00:00:00 --- log: started forth/07.02.28
00:03:22 --- join: ecraven (i=nex@eutyche.swe.uni-linz.ac.at) joined #forth
00:11:51 --- join: snowrichard (n=richard@65.125.86.66) joined #forth
00:11:58 <snowrichard> hi
00:12:34 --- quit: virsys ("bah")
00:13:20 --- quit: snowrichard (Remote closed the connection)
00:14:52 --- join: virsys (n=virsys@or-71-54-194-74.dhcp.embarqhsd.net) joined #forth
00:50:53 --- quit: Baughn (Read error: 145 (Connection timed out))
00:54:14 --- join: Zarutian (n=Zarutian@194-144-84-110.du.xdsl.is) joined #forth
00:57:58 --- quit: Zarutian__ (Read error: 104 (Connection reset by peer))
01:00:24 --- join: Zarutian__ (n=Zarutian@194-144-84-110.du.xdsl.is) joined #forth
01:00:24 --- quit: Zarutian (Read error: 104 (Connection reset by peer))
01:29:53 --- join: Baughn (n=svein@195134062077.customer.alfanett.no) joined #forth
02:58:40 --- quit: slava ()
04:00:54 --- join: Crest (n=crest@p54896FA4.dip.t-dialin.net) joined #forth
04:07:29 --- quit: JasonWoof ("leaving")
04:28:23 --- join: ygrek (i=user@gateway/tor/x-ae60eaeffb6c4a0b) joined #forth
04:58:11 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth
04:58:11 --- mode: ChanServ set +o tathi
05:29:34 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #forth
05:37:27 --- nick: Zarutian__ -> Zarutian
05:41:43 --- quit: Crest ("This computer has gone to sleep")
05:44:08 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #forth
05:44:08 --- mode: ChanServ set +o Quartus_
05:59:19 --- join: Crest (n=crest@p54896FA4.dip.t-dialin.net) joined #forth
06:01:30 --- quit: arke ("Konversation terminated!")
06:02:07 --- join: edrx (n=Eduardo@200.217.105.248) joined #forth
06:09:58 <Quartus_> Hi.
06:10:31 <tathi> Hey.
06:14:04 <Quartus_> how goes the metacompilation?
06:15:58 <tathi> ok
06:16:33 <tathi> so many little details...
06:17:06 <Quartus_> any major hurdles?
06:17:55 <tathi> Nothing too major.  Still seems like every time I code something new for the target, I find something new to tune in the cross-compiler's behavior.
06:18:15 <tathi> I'm glad I didn't realize when I started how tricky it is to make a good cross-compiler. :)
06:18:32 <Quartus_> heh
06:19:30 <tathi> Actually, at this point, it's mostly just improving stuff that I knew had a quick-and-dirty implementation.
06:19:59 <Quartus_> almost ready for prime time?
06:20:15 <tathi> heh
06:21:36 <tathi> No, I haven't been working on it as hard as I'd like.
06:21:49 <tathi> I thought I would be a lot farther along by now.
06:22:13 <tathi> Still, I'm having enough fun that I keep plugging away at it.
07:16:25 <Quartus_> it snowballs eventually
07:21:13 <tathi> sure
07:29:16 --- quit: Crest ("This computer has gone to sleep")
07:54:26 --- quit: Quartus_ (Read error: 60 (Operation timed out))
08:01:31 --- join: Quartus_ (n=Quartus_@209.167.5.2) joined #forth
08:01:31 --- mode: ChanServ set +o Quartus_
08:10:43 --- join: Crest (n=crest@p54896FA4.dip.t-dialin.net) joined #forth
09:07:40 --- quit: edrx (Read error: 113 (No route to host))
09:12:34 --- join: arke (n=chris@pD9E068B7.dip.t-dialin.net) joined #forth
09:12:34 --- mode: ChanServ set +o arke
09:14:14 --- quit: ecraven ("bbl")
09:28:04 --- quit: ygrek (Remote closed the connection)
09:30:23 --- join: ygrek (i=user@gateway/tor/x-496f41e8c5cb4f02) joined #forth
10:03:59 --- join: iano (n=iosgood@gw-pdx.extensis.com) joined #forth
10:16:49 --- join: JasonWoof (n=jason@c-71-192-30-169.hsd1.ma.comcast.net) joined #forth
10:16:49 --- mode: ChanServ set +o JasonWoof
10:21:33 <Quartus_> hey jw
10:36:14 <JasonWoof> hi :)
10:39:25 --- quit: Crest ("Leaving")
11:27:08 --- join: Crest (n=crest@p54896FA4.dip.t-dialin.net) joined #forth
11:57:10 --- join: arke_ (n=chris@pD9E0721E.dip.t-dialin.net) joined #forth
12:12:29 --- quit: arke (Read error: 110 (Connection timed out))
12:14:28 --- quit: ygrek ()
12:37:16 --- join: mark4 (n=mark4@70.102.202.162) joined #forth
12:40:34 --- quit: madwork (Read error: 54 (Connection reset by peer))
12:47:11 --- quit: mark4 ("Leaving")
13:31:43 --- quit: timlarson_ (Read error: 104 (Connection reset by peer))
13:36:48 --- join: madwork (n=foo@204.138.110.15) joined #forth
13:39:02 --- quit: Quartus_ (Read error: 104 (Connection reset by peer))
14:00:36 --- nick: arke_ -> arke
14:00:36 --- mode: ChanServ set +o arke
14:10:19 --- join: edrx (n=Eduardo@200.217.105.18) joined #forth
14:13:12 --- quit: tathi ("leaving")
14:48:54 --- quit: Zarutian (Read error: 104 (Connection reset by peer))
14:49:15 --- join: Zarutian (n=Zarutian@194-144-84-110.du.xdsl.is) joined #forth
15:22:21 --- join: slava (n=slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
15:22:21 --- mode: ChanServ set +o slava
15:30:53 <Quartus> hey slava
15:30:56 <slava> hey
15:31:02 <Quartus> werty
15:31:05 <Quartus> werty werty werty
15:31:56 <slava> Then you may find strong drugs pleasant , also ? 
