00:00:00 --- log: started forth/06.11.10
00:03:22 --- join: Cheery (n=Cheery@a81-197-54-146.elisa-laajakaista.fi) joined #forth
00:33:23 --- nick: arke_ -> arke
03:24:31 --- join: zpg (n=user@smaug.linux.pwf.cam.ac.uk) joined #forth
03:24:55 <zpg> morning
03:25:03 <JasonWoof> mornin
03:25:27 <JasonWoof> how's things?
03:25:30 <zpg> hi JasonWoof 
03:25:39 <zpg> not bad thanks, a little overtired but that aside, going well. yourself?
03:25:46 <JasonWoof> I'm just cooling off from being mad at sqwebmail
03:25:52 <zpg> heh
03:25:57 <zpg> how're the projects coming along?
03:26:07 <JasonWoof> don't see why they can't have documentation
03:26:09 <JasonWoof> and a config file
03:26:45 <JasonWoof> I haven't done crap on my fun programming projects
03:26:46 <zpg> i think an old email address of mine had a web interface based on that (unless i'm confusing names)
03:26:57 <JasonWoof> been putting lots of energy into moving my hosting stuff to a new server
03:27:07 <JasonWoof> I planned on finishing that around now
03:27:15 <JasonWoof> but I got distracted trying to fix sqwebmail
03:27:20 <JasonWoof> I might put it off until tomorrow night
03:27:46 <JasonWoof> I switched everybody over to squirrelmail when I moved them to the new server
03:27:51 <JasonWoof> but one of my clients freaked out
03:27:57 <JasonWoof> so I installed sqwebmail
03:28:09 <JasonWoof> but didn't bother fixing the paths for the icons and css file
03:28:13 <zpg> data loss?
03:29:02 <JasonWoof> which I thought would be fine for a while, but it had two problems: 1) IE wouldn't display the alt tags for the icons 2) it displayed one useful bit of information in black on black
03:29:18 <JasonWoof> no data loss
03:29:26 <zpg> just irritation.
03:30:17 <JasonWoof> now it's 6:30am and I haven't even gotten the mailing list software installed, let alone set up
03:30:41 <JasonWoof> I'm tired, so I'm afraid if I just go ahead and do it I'll do somethething stupid
03:33:26 <zpg> pulled an all nighter?
03:33:55 <zpg> heh, http://dec.bournemouth.ac.uk/forth/euro/intro.html ==> "Delegates from all parts of Europe including Eastern Europe and the Former Soviet Union are common"
03:34:27 <zpg> erm, okay -- that's the first sign of tiredness. missing out on the word Former.
03:42:59 <JasonWoof> yeah, I like to move websites in the middle of the night
03:43:12 <JasonWoof> then I sleep for the morning and some of the afternoon
03:43:21 <JasonWoof> kinda sucks now that the sun goes down at 4:30 though
03:43:25 <JasonWoof> I don't see it much
03:44:06 <JasonWoof> seems like it went from 7pm to 4:30pm pretty quick
03:44:44 <zpg> yeah, that's a common occurrence here too. getting more overcast by the day of course, though yesterday was pretty chipper.
03:45:14 <JasonWoof> ok, I'm definitely going to sleep
03:45:18 <JasonWoof> none of this installing crap
03:45:40 <JasonWoof> for my sanity, I need to get this thing done
03:45:44 <JasonWoof> but not right now
03:45:46 <JasonWoof> now I need sleep
03:46:13 <JasonWoof> heh, also for my sanity :)
03:47:54 <zpg> it would seem so.
03:47:54 * zpg winks
03:51:56 --- quit: JasonWoof ("off to bed")
03:52:33 <zpg> best be off too.
03:52:34 --- quit: zpg ("ERC Version 5.0.4 $Revision: 1.726.2.19 $ (IRC client for Emacs)")
04:11:52 --- nick: crest_ -> Crest
04:16:53 --- quit: virsys (Read error: 54 (Connection reset by peer))
04:31:26 --- join: vatic (n=chatzill@pool-162-84-178-20.ny5030.east.verizon.net) joined #forth
05:17:17 <vatic> Quartus: http://www.wulfden.org/PRC/
05:59:03 --- join: marble (n=glass@cpc1-bolt6-0-0-cust18.manc.cable.ntl.com) joined #forth
06:14:17 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth
06:18:28 --- join: erider (n=erider@unaffiliated/erider) joined #forth
06:19:50 --- nick: Raystm2 -> nanstm
06:22:58 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth
06:24:00 <Ray_work> Good morning, everybody! It's FRIDAY!
06:24:09 <erider> good morning 
06:27:56 <marble> good afternoon
06:32:01 <erider> hi marble
06:32:18 <marble> hey erider 
06:32:42 <erider> whats going to be the topic of the day 
06:33:47 <marble> still just reading stuff
06:33:51 <vatic> Ray_work: TGIF Buster!
06:34:32 <marble> Starting Forth and Thinking Forth are pretty damn good -- I would love to have in print
06:36:03 * erider hasn't got to Thinking Forth yet still on Forth Primer
06:49:46 <Ray_work> hi vatic :0
06:49:51 <Ray_work> hi marble.
06:49:56 <Ray_work> hi erider.
06:50:20 <vatic> Ray_work: hey yo!
06:50:23 <Ray_work> I've got Starting Forth, from Quartus.
06:50:37 <Ray_work> You could print out a Thinking...
06:54:55 <marble> it wouldn't be the same
06:55:01 <marble> say, are you actually at work?
06:55:35 <marble> (and does this mean an office, or from home?)
