint without any special formats even with befunge code
11:58:07 <tusho> N ( name -- ) set the name of the new one
11:58:19 <tusho> C ( 0"gnirts" name -- ) add command
11:58:22 <tusho> S ( -- ) save fingerprint
11:58:33 <tusho> AnMaster: lets you make a fingerprint in befunge code
11:58:35 <tusho> so that you can do
11:58:41 <AnMaster> tusho, sounds like MACR kind of
11:58:44 <tusho> "FGPT"(...)"LOOC"(...)
11:58:52 <tusho> where you used the FGPT bit to define COOL
11:59:27 <AnMaster> anyway: dynafing/mini-funge/fungelib doesn't always work...
11:59:36 <AnMaster> you couldn't implement FILE using that
12:00:38 <tusho> AnMaster: but the thing is, all of those don't let you run them from in befunge code
12:00:46 <tusho> this would just be a regular fingerprint that lets you make a new fingerprint
12:01:12 <AnMaster> tusho, anyway this sounds almost like MACR
12:01:28 <AnMaster> you may want to take a look at the new MACR in RC/Funge
12:01:43 <AnMaster> but probably useful for you anyway
12:02:06 <oklopol> how many fingerprints are there?
12:02:13 <oklopol> if it's more than 20, i'll make one more
12:02:17 <tusho> oklopol: like 10-20
12:02:29 <tusho> oklopol: they're implementation specific though
12:02:35 <tusho> oh you mean just spec it
12:02:41 <AnMaster> but no implementation have them all
12:02:43 <tusho> and let AnMaster and Deewiant and MikeRiley implement it?
12:03:06 <oklopol> tusho: yes just spec it, or alternatively use a fingerprint-generating fingerprint to make it in befunge
12:03:33 <tusho> oklopol: it's more fun just to spec it
12:03:36 <tusho> and make it a little vague
12:03:38 <AnMaster> tusho, for example only ais' addition to cfunge got IFFI
12:03:46 <tusho> then you get to watch AnMaster, Deewiant and MikeRiley argue aebout it
12:03:53 <oklopol> tusho: not necessarily, i love coding in befunge
12:03:59 <tusho> and maybe get it in to mycology, if Deewiant isn't lazy
12:04:00 <AnMaster> tusho, well since oklopol made it we will argue with him
12:04:09 <tusho> oklopol: solution - spec the fingerprint, then write programs using it
12:04:22 <AnMaster> anyway I may or may not implement it
12:04:37 <tusho> oklopol: make sure to use it to write something amazing
12:04:42 <tusho> so that they just have to implement it
12:04:43 <oklopol> and i can have all kinds of interesting side-effects then.
12:05:02 <AnMaster> tusho, if he wrote a good and *fast* ray tracer in Befunge....
12:05:10 <AnMaster> that would be about the only thing ;P
12:05:23 <oklopol> for instance, i doubt s/.../.../ for the code would be possible if you wrote it in befunge
12:05:41 <AnMaster> replace any 3 chars with 3 dots?
12:05:51 <oklopol> replace something in code with something else
12:05:58 <oklopol> for all instances of something
12:06:00 <AnMaster> oklopol, see REXP for regex, but I don't think it does replacement yet
12:06:05 <tusho> AnMaster: no, in the program
12:06:09 <tusho> he wants to be able to run a regexp on the program
12:06:10 <tusho> from inside the program
12:06:14 <AnMaster> oh you mean search and replace on funge space
12:06:35 <AnMaster> but would be quite complex I think
12:06:36 <oklopol> as something that would be easy to spec, but hard to write as a befunge module.
12:07:02 <tusho> AnMaster: no, no, he wants you to implement it see ;)
12:07:03 <AnMaster> oklopol, it *could* be done in dynafing/mini-funge
12:07:18 <AnMaster> tusho, cfunge doesn't do dynafing or such yet
12:07:34 <oklopol> AnMaster: probably possible, although i don't know how you'd do it
12:08:12 <AnMaster> oklopol, well.... dynafing is appendix C in http://rage.kuonet.org/~anmaster/funge-108/funge108.pdf
12:08:25 <tusho> oklopol: don't use dynafing
12:08:28 <tusho> it's just AnMaster's toy invention
12:08:33 <tusho> RC-Funge has something similar that is actually implemented
12:08:47 <tusho> AnMaster: fine, well, !Befunge is kind of useless
12:08:59 <AnMaster> tusho, it is more powerful than mini-funge
12:08:59 <tusho> 1. broken 2. you basically can't run it on modern systems
12:09:07 <tusho> AnMaster: and more powerful == less fun to do stuff in
12:09:37 <oklopol> of course it would be trivial if it, for instance, kinda temporarily pastes the whole fingerprint source where the instruction was, and you'd have G and P for accessing the fingerprint code, and g and p for the original code
12:09:44 <oklopol> and X for exiting the fingerprint
12:09:48 <AnMaster> mini-funge is almost a subset of dynafing anyway
12:10:09 <AnMaster> the different is G/P and g/p have interchanged meanings
12:10:24 <AnMaster> Feature Dynafing 1.3 RC/Funge’s mini-funge
12:10:24 <AnMaster> Instruction to get and set in G and P g and p
12:10:24 <AnMaster> Instruction to get and set in g and p G and P
12:10:34 <oklopol> i'm actually not sure how it'd work to call shit in the fingerprint itself.
12:10:49 <oklopol> but anyway it's either trivial or impossible, i think.
12:11:05 <AnMaster> oklopol, write the specs first
12:11:08 <oklopol> but anyway, there will always be stuff you can't write as a befunge module
12:11:12 <AnMaster> then I'll see what I think of it
12:11:15 <oklopol> AnMaster: i'm not going to make s///
12:11:23 <AnMaster> oklopol, indeed there will be stuff
12:11:30 <oklopol> "there will always be stuff you can't write as a befunge module" is my point
12:11:54 <AnMaster> oklopol, TRDS: time travel in funge space for IPs
12:13:08 <AnMaster> maybe I should make two fingerprints that only differ in case
12:13:19 <AnMaster> would be fun to see what Deewiant would say ;P
12:13:36 <AnMaster> "ARGH CAN'T MAKE THE SOURCE CODE FILES AS WINDOWS IS CASE INSENSITIVE!"
12:15:30 <Deewiant> or maybe it'll be so full of posix-only features that I can't implement it anyway ;-P
12:16:06 <tusho> he'd just call it dirf_lower
12:16:07 <AnMaster> Deewiant, could ccbi handle weirdly named fingerprints?
12:16:32 <tusho> Deewiant: probably
12:16:35 <tusho> he'd just not name the file that
12:17:08 <AnMaster> cfunge's fingerprint.h generation scripts would need some changes in order to support it
12:17:58 <Deewiant> CCBI requires no changes for such
12:18:16 <AnMaster> Deewiant, could CCBI handle a fingerprint 0?
12:18:36 <Deewiant> I debated about making a fingerprint 0 just for Mycology
12:18:49 <AnMaster> Deewiant, well cfunge will never try to load 0
12:18:57 <AnMaster> it just use it as a end of list marker
12:18:59 <Deewiant> yes, and hence cfunge doesn't conform to the spec.
12:19:15 <AnMaster> Deewiant, it just don't implement 0
12:19:27 <AnMaster> Deewiant, loading AAAA will reflect as well as loading 0
12:19:45 <AnMaster> so it totally conforms to the specs
12:20:24 <AnMaster> Deewiant, I can't see how you end up with "doesn't conform" here
12:20:25 <Deewiant> maybe I should make 0 a combination of useful stuff from various fingerprints
12:20:45 <Deewiant> just so that you can start a befunge program with ( and have A-Z map to handy things
12:20:54 <AnMaster> Deewiant, then it would be a 3 line change to use sizeof and some stuff to get the size of the array instead
12:21:09 <Deewiant> so why don't you change it :-)
12:21:14 <AnMaster> + a few lines of change to generation of fingerprint.h script
12:21:28 <AnMaster> Deewiant, because it works, and because I don't need it yet
12:21:42 <AnMaster> as I said, I could change it easily enough if I needed to
12:22:14 <Deewiant> hmm, filenames can't contain a null byte can they
12:22:31 <AnMaster> indeed that would be another change, in that case to the generation scripts
12:22:39 <AnMaster> Deewiant, you could call the file null.d?