15:32:22 <slava> pot, kettle, black
15:32:41 <Quartus> I think he would find strong drugs useful.
15:37:31 --- quit: edrx (No route to host)
15:37:50 <crc> good evening
15:37:57 <Quartus> Hi crc.
15:46:26 <Quartus> What's new?
15:46:34 <crc> again, nothing much
15:47:15 <Quartus> Again?  ("Watch me pull a rabbit out of my hat!")
15:56:46 --- quit: Crest (Read error: 110 (Connection timed out))
16:19:41 --- quit: Zarutian (Read error: 104 (Connection reset by peer))
16:19:50 --- join: Zarutian (n=Zarutian@194-144-84-110.du.xdsl.is) joined #forth
16:21:26 <iano> yet another integer sqrt: http://paste-bin.com/4015
16:43:36 <JasonWoof> : sqrt-closer 2dup / over - 2 / ;
16:43:37 <JasonWoof> : sqrt 1 begin sqrt-closer dup while + repeat drop nip ;
16:44:35 <JasonWoof> from: http://jasonwoof.org/sqrt.fs
16:44:46 <iano> which is faster?
16:46:26 <JasonWoof> dunno, I'm not farmilliar with the method in your paste
16:46:32 <JasonWoof> time it
16:49:09 <JasonWoof> mine is a direct translation of the newton method
16:49:15 --- quit: Quartus (Read error: 110 (Connection timed out))
16:49:38 <JasonWoof> for maximum speed and minimal readability, inline sqrt-closer
16:50:14 <iano> is it guaranteed to converge?
16:50:38 --- join: Quartus (n=neal@CPE0001023f6e4f-CM013349902843.cpe.net.cable.rogers.com) joined #forth
16:50:38 --- mode: ChanServ set +o Quartus
16:54:12 <JasonWoof> : foo 10000 1 do i sqrt . loop ; foo
16:54:14 <JasonWoof> that went well
16:54:54 <JasonWoof> interestingly, it will sometimes go into an infinite loop if you replace the "2 /" with "2/"
16:55:16 <JasonWoof> I was very suprised to discover that "2/" gave different results than "2 /" for some inputs
16:55:26 <JasonWoof> most notably (and problematically) -1
16:56:32 <Quartus> difficult to take the square root of a negative number
16:56:44 <slava> i presume 2/ is a right shift
16:56:52 <slava>  -1 sqrt .
16:56:53 <slava> C{ 0.0 1.0 }
16:57:45 <Quartus> 2/ is a high-bit-preserving right shift
16:57:57 <JasonWoof> Quartus: I'm not attempting that. the difference between the most recent guess (squared) and the actual square is sometimes negative
16:58:04 <JasonWoof> right
16:58:41 <JasonWoof> I posted a different version that uses herkforth style, and someone submitted the version I posted above with the WHILE loop in it
16:59:06 <JasonWoof> except they replaced "2 /" with "2/" in sqrt-closer, which broke the algorithm for some inputs
17:02:10 <JasonWoof> as far as speed... I suspect that my algorithm is faster for small inputs
17:02:27 <JasonWoof> hmm...
17:02:35 <JasonWoof> now I'm curious how many iterations it takes for various inputs
17:06:24 <JasonWoof> up to 100, no more than 6 iterations, up to 10,000 no more than 10
17:08:30 <JasonWoof> up around 2,000,000,000 it takes 19 or 20 iterations
17:08:46 <iano> nice, that is quite efficient
17:11:10 <JasonWoof> the rounding is decent too
17:12:35 <JasonWoof> it mostly rounds to the nearest
17:13:13 <JasonWoof> except it has a slight tendancy to round up when it's close, like 5.47
17:14:05 <Quartus> I missed the paste I think.