06:56:30 --- quit: Cheery (Read error: 104 (Connection reset by peer))
06:57:14 --- join: Cheery (n=Cheery@a81-197-54-146.elisa-laajakaista.fi) joined #forth
07:06:33 --- quit: erider (Read error: 110 (Connection timed out))
07:07:10 --- join: erider (n=erider@unaffiliated/erider) joined #forth
07:28:45 --- quit: erider (Read error: 110 (Connection timed out))
07:29:07 --- join: erider (n=erider@unaffiliated/erider) joined #forth
08:00:25 <Razor-X> How about Lulu for publishing?
08:14:31 <erider> hi all 
08:18:31 --- quit: vatic ("*poof*")
08:30:07 --- quit: virl (zelazny.freenode.net irc.freenode.net)
08:30:07 --- quit: warpzero (zelazny.freenode.net irc.freenode.net)
08:30:07 --- quit: ohub (zelazny.freenode.net irc.freenode.net)
08:30:07 --- quit: Quartus (zelazny.freenode.net irc.freenode.net)
08:30:22 --- quit: marble (zelazny.freenode.net irc.freenode.net)
08:30:23 --- quit: larsb (zelazny.freenode.net irc.freenode.net)
08:30:27 --- join: Quartus (n=trailer@CPE0001023f6e4f-CM013349902843.cpe.net.cable.rogers.com) joined #forth
08:30:27 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth
08:30:27 --- join: warpzero (n=warpzero@wza.us) joined #forth
08:30:27 --- join: ohub (n=oherrala@sikw1.oulu.fi) joined #forth
08:30:27 --- mode: irc.freenode.net set +o Quartus
08:30:41 --- quit: Ray_work (zelazny.freenode.net irc.freenode.net)
08:30:41 --- quit: cmeme (zelazny.freenode.net irc.freenode.net)
08:30:42 --- quit: Snoopy42 (zelazny.freenode.net irc.freenode.net)
08:30:42 --- quit: Crest (zelazny.freenode.net irc.freenode.net)
08:30:45 --- join: marble (n=glass@cpc1-bolt6-0-0-cust18.manc.cable.ntl.com) joined #forth
08:30:45 --- join: larsb (i=lars@1-1-14-10a.kt.gbg.bostream.se) joined #forth
08:31:03 --- quit: arke (zelazny.freenode.net irc.freenode.net)
08:31:03 --- quit: madwork (zelazny.freenode.net irc.freenode.net)
08:31:04 --- quit: ccfg (zelazny.freenode.net irc.freenode.net)
08:31:18 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth
08:31:18 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth
08:31:18 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-121-213.pools.arcor-ip.net) joined #forth
08:31:18 --- join: Crest (n=crest@p5489780A.dip.t-dialin.net) joined #forth
08:31:23 --- join: arke (n=Chris@pD9E06DE3.dip.t-dialin.net) joined #forth
08:31:23 --- join: madwork (n=foo@204.138.110.15) joined #forth
08:31:23 --- join: ccfg (n=ccfg@dsl-roigw1-fe8ade00-21.dhcp.inet.fi) joined #forth
08:44:54 <Crest> why did freebode-connect send me a ctcp version?
08:45:11 <Crest> *freenode-connect
08:45:19 --- join: virsys (n=virsys@or-71-53-68-118.dhcp.embarqhsd.net) joined #forth
08:48:30 <marble> large netsplit
08:49:41 <Crest> i just don't know what use the information a ctcp version is worth that for
08:51:32 <segher> it's the open proxy scanner
09:02:25 <Crest> yes but every one can set his response to a ctcp version request
09:08:20 <Ray_work> marble: ya I'm actually at an office away from the house.
09:09:28 <Ray_work> I'm in materials handling.
09:10:12 <Ray_work> Sorry for delay, I often get to where I actually have to work.
09:11:03 --- quit: erider (Read error: 110 (Connection timed out))
09:11:59 --- join: erider (n=erider@unaffiliated/erider) joined #forth
09:14:01 <marble> No worries -- I was just curious, is all
09:14:17 <marble> I only IRC'd at work during the lunch hour
09:18:47 --- join: zpg (n=user@user-514d7663.l2.c2.dsl.pol.co.uk) joined #forth
09:19:00 <Ray_work> hehe, I get a way from my desk on the lunch hour and don't IRC :)
09:19:08 <zpg> hi
09:19:16 <Ray_work> zpg howdy.
09:19:26 <zpg> greetings Ray
09:20:41 * Ray_work is so glad that it's Friday.
09:21:34 <Ray_work> oh marble: I'm also Raystm2 in these chats when I'm home. My wife changes my nick to nanstm when she's on.
09:25:45 <zpg> yep, the Friday angle has really improved the week. I can finally catch up on all that abbreviated sleep.
09:26:44 <Ray_work> <Homer><drooling>"mmm sleep"
09:27:36 <zpg> quite.
09:30:30 <zpg> Ray_work: what've you been up to, outside of work?
09:31:21 <Ray_work> zpg: out side of work I'm working on a few projects. 
09:31:25 <Ray_work> 1) RxChess
09:31:40 <Ray_work> 2) Palm ChuckBot the Cursor.
09:32:07 <Ray_work> 3) my FolEndar (pronounced folly ender) organizer invention
09:32:13 <zpg> what's ChuckBot when it's at home?
09:32:21 <Ray_work> 4) colorforth.info
09:32:31 <Ray_work> best example is...
09:32:46 <Ray_work> http://www.mtsu.edu/~untch/karel/  but for forth
09:34:09 <zpg> ah i see.