12:23:09 <AnMaster> Deewiant, even / is valid in filenames I think
12:23:29 <Deewiant> I think NUL and / are the only two disallowed in *nix
12:23:36 <Deewiant> in windows, some others as wel
12:23:45 <AnMaster> Deewiant, well "who cares about windows" *ducks*
12:24:12 <Deewiant> this means that mini-funge can't load fingerprints that contain NUL
12:24:25 <AnMaster> Deewiant, well the Funge-108 variant of mini-funge could
12:24:39 <AnMaster> Deewiant, file name doesn't matter there
12:24:56 <AnMaster> ==U:<URI> URI ==U:http://example.com/test
12:24:56 <AnMaster> ==F:<string> Legacy fingerprint string =F:TEST
12:25:42 <AnMaster> Deewiant, see how useful that is? :)
12:25:52 <tusho> AnMaster: for legacy names
12:26:03 <Deewiant> yeah, if we ever make more than 2^32 fingerprints
12:26:15 <AnMaster> Deewiant, no I mean that way to declare the names
12:26:15 <tusho> AnMaster: funge-fingerprint-legacy://TRDS
12:26:20 <tusho> just use an unregistered URI secheme
12:26:23 <tusho> technically they're not valid URIs
12:26:24 <AnMaster> Deewiant, allows all those chars you mentioned
12:26:28 <tusho> but in practice it'd be a lot easier
12:26:31 <Deewiant> AnMaster: yeah, exactly, useful if we ever run out of fingerprint names
12:26:32 <tusho> (than supporting both)
12:26:59 <AnMaster> tusho, it is rather easy anyway: "if loading it as uri fails, fall back to old loading scheme"
12:27:18 <AnMaster> Deewiant, and to avoid name collisions
12:27:22 <tusho> AnMaster: yes, but for defining it
12:28:11 <Deewiant> AnMaster: 0 can be defined anyway, just not doable in minifunge :-)
12:28:28 <tusho> Deewiant: here's how you should define 0 for mycology
12:28:33 <AnMaster> Deewiant, now it could be done in mini-funge
12:28:37 <tusho> A-Z are all different parts of the test
12:28:44 <tusho> and when you wanna test 0 in mycology
12:28:51 <AnMaster> tusho, you still don't have to implement 0
12:28:53 <tusho> 10(ABCDEFGHIJKLMNOPQRSTUVWXYZ)
12:29:09 <tusho> AnMaster: 10 = 1"\0"
12:29:27 <AnMaster> tusho, I was confused there :P
12:29:47 <Deewiant> you don't even need 01, just 0
12:29:48 <AnMaster> actually do you test that Deewiant, loading 0(
12:30:16 <Deewiant> AnMaster: I think that's what I use to test ( and )
12:30:38 <AnMaster> Deewiant, well those tests would need to be rewritten if someone implements 0
12:30:48 <tusho> AnMaster: you know what you must now do ;)
12:30:53 <tusho> find a bug in mycology! go implement 0
12:30:59 <Deewiant> AnMaster: yes, because there is no resereved unimplemented fingerprint
12:31:16 <AnMaster> Deewiant, wait you could make one, define it as *never implemented*
12:31:23 <tusho> AnMaster: make it print out "AnMaster 0wned Mycology!!! greetz to mikeriley"
12:31:26 <tusho> and exit the program
12:31:47 <tusho> AnMaster: itym "very funny"
12:32:43 <fizzie> Mostly used in the construction "ITYM 'foo' HTH HAND".
12:33:06 <fizzie> Happy to help, have a nice day.
12:33:37 <tusho> I use it sincerely
12:33:40 <tusho> and most people I know do
12:34:02 <tusho> i was pointing out that anmaster probably meant "very funny"
12:35:48 <AnMaster> hm maybe define UDEF as "this fingerprint is never implemented"
12:36:18 <AnMaster> then require implementations to support UDEF for mycology to be able to finish testing
12:36:25 <Deewiant> "N/A" - "this fingerprint is not available"
12:36:46 <tusho> {Kim Cameron, Microsoft’s chief architect of identity, is an enthusiastic advocate of information cards, which are not only vastly more secure than a password-based security system, but are also customizable, permitting users to limit what information is released to particular sites. “I don’t like Single Sign-On,” Mr. Cameron said. “I don’t believe in Single Sign-On.” }
12:36:48 <Deewiant> supporting it means to not support it, so it's not exactly a requirement :-P
12:36:54 <tusho> (from a stupid new york times article about openid via reddit)
12:37:03 <tusho> you can limit what sites can get info from your openid too!
12:39:28 <AnMaster> hm writing a funge interpreter in a purely functional language would be hard I think
12:39:50 <AnMaster> Deewiant, fingerprints need to modify global state
12:39:53 <Deewiant> depends on the language itself, of course :-P
12:40:20 <Deewiant> erlang isn't purely functional, is it?
12:40:26 <AnMaster> I guess haskell too, but I don't know haskell at all, I do know some erlang
12:40:36 <AnMaster> Deewiant, almost purely functional I think
12:40:57 <AnMaster> for example you can only "bind" a variable once
12:41:07 <AnMaster> which makes it hard to do a global state
12:41:14 <Deewiant> in any case, you can just pass around whatever it is you want to "modify"
12:41:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | !1 2%%3 4...
12:41:32 <Deewiant> in haskell, this is made somewhat easier by things like the State monad
12:41:52 <Deewiant> in erlang, I'd imagine it's easier for some other reason because it's not pure ;-)
12:45:10 <AnMaster> Deewiant, I don't "get" monads
12:45:39 <Deewiant> it's easier to "get" individual monads, like Maybe and State
12:45:50 <Deewiant> and then later "get" what they have in common
12:46:09 <AnMaster> well I'm learning erlang currently
12:52:54 <tusho> AnMaster nobody gets monads
12:52:56 <tusho> apart from like 5 people
12:53:02 <tusho> (i am one of them)
12:55:11 <oklopol> monads mean that you like have IO but it's not like side-effect!
12:55:33 <tusho> oklopol: not really
12:55:48 <oklopol> what was that sarcasm-tag again?
12:55:54 <tusho> oklopol: you missed mine, I see
12:56:01 <tusho> anyway I am an expert on NOMADS
12:56:08 <tusho> if anyone has any questions about NOMADS!!!!!!!!!!!!!!!!!!!!!!! I can answer them
12:56:24 <oklopol> tusho: i did miss it, where is it?
12:56:33 <tusho> oklopol: after 'not really'
12:56:58 <oklopol> or have i been nnscripted..?
12:57:55 <tusho> AnMaster: they are NOMADS
12:57:58 -!- Judofyr has joined.
12:57:58 <tusho> they have the following operations
12:58:13 <tusho> RTEURN :: M A -> A
12:58:19 -!- jix has quit ("This computer has gone to sleep").
12:58:23 <tusho> BIDN :: A -> (M A -> B) -> B
12:58:26 <AnMaster> tusho, are they in any way related to monads?
12:58:29 <tusho> FIAL :: M STRING -> A
12:58:32 <tusho> they are NOMAAAAAAAAAAADS
12:58:55 <oklopol> AnMaster: no those are for monads
12:59:03 <AnMaster> oklopol, oh? I don't get monads
12:59:44 <tusho> AnMaster: they are NOMADSSSSSSS
13:00:06 <oklopol> basically a monad is kind of a container with a few operations for sequencing computation, and optionally additional operations for tinkering with its state.
13:00:40 <AnMaster> oklopol, I see, so how do you manage to do write/read/seek on a file stream in haskell?
13:00:58 <oklopol> AnMaster: IO is not necessarily the best way to get into monads as a concept
13:01:17 <tusho> AnMaster: i think you'd have to grasp functional programming more to get it
13:01:23 <tusho> it's not easy to explain to a c programmer.