17:14:14 <JasonWoof> the other nice thing, is that when you raise the operand, you always seem to get a higher result
17:14:31 <JasonWoof> : sqrt-closer 2dup / over - 2 / ;
17:14:35 <JasonWoof> : sqrt 1 begin sqrt-closer dup while + repeat drop nip ;
17:15:11 <JasonWoof> Quartus: and iano pasted this link: http://paste-bin.com/4015
17:15:33 <iano> they are also here: http://wiki.forthfreak.net/index.cgi?SquareRoot
17:19:32 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-130-046.pools.arcor-ip.net) joined #forth
17:19:38 <Quartus> ta
17:19:44 <iano> yours is for i > 0;  0 sqrt 1 =
17:22:53 <JasonWoof> true
17:24:55 <JasonWoof> here's a nice word to test your sqrt implementations: : sqrt-test 2000000000 1 do i dup . sqrt . cr i 100 / 1+ +loop ;
17:25:59 --- quit: iano ()
17:26:26 <Quartus> you shocked him
17:30:29 --- join: imaginator (n=George@georgeps.dsl.xmission.com) joined #forth
17:34:49 <JasonWoof> hi imaginator
17:35:08 <imaginator> hi JasonWoof 
17:35:40 <imaginator> so what's new in the Forth world?
17:36:35 <JasonWoof> just found out why I was getting reports that my forth SQRT word posted on my wiki had a bug
17:36:48 <JasonWoof> someone posted an ANS variant below mine, and that one had a bug
17:37:07 <Quartus> bastards!
17:37:07 <JasonWoof> they replaced "2 /" in mine with "2/" and didn't bother testing it much
17:37:32 <JasonWoof> with that change, it hangs with inputs such as 3, 8, 15, etc
17:38:50 --- quit: Snoopy42 (Read error: 110 (Connection timed out))
17:39:03 --- nick: Snoopy_1711 -> Snoopy42
17:39:59 <imaginator> that seems to happen on wikis.  sometimes you get improvements, and sometimes bugs.
17:40:31 <JasonWoof> yeah
17:40:53 <JasonWoof> I would have figured it out months ago if I knew it was in that variant
17:41:16 <JasonWoof> ahh well, progress is progress, no matter how slow
17:42:36 <imaginator> I should keep that in mind with my big NexX project.  I feel like it's going very very slowly, but some day I'll replace X for all my uses at least, and games.
17:44:24 <slava> nexx?
17:44:44 <Quartus> xell?
17:45:54 <Quartus> Why replace X?
17:46:55 <imaginator> slava: it's my replacement for X11 that currently runs on top of NetBSD, though OpenBSD should work too.  It has a wscons mouse driver, a VESA video driver, window rotation, full 32-bit colors with alpha blending, an X11 test server.
17:47:12 <slava> x11 compatible?
17:47:12 <Quartus> to what end? To eventually reproduce X11?
17:47:18 <imaginator> Quartus: because I hate X
17:47:31 <Quartus> You're creating a clone.
17:47:51 <slava> any of you guys heard of NeWS?
17:47:54 <imaginator> slava: it will be eventually.  I'm going to hack Xvfb to allow me to embed X11 windows
17:47:55 <slava> clone that instead.
17:48:05 <imaginator> Quartus: it sure as hell isn't a clone
17:48:05 <Quartus> Why create a duplicate of something you hate?
17:49:31 <Quartus> Is there another word you prefer that means functionally-identical reproduction?
17:50:14 <imaginator> all an application sees is an RGBA buffer, the client-side libraries such as nx_text (based on FreeType) render into that buffer.   Each RGBA buffer is shared with the server over a RAM-based file system, which the client and server mmap.  Some protocol requests go over sockets and file system fifos.
17:50:33 <slava> how does it compose window contents onto the screen? opengl?
17:51:10 <imaginator> just basic C code.  as there aren't good OpenGL drivers yet for my Nvidia card that run in NetBSD it's rather hopeless to target OpenGL
17:51:23 <imaginator> but that doesn't make it slow.  it's darn fast
17:52:40 <imaginator> for instance nexx_server which runs under X11 with the Xv (X video extension) is really very responsive.  Even with many rotated windows, because of my treecache design
17:53:29 <imaginator> then the nexx_vbe_server gets about 40 FPS for second, which isn't bad.  It's quite usable with the test clients that have rotation and complex alpha blending.
17:54:23 <imaginator> I'm using SSE asm for the VESA/vbe driver
17:54:56 <imaginator> I also set an MTRR register to optimize the transfers.
17:55:57 <imaginator> hehe s/FPS for second/FPS/
17:57:20 <JasonWoof> what's vbe again?
