00:01:20 <GregorR-L> And no f***ing way I'm taking a Java class.
00:02:07 <pikhq> If that's a requirement at a college, I'll leave.
00:02:21 <GregorR-L> If you learn C++, you basically learn Java by default, just dumb down all your knowledge to what a child with down syndrome could understand.
00:02:25 <pikhq> I don't care if it's friggin' MIT, I'll leave. . .
00:03:59 <RodgerTheGreat> you can complain about languages all you want, but in the end that solves nothing. As a rule, I never piss and moan about the languages I need to use.
00:04:44 <pikhq> I piss and moan about the languages I *don't* need to use. :p
00:06:04 <RodgerTheGreat> and I'd be willing to bet that you two are just complaining about how it "isn't like C" and it doesn't compile to machinecode, rather than something meaningful to criticize about it as a language. You're just whining about Java because that's what everyone does.
00:06:54 <GregorR-L> I'm whining about Java because I learned it, and it pains me.
00:07:08 <GregorR-L> Java has a large number of silly issues, many spanning from its ridiculous dependency on OOP.
00:07:42 <GregorR-L> Yes - and it puts a dependency on your code.
00:07:44 <RodgerTheGreat> I think it's a more conceptually pure language than C for that very reason.
00:08:04 <GregorR-L> Yes - but conceptual purity is bad at the cost of flexibility.
00:08:11 * pikhq actually doesn't like C that much, either
00:08:46 <GregorR-L> Attempts at purity box programmers into corners and require them to use paradigms that may or may not actually fit what they're coding well.
00:08:46 <pikhq> C has the following going for it: it gets the job done.
00:08:50 <pikhq> That's about it. . .
00:09:22 <RodgerTheGreat> the point of java is not to be powerful, it's to enforce strict object-oriented design, which is why it's a good language for teaching algorithms and so on with.
00:09:23 <GregorR-L> C is a very pure abstraction of ASM, which is what it has going for it.
00:09:35 <pikhq> GregorR-L: That's *why* it gets the job done.
00:09:50 <GregorR-L> Mind you, I prefer other languages to C ;)
00:09:54 <pikhq> RodgerTheGreat: Lisp is a better language for teaching algorithms.
00:10:10 <GregorR-L> Lisp is a better language for teaching /particular/ algorithms.
00:10:19 <GregorR-L> If you're going to teach OO concepts with Lisp, you're an idiot.
00:10:24 <pikhq> GregorR-L: Well, true. . .
00:10:29 <RodgerTheGreat> true, but it isn't a very good language for teaching OOP (even though YES, you can use it for that purpose)
00:10:35 <jix> java isn't really object orientated
00:10:57 <pikhq> GregorR-L: His point was teaching *algorithms* with Java was a good use of it (which it isn't particularly good for). . .
00:10:57 <jix> there are things which arn't objects
00:11:20 <jix> AFAIK ints arn't objects
00:11:25 <jix> (you have to use wrapper classes)
00:11:28 <GregorR-L> pikhq: I agree with RodgerTheGreat on that point.
00:11:33 <jix> RodgerTheGreat: why?
00:11:37 <GregorR-L> But I managed to learn OO before Java.
00:11:38 <jix> that isn't very OOish
00:11:39 <pikhq> Python, I *think*, is a good deal better for OOP (although I've not used it, so take that with a pinch of salt).
00:11:50 <jix> pikhq: yes it is
00:36:26 <pikhq> When the program starts:
00:36:33 <pikhq> There is a mathematician named Gregor
00:36:43 <pikhq> There is a language called COBOL.
00:36:55 <pikhq> Gregor's first argument is 1.
00:37:07 <pikhq> Gregor's second argument is COBOL.
00:37:43 <pikhq> ORK is to become Gregor's argument.
00:38:55 <pikhq> Asztal: ORK is more verbose than COBOL; deal with it.
00:39:18 <pikhq> Thus why it should be either that or ADD 1 TO COBOL GIVING COBOL.
00:40:53 <Asztal> I don't mind the verboseness, I just wondered about what Mathematician and Language are meant to mean :)
00:42:35 <pikhq> We created two objects: an instance of "mathematician" called Gregor, and an instance of "language" called COBOL.
00:42:44 <pikhq> "mathematician" is an object for doing arithmetic.
00:42:52 <pikhq> "language" is an object I made up.
00:46:11 <pikhq> I prefer Glass; less verbose, more object oriented.
00:46:28 <pikhq> Although ORK has one advantage. . . I can understand how its compiler works! :p
00:53:19 <GregorR-L> Idonno, I remember almost nothing about it :P
00:54:40 <jix> http://monome.org/pages/40h << this is really cool but expensive as hell
00:54:50 <jix> idea: do it yourself for < 50$
00:54:53 <jix> should be possible
00:56:07 <jix> my current circuit (just idea nothing built) consists of 64 leds 64 buttons 8 8bit latches an atmega32 micro controller and some usb gpio chip
00:56:44 <Razor-X> Can the Linux kernel hand you malformed UDP packets, or is it always packet arrives or no arrives?
00:57:16 <pikhq> The UDP *spec* doesn't allow you to receive malformed UDP packets.