13:01:29 <oklopol> the state there would be the world's state
13:01:37 <AnMaster> tusho, I do know some erlang now, and also some elisp
13:01:41 <tusho> AnMaster: still. :)
13:01:45 <oklopol> you have a few operations for playing with the state, like reading files and writing to the printer
13:01:56 <tusho> oklopol: i think we should wait a bit
13:02:05 <tusho> it'll be easier once he's learned erlang
13:02:12 <AnMaster> if I'd write a erlang program I would write it in C in my head, then convert to erlang
13:02:13 <tusho> and how to do state in that
13:02:23 <Deewiant> AnMaster: that's probably not a good idea
13:02:24 <oklopol> well he definitely shouldn't do like me and not touch haskell at all and just read tutorials on monads :D
13:02:24 <tusho> AnMaster: yeah ... don't write any erlang programs ;)
13:02:30 <Deewiant> I expect the code would be quite unidiomatic
13:02:36 <tusho> Deewiant: 'quite'?
13:02:37 <AnMaster> but I'm still learning erlang :P
13:02:41 <tusho> AnMaster would be a human compiler
13:02:47 <tusho> one that makes everything slower to boot
13:02:55 <Deewiant> tusho: well, I don't know Erlang so I don't know what idiomatic Erlang should look like :-)
13:03:28 <AnMaster> when I began with C I was thinking in bash, when I began with bash I was thinking in pascal
13:03:40 <AnMaster> the first few weeks the code doesn't look good at all of course
13:03:47 <Deewiant> so actually you've been thinking in the same language all the time, you just changed the syntax ;-)
13:04:04 <AnMaster> Deewiant, you *learn* to "think" in the language
13:04:04 <tusho> Deewiant is pretty much right
13:04:13 <tusho> AnMaster: i'm afraid bash, C and pascal are basically the same
13:04:20 <tusho> single paradigm, not that many differences
13:04:27 <tusho> so you really only know one language :P
13:04:40 -!- RedDak has joined.
13:04:41 <tusho> heh, funny to think that AnMaster would use a propietary language on a propietary system
13:04:50 <AnMaster> tusho, I used Mac OS 7 ages ago
13:05:05 <tusho> Deewiant: i can answer that
13:05:06 <AnMaster> Deewiant, no, but I have read about it, it smalltalk looks very interesting
13:05:15 <AnMaster> but I can only do one thing at a time
13:05:35 <AnMaster> tusho, anyway I also coded C#, so yes I can think in object orientation as well
13:05:46 <tusho> C# and object pascal aren't real oo AnMaster :)
13:05:57 <tusho> they break several core principles
13:05:59 <Deewiant> they're mostly imperative with OO on the top
13:06:10 <tusho> smalltalk's object model obeys it because it invented it :P
13:06:15 <tusho> C++ is ... no way OOP
13:06:32 <tusho> obj-c is very close to real OO
13:06:36 <tusho> because it's basically C with smalltalk on top
13:06:49 <tusho> it has some funky @ and - syntax for declaring classes
13:06:55 <tusho> but the actual message sends?
13:07:06 <tusho> [obj someMsg:YES withX:[obj2 test:NO]]
13:07:08 <AnMaster> tusho, anyway I also know befunge which is a different programming model yet
13:07:16 <AnMaster> and I think in befunge when coding it
13:07:18 <tusho> befunge is still imperative though
13:07:19 <Deewiant> no, Befunge is also imperative :-P
13:07:30 <AnMaster> but it is not like C at all still :P
13:07:31 <Deewiant> and you don't even write self-modifying code
13:08:29 <Deewiant> so... not at all self modifying :-P
13:08:42 <AnMaster> Deewiant, true, but most time you don't need it
13:08:55 <AnMaster> Deewiant, how would you write that code using self modification in some constructive way?
13:09:18 <AnMaster> wait I can see one way, write the value to a cell then put a ' in front
13:10:11 <Deewiant> but yeah, I was just saying that Befunge is really imperative, if your code is very self-modifying then it might be considered as somewhat of a a different mindset but probably not :-)
13:11:18 <AnMaster> Deewiant, anyway erlang is a totally different mindset
13:11:40 <AnMaster> and look at my def-c low level compiler, it is not classic C
13:11:44 <Deewiant> quite, if you use it correctly ;-)
13:12:03 <AnMaster> def-c, the def-bf compiler that is
13:12:15 <AnMaster> Deewiant, it uses recursion a lot
13:12:29 <AnMaster> not functional, but not classical C etierh
13:12:43 <Deewiant> recursion in itself doesn't mean it's not somehow normal C :-)
13:13:00 <AnMaster> Deewiant, recursion isn't very common in C programs
13:13:16 <Deewiant> yeah, because C programmers generally can't think recursively
13:13:23 <AnMaster> I bet 99% of the functions in a normal C program aren't recursive
13:13:27 <Deewiant> so they mash something into a really complicated loop when recursion would serve them better
13:13:42 <AnMaster> Deewiant, but my Brainfuck parser and optimizer are both recursive
13:14:06 <Deewiant> which is because it makes more sense to do so than to loop :-P
13:14:13 <Deewiant> not because you're writing in a non-C style
13:14:25 <Deewiant> if you wanted to loop, you'd have to manage your own stack
13:14:31 <AnMaster> Deewiant, my BASE.c binary() is recursive
13:14:35 <Deewiant> so it'd be just like recursion anyway
13:24:39 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | 3PLN = $1 ;d.
13:24:43 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | argh.
14:10:30 -!- RedDak has quit (Remote closed the connection).
14:12:03 <tusho> think i've basically hit the most minimum design here
14:12:39 <tusho> removing any more would add extra whitespace (enough to be ugly), or make the edit box too small, or remove the tiny bit of fonts I have
14:19:38 <Deewiant> AnMaster: mail coming your way
14:20:58 <AnMaster> Deewiant, blergh -- makes thunderbird think everything below is a signature
14:21:42 <AnMaster> Deewiant, yes there is a space there in your stuff
14:21:47 <AnMaster> Just all kinds of stuff that came to mind based on the draft version 2.
14:21:47 <AnMaster> -- You often say "IP index" when you mean "IP ID". I think it would be clearer to just use the latter all the way through. --
14:21:48 <tusho> LOL. whatismyip.org is down
14:22:00 <Deewiant> yeap, I typoed in the first --
14:22:01 <AnMaster> tusho, it isn't hard to get your ip..
14:22:13 <tusho> AnMaster: yes, but 'whatismyip.org[enter]' is instinctive
14:22:16 <Deewiant> ah well, I can resend it if you want ;-P
14:22:30 <tusho> meanwhile http://91.105.115.23/on/FooBarBaz:Sandbox behold the painful minimalism!
14:23:05 <AnMaster> tusho, nice, but I prefer sans-serif on screen
14:23:16 <tusho> AnMaster: The header and the footer are sans-serif. :P
14:23:23 <tusho> Also the headings.
14:23:40 <tusho> AnMaster: yea, I was trying to differenciate between the auxillary debris and the actual text
14:23:51 <tusho> and my reasoning was that the actual text is more formal than the clutter around it
14:24:02 <tusho> (even though that clutter is 2 lines, whatever)
14:24:05 <AnMaster> tusho, what about using <hr />?
14:24:19 <tusho> <hr /> is unsemantic, I'd add a bottom border
14:24:28 <tusho> i generally prefer typography & whitespace to seperate sections
14:24:46 <AnMaster> tusho, anyway I wouldn't read a wiki with that font for the main text
14:25:12 <tusho> my personal wiki of opinion will surely fail because of the money you deprive it by not clicking on its huge swaths of ads, I'm sure
14:25:24 <tusho> do you refuse to read any website that's in a serif font?
14:25:52 <tusho> AnMaster: it has 85 ads
14:25:55 <tusho> flashing all over the page
14:25:59 <tusho> you are depriving me of tons of money!
14:26:00 <AnMaster> tusho, and no I just think it is painful to read serif on screen
14:26:01 <tusho> i need your readership!
14:26:46 <tusho> AnMaster: I am homeless and cannot afford food. You are depriving me money I need to stay alive by not clicking on the huge numbers of ads on my site, which you refuse to read because it's serif.