17:57:31 <imaginator> VESA BIOS Extensions 
17:57:41 <JasonWoof> oh
17:57:44 <JasonWoof> not something I have
17:57:47 <JasonWoof> ok
17:57:51 <imaginator> you probably do, without knowin git
17:57:53 <JasonWoof> what's the point of rotated windows?
17:57:56 <JasonWoof> I've got a mac
17:58:01 <imaginator> ah
17:58:40 <JasonWoof> you can very easilly get a memory address for a framebuffer
17:58:41 <imaginator> well, the rotated windows are fun, and you can rotate objects on the screen, say for games, such as wheels in a game.
17:58:43 <JasonWoof> but it's slow as hell
17:59:01 <JasonWoof> so it's not just in increments of 90 degrees then
17:59:13 <imaginator> no, it's full [-360,360] 
17:59:27 <imaginator> with 1 loss per rotation.  it rotates a copy of the original
17:59:28 <JasonWoof> ok, so you have a sprite engine built into the window system
17:59:31 <JasonWoof> does it do scaling too?
17:59:38 <imaginator> yes, I plan to have that too
17:59:44 <imaginator> I've got the code in there, but it doesn't do anything yet
18:00:10 --- mode: Quartus set -b *!*@chello062178085149.1.12.vie.surfer.at
18:00:17 <Quartus> (added to auto)
18:00:44 <JasonWoof> it makes sense to me to have that stuff in a sprite engine that does things within a window
18:01:04 <JasonWoof> but I can't think why you'd ever want to do that to the whole window
18:01:27 <JasonWoof> perhaps for visual glits like Apple's "genie" minimize effect
18:01:35 <imaginator> well, it's interesting.  I've shared the prototype with some people and they seem to enjoy rotating images, and tinkering with it.
18:01:53 <JasonWoof> yeah, glitz is fun for 10 minutes
18:02:01 <Quartus> JasonWoof, it's the newest gimmick in window managers, to turn windows edgewise.  It comes of excess graphics capacity combined with a frantic need to do something different.
18:02:20 <JasonWoof> but in my experience, once the novelty wears off, it just gets annoying and I want it to skip the animation and just do what I said
18:02:40 <imaginator> well, if you could have anything in a window system, what would you have?
18:02:45 <JasonWoof> Quartus: yep, gimmick is the word
18:02:51 <Quartus> imaginator, fewer gimmicks.
18:02:53 <imaginator> I know I want scaling.  That's definitely useful for people with visual problems.  
18:03:00 <JasonWoof> imaginator: ion3 works beautifully for me
18:03:18 <JasonWoof> the most important feature for me in a window manager is that it automatically uses the whole screen for ALL windows
18:03:26 <JasonWoof> except where I have to make exceptions
18:03:31 <imaginator> interesting
18:03:33 <JasonWoof> because of annoying gui design
18:03:39 <JasonWoof> like the gimp and xmms
18:03:40 <imaginator> I once wrote a WM called Panache that was like that
18:03:46 <imaginator> transient windows?
18:03:48 <JasonWoof> there are quite a few
18:04:02 <JasonWoof> transient windows are irritating no matter what window manager you have
18:04:17 <imaginator> well they're good in the WM I wrote called Whim
18:04:21 <imaginator> for the most part
18:04:39 <JasonWoof> transient windows are generally used for dialog boxes
18:04:41 <JasonWoof> which are always annoying
18:04:50 <imaginator> it positions based on the XID resource allocated to a window, so it groups automatically (if wanted)
18:05:03 <JasonWoof> or, they are used when people are too lazy to make it so you can use the app while you have the preferences panel open
18:05:19 <imaginator> how would you implement a dialog-like box?   would you have a constant keyboard input area?
18:05:32 <JasonWoof> I wouldn't
18:05:35 <JasonWoof> I hate them
18:05:37 <imaginator> what would you do?
18:05:48 <JasonWoof> read The Humane Interface
18:05:49 <imaginator> integrate the state into the GUI?
18:06:04 <imaginator> I tried Archy and I've read some of the author's works.
18:06:05 <JasonWoof> state gets you into trouble
18:06:32 <JasonWoof> I don't know if you can go completely state-less
18:06:44 <JasonWoof> since nowadays you need to work with things other than text
18:06:52 <imaginator> you can't, and the author goes into that IIRC.  some things he considered to be not-state that I would.
18:07:18 <JasonWoof> I like the test "can you start typing without looking at the monitor first?"
18:07:33 <imaginator> Archy was really disappointing to me.  I started out with a great goal, but last I saw it's been failure.
18:07:37 <slava> yeah, its great for those times when your monitor just won't work, or you're unable to turn your head.
18:07:48 <imaginator> lol
18:08:12 <JasonWoof> slava: it's great for any time when you want to think about something besides the state of your software
18:08:19 <Raystm2> Hi guys. Question for Quartus. Come in Quartus, are you recieving?