00:57:28 <Razor-X> Aha. I haven't read the RFC for UDP just yet.
00:57:43 <pikhq> Not sure if Linux actually follows the RFC, but that's what the RFC says. ;)
00:58:11 <Razor-X> Linux takes it a step further. If you use SOCK_DGRAM, it autoverifies the checksum and only then passes on the packet.
00:58:47 <jix> but no one cares!
00:59:03 <pikhq> Um. . . That's the defined UDP behavior. . .
00:59:06 <jix> about my do it yourself 40h thing?
00:59:31 <Razor-X> pikhq: But you're supposed to implement the checksum checking on your own.
00:59:45 <Razor-X> You do, if you use SOCK_RAW, but I don't want to go *that* low level.
01:00:00 <pikhq> Razor-X: That's an issue with the BSD sockets API, not the UDP spec. . .
01:00:14 -!- Asztal has quit ("Chatzilla 0.9.72-rdmsoft [XULRunner 1.8.1b2/0000000000]").
01:05:11 -!- jix has quit ("Bitte waehlen Sie eine Beerdigungnachricht").
01:26:04 -!- kipple_ has quit (Read error: 110 (Connection timed out)).
01:39:42 -!- GregorR-L has quit ("Leaving").
02:14:32 -!- ivan` has joined.
02:52:22 * Sgeo is implementing http://esoteric.voxelperfect.net/wiki/Brainscrambler
02:52:27 <Sgeo> I need more details though
02:52:42 <Sgeo> I'm already assuming I can't + or - without a * somewhere
02:52:56 <Sgeo> Is the number output as ASCII
02:55:07 <Sgeo> And when I move the current number, do I rotate? Is the version of curnum on the original stack deleted?
02:58:21 <Sgeo> Oh, and I'm making numbers below -1 illegal
02:59:38 <pikhq> Just make numbers wrap.
03:01:35 <Sgeo> Not that I'm doing it >.<
03:16:56 <Sgeo> I can finish up a non-looping demo now >.<
03:18:16 -!- wooby has joined.
03:18:43 -!- wooby has quit (Client Quit).
03:22:56 * Sgeo needs some test code
03:25:49 <Sgeo> It's a non-looping demo version
03:25:59 * Sgeo wrote some test code
04:05:48 -!- Arrogant has joined.
04:07:18 -!- Arrogant has quit (Read error: 104 (Connection reset by peer)).
04:07:18 -!- Arrogant has joined.
04:07:21 -!- Arrogant has quit (Read error: 54 (Connection reset by peer)).
04:08:52 -!- Arrogant has joined.
04:09:21 -!- Arrogant has quit (Read error: 104 (Connection reset by peer)).
04:11:08 -!- Arrogant has joined.
04:11:19 -!- Arrogant has quit (Read error: 104 (Connection reset by peer)).
04:24:41 <Sgeo> http://sgeo.diagonalfish.net/esoteric/brainscrambler_noloop.py.txt
04:25:54 <Sgeo> Any comments before I go to sleep?
04:36:11 -!- Sgeo has quit ("Ex-Chat").
04:47:49 -!- wooby has joined.
04:48:18 -!- Arrogant has joined.
04:48:22 -!- wooby has quit (Client Quit).
05:06:54 -!- Eidolos has joined.
05:50:36 -!- GregorR-L has joined.
06:07:48 -!- ghostless has joined.
06:09:03 -!- oerjan has joined.
06:09:57 -!- calamari has quit ("Leaving").
06:15:35 -!- Arrogant has quit ("Leaving").
06:16:52 -!- Arrogant has joined.
06:45:37 -!- Arrogant has quit ("Leaving").
06:49:28 -!- wooby has joined.
06:57:17 -!- Arrogant has joined.
07:23:27 -!- wooby has quit.
07:34:58 -!- oerjan has quit ("Leaving").
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:17:13 -!- GregorR-L has quit (Read error: 110 (Connection timed out)).
08:25:39 -!- Arrogant has quit ("Leaving").
09:56:04 -!- ivan` has quit (Read error: 104 (Connection reset by peer)).
09:58:31 -!- jix has joined.
10:18:11 -!- wooby has joined.
11:37:39 -!- wooby has quit.
12:48:43 -!- _milo has quit (Read error: 110 (Connection timed out)).
13:16:13 -!- _milo has joined.
13:33:23 -!- _Ann_ has joined.
13:34:27 -!- _Ann_ has left (?).
14:22:30 -!- sekhmet has quit ("leaving").
14:23:34 <pikhq> Mmkay. . . Discovered that LostKng.b, with line breaks, compiles correctly on basm.b. . .
14:24:00 <pikhq> No doubt the issue is that I was using a version with line breaks stripped, causing the pointer's movement to work somewhat oddly.
14:27:09 <pikhq> And there is *one* other issue with the compiler; it will output 0x00 before 1 or 2 digit numbers (easier than using conditional output there. . .).
14:27:16 -!- sekhmet has joined.
14:27:31 <pikhq> This, fortunately, has only one side effect: GCC gives a lot of warnings.
14:30:24 -!- sekhmet has quit (Client Quit).