09:34:32 <zpg> dynamic languages are obviously seem better suited to Karel than the C(++) of the example code.
09:35:01 <zpg> haven't seen the .info site before; first comment, the colours blinded me
09:35:41 <Ray_work> Sorry, I've got a new page in the works going up very soon.
09:35:54 <Ray_work> It's supposed to look just like the colorForth environment.
09:36:45 <Ray_work> you haven't seen the site because i'm not finished or even a good start on it and I'm not currently advertising it till I do.
09:37:27 <zpg> ah ok.
09:37:36 <zpg> what's the status of these various projects then?
09:38:27 <Ray_work> RxChess, working, needs better board, better pieces, better def of abs, AI, and some sort of communication.
09:39:01 <Ray_work> Palm chuckBot: working, needs walls beepers and a header display.
09:39:24 <Ray_work> Needs to use the keys on the palm as well.
09:40:05 <Ray_work> FolEndar, Working, I use it daily again. Sent several units to friends for review. working on the manual.
09:40:28 <Ray_work> you know about colorforth.info. 
09:40:38 <Ray_work> I'm sure i'm missing something. 
09:40:40 <zpg> good stuff!@
09:41:17 <Ray_work> OH yeah I want to do Guitar instruction videos online in a private pay-for-cheaply-by-the-month subscription model.
09:41:54 <Ray_work> I can teach any beginner to play a favorite song in under 15 minutes, the rest builds from there.
09:42:59 <zpg> neat
09:43:00 --- join: crest_ (n=crest@p54897475.dip.t-dialin.net) joined #forth
09:43:09 --- quit: Crest (Read error: 145 (Connection timed out))
09:43:18 <zpg> so, you say your wife is a freenoder too?
09:44:33 <Ray_work> ya, but she's just sorta interested in the guys that i chat with because i'm always telling her about y'all.
09:44:53 <Ray_work> We have a private channel to keep up with eachother.
09:45:44 <Ray_work> She works the MidShift, we are often two ships passing in the night, iycmd.
09:46:35 <zpg> javol
10:03:25 <Ray_work> medication time
10:13:24 <zpg> hoorah
10:15:47 * Ray_work is diabetic and ADHD if you couldn't tell that last part.
10:20:53 --- nick: crest_ -> crest
10:21:15 --- nick: crest -> Crest
10:26:56 <zpg> you've mentioned the latter before; didn't know about the diabetes. which type?
10:34:28 --- join: forther (n=forther@c-67-180-209-27.hsd1.ca.comcast.net) joined #forth
10:35:24 <Ray_work> 2
10:35:49 <Ray_work> just found out myself in sept. I was 8x the norm.
10:35:58 <Ray_work> prob'ly for years.
10:37:55 <erider> hey Ray_work
10:38:59 --- join: snoopy_1611 (i=snoopy_1@dslb-084-058-108-180.pools.arcor-ip.net) joined #forth
10:40:19 <zpg> so for type 2, do you just have to adjust in terms of diet? or is insulin still involved.
10:40:35 <zpg> +?
10:47:26 --- quit: Snoopy42 (Read error: 145 (Connection timed out))
10:47:28 --- nick: snoopy_1611 -> Snoopy42
10:47:48 <Ray_work> insulin is involved. I use a sliding scale. 
10:47:56 <Ray_work> I test 4 times a day.
10:48:53 <zpg> yikes.
10:54:54 <Ray_work> Diet is the main control, that and exercise.
10:55:38 <Ray_work> Now, when I first started, I shot 4 times a day. Now it's more like every couple of days I might have to take a shot once that day.
10:57:09 <Ray_work> I wanna ( ive seen some already) do a diabetes thingy for the Palm, you know, record meals exercize medication intake ...
10:57:24 <zpg> yeah, sounds like a good application to write.
11:25:07 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth
11:28:22 --- quit: segher (Nick collision from services.)
11:28:36 --- join: segher (n=segher@dslb-084-056-138-094.pools.arcor-ip.net) joined #forth
12:05:56 <Ray_work> LUNCH!
12:06:10 <Ray_work> or is that launch?
12:06:19 <Ray_work> no it's definately LUNCH! 
12:18:52 <erider> Ray_work: its holiday 
12:24:22 --- quit: madwork (Read error: 60 (Operation timed out))
12:24:32 --- join: madwork (n=foo@204.138.110.15) joined #forth
12:25:34 --- join: vatic (n=chatzill@pool-162-84-178-20.ny5030.east.verizon.net) joined #forth
12:28:19 --- join: erider_ (n=erider@unaffiliated/erider) joined #forth
12:28:57 --- quit: erider (Read error: 104 (Connection reset by peer))
12:37:56 --- nick: erider_ -> erider
12:38:46 <zpg> la de da
12:41:17 <zpg> hey erider, vatic 
12:42:14 <Quartus> what's up?
12:43:08 <zpg> oh hey Q. not too much, unresponsive brain today i'm afraid.
12:43:15 <Quartus> WAKE UP!
12:43:20 <zpg> aagh!
12:43:56 --- quit: jackokring (Read error: 113 (No route to host))
12:45:18 <zpg> so i got some of that assoc-array based code working by pre-assigning everything. in a situation where you're planning on accepting user-data, saving the string and pushing it to a list, i'm guessing the best option in QF would be ALLOCATE? that HERE stuff we where discussing yeseterday still intrigues me, especially as a lot of the 'here allot' stuff would need revising otherwise.