14:26:48 <tusho> Have you no heart? :(
14:27:04 <AnMaster> tusho, there are no ads, I checked in another browser
14:27:12 <tusho> AnMaster: I CAN'T AFFORD FOOD
14:27:13 <AnMaster> Deewiant, well use a single - or whatever then :P
14:27:16 <Deewiant> AnMaster: so I just used 3 dashes now
14:27:18 <tusho> BECAUSE OF PEOPLE LIKE YOU
14:28:19 <AnMaster> tusho, I would accept ads if they weren't animated, too many ads are animated, result: I block all ads
14:28:29 <tusho> AnMaster: DO YOU WANT ME TO STARVE?!?!?!!
14:29:26 <AnMaster> tusho, do you want people suffering from epilepsy to get a seizure when viewing your site?
14:29:41 <AnMaster> sure I don't have epilepsy, but I can't stand animated ads
14:29:43 <tusho> AnMaster: AS LONG AS I CAN AFFORD FOOD
14:29:48 <tusho> WHICH I CAN'T RIGHT NOW
14:29:49 <AnMaster> THERE ARE NO ADS ON YOUR FUCKING WEBSITE
14:29:55 <tusho> I HAVE TO PICK UP SCRAPS FROM THE TRASH CANS
14:29:56 <AnMaster> I CHECKED WITH ANOTHER BROWSER
14:30:02 <tusho> THEY HAVE DISEASES
14:30:05 <AnMaster> ALSO YOU LIVE AT HOME SINCE YOU ARE 12
14:30:15 <tusho> hahahaha you're like the easiest person to troll ever :D
14:30:25 <AnMaster> tusho, I'd say the same about you
14:30:45 <tusho> 4 minutes from naught to capitals and swearing
14:31:26 <tusho> ah yes but I successfully annoyed you
14:33:03 <AnMaster> tusho, you did, becuase I was trying to read about function references in erlang
14:33:29 <tusho> cmeme: can you go away; ircbrowse is dead
14:34:06 <tusho> AnMaster: ircbrowse.com
14:34:21 <AnMaster> tusho, or it is temp down simply
14:34:33 <tusho> its been like that for months
14:34:39 <tusho> and before that cmeme kept crashing
14:34:46 <tusho> even when it was in here
14:36:01 <tusho> aardappel is the channel owner
14:36:05 <tusho> but he hasn't come here since 2002 iir
14:36:19 <tusho> (aardappel = FALSE)
14:36:46 <AnMaster> well someone could make a group registration asking for take over of esoteric
14:36:59 <AnMaster> as a nice side effect you could get esoteric/ cloaks then :P
14:37:08 <tusho> this channel name is incorrect
14:37:10 <tusho> it should be ##esoteric
14:37:14 <tusho> but it was made before that policy
14:37:21 <AnMaster> * You have been kicked from ##esoteric by ChanServ (Invite only channel)
14:37:22 <tusho> and there's no point changing it now
14:37:34 <tusho> AnMaster: probably for the religious/magic kind of esoterica
14:37:46 <AnMaster> tusho, also there was no one else there
14:37:51 <tusho> AnMaster: http://en.wikipedia.org/wiki/Esoteric
14:37:53 <AnMaster> chanserv joined a second after me
14:38:22 <tusho> we are not changing to ##esoteric
14:38:24 <tusho> it's not happening
14:38:42 <tusho> it's been like this since 2002 and freenode are too busy whining for money to do anything about the tos violation :p
14:41:35 -!- kar8nga has joined.
14:45:43 <fizzie> Uh, yes, I think I have a reasonably wide set of privileges in chanserv for #esoteric.
14:50:39 <tusho> fizzie: i won't believe it until you ban me
14:50:51 <tusho> that's what every op must do in here, otherwise they are not an op. it's simple logic, really.
14:50:57 <tusho> lament's the only one so far
14:52:19 <fizzie> Hmm, are freenode's services down? "ChanServ: No such nick/channel"
14:53:03 <Ilari> Just netsplitted...
14:54:43 <tusho> everyone leave #esoteric
14:54:47 <tusho> I need to steal ops while services are away
14:56:38 -!- ChanServ has set channel mode: +o fizzie.
14:56:53 <fizzie> Did you really need the +b, too, or can I just deop myself now?
14:57:54 <tusho> i also need a kick
14:58:11 <tusho> fizzie: it is very important
14:58:23 -!- fizzie has set channel mode: +b tusho!n=tusho@91.105.115.23.
14:59:23 -!- fizzie has set channel mode: -b tusho!n=tusho@91.105.115.23.
14:59:27 -!- fizzie has set channel mode: -o fizzie.
14:59:33 <fizzie> Can't really say I saw the point in that.
15:00:33 <fizzie> Well, he looks like one.
15:02:08 <fizzie> I wonder if the ban+kick thing was some sort of crafty "I will frame them to be horrible tyrants and complain to freenode people" ploy.
15:02:37 <fizzie> optbot: What do you think? Didn't tusho write you?
15:02:37 <optbot> fizzie: or well, one can't join graue's forum, but i meant with that which one to start using more
15:03:25 -!- unrelatedguy has joined.
15:04:03 <fizzie> I think the word "brother" already implies that.
15:04:46 <fizzie> My wife's sister is a boy, actually, so I guess that sort of stuff happens.
15:04:58 -!- KingOfKarlsruhe has quit (Remote closed the connection).
15:06:27 <fizzie> Because it seemed to be his life-long ambition.
15:07:05 <fizzie> Well, you probably already read the logs?
15:07:48 <fizzie> That can't be true. There were no smileys anywhere. It's not a joke if there's no ":)".
15:08:41 <unrelatedguy> now can you please unban my brother i don't think that was very nice
15:08:53 <fizzie> I already did, I think.
15:09:28 -!- tusho has joined.
15:09:49 -!- unrelatedguy has quit ("leaving").
15:10:02 <Deewiant> I fail to see how this was very important
15:10:25 <fizzie> Incidentally, how can your brother be "unrelated" to you? Or maybe the nickname isn't related to this case.
15:11:01 <tusho> that's the nick he uses on irc...
15:11:34 <tusho> on a completely different note, I don't have a brother
15:11:58 <fizzie> Not even a female one? Gasp! Shock, horror.
15:12:10 <tusho> i know it's shocking
15:12:23 <tusho> fizzie: hey i think you missed the ":)"
15:13:25 <fizzie> I guess that's one way to get more chatter on the channel.
15:14:08 <optbot> tusho: 72 bottles of beer on the wall, 72 bottles of beer.
15:14:15 <tusho> optbot: take one down and pass it around
15:14:16 <optbot> tusho: because youre at work?
15:14:20 <tusho> optbot: yes. can't drink at work.
15:25:23 -!- Corun has joined.
15:51:32 -!- Ilari has quit (""Won't be back for a while..."").
16:05:52 -!- Judofyr has quit (Read error: 104 (Connection reset by peer)).
16:06:23 -!- Judofyr has joined.
16:54:51 <AnMaster> new version of the draft of funge-108
16:55:47 <Deewiant> if it's just some typo fixes it's not important enough to ping me ;-)
17:14:47 -!- Hiato has joined.
17:57:39 -!- MikeRiley has joined.
18:00:45 <Deewiant> MikeRiley: another question btw, the matrix-creating instructions in 3DSP (PRSTY) don't make sense to me: P for instance is (Vdm Vsm -- ) - where's the result stored
18:02:13 <MikeRiley> stored on the stack, will have to fix the documentation one for that
18:02:32 <Deewiant> I guess it should be (Vsm -- Vdm) or something then?
18:02:59 <Deewiant> and how are vectors/matrices stored, x y z / x1 y1 z1 x2 y2 z2 x3 y3 z3 or what?
18:04:03 <MikeRiley> matrices x1 y1 z1 x2 y2 z2 x3 y3 z3
18:04:19 <MikeRiley> never claimed to be good at documenation!!! eheheeheheheheheeh
18:04:28 <MikeRiley> really need peer review for that stuff!!!! eheheheheheheeh
18:05:12 <Deewiant> for subtraction, one never knows if it's "a - b" or "b - a"... the latter is standard in Funge but I'd document it anyway
18:05:36 <Deewiant> (applies to 3DSP's B, CPLI's S, probably FPSP/FPDP and others also)
18:05:43 <MikeRiley> actually,,,,hold on....need to take a look at the source....the matrices themselves may be in funge-space
18:06:07 <Deewiant> aha, so Vdm is a pointer to funge-space
18:06:19 <Deewiant> that was not at all obvious ;-)
18:06:50 <Deewiant> and then, they're stored x1 y1 z1 x2 y2 z2 x3 y3 z3 within funge-space, taking up 9 cells thus?