14:39:57 -!- bsmntbom1dood has joined.
14:42:22 <pgimeno> pikhq: just wondering... wouldn't it be easier to generate octal numbers rather than decimal?
14:46:51 <pgimeno> they could be fixed length that way
14:50:30 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)).
15:00:58 -!- calamari has joined.
15:05:13 -!- sekhmet has joined.
15:09:52 <calamari> Looks like a Pentium 166 is too slow to render php / python wiki pages
15:39:47 -!- calamari has quit ("Leaving").
16:34:33 -!- kipple_ has joined.
16:39:54 -!- bsmntbom1dood has changed nick to bsmntbombdood.
17:10:48 -!- tgwizard has joined.
17:44:57 <pikhq> pgimeno: You want to write the macro for that? :p
17:45:44 <pikhq> pgimeno: Besides, *all* that happens is "WARNING: Misplaced null char" from GCC; tr -d /0 or just ignoring it leaves it still working.
17:46:27 * pikhq has confirmed that the issue with his compiler is with an overflow on *some* counter; compiling the newlined version works
17:46:31 <pikhq> pgimeno: Probably.
17:46:39 <pikhq> It wouldn't be *that* hard to fix, anyways. . .
17:46:57 <pikhq> The cells are cleared after outputting it, anyways.
17:48:40 <pgimeno> if you have a divide-by-ten it wouldn't be hard to do a divide-by-eight
17:50:11 <pikhq> It's a generalised div and mod. . .
17:50:53 <pikhq> I could probably A) switch to divide-by-eight B) make it loads shorter by doing divide-by-eight specifically. . .
17:51:15 <pgimeno> you can generate 4-digit numbers from 0000 to 0377 and forget about length etc
17:51:24 <pikhq> But, then, I don't feel like it at the moment. :p
17:52:50 <pikhq> basm.b seems fairly fast when compiled to C. . .
17:52:58 <pikhq> Compiles LostKng.b in 5 seconds.
17:55:16 <pikhq> . . . Unfortunately, GCC spends a good deal more time working on it.
18:29:31 -!- ivan` has joined.
18:32:12 -!- fr34k has joined.
18:38:05 -!- fr34k has changed nick to boblol.
18:57:31 * pikhq starts rewriting major chunks of basm.bfm
19:05:27 -!- boblol has quit (Read error: 110 (Connection timed out)).
19:13:22 <pikhq> This is going to be a big job. . .
19:15:32 <pikhq> One not at all helped by the massiveness of what needs to be put into memory.
19:36:16 -!- Asztal has joined.
19:36:29 * pikhq has 18 characters done for the new set of strings. . . Dear God, that's a lot. :/
19:36:48 <pikhq> 18 down, 140 to go. :'(
19:38:33 * pikhq is going elsewhere, for sanity's sake
20:23:48 -!- Sgeo has joined.
20:25:36 <Sgeo> Any comments on my code
20:28:47 -!- tokigun has quit (Read error: 110 (Connection timed out)).
20:31:45 <dbc> Leaving aside malloc vs. calloc, why not just declare the array as a global, "char a[30000], *p=a;"?
20:33:59 <dbc> If the memory usage is fixed, why allocate it dynamically?
20:43:36 <jix> because calloc fills it with zeros? .... char a[30000] gets filled with zeros too but i'm not sure if that is guaranteed...
20:44:24 <dbc> It is if it's a global.
20:44:37 <jix> when the c specs say so
20:51:06 -!- tokigun has joined.
21:22:11 -!- kipple_ has quit ("See you later").
21:41:31 -!- kipple_ has joined.
22:07:50 -!- tgwizard has quit ("Leaving").
22:26:21 <pikhq> dbc: Gregor complained when I did it that way. XD
22:26:55 <pikhq> Damned if I do, damned if I don't.
22:39:33 <GregorR> The memory usage should not be fixed.
22:39:42 <GregorR> If the memory usage is fixed, sure, do it globally.
22:39:49 <GregorR> But if it's fixed, you've written a worthless language.
22:55:53 <ghostless> i just started using the fungus IDE and writing befunge...and whenever i try to use a & for number input it skips my next instruction...i end up having to put a number after it just to correct it...thats not normal right?
22:59:55 -!- calamari has joined.
23:44:14 -!- anonfunc has joined.
23:56:27 <Razor-X> My take on the CS world for the week/day/moment: Why the emphasis on OO? Most OO concepts are useless in all but the largest projects, and only in certain ways.
23:56:45 <Razor-X> *and only useful in certain styles of projects.
23:58:05 <Razor-X> Anything past Classes, Methods, public/private model, and basic inheritance is useless complication IMO.
23:58:19 <Razor-X> And constructors and destructors, of course.
23:58:23 <RodgerTheGreat> well, it makes it easy to re-use code or work as a group.
23:59:01 <Razor-X> I've looked at both Java and C++ from an OOP point of view. I'll have to disagree with pikhq and say that Java's implementation of OOP is LEAGUES better. But it's still, IMO, unneccessary programmer overhead.
23:59:24 <Razor-X> I've never worked with code that needs anything more than what I listed above.