12:45:38 <erider> hi zpg
12:46:02 <zpg> hey
12:46:16 <Quartus> zpg, depends entirely on how much data you think you'll be dynamically using.
12:46:46 <zpg> well, substantially less than the normal QF heap size.
12:47:18 <zpg> i'm not sure how that size is affected when MakePRC is called to produce a standalone.
12:47:18 <Quartus> So you could go either way with it.
12:47:31 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth
12:47:44 <Quartus> Dataspace is static, and allot won't expand it at run-time. So if you need available space, you have to pre-allot it.
12:47:48 <zpg> okay, so could you tell me again about the hereptr stuff you were suggesting yesterday?
12:47:55 <Quartus> Do you still have it logged?
12:48:07 <zpg> not locally, but let me look on the HTTP log....
12:48:30 <Quartus> That's the whole thing. It's just a positive allot before the makeprc, that the main word rolls back.
12:49:06 <zpg> 19:43:30 <Quartus> The other method is the one I outlined.       variable hereptr    : main hereptr @ here allot ... ;      here hereptr !  10000 allot  ... ' main MakePRC
12:49:06 <zpg> 19:44:46 <Quartus> er,         : main hereptr @ here - allot ... ;
12:49:13 <Quartus> That's it.
12:49:25 <zpg> oh i see, i missed the resetting of HERE
12:50:27 <zpg> so if i'm calling words that are ALLOTing from within main, this code will handle the situation?
12:50:33 <Quartus> Yes.
12:50:44 <Quartus> Up to 10000 bytes, in that example.
12:50:51 <zpg> let me try it...
12:50:56 <Quartus> The downside is your standalone app will be 10K bigger.
12:51:03 * zpg nods
12:52:39 <Quartus> Alternatively, use ALLOCATE, or use the Palm OS memory calls and work with 32-bit pointers (this last one is the most flexible).
12:53:11 <zpg> yep, i'm clear about ALLOCATE, this is just for a demo. i can make a prc then transfer it and get a screenshot.
12:53:30 <zpg> what's the purpose of: hereptr @ here  ?
12:53:38 <Quartus> hereptr @ here -
12:53:42 <zpg> oh, sorry
12:53:56 <zpg> that was the "er," line
12:53:56 <Quartus> It'll result, in this case, in -10000
12:54:28 <Quartus> It's flexible, if you change the 10000 in the makeprc it'll just work.
12:54:33 <zpg> neat.
12:57:38 <zpg> curious, still getting a Fatal Exception when I run the PRC
12:58:21 <Quartus> I'll get my Magic 8-Ball, maybe it can see your code better than I can. :)
12:58:26 <zpg> heh
12:59:34 <vatic> Hi zpg, Quartus!
12:59:40 <Quartus> Hey.
13:00:03 <zpg> yo
13:00:25 <zpg> Quartus: i'd show you, but i'm dockless at the moment.
13:01:14 <Quartus> well, make sure you're doing all of the steps.
13:01:25 * zpg nods
13:02:54 <vatic> Quartus: I wanted to explore this business about (not) deferring some more. If I have a word, say: "update" that at various times should either increment or decrement a value, how could I assign those behaviors at the appropriate moments? 
13:03:40 <Quartus> vatic, I doubt I'd do it by vectoring the behaviour of the word.  I'd pass an appropriate value into the update word.
13:04:40 <zpg> a crude non-vectoring way would be to have a variable as flag for the update process
13:04:58 <Quartus> That'd be a crude way indeed.  
13:04:59 <zpg> if it's a binary process, you could do update-flag @ if ... then
13:05:26 <zpg> vatic: what determines  'at certain times'?
13:05:28 <Quartus> If your app knows enough to revector a word to make it either increment or decrement, it knows it wants to increment or decrement, and can just go ahead and do the ding-dang thing.
13:07:15 <zpg> ooh, this pre-ALLOT is working now, I think.
13:07:51 <vatic> I was assuming that the word would be re-entrant and that a variable would indicate whether to increment or decrement. Different processes might be doing different things at any one time.
13:08:58 <Quartus> vatic, too many ambiguities there.  I don't know what the problem space you're describing really is.  I know that vectoring words is something I rarely do, as it's a curious way to program, but if it works for you, go crazy. :)
13:14:44 <Quartus> This is the mobile client I use, zpg: http://jmirc.sourceforge.net/
13:16:22 <zpg> ah, neat.
13:16:27 <zpg> incidentally, is there a cleaner way of saying:
13:16:32 <zpg> POSTPONE X POSTPONE Y POSTPONE Z
13:16:35 <zpg> i.e.
13:16:44 <zpg> P[ X Y Z ]P -- something like that?
13:17:07 <Quartus> One alternative is s" x y z" evaluate, but that's late-bound.
13:17:14 <zpg> and btw, HEREPTR works a charm.
13:17:19 <Quartus> Good.
13:17:56 <Quartus> If you want to layer syntactic-sugar overtop of POSTPONE, you could -- probably better to step back and ask why you're doing POSTPONE X POSTPONE Y POSTPONE Z so often that you feel a need to shorten it up.
13:18:38 <zpg> that's true.
13:19:21 --- quit: cmeme (Remote closed the connection)
13:19:33 <zpg> to reprise our immediate/postpone discussion -- could i place x y z in a function, say an immediate one, then postpone a call to that?
13:19:43 <Quartus> Can you be specific?
13:19:56 <zpg> something like "sliteral type cr"
13:20:03 <zpg> for example
13:20:12 <Quartus> sliteral is immediate.  Can you show an actual example of what it is you want to do?