18:07:05 <MikeRiley> just looked at source,,,,matrices are in funge-space,,,,so where matrices are arguments they are vectors to funge-space
18:07:16 <MikeRiley> let me check the format,,,,hold on
18:07:44 <MikeRiley> stored in funge-space in 2 dimensions...
18:09:04 <MikeRiley> i guess that fingerprint can use some better documentation!!!! eheheheheheheheeheh
18:12:51 <MikeRiley> that functions maps a 3d point to a 2d point, if you hold on, can get you the formula for it??
18:13:15 <Deewiant> there are many ways of doing it
18:13:18 <MikeRiley> idea of that fingerprint is to be able to work with objects in 3 dimensional space, but be able to draw them on the screen in 2 dimenstions...
18:13:24 <MikeRiley> yes there are ,,,, let me get my formula...
18:13:31 <Deewiant> so document which way it is ;-)
18:13:35 <MikeRiley> which came out of a raytracing book, if i remember....
18:13:53 <MikeRiley> case EX_3DSP+14:Float.i=Pop(cip); b.z=Float.f; /* V */
18:13:53 <MikeRiley> if (b.z==0) Float.f=b.x; else Float.f=b.x/b.z;
18:13:54 <MikeRiley> if (b.z==0) Float.f=b.y; else Float.f=b.y/b.z;
18:14:50 <Deewiant> so it's just (x,y,z) -> (x/z, y/z)
18:17:06 <Deewiant> hmm, what's with the zero checks
18:17:17 <MikeRiley> to prevent division by zero errors...
18:17:28 <Deewiant> floating point division by zero is well-defined :-P
18:17:34 <MikeRiley> z=0 is the same plane as the screen
18:17:57 <MikeRiley> positive z is away from viewer,,,negative z is towards viewer
18:18:13 <MikeRiley> no,,,since z=0 is the reference poitn for everything else
18:18:48 <MikeRiley> so with z=0, no perspective calculation needs to be done, just use the x and y as is
18:31:27 <MikeRiley> important point,,,Rc/Funge-98 source code for 3DSP is a bad example, it is ALL WRONG!!!! looks like it is working with interger values instead of floating point.....i have to rewrite that module....
18:31:39 -!- Hiato has quit ("Leaving.").
18:32:14 <MikeRiley> all of the numbers used in 3DSP need to be floating point,,,same as in FPSP...
18:32:31 <Deewiant> MikeRiley: hmm BTW, translation matrices need to be 4x4 don't they??
18:33:07 <Deewiant> or is that meant to be handled in X
18:33:08 <MikeRiley> Rc/Funge-98 is using 4x4 matrices....
18:34:13 <MikeRiley> so revise what i said earlier about matrices in funge-space,,,they are stored 4x4...
18:35:48 <MikeRiley> the functions i was using for matric storage:
18:35:59 <MikeRiley> struct MATRIX Get_Matrix(int x,int y,int z)
18:36:24 <MikeRiley> void Put_Matrix(struct MATRIX *a,int x,int y,int z)
18:36:54 <MikeRiley> which also look wrong actually!!!! eheheheheheheeheheheh
18:37:34 <MikeRiley> that whole module looks like it needs to be scrapped and rewritten!!!! eheheheheheheeheh
18:38:13 <MikeRiley> shoult be Get_Funge(x+fx,y+fy,z) and Put_Funge(x+fx,y+fy,z,f.i)
18:38:45 <Deewiant> MikeRiley: X is just matrix*vector, right?
18:39:59 <MikeRiley> starting to wonder how my test program for that fingerprint ever worked!!!
18:40:32 <MikeRiley> but running it,,,,does come up with the right values....hmmmmmmmm
18:41:30 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | The 7th level of hell is reserved for those violent against people and property..
18:41:55 <MikeRiley> think i will write a new test program for that module that is more rigourous,,,
18:47:02 <Deewiant> MikeRiley: what order should Y multiply in
18:47:11 -!- Hiato has joined.
18:48:16 <MikeRiley> actually i see how my code works now....it is correct....do need to fix Get_Matrix and Put_Matrix tho....rest is correct,,,,probably will not work right on a 64-bit machine tho...
18:49:11 -!- kar8nga has left (?).
18:49:49 <MikeRiley> single is 32-bits, so fits in the 32-bit cells of Rc/Funge-98....using the integer protion of a union to get/put cells, but using the float portion of the union for the calculations...
18:50:19 <MikeRiley> not as much rewriting as i thought,,,but will need to modify the code to be 64-bit safe...
18:58:34 -!- RodgerTheGreat_ has quit.
19:07:11 <Deewiant> MikeRiley: just read your e-mails
19:07:22 <Deewiant> about MODU, good point, I forgot it was cat's-eye's, my bad
19:08:01 <Deewiant> about DATE, yes, some people don't use the gregorian, such as some asian countries
19:08:17 <Deewiant> about EMEM, so "addr" is an offset into the malloced array if I got you correctly?
19:08:56 <Deewiant> for IMAP, great, but why not go all the way and allow all values that a cell can hold
19:11:48 <MikeRiley> yes addr in EMEM referes to the address in the malloc'ed block,,,
19:12:24 <MikeRiley> i suppose IMAP could do that, just complicates the code...right now i just use a 256 element array so quick conversion,,,could go to something else...
19:13:34 <Deewiant> start by documenting what IMAP is supposed to do :-)
19:13:36 <MikeRiley> in EMEM addr=0 is the first byte of alloc'ed memory,,,the actual memory pointer is not exposed to the funge program itself
19:13:51 <MikeRiley> i have changed IMAP in the documentaion, just not posted it yet
19:14:12 <MikeRiley> even hnd is not necesarily the actual pointer returned my malloc,,,
19:14:17 <Deewiant> and I'm not sure I'm going to implement EMEM, it seems totally pointless to me :-P
19:14:35 <MikeRiley> i doubt i will ever use that one myself...
19:14:57 <Deewiant> if I were you I'd just remove it entirely, it has practically zero use
19:14:57 <MikeRiley> on a system where the pointer returned from malloc does not fit into the cell size, then some other mechanism must be used
19:15:11 <Deewiant> since we already have two data storage locations
19:15:15 <MikeRiley> agree,,,,near zero use....but already written....
19:15:54 <Deewiant> the only point for EMEM that I can see is if you're running on a /very/ low-memory system
19:15:57 <MikeRiley> not sure why i even defined EMEM actually,,,probably was bored that day and decided to do somethign strange...
19:16:04 <Deewiant> but... you're not going to be running funge on such a system anyway
19:16:15 <Deewiant> and even if you are, you can't do many useful things even with EMEM
19:17:02 <Deewiant> MikeRiley: and as for "already written", I'm going to quote from another channel
19:17:05 <Deewiant> 2008-08-10 18:56:16 ( gwern) for every line of code you delete, god resurrects a kitten
19:17:55 <olsner> hmm, I wonder if the reverse also holds... that for every line of code you write, god kills a kitten
19:18:16 <Deewiant> only if somebody asks you to delete it ;-)
19:19:50 <tusho> every time you program in visual basic a kitten kills god
19:21:58 <MikeRiley> chances are EMEM was the result of being bored and running out of fingerprint ideas!!! eheheheeheh
19:22:33 <Deewiant> meh, you don't have to churn out a fingerprint every day :-P
19:22:42 <Deewiant> especially if it's just a pointless one >_<
19:23:04 <MikeRiley> i know....but when i get bored,,,got to do something...eheheheeheheheh
19:23:19 <MikeRiley> and since funge is my current focus these days,,,that is what gets worked on...
19:23:43 <Deewiant> have you got a funge-108 mode yet, do that instead :-P
19:24:01 <MikeRiley> i have most of one...not too much more to do and then it should work,,,
19:24:15 <MikeRiley> i have all the simpler stuff,,,,like reflects on negatives for various instructions....