18:08:26 <JasonWoof> or if you can't type, and have to look at the keyboard
18:08:32 <Raystm2> or receiving even?
18:08:33 <Quartus> hey Raystm2
18:08:43 <Quartus> distant ships smoke on the horizon
18:08:49 <JasonWoof> I can't touch-type qwerty anymore. I can't tell you how many times on other people's computers I've opened a web-browser, and started typing a url.
18:09:01 <JasonWoof> when I look up, I found that it's stopped accepting keystrokes after a few characters
18:09:02 <Quartus> You have ghettoized your typing skills?
18:09:22 <JasonWoof> I switched to dvorak in highschool
18:09:33 <Quartus> that'll do it
18:09:35 <JasonWoof> haven't had problems with wrist or finger strain since
18:09:40 <imaginator> I'm used to the pdksh vi-mode, so when I use entry boxes I find I expect the same kind of behavior
18:10:27 <JasonWoof> I'm still a bit torn on how best to handle modes/modelessness
18:10:38 <JasonWoof> I really really like vim
18:10:46 <Quartus> I have never had problems with wrist and finger strain.  I expect your typing posture when you used qwerty was problematic.
18:10:48 <Raystm2> Quartus: Currently parusing the Wikipedia Forth article, specifically, a language comparison with python that is not really a good comparison for what the associated text represents... ( build up over -- question follows) 
18:11:04 <JasonWoof> iirc I used emacs for over a year, and though I screw up the modes a lot, I still vastly prefer vim
18:11:11 <Quartus> Raystm2, I haven't seen the article correctly.  How bizarrely pointless to include a python comparison.
18:11:12 <JasonWoof> so I definitely see the advantage of modes
18:11:27 <Raystm2> Where the heck is that assume c compare at?
18:11:35 <Raystm2> The one on your site. 
18:11:35 <Quartus> assume c compare?
18:11:39 <JasonWoof> Quartus: perhaps. and small hands doesn't help
18:11:54 <Raystm2> Been searching here for a while. I know I saw it there, right?
18:11:56 <Quartus> oh, the Palm coding one.  Hang on
18:12:00 <Raystm2> ya ya.
18:12:06 <imaginator> JasonWoof: I wrote my own editors, though I still use nvi.  my editors save automatically, some have syntax highlighting, or brace matching..  
18:12:08 <Quartus> http://quartus.net/forth  the link is slightly down the page
18:12:25 <imaginator> JasonWoof: one I use a lot now has indentation-based highlighting
18:12:29 <Raystm2> I'll add the link to the discussion and see if anybody wants to include that instead..
18:12:33 <JasonWoof> Quartus: I don't think my posture is excelent now. I have a nice big callous on my left arm just below the elbow where I mash it into the edge of my desk
18:13:14 <Quartus> JasonWoof, it's more about where the heels of your hands rest, and how you hold your arms relative to your wrists and hands
18:13:19 <Quartus> applies to any keyboard layout
18:13:24 <Raystm2> Tape a pad to that part of the desk, then heal. 
18:13:38 <JasonWoof> Quartus: that makes sense
18:13:57 <JasonWoof> Quartus: I think I do angle my hands in (at the finger end) more now
18:14:09 <Quartus> you could switch back to qwerty without any strain, and deghettoize yourself
18:14:14 <JasonWoof> kinda like the "ergonomic" keyboards help you do
18:14:17 <Quartus> Just be mindful of your hand positioning.
18:14:27 <JasonWoof> QWERTY is a pain
18:14:50 <JasonWoof> I much prefer having the common letters on the home row
18:15:03 <Quartus> The pain for me would come from being unable to use 99.9% of the world's computers.
18:15:05 <imaginator> this is one of my weird editors http://www.xmission.com/~georgeps/tmp/ed6.png
18:15:06 <JasonWoof> I don't use other people's computers much
18:16:19 <JasonWoof> imaginator: interesting pink boxes
18:16:45 <JasonWoof> that's cool you've written editors
18:16:47 <Raystm2> Quartus: I don't think that learning a second keyboard layout, and switching between them often is a real problem. ( re. 99.9%)
18:16:51 <JasonWoof> I'm too lazy
18:17:06 <JasonWoof> I know people who can switch
18:17:19 <Raystm2> I do it all the time.
18:17:26 <Quartus> Raystm2, my comparison page shows what Palm advocates as a programming model vs. what Quartus Forth abstracts more simply.  It isn't really a comparison between the two languages, as much as a comparison of the two approaches.
18:17:27 <imaginator> this is the original (with blue) http://www.xmission.com/~georgeps/implementation/software/Fed_Builder/ed2.png
18:17:29 <JasonWoof> it might be very hard for me though, even if I put my mind to it and practice
18:17:36 <JasonWoof> I'm "severely dyslexic"
18:17:46 <Quartus> Raystm2, I have never met anyone who was proficient at more than one layout.