13:20:29 --- join: cmeme (n=cmeme@boa.b9.com) joined #forth
13:20:37 <zpg> print a string with a bit of sugar. 
13:20:47 <Quartus> Right. I mean can you show code?
13:20:49 <zpg> but within or without a definition. so ." " with a bit more.
13:21:00 <zpg> again, it's on the palm. hang on, i'll hack up something similar in gforth.
13:21:40 <Quartus> It's difficult to answer questions in the form of "what if I want to do X and then Z and then an immediate version of Y, except re-entrantly?" :)
13:22:34 <zpg> : x" [char] " parse save-mem state @ if postpone 2literal postpone type postpone cr else type cr then ; immediate redefined x"   ok
13:22:41 <zpg> : blah x" hi there" ;  ok
13:22:43 <zpg> blah hi there
13:22:46 <Quartus> well, that's got to win a prize for longest word evar.
13:22:51 <zpg> that's just a ." ..."
13:23:04 <zpg> yeah, i know. I added to an existing definition. hence the question :)
13:23:26 <zpg> wihtout the state stuff, of course, it's:
13:23:38 <Quartus> If it's ." you want, use .".  Simple answer. :) If it's ." you want to replicate, use SLITERAL, instead of building your own out of save-mem and 2literal.
13:23:44 <zpg> : y" [char] " parse type cr ;   ok
13:23:44 <zpg> y" hi there" hi there
13:24:25 <zpg> hmm okay. but what would you suggest for this type/cr stuff -- can i place that elsewhere and call it more cleanly
13:24:27 <zpg> ?
13:24:43 <Quartus> You want a ." that does a CR?
13:24:44 <zpg> (it's not just ." ..." functionality, i'm aiming to modify the string before typing -- it's a test)
13:25:35 <zpg> my actual word is named "heading:", which prints a string then a cr then a set of dashes.
13:25:42 <Quartus> : do-something-with-string  type cr ;   : ."   [char] " parse  postpone sliteral  postpone do-something-with-string ;
13:25:54 <zpg> ah okay, it's that simple. neat.
13:26:35 <Quartus> Unless you have a pressing need for a state-smart word, don't bother.
13:32:03 <Quartus> And in the case of your specific requirement, I'd write        : .heading ( c-addr u -- )  type cr ." ---------------" ;
13:32:12 <Quartus> Then in use,     ... s" Heading 1" .heading
13:37:16 <Quartus> You can go nuts with the state-smarts and the forward parsing, but it's rarely worth the added effort and complexity.
13:39:36 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #forth
13:49:58 <Razor-X> Quartus: Do you do all of your Palm development via Graffiti?
13:52:31 <Quartus_> Mostly. I use a keyboard on occasion.
13:53:44 <Razor-X> Wow.
13:54:04 <Quartus_> ?
13:54:23 <Razor-X> Heh. My Graffiti speed compared to my typing speed is horribly slow.
13:54:42 <Razor-X> Keyboard typing, of course.
13:54:48 <Quartus_> but your programming speed is probably slower than your graffiti speed.
13:55:02 <Razor-X> Not yet it isn't, since I'm awfully slow at Graffiti.
13:55:08 <Razor-X> I guess that'll just come with practice.
13:55:43 <Razor-X> I'm having a bunch of fun playing with QF though.
13:55:53 <Quartus_> it encourages thinking before writing, never a bad idea. :)
13:55:59 <Razor-X> Heh.
13:56:57 <Quartus_> glad you're having fun!
13:57:16 <Razor-X> Yup :)
13:59:55 <erider> hi all 
13:59:57 * marble wonders what wrong with the palmos emulator
14:00:23 --- part: marble left #forth
14:00:31 <Quartus_> having trouble with it?
14:04:22 <Quartus_> perhaps he's suggesting you should use the emulator instead of the actual gadget.
14:04:41 <Razor-X> Oh. Possibly.
14:04:53 <Razor-X> But, there's a certain amount of fun on actually being in front of what you're programming for.
14:05:05 <Quartus_> which is fine, but not as cool. :)
14:05:10 <Quartus_> right.
14:09:29 --- join: JasonWoof (n=jason@unaffiliated/herkamire) joined #forth
14:09:29 --- mode: ChanServ set +o JasonWoof
14:09:42 <Quartus_> hey Jasonwoof
14:09:54 <JasonWoof> hey :)
14:11:33 --- quit: neceve (Read error: 145 (Connection timed out))
14:12:11 <zpg> Quartus_: ah, thanks for the input.
14:12:52 <Quartus_> sure. S" is already factored, so leveraging it makes sense.
14:13:42 <Quartus_> if you like,  : .heading"  postpone s" postpone .heading ; immediate
14:15:06 <Quartus_> that's a reasonable factoring, if the icing is important to you.
14:16:51 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #forth
14:57:15 <zpg> neat
14:57:38 <zpg> i like that factoring -- will try it later on.
14:57:44 <zpg> must dash for now, thanks again.
14:57:45 --- quit: zpg ("ERC Version 5.1.3 (IRC client for Emacs)")
15:05:09 --- quit: erider ("I don't sleep because sleep is the cousin of death!")
15:08:41 --- join: erider (n=erider@unaffiliated/erider) joined #forth
15:09:47 --- quit: Ray_work (Read error: 131 (Connection reset by peer))
15:10:30 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/")
15:27:21 --- join: I440r (n=spam@67.135.84.40) joined #forth
15:39:41 --- quit: jackokring (Read error: 110 (Connection timed out))
15:57:15 <I440r> I must Create a System, or be enslav'd by another Man's;
15:57:16 <I440r> I will not Reason and Compare; my business is to Create.