19:24:22 <Deewiant> so do that instead of creating useless fingerprints >_<
19:24:28 <MikeRiley> have a changed version of the new fingerprints....
19:24:49 <MikeRiley> what is wrong with useless fingerprints???? half the fingerprints i have made are probably useless!! eheheehheeheheh
19:25:13 <Deewiant> some are at least interesting, such as TRDS
19:25:27 <Deewiant> but EMEM is useless /and/ boring :-P
19:25:32 <MikeRiley> that one i think was born out of insanity!!!! eheheheheheeheh
19:25:54 <Deewiant> I guess you pass mycotrds, or?
19:26:31 <MikeRiley> no,,,,it passes my own diagnostics now....but mycotrds exits normally before it should...i sent you an email about it,,,wondering what should have happened....never got a response...
19:26:44 <MikeRiley> so trds is on hold at the moment while i define some more useless fingerprints!!! eheheheheheeheh
19:27:16 <psygnisfive> whooo so im using this here dvorak keyboard and im getting faster and faster :o
19:27:36 <Deewiant> MikeRiley: you can always look at CCBI's results to see what should happen
19:28:16 <MikeRiley> i have looked at CCBI's results,,,,but still not quite sure what the code was doing....did not feel like diving into it....
19:28:18 <Deewiant> MikeRiley: the problem with TRDS is that it's never obvious what exactly is going on and where, so I can't just tell you where the problem is from your output :-P
19:28:36 <MikeRiley> should just crank up the debugger and single step through that section...
19:28:46 <Deewiant> so how can you be bored if you haven't resolved this? ;-P
19:28:55 <Deewiant> more interesting than EMEM at least...
19:29:07 <MikeRiley> true,,,would be more interesting than EMEM....
19:29:31 <MikeRiley> shall i just mark EMEM as deprecated and mention that nobody should implement it??? eheheeheheheheheheheheheheh
19:29:51 <Deewiant> how about, remove it and say it was a stupid idea ;-P
19:30:11 <MikeRiley> kinda like PNTR,,,,very surprised that you have a test for that since PNTR was renamed to INDV very very early...
19:30:29 <Deewiant> if you renamed it very very early, why'd you keep it in the readme :-P
19:30:54 <MikeRiley> remember,,,,not terribly good with documentation!!! eheheheheeheheheh
19:30:55 <Deewiant> so how about removing that, too
19:31:33 <MikeRiley> i agree that the PNTR should not exist,,,in the original registry it actually appeared as INDV,,,no sense in having 2 names for the same module...
19:31:53 <MikeRiley> i will remove it from all my documentation if you remove teh tests for it in mycology....
19:32:11 <Deewiant> AnMaster: if you implement PNTR, don't.
19:32:48 <Deewiant> and an accident that it still exists
19:32:53 <MikeRiley> PNTR was renamed to INDV long long ago....at chris's request...
19:33:31 <MikeRiley> nothing should have ever implemented PNTR...only INDV....
19:33:32 <AnMaster> Deewiant, do someone want to use PNTR for something else?
19:33:43 <AnMaster> anyway I just map it as an alias to INDV
19:34:04 <MikeRiley> that is how it is in my code now....a pointer to INDV,,,,and that did not happen until mycology wanted to test PNTR....
19:34:08 <AnMaster> why? it might break code compatiblity
19:34:20 <MikeRiley> there should not be any code that uses PNTR....
19:34:50 <MikeRiley> then the code needs to change PNTR to INDV....
19:35:18 <AnMaster> MikeRiley, anyway why is EMEM a stupid ide?
19:35:41 <MikeRiley> did not say it was a stupid idea....only that it does not have much use...
19:36:06 <Deewiant> SMEM I don't know about, but EMEM is utterly pointless
19:36:11 <AnMaster> MikeRiley, don't you need to allocate with EMEM first to use it with SMEM?
19:36:29 <AnMaster> Deewiant, not really, for interaction with native code it seems very useful
19:36:31 <MikeRiley> no you do not...are you familiar with what SysV shared memory is???
19:36:44 <Deewiant> AnMaster: ... how would you use EMEM to interact with native code
19:36:47 <AnMaster> if I ever implement the C<->Befunge FFI I planned EMEM seems like a good idea
19:37:00 <Deewiant> AnMaster: you do realize that EMEM would be incompatible with such an FFI
19:37:01 <AnMaster> Deewiant, you need to pass pointers to native code sometimes
19:37:08 <Deewiant> AnMaster: and where do you get the pointer from
19:37:18 <Deewiant> AnMaster: you can't get the pointer from EMEM.
19:37:33 <Deewiant> you only get an abstract handle.
19:37:36 <MikeRiley> depends how you implement EMEM,,,,
19:37:39 <AnMaster> well maybe the FFI would have an "extract pointer from EMEM using handle"
19:37:40 <Deewiant> if you use it as a pointer your code is INCORRECT.
19:37:53 <MikeRiley> even as an abstract pointer,,,,for my implementation,,,the real pointer is obtainable from the IP data block
19:38:20 <Deewiant> AnMaster: why do that when you can just call malloc directly from the FFI.
19:38:26 <MikeRiley> the data structure that holds all the info for an IP,,,,position, delta, etc...
19:38:42 <AnMaster> More fingerprints = more features
19:38:53 <AnMaster> more features = more potential users
19:39:20 <Deewiant> AnMaster: so you think it's a good idea to have a huge fingerprint tree, only one of which does something useful, the others are only supporting instructions for that one? :-P
19:39:45 <AnMaster> "the vibrant funge community" is still far from a reality
19:39:47 <Deewiant> EMEM would be like that, useless in itself
19:40:28 <MikeRiley> i do not believe that you should have a bunch of fingerprints only to support some other fingerprint....
19:40:42 <MikeRiley> each fingerprint should have its own uses...
19:40:44 <AnMaster> I have been thinking of writing a Funge-93 interpreter in erlang
19:40:51 <AnMaster> mainly as a nice project to learn erlang
19:40:55 <MikeRiley> as it is,,,,none of my fingerprints require some other fingerprint to be loaded...
19:41:12 <Deewiant> which is why I don't like it very much
19:41:24 <MikeRiley> not really,,,it just added features to SOCK,,,
19:41:36 <AnMaster> MikeRiley, idea for a good fingerprint however:
19:41:36 <MikeRiley> just like JSTR changed features of STRN
19:42:04 <Deewiant> JSTR can be used without STRN, SCKE can't be used without SOCK
19:42:07 <AnMaster> or maybe it should become part of SOCK, but I don't like changing existing fingerprints
19:42:14 <MikeRiley> i agree SCK6 would be a good addition
19:42:49 -!- dogface has joined.
19:42:52 <MikeRiley> i agree about changing fingerprints,,,especially ones that have been around a long time....
19:42:59 <MikeRiley> so SCKE makes sense for what it added...
19:43:32 <MikeRiley> changing a fingerprint that somebody might be using,,,can break code...better to make a new one....
19:43:33 <psygnisfive> ========================-=-----------------------
19:43:44 <AnMaster> MikeRiley, of course with Funge-108 that would be easy, if you had an uri (gopher://example.org/0/funge/my-hand/my-finger/1.0) then you could make a 2.0 and so on
19:44:06 <AnMaster> the author could say that gopher://example.org/0/funge/my-hand/my-finger should map to the last one that is implemented
19:44:23 <MikeRiley> well...even tho i do not like the uri for the fingerprint name,,,,still nothing but a string compare as far as i am concerned,,,and so yes....could be versioned that way....
19:44:55 <AnMaster> MikeRiley, string compare with some twists, uris are case sensitive, but the domain name isn't
19:45:06 <AnMaster> there are more details in the rfc
19:45:15 <MikeRiley> you can use case insensitive string cmpare....strcasecmp.....
19:45:24 <AnMaster> well the rest of it is case sensitive
19:45:31 <AnMaster> I will need to read up on the details myself
19:47:07 <AnMaster> what license is RC/Funge under?
19:47:42 <MikeRiley> my own....modify it, use it, copy, distribute,,,no commercial use without my permission,,,,my copyright notices must remain inteact....
19:48:05 <MikeRiley> very close to bsd license,,,,but i disallow commercial use...