18:17:48 <Raystm2> Sort of like learning variations of musical instruments.
18:18:27 <Raystm2> True, I don't think that i'm completely proficient in both, I have much less practice with one.
18:18:39 <JasonWoof> mmmm blue
18:18:54 <Quartus> In fact I believe switching between layouts will hamper proficiency in any single one.
18:19:00 <JasonWoof> seems to me the indentation hilighting would be more useful if the boxes spanned vertically, instead of horizontally
18:19:29 <Raystm2> I'm dyslexic, but not severely. I was able to realize it myself, early in my schooling, and figured out how to compensate on my own.
18:19:53 <JasonWoof> imaginator: funky language
18:20:38 <imaginator> yah, it's Tcl
18:20:55 <JasonWoof> Raystm2: yeah, I do pretty well, but some things like dealing with multiple keyboard layouts seem like more trouble than their worth
18:21:08 <JasonWoof> I usually hate the GUIs on other people's computers anyway
18:21:21 <JasonWoof> and windows doesn't even come with an ssh client anyway
18:21:34 <Quartus> putty is your friend
18:21:55 <JasonWoof> and even if it did, I'm don't really like the idea of typing my shell passwords into such a virus-magnet OS
18:22:44 <JasonWoof> yeah, putty is good
18:22:47 <Quartus> you can always do what I do, carry your keys and putty around on a USB stick
18:23:04 <Quartus> Then you're not typing in your shell password.
18:23:43 <JasonWoof> yeah, then I'm just connecting my keys to the infected box
18:23:52 <JasonWoof> nah, I prefer just taking my laptop with me when I need it
18:24:15 <imaginator> I found that with some large projects I used to get lost a lot in editors like vi, even with ctags.  I'd also find myself creating new files, or moving things around a lot.
18:24:18 <JasonWoof> if I didn't have that, I'd consider carying my FSF Associate Membeship Card
18:24:20 <imaginator> Now I don't do that anymore
18:24:29 <Quartus> the keys themselves are password-protected.
18:24:40 <JasonWoof> (which is a bootable business-card CD)
18:25:15 <JasonWoof> tags are awesome
18:25:21 <JasonWoof> although I must admit that I don't use them that much
18:25:27 <JasonWoof> Quartus: oh, right
18:26:07 <Quartus> You'd have to have one seriously-well-tailored virus to snag your credentials.
18:26:17 <imaginator> JasonWoof: you mentioned the scaling of windows... This is a picture of editing the code for that: http://www.xmission.com/~georgeps/tmp/Fed_Builder.png
18:26:20 <JasonWoof> yeah, I forgot about encrypting the keys
18:26:27 <Quartus> Expensive.  Cheaper to drag you into a van and beat your passwords out of you.
18:30:12 <JasonWoof> imaginator: looks good. except it would probably speed things up (in the case where there's already a treecacheimage of the correct dimentions) to just blank the existing treecacheimage
18:30:25 <JasonWoof> or is blanking even nessesary?
18:30:44 <imaginator> hmm, good idea.  
18:30:59 <JasonWoof> if you're blitting over the whole thing, you wouldn't even need to blank it
18:34:16 <JasonWoof> imaginator: did you write that editor (fed builder)?
18:36:19 <imaginator> yes
18:36:27 <JasonWoof> looks cool
18:36:30 <imaginator> thanks
18:37:02 --- join: nighty^ (n=nighty@61.237.235.72) joined #forth
18:37:20 <JasonWoof> why don't you have an editable prototype?
18:37:46 <imaginator> what do you mean?
18:38:35 <JasonWoof> I'm just curious why you decided to split the prototype into two fields, and leave out the name completely from the input boxes
18:38:46 <JasonWoof> (two fields being return type and arguments)
18:39:25 <JasonWoof> man, I've been spacy ever since my nap
18:41:14 <imaginator> ah, well the name of the function is created by right-clicking in the listbox and doing Add Function
18:41:40 <imaginator> there's also a Rename Function via right-click too
18:41:46 <JasonWoof> right
18:41:52 <JasonWoof> that's what I was hoping
18:42:15 <imaginator> I split them that way, because sometimes it doesn't all fit visually otherwise, and this was logical
18:42:24 <JasonWoof> ahh
18:42:42 <JasonWoof> is your screen really that shape?