15:57:16 <I440r>                 -- William Blake, "Jerusalem"
15:57:24 <I440r> :)
15:59:19 --- nick: nanstm -> Raystm2
16:02:50 --- join: slava (n=slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
16:02:50 --- mode: ChanServ set +o slava
16:02:57 <I440r> hi slava
16:03:01 <slava> yo
16:03:06 <slava> regarding optimizing compilers
16:03:14 <slava> i'm debugging an issue where factor crashes on windows if compiled with -O3
16:03:15 <slava> but not -O2
16:03:18 <slava> gcc sucks
16:03:24 <I440r> i rest my case
16:03:31 <slava> actually -O2 -finline-functions triggers the crash
16:03:36 <slava> why the hell would inlining cause issues
16:04:00 <I440r> hrm dunno
16:04:05 <slava> I440r: factor is about 8000 lines of C, and i've found numerous gcc issues
16:04:15 <I440r> other than refering you back to your "gcc sucks" statement
16:04:20 <slava> segfaults gcc < 3.3 completely
16:04:29 <I440r> have you reported them to the developers ?
16:04:31 <slava> the C code is very simple
16:04:32 <slava> yes
16:04:38 <slava> everything's been fixed so far
16:04:45 <I440r> cool
16:04:52 <slava> i think gcc has too many switches
16:04:53 <I440r> at least they are listening to their user base
16:05:04 <I440r> its part of whats wrong with C
16:05:04 <slava> it should be 'optimization on' and 'optimization off, debug'
16:05:14 <slava> but they have 3423423 -ffuck-with-my-code options
16:05:14 <I440r> its a very complex solution to a very simple problem
16:05:21 <Quartus_> have you looked at the generated assembly to see what the issue might be?
16:05:30 <slava> that's what i'm doing now
16:06:30 <Quartus_> optimization is a hard problem. Tough to test all permutations.
16:24:47 --- quit: Quartus_ (Read error: 104 (Connection reset by peer))
16:26:25 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #forth
16:26:25 --- mode: ChanServ set +o Quartus_
16:37:03 <Quartus> That's the benefit of having all the flags; you can selectively disable elements of the optimization to test interactions if you have a problem.
16:37:20 <slava> yeah but making sure each combination of flags works is itself a daunting task
16:37:38 <Quartus> Well, yes, but all the worse if you could only switch them all off, or all on.
16:37:59 <slava> in factor the optimizer is always on
16:38:06 <slava> this hasn't caused any problems
16:38:28 <slava> there's only one code path to test, so to speak
16:38:43 <Quartus> Sure. It's less of an issue with Forth than it is with Algol60-derivative stuff like C.
16:40:17 <Quartus> Perhaps it's that optimization for register-based abstract machines is more complex.
16:41:40 <slava> yes, it is
17:12:36 --- quit: erider (Read error: 110 (Connection timed out))
17:13:19 --- join: erider (n=erider@unaffiliated/erider) joined #forth
18:00:21 --- quit: vatic ("*poof*")
18:10:14 --- quit: madgarden ("?OUT OF DATA  ERROR")
18:11:27 --- join: madgarden (n=madgarde@bas2-kitchener06-1096620860.dsl.bell.ca) joined #forth
19:08:53 --- quit: segher (Read error: 104 (Connection reset by peer))
19:12:05 --- join: segher (n=segher@dslb-084-056-138-094.pools.arcor-ip.net) joined #forth
20:46:36 <Quartus> Any life out there?
20:49:31 <erider> yup
20:50:14 <slava> hi
20:50:48 <forther> sure
20:51:57 <erider> :)
20:53:50 <Quartus> Hey. What's up?
20:59:17 <slava> debugging
20:59:49 <Quartus> Progress?
21:00:02 <slava> oh, i put the windows issue aside for now
21:03:39 <erider> looking over Icon
21:05:39 <erider> Quartus what do you think about icon well anyone for the matter 
21:06:09 <Quartus> You asked me this before. I've had very little experience with Icon. It's derived from SNOBOL, which I have used.
21:06:39 <slava> it always somewhat pleasures me to find a word which is no longer called from anywhere, and i can remove
21:06:52 <Quartus> slava -- heh. You should have an automatic process to find those.
21:07:01 <slava> no, because some words are useful interactively
21:07:04 <slava> i doubt .s is called anywhere in the code
21:07:18 <Quartus> Those must be easily identified, though.
21:08:24 <slava> good point
21:08:29 <erider> Quartus how is the book coming?
21:08:30 <slava> how do you identify interactive words in quartusforth?
21:08:58 <Quartus> I don't specially identify them, slava -- could do, I suppose.  I can't think of a use for it.
21:09:05 <Quartus> erider, coming along nicely.
21:09:07 <slava> for finding unused words, perhaps.
21:09:28 <Quartus> I find unused words by finding only those that are used, when generating stand-alone apps.
21:09:43 <erider> cool! are you using awk to do some editing for you?
21:09:50 <Quartus> editing?
21:10:09 <Quartus> awk for editing? What do you mean?
21:10:22 <erider> text stuff 
21:10:34 <Quartus> I know what awk is for, I'm just not understanding your question.
21:10:57 <erider> errors the you make on the book 
21:11:07 --- part: forther left #forth
21:11:37 <Quartus> erider, how would you think awk would apply to finding errors in a manuscript?
21:11:58 <slava> you're writing a book?
21:12:10 <Quartus> Yes, a book on Forth.