19:48:13 <AnMaster> MikeRiley, you are interested in section 6 of http://tools.ietf.org/rfc/rfc3986.txt only
19:49:16 <AnMaster> much much less than the around 72 pages in total
19:50:03 <AnMaster> determination of equivalence or difference of URIs is based on string
19:50:03 <AnMaster> comparison, perhaps augmented by reference to additional rules
19:50:03 <AnMaster> provided by URI scheme definitions"
19:50:20 <AnMaster> those additional rules may be a tiny problem
19:50:47 <MikeRiley> that is partly why i disagree on uris,,,,solves the problem in too complicated a way...
19:51:07 <AnMaster> MikeRiley, it certainly avoids collision issue though
19:51:14 <MikeRiley> the 4 character one was lousy,,,but i think the uri approach is doing too much...
19:51:21 <AnMaster> MikeRiley, you see I had some collision issues with tusho here
19:51:23 <MikeRiley> i agree on the goal...to avoid collisions....
19:51:31 <MikeRiley> just think there are better ways to go about it
19:51:43 <tusho> MikeRiley is wrong
19:51:44 <AnMaster> and same could easily happen for fingerprint
19:51:50 <tusho> URIs are specifically designed to be universal identifiers
19:51:56 <tusho> and should be used for this purpose whenever possible
19:52:00 <MikeRiley> i understand that...that was the problem with the original method...
19:52:18 <tusho> and URIs are the right way to solve it
19:52:25 <tusho> because they are specifically designed for it
19:52:28 <MikeRiley> maybe so tusho....but i am a believer in the keep it simple philosophy....
19:52:32 <tusho> and interoperability between technologies is a good thing
19:52:35 <tusho> and URIs are simple
19:52:39 <tusho> you don't implement them yourself
19:52:40 <MikeRiley> and uris do not provide a simple solution to the problem
19:52:48 <tusho> but you appear to ignore me whenever I say "use a library"
19:52:59 <MikeRiley> very against the use of code tht is not mine
19:53:11 <MikeRiley> unless they are very standardized...
19:53:20 <MikeRiley> and even then,,,,i tend not to use them...
19:53:29 <tusho> don't use others code = you suffer whenever something non-trivial comes up
19:53:46 <tusho> "implementing it -from scratch- is complicated" does not mean something is complicated.
19:53:54 <MikeRiley> i do not have a problem with that...
19:54:02 <MikeRiley> somebody else's code also means someobyd else's bugs...
19:54:07 <AnMaster> MikeRiley, cfunge use no external libraries that aren't very common (ncurses) or optional (Boehm-GC)
19:54:10 <tusho> for URIs, most people will do uri_parse("..."), then uri_equal(x, y)
19:54:18 <MikeRiley> at least with my code...i can understand and fix problems
19:54:19 <AnMaster> MikeRiley, it does however use some small libraries where the source is included
19:54:22 <tusho> and comes with a good standard designed exactly for this behind it
19:54:26 <tusho> and interoperability
19:54:55 <MikeRiley> Rc/Funge-98 uses no libraries outside of what can be found in any unix installation...
19:55:07 <tusho> uri libraries are small
19:55:07 <AnMaster> MikeRiley, so include the code?
19:55:10 <tusho> just include it as a c file
19:55:17 <MikeRiley> no additional libararies are required...and therefore easier for somebody else to compile and get running....
19:55:22 <AnMaster> MikeRiley, I use a small 2 file hash library for example
19:55:26 <Deewiant> MikeRiley: such as X, which isn't found in most of my unix installations...
19:55:27 <AnMaster> I include the source *inside* cfunge
19:55:31 <MikeRiley> nothing i hate more than program where i have to go and find 20 other libraries to get working...
19:55:40 <tusho> MikeRiley: so INCLUDE THE SOURCE
19:55:43 <tusho> in your source tree
19:55:45 <tusho> it's only one file
19:55:47 <Deewiant> MikeRiley: package managers and such exist
19:56:01 <MikeRiley> as far as X,,,do not include those modules...only TURT and WIND use x....
19:56:03 <AnMaster> ANYWAY WHY NOT JUST INCLUDE THE C FILE IF IT IS SMALL
19:56:09 <MikeRiley> and if you do not have x....those those are not usefull anyways...
19:56:21 <Deewiant> AnMaster: http://en.wikipedia.org/wiki/X_Window_System
19:56:34 <AnMaster> I got it on this system though
19:56:36 <tusho> welp, seems MikeRiley is just simply pretending the lines that answer his question don't exist...
19:56:37 <Deewiant> MikeRiley: I might have something else instead, though :-P
19:56:38 <MikeRiley> then comes licensing issues when using other people's code....i may not agree with their licensing terms...
19:56:52 <tusho> MikeRiley: there are BSD/MIT licensed uri libs...
19:57:03 <tusho> you can just include them in your source tree
19:58:08 <MikeRiley> again,,,do not like including other code within my projects...
19:58:16 <MikeRiley> will come up with my own solutions instead...
19:58:23 <tusho> MikeRiley: then have fun implementing URIs
19:58:28 <tusho> but we've offered perfectly good solutions
19:58:30 <tusho> you just don't like them
19:58:33 <MikeRiley> will not implement uris,,,,and do not need to....
19:58:43 <tusho> ok then you won't conform to funge-108
19:58:47 <tusho> so have fun with that
19:58:57 <MikeRiley> as long as i can load fingerprints for funge-108,,,then it conforms....
19:59:10 <tusho> you won't be able to if you don't support uris
19:59:11 <MikeRiley> and string comparison will accomplish that...
19:59:12 <Deewiant> and how can you, without implementing them...
19:59:17 <tusho> MikeRiley: no it won't
19:59:19 <tusho> uris have case rules
19:59:29 <tusho> and there's various things that cancel each other out
19:59:36 <tusho> once you implement them all you just implemented URIs.
20:00:04 <MikeRiley> i may do something simpler...case may not really matter...will see when somebody starts making fingerprint names with uris....
20:00:20 <MikeRiley> until then,,,it is a moot point...
20:00:24 <tusho> Deewiant: I hope mycology-108 will test uri equivilence thoroughly
20:00:33 <Deewiant> tusho: I won't be the one to write it :-P
20:00:43 <tusho> Deewiant: hehe, AnMaster might
20:00:54 <tusho> AnMaster: mycology-108
20:01:01 <tusho> to test for things like uri equivilence works for loading fingerprints
20:01:03 <Deewiant> I don't like funge-108 as a whole anyhoo, -98 is good enough
20:01:13 <tusho> MikeRiley is claiming that he can implement uri equivilence wrongly and still be conformant
20:01:16 <tusho> but of course it won't be
20:01:21 <tusho> since he will consider the same URI as seperate
20:01:26 <tusho> and thus a bug in fingerprint loading
20:01:31 <tusho> so mycology-108 should test that
20:01:39 <MikeRiley> i will make sure that i can load defined fingerprints,,,and will not claim funge108 complience,,,even tho it mostly will be...
20:01:47 <AnMaster> at least I will write a few test suites for it
20:01:57 <tusho> MikeRiley: ok. if you just want to be wrong and even though we tell you easy ways to be right, then just fine
20:01:59 <AnMaster> tusho, I already got some small test programs to test some parts
20:02:04 <tusho> MikeRiley: your loss
20:02:22 <tusho> cfunge will be a "funge-108 interpreter" and yours will be a "broken partly-funge-108 interpreter"
20:02:27 <AnMaster> heck *I* will use an URI library
20:02:30 <MikeRiley> that is fine....you are right,,,,my loss....since i am probably the only user of Rc/Funge-98 anyways...does not really matter does it???
20:02:43 <tusho> no, but by that argument I don't even know why you are bothering changing it
20:03:22 <MikeRiley> and if somebody writes funge108 programs,,,i would like to be able to run them...
20:03:32 <tusho> MikeRiley: but what if they happen to rely on that weird uri behaviour
20:03:36 <Deewiant> even though you won't be able to, if you're not compliant...
20:03:38 <tusho> because they're insane - after all this is befunge
20:03:48 <tusho> You underestimate esolang programmers.