18:42:45 <imaginator> yes
18:42:53 <imaginator> 1024x1280
18:43:13 <imaginator> at least my left screen is
18:43:31 <imaginator> hey, thanks again for noticing that optimization.  It seems to have improved performance
18:43:55 <JasonWoof> that image is 856x1008
18:44:14 <JasonWoof> np
18:45:11 <JasonWoof> your x replacement thing sounds very cool
18:45:25 <imaginator> thanks
18:45:38 <JasonWoof> I plan to have my next desktop computer run FreeBSD
18:46:13 <JasonWoof> but that'll probably wait until this one breaks and/or I have money
18:46:57 <imaginator> yah, I mostly run NetBSD, and the window system is easier to build the lower layers for in netbsd or openbsd, because there's a generic mouse, and keyboard layer called wscons.
18:47:23 <imaginator> I think FreeBSD might have something like that, but I'm not sure
18:48:16 <imaginator> otherwise I'd have to support various drivers (at least based on my study of the other code in X11 for Linux and other systems)
18:48:43 <JasonWoof> hmmm... netbsd would probably run on my computer
18:48:52 <JasonWoof> I think tathi was experimenting with that
18:50:56 <JasonWoof> I guess I'm afraid to use it, because I don't think the packages will have gotten enough testing on ppc
18:51:15 <imaginator> hmm that's a possibility
18:51:29 <JasonWoof> I don't know if that's really a problem
18:51:36 <imaginator> pkgsrc generally has a lot of activity
18:51:59 <imaginator> but I'm not sure how many ppc users there are
18:52:01 <JasonWoof> does netbsd have something like portupgrade?
18:52:12 <imaginator> I usually run make update 
18:52:33 <imaginator> I don't know if that's like portupgrade, but I assume it's similar
18:53:10 <JasonWoof> on FreeBSD there's a command to tell you everything that can be upgraded: portversion -v -L =
18:53:33 <JasonWoof> and another to upgrade them all, saving a tarbal (or something) of the old version of each: portupgrade -b -a
18:54:25 <imaginator> I'll check out the pkgsrc docs to see if there's a pkgtools package that does that
18:55:16 <JasonWoof> I had a hell of a time finding much documentation on how to use the package manager on FreeBSD
18:55:39 <JasonWoof> I found a bunch of tutorials on getting it installed, and using it to install something
18:56:01 <JasonWoof> but I couldn't find docs on doing other basic tasks, like uninstalling something, upgrading, finding out what package installed a given file, etc
18:56:44 <imaginator> http://www.netbsd.org/Documentation/pkgsrc/
19:00:21 <imaginator> I usually update packages when audit-packages reports something important enough to warrant it.  I'm behind a firewall, but I have a few ports exposed.
19:01:43 <JasonWoof> yeah
19:01:48 <JasonWoof> I run a server
19:01:56 <JasonWoof> this time around I decided to keep everything up to date
19:02:39 <JasonWoof> I update every week or so, or whenever portaudit gets excited, whichever comes first
19:03:41 <JasonWoof> most of it probably doesn't matter, but I like having spamassassin as up to date as it can be
19:04:41 <JasonWoof> I'm glad to hear NetBSD has an audit thing
19:05:08 <JasonWoof> it is amazingly useful
19:05:31 <JasonWoof> now that I've heard of it, I can't imagine why I've never heard of such a thing for linux
19:06:00 <imaginator> with Linux I also use pkgsrc, so the same vulnerability list works and I can also use audit-packages
19:06:27 <JasonWoof> do you need solaris for that?
19:06:52 <imaginator> no
19:08:51 <imaginator> www.pkgsrc.org has the details on all of the supported systems.  I last used it with Slackware
19:09:02 <JasonWoof> I'm still a little boggled by the idea of using a package manager from another OS
19:10:05 <JasonWoof> but it happens
19:10:30 <JasonWoof> gentoo's portage is working on Mac OS X now I think
19:10:38 <imaginator> well, they recognized a need to have a portable package system, and it gets them more developers, so it's good.
19:11:14 <JasonWoof> oh, sure
19:11:19 <JasonWoof> it's a good idea
19:11:25 <JasonWoof> I just didn't realize it would really work
19:11:42 <JasonWoof> because I thought of packages as the system-specific stuff required to install the given piece of software
19:12:30 <JasonWoof> but the more thoughts and experience I have, the more I see that it's mostly stuff specific to the piece of software, rather than the system
19:16:41 <JasonWoof> I have this idea that I hate writing user interfaces
19:16:51 <JasonWoof> which may be the primary thing stopping me from finishing Fronds
19:16:58 <imaginator> what's Fronds?
19:17:08 <JasonWoof> my IDE/OS
19:17:11 <imaginator> cool
19:17:18 <JasonWoof> a little self-contained programming environment
19:17:18 <imaginator> is it Forth powered? 