21:12:18 <erider> hmm I thought you were writing it on your computer 
21:12:34 <Quartus> erider, yes, I'm writing it on my computer.  Awk is a programming language.
21:13:31 <erider> I was think like you know some people use perl to find double words or repeats in there text 
21:13:59 <erider> s/think/thinking 
21:14:14 <Quartus> I rely on human proofreading.
21:14:48 <erider> ah ok sorry about the confusing 
21:15:06 <Quartus> That's ok.
21:18:57 <erider> good night guys 
21:19:03 <Quartus> ciao
21:19:13 <erider> hasta luego 
21:19:24 --- quit: erider ("I don't sleep because sleep is the cousin of death!")
21:23:38 --- join: arke_ (n=Chris@pD9E059E0.dip.t-dialin.net) joined #forth
21:31:47 --- quit: arke (Read error: 60 (Operation timed out))
21:44:49 <JasonWoof> computers aren't so good at finding that you typed "there" when you meant "their".
21:45:19 <Quartus> And there's a lot more to editing than just finding typos, anyway.
21:45:27 <JasonWoof> yep
21:45:36 <Quartus> I'll take well-written with typos over crap that has no typos any day.
21:46:15 <JasonWoof> I just think it's funny when people make horribly grammar errors when talking about grammar
21:46:29 * slava 's brain explodes
21:46:32 <Quartus> heh. Well, maybe awk can help.
21:46:41 <JasonWoof> I just finished reading the Myth series. The last few books have a lot of errors
21:47:15 <JasonWoof> I don't think I'm a terribly good proofreader, but i found a lot of errors just reading through it once
21:47:18 <Quartus> Myth? Aspirin?
21:47:28 <JasonWoof> asprin
21:47:29 <JasonWoof> yes
21:47:50 <Quartus> He's a dreadful hack. I'm not surprised.
21:47:56 <Quartus> Hard to find proofreaders for prose that bad. :)
21:48:06 <JasonWoof> yeah, I don't get it
21:48:10 <JasonWoof> the money must have been there
21:48:29 <Quartus> Those big absorbent 10-part novel series make a certain amount of money.
21:50:28 <JasonWoof> you'd think after the first nine books they'd be able to afford a proofreader for the next three
21:50:53 <Quartus> Likely the returns on each successive one were smaller and smaller.
21:51:09 <JasonWoof> why?
21:51:11 <Quartus> How many people are going to hold on through a crappy premise stretched that thin?
21:51:45 <Quartus> Takes grim determination.  So people might pick up book one, a lot of them won't finish it.  Of those that do, not all will buy book two.  Etc.
21:51:47 <JasonWoof> oh, I was thinking that most of the books would be bought by people who like them
21:52:14 <JasonWoof> are you figuring most people will start with book one?
21:52:18 <Quartus> Wouldn't you?
21:52:37 <JasonWoof> well, I did because my dad owned them all
21:52:39 <Quartus> Don't forget the used market, too.  Only the most rabid fan will want every new volume in hardcover, let alone in paperback.
21:52:45 <JasonWoof> but I think people start at whatever one they run into
21:52:57 <Quartus> I've never started a series partway in.  
21:53:10 <JasonWoof> some series it doesn't really matter
21:53:45 <Quartus> Maybe.  I avoid the multi-volume series at any rate.  Too many bad experiences with 'em
21:53:51 <JasonWoof> that's why he doesn't make much of a deal about it being in a series
21:54:06 <JasonWoof> I like to try books from long series
21:54:11 <JasonWoof> because if I like it I know there's lots more
21:54:27 <JasonWoof> whereas if I like an individual book, that's it.
21:54:32 <JasonWoof> I have to go looking again
21:54:38 <Quartus> Most fiction authors have maybe one book in them.  Some know enough to stop there.
21:55:00 <JasonWoof> and some write lots of fun stories
21:55:16 <Quartus> A few.
21:55:24 <JasonWoof> who do you like?
21:55:47 <Quartus> Some hacks keep churning them out until the last tree is felled.
21:56:46 <Quartus> Donald Westlake.  Philip K. Dick.  Harlan Ellison.  Robert Heinlein.  Robert Anton Wilson.  Issac Asimov.  Robert Sheckley.  John Brunner.  To name a few.
21:57:55 <Quartus> E. E. Smith. Theodore Sturgeon.  Walter Tevis.
21:59:22 <Quartus> Fritz Leiber.
22:00:02 --- join: jdrake (n=jdrake@d57-164-139.home.cgocable.net) joined #forth
22:00:40 <Quartus> Cordwainer Smith.
22:01:24 <Quartus> van Vogt, Bester, Hoyle.  I'll stop. :)
22:02:10 <jdrake> Hallowed are the Machine Stacks
22:02:59 <Quartus> Hallowed, eh?
22:03:15 <jdrake> It is so very hard to find channels pertaining to fortran
22:04:02 <Quartus> I see you've formed one with Mr_Science.
22:06:32 <jdrake> he just happens to be there when I joined :p
22:06:48 <jdrake> I am curious enough to learn fortran
22:07:28 <Quartus> Well, there's fortran, and there's fortran.
22:07:37 <jdrake> really
22:07:40 <Quartus> The latest incarnation bears very little resemblance to the early versions.
22:07:52 <jdrake> That is ok 
22:08:07 <Quartus> As in, you'd never identify the new stuff as Fortran if you'd ever seen the old stuff.
22:08:26 <jdrake> which is fine
22:09:00 <Quartus> It's fine, whatever that means, but it does mean you have to pick your target.  If you're looking for vacuum-tube charm, you want Ye Olde Fortran.