20:04:01 <tusho> If your impl is broken due to some obscure cornercase, you can bet someone will write a program that will break because of it
20:04:04 <tusho> but work on compliant impls
20:04:14 <AnMaster> MikeRiley, %63 is same as c I think for exmaple
20:04:20 <MikeRiley> in which case,,,i will fix the corner case...
20:04:30 <tusho> MikeRiley: and eventually we'll cover them all
20:04:33 <tusho> and you'll have implemented URIs
20:04:50 <MikeRiley> well...then i will have implemented URIs!!!! the hard way!!!!! eheheheheheheheeheh
20:05:01 <tusho> so, we were trying to offer the easy way
20:05:15 <tusho> that also lets you say 108-compliant. :p
20:05:22 <MikeRiley> yeah,,,but one person's easy way is not necessarily good for anotgher...
20:05:42 <tusho> have fun implementing URIs ... over months ...
20:05:50 <MikeRiley> different people....different philosophies...
20:06:06 <MikeRiley> i guarantee you,,,i will have fun at it!!! even if it takes years....
20:07:08 <MikeRiley> my fun is working on the interpreter....find things that break it,,,,means more fun for me fixing it....
20:07:42 <MikeRiley> mycology was the greatest thing ever for my funge fun....found all kinds of problems that needed to be fixed!!!! eheheheheheehh
20:08:08 <tusho> AnMaster: can I help write mycology-108
20:08:11 <tusho> I wanna include fuzz testing in it
20:08:18 <Deewiant> MikeRiley: what does the UNDEF test in INDV say for RC/Funge-98
20:08:32 <AnMaster> tusho, hah, no fuzz testing would be a separate program
20:08:41 <tusho> AnMaster: no no no it wouldn't be for no-crashness
20:08:47 <tusho> it would be to actually test funge 108 compliance
20:08:51 <Deewiant> MikeRiley: you don't need a new mycology
20:08:53 <AnMaster> tusho, yes like mycorand is a separate part
20:09:00 <MikeRiley> Testing fingerprint INDV... loaded.
20:09:01 <MikeRiley> UNDEF: V uses illogical RC/Funge-98 order
20:09:05 <tusho> AnMaster: well, presumably mycology-108 will be like 5-7 parts
20:09:12 <tusho> and mycology108.bf will just do each one
20:09:12 <Deewiant> AnMaster: the only reason mycorand is separate is for befunge-93
20:09:15 <AnMaster> tusho, I think that is said to SHOULD not MUST in most cases
20:09:27 <tusho> AnMaster: not just "give the input crap see what it does"
20:09:30 <tusho> but a constrained range of crap
20:09:31 <Deewiant> MikeRiley: so have you thought about switching to the logical order, and maybe specifying it like that? ;-)
20:09:34 <tusho> and to test for behaviour
20:09:38 <AnMaster> tusho, the software may not implement i and o
20:09:46 <MikeRiley> actually,,,yes i have thought about it,,,and will actually do that...
20:09:53 <AnMaster> tusho, and yes that makes sense
20:09:54 <tusho> AnMaster: well yes, it wouldn't use them
20:10:23 <AnMaster> anyway atm I'm googling for an uri library
20:10:56 <AnMaster> http://www.w3.org/Library/src/HTParse.html
20:11:04 <AnMaster> will need to take a closer look
20:11:14 <tusho> yes use that one AnMaster
20:11:17 <tusho> since its from the w3c
20:11:21 <tusho> they are pretty good with uris
20:11:37 <MikeRiley> got to run for now...church to go to....thanks for the stimulating conversation!!! and hope my devil's advocate approach does not offend anybody.....bye for now...
20:12:07 <tusho> MikeRiley: he was making a joke
20:12:09 <tusho> (a rather lame one)
20:12:16 <AnMaster> isn't that for lambda calculus?
20:12:21 -!- MikeRiley has left (?).
20:27:23 -!- kirarinsnow has quit (Read error: 110 (Connection timed out)).
20:35:12 -!- tusho has changed nick to tusho|away.
20:38:51 <oklopol> AnMaster: do you understand the concept of lambda?
20:39:38 <AnMaster> oklopol, yes, I seen it in a few languages
20:39:42 -!- Sgeo has joined.
20:39:51 <oklopol> okay, lambda calculus is about just having those.
20:40:06 <AnMaster> oklopol, and no + or - and so on iirc
20:40:10 -!- olsner has quit ("Leaving").
20:40:15 <oklopol> turns out we can define things like integers and lists as functions returning other functions
20:40:35 <AnMaster> oklopol, what does the innermost function return?
20:40:55 <oklopol> well it will return a function.
20:41:46 <oklopol> 2=\fx.f(fx) <<< this is just a function that takes a function f and a function x, then calls f with f called with x
20:41:55 <oklopol> so lambda f: lambda x: f(f(x))
20:42:16 <oklopol> takes a function f, returns a function that takes a function x and then calls f with f(x)
20:42:59 <oklopol> this is one way to define numbers in lambda calculus
20:43:29 <oklopol> a number N in lambda calculus is a function that takes two functions and calls the first one with the latter one N times
20:43:36 -!- tusho|away has changed nick to tusho.
20:44:54 <oklopol> bad explanation, better look at what tusho said there
20:45:30 <tusho> oklopol: AnMaster thinks in c
20:45:36 <tusho> at present he has no chance of getting LC
20:46:01 <oklopol> i could easily explain it irl
20:46:28 <tusho> oklopol: except AnMaster generally doesn't grasp anything non-imperative
20:46:31 <tusho> I've tried to explain lc
21:03:10 -!- sebbu2 has joined.
21:05:55 <AnMaster> sorry was afk, had to mend a broken printer
21:16:49 <AnMaster> tusho, about that w3c library, seems it only does normalizing urls
21:17:01 <AnMaster> maybe comparsion is somewhere else in libwww
21:17:06 <tusho> compare = string_equality . normalize
21:17:29 <AnMaster> what about uri protocol differences
21:19:01 -!- jix has joined.
21:20:40 -!- sebbu has quit (Read error: 110 (Connection timed out)).
21:33:14 -!- RedDak has joined.
21:45:55 -!- psygnisfive has quit (Read error: 110 (Connection timed out)).
21:52:20 -!- Mony has quit ("À vaincre sans péril on triomphe sans gloire...").
21:57:50 -!- jix has quit ("This computer has gone to sleep").
22:01:36 -!- Hiato has quit (Read error: 104 (Connection reset by peer)).
22:06:55 -!- SimonRC has quit (Remote closed the connection).
22:06:59 -!- SimonRC has joined.
22:16:50 -!- psygnisfive has joined.
22:17:32 -!- kar8nga has joined.
23:03:52 -!- kar8nga has quit ("Leaving.").
23:07:49 -!- dogface has changed nick to ihope.
23:07:55 -!- ihope has changed nick to dogface.
23:35:29 <tusho> AnMaster-logread-note:
23:35:34 <tusho> Funge-108 is based on Funge-98 (Pressey, 1998) that is curre
23:35:41 <tusho> The Funge-98 (Pressey, 1998) standard got se1
23:35:54 <tusho> This document will be released under either CC-by-sa2 or GFDL when it is finished. As it is not
23:35:55 <tusho> yet decided on this document is currently under “copyright” however.
23:36:03 <tusho> -> if you want ESO standardisation use CC-pd
23:36:16 <tusho> will likely have more comments later
23:38:22 -!- olsner has joined.
23:38:29 * tusho ponders what a fractional shiftwidth would look like
23:38:41 <tusho> Presumably it'd start off with three spaces.
23:40:09 <olsner> and then approx. 1/7th of a column? it's just a matter of allowing non-integer indents, and approximating to the nearest pixel should probably be acceptable?
23:40:38 <tusho> olsner: Wouldn't it go 3, 3.1, etc?
23:40:51 <tusho> Or would it really be 3, .01, .004 - because that'd be quite unusable, obviously
23:42:06 <olsner> I thought you meant tabstops at 3.14, 6.28, 9.42 etc
23:45:22 <tusho> olsner: it actually works quite well
23:46:20 <tusho> olsner: http://pastebin.ca/raw/1160950 <- the line feels natural
23:50:42 -!- Judofyr has quit.
23:53:02 <tusho> anyone have thoughts?