19:17:22 <JasonWoof> kinda
19:17:29 <JasonWoof> it's very forth-like
19:17:31 <JasonWoof> but it's not text based
19:17:45 <JasonWoof> or, at least not based on text files
19:17:49 <JasonWoof> all word names are text, and you type
19:17:56 <JasonWoof> but there's no parsing at compile time
19:18:11 <JasonWoof> that kind of stuff is in the editor
19:19:02 <JasonWoof> I've got it displaying code well, and you can browse around
19:19:20 <JasonWoof> I just need to finish the editing functionality
19:19:52 <imaginator> sounds interesting
19:20:14 <JasonWoof> yeah
19:20:21 <JasonWoof> it's my little computer fantasy
19:20:43 <JasonWoof> it basically came out of my thoughts like "I wish computers worked like this..."
19:21:11 <JasonWoof> my last forth (herkforth) compiles directly to a ppc binary
19:21:20 <imaginator> ah, I remember your herkforth
19:21:44 <JasonWoof> I was pretty dissolusioned when apple decided to drop ppc
19:21:49 <slava> why?
19:22:00 <TreyB> Effort invested, I'd guess.
19:22:07 <JasonWoof> PPC was supposed to be the processor of the future
19:22:12 <slava> you shouldn't target a compiler at a single cpu.
19:22:19 <JasonWoof> iirc microsoft said they'd switch to it
19:22:22 <JasonWoof> then they didn't 
19:22:32 <JasonWoof> then apple switched to x86 and microsoft is selling ppc boxes
19:22:35 <slava> takes me about a week of work to port factor to a new cpu. *shrug*
19:23:11 <JasonWoof> I don't like dealing with differences between people's machines
19:23:28 <JasonWoof> so this time around I dediced to code a simple virtual machine
19:23:37 <JasonWoof> that works the way I wish computers did
19:24:57 <slava> right, but you have to port your VM to every platform too.
19:25:59 <JasonWoof> then I can program for that, and program (to quote some cvs documentation) from the comfort of my computer generated reality
19:26:32 <JasonWoof> I wrote my VM in C using just SDL
19:26:38 <slava> well sure, any portable program should have a well-defined portability layer.
19:26:42 <JasonWoof> it should be very easy to get working on most systems
19:26:51 <slava> if you want portability, your high level code shouldn't use inline assembly or directly make system calls
19:26:53 <JasonWoof> besides, it's simple, and therefor easy to rewrite
19:27:02 <JasonWoof> and it's well-documented, so other people could do that
19:27:11 <slava> but you might find sdl is too limiting.
19:27:15 <slava> no clipboard access or multiple windows.
19:27:24 <JasonWoof> we've had this conversation before
19:27:33 <slava> i know.
19:27:41 <JasonWoof> I hate multiple windows, but clipboard access would be nice
19:28:06 <slava> i wonder if there's an ui paradigm that will allow me to make effective use of my screen without multiple windows.
19:28:12 <slava> some kind of tiling perhaps, but it could get painful
19:28:25 <imaginator> slava: I think that's what ion is like, and my Panache was like that too
19:28:39 <slava> but at 2560x1600?
19:28:43 <JasonWoof> slava: lots of programs manage screen space
19:28:57 <JasonWoof> slava: the major browsers, gqview
19:29:46 <JasonWoof> slava: sheesh, well if you're screen's that big you'll need to split in vertically
19:31:31 --- join: iano (n=iosgood@host-226-157.dhcp.pdx.edu) joined #forth
19:31:36 <JasonWoof> wb
19:32:28 <slava> most of my screen is taken up by three vertical editors but i have a ton of overlapping terminals, etc
19:32:32 <slava> i don't want them all to be tiled
19:40:12 <JasonWoof> good for you
19:41:57 <JasonWoof> I think my screen is a nice size for windows to be
19:42:10 <JasonWoof> I don't see paying a thousand dollars for more pixel that my wm doesn't use
19:42:21 <JasonWoof> s/pixel/pixels/
19:42:53 <JasonWoof> I have one "desktop"/"workspace" that has the screen split up into tiles for the gimp
19:43:07 <JasonWoof> and one funky one that lets xmms do it's stupid thing
19:43:22 <JasonWoof> but asside from that, everything is full screen, which is a nice size for me
20:24:20 --- quit: nighty^ ("Disappears in a puff of smoke")
20:42:14 --- quit: iano ()
20:48:25 --- join: iano (n=iosgood@host-226-157.dhcp.pdx.edu) joined #forth
21:35:44 --- join: Quartus_ (n=Quartus_@209.167.5.2) joined #forth
21:35:44 --- mode: ChanServ set +o Quartus_
22:10:16 <JasonWoof> vim is my personal hero
22:10:46 <JasonWoof> the vast majority of the time when I want it to do something, it has that feature
22:10:51 <JasonWoof> and it's easy to get it to work the way I want
22:11:26 <JasonWoof> eg just now I decided to see if there was a commandline switch to display all files specified in the commandline at once (by splitting the screen)
22:11:30 <JasonWoof> and sure enough -o
23:08:21 --- quit: iano (Read error: 113 (No route to host))
23:59:59 --- log: ended forth/07.02.28