22:09:22 <jdrake> I am looking for modern stuff
22:11:06 <jdrake> I am choosing to try fortran rather than relearn a functional language
22:11:22 <Quartus> functional language such as...?
22:11:25 <slava> what do you want to use fortran for?
22:11:53 <jdrake> Quartus, ocaml, sml, or haskell
22:12:11 <jdrake> slava, I am attempting to learn one of the methods for n-body simulations.
22:12:30 <jdrake> Being that it involves numbers, fortran sounded like a decent application
22:12:33 <Quartus> Certainly Fortran isn't functional.  The latest incarnations have OO, though.
22:12:35 <slava> no point doing that in an archaic language
22:13:11 <Quartus> Fortran has no particular strength with numbers.
22:13:38 <jdrake> As I recall its primary application is numerical processing
22:14:23 <Quartus> Originally, the primary application of computers was numerical processing; Fortran was one of the very first languages.  It may well have been the only logical choice for numerical processing anywhere from 1956-1960.
22:14:36 <slava> up to the late 80's even
22:15:15 <slava> damnit, now i've found a bug in Cocoa.
22:15:18 <Quartus> Its advantage was not that it could do numerical work faster or better, but that it allowed the representation of formulas inside a program in a format somewhat similar to that found in mathematical texts.
22:16:09 <Quartus> Algol60 came along and had that too, though, as does virtually every derivative of it.
22:19:27 <Quartus> Fortran was sui generis for years, and heavily pushed by IBM along with their computers, so it obviously got used a lot, in spite of whatever limitations it had.
22:19:42 <jdrake> All systems have limitations
22:20:23 <Quartus> Fortran had more than the usual allotment, not least because it was the first language an optimizing compiler was ever written for.  The field of computer language design didn't even exist at the time.
22:24:22 <Quartus> It quickly came into being, partly in reaction to the problems in Fortran (and the slightly later monstrosity COBOL).  Early products of the field were Algol60 and Lisp, from which most other computer languages are descended.
22:24:50 <jdrake> So what you two are generally saying is pick another language
22:25:37 <Quartus> I would.
22:26:18 <jdrake> I already know I don't want to use C for this, but I also don't want to use an interpreted language.
22:26:31 <slava> use ocaml or common lisp
22:27:18 <jdrake> I don't think I would want to risk my soul on CL
22:27:28 <jdrake> Why do you choose ocaml specifically otherwise?
22:27:43 <slava> because the ocaml implementation is very fast, and its a nice high level language, just like CL
22:28:11 <Quartus> ML is descended of Lisp anyway.
22:28:21 <slava> if you reject C and CL, your only choices for a high performance free compiled language is Ocaml and Java
22:29:31 <jdrake> How is ML descended from Lisp?
22:29:42 <slava> it isn't, really
22:30:16 <Quartus> If it can be said to have any lineage at all, it was influenced by ISWIM, which has its roots in Lisp.
22:35:40 <Quartus> Landin said "Iswm can be looked on as an attempt to deliver Lisp from its eponymous commitment to lists, its reputation for  hand-to-mouth storage allocation, the hardware dependent flavor of its pedagogy, its heavy bracketing, and its compromises with tradition."
22:35:51 <Quartus> ISWIM, that is.
22:38:33 <Quartus> So ISWIM is a direct descendant of Lisp, and ISWIM is the cited precursor to ML.
22:58:33 --- quit: slava ()
23:06:41 --- join: fission (n=fission@pdpc/supporter/active/fission) joined #forth
23:35:52 <JasonWoof> ok, I'm thoroughly bored with mailman
23:36:13 <fission> heh
23:36:23 <JasonWoof> I got a list all set up and tested for myself
23:37:00 <JasonWoof> but I still have to figure out how to copy my clients list (the settings and archive, etc) to this new install (on my new server)
23:37:24 <JasonWoof> I can't think of a way to test his list without spamming everybody
23:37:38 <fission> what mta do you use?
23:37:42 <JasonWoof> qmail
23:37:46 <fission> oh hehe
23:37:51 <JasonWoof> what?
23:37:54 <fission> I mean... er, sorry I can't help you then :D
23:38:00 <JasonWoof> nothing to do with qmail
23:38:04 <JasonWoof> qmail works great
23:38:28 <fission> oh I wasn't saying that it was
23:38:37 <JasonWoof> only problem with qmail is it doesn't do valid-user-checking before accepting messages into the queue
23:38:52 --- quit: jdrake ("Leaving")
23:38:58 <JasonWoof> I just haven't seen any instructions on how to move a mailman install to a new server
23:39:10 <fission> I see
23:39:29 <JasonWoof> so I'm just going to have to guess, and copy any folders that seem to contain list-specific data, and hope it works
23:39:55 <fission> does mailman use the local MTA to send mail?
23:40:09 <JasonWoof> dunno
23:40:15 <fission> well, what I was going to suggest...
23:40:29 <JasonWoof> I think I saw something that said that your local MTA must allow relaying for localhost
23:40:30 <fission> if you temporarily stop the MTA, you should be able to examine the queue
23:40:42 <JasonWoof> so it might connect to 127.0.0.1 instead of calling /usr/bin/sendmail
23:40:44 <fission> or, in the case of, say, postfix, move mail to the held queue
23:40:50 <fission> I see, I see
23:41:06 <JasonWoof> oh, I see
23:41:09 <JasonWoof> I don't know how to do that
23:41:23 <JasonWoof> well, I could certainly kill the process that sends things
23:41:31 <fission> yeah that's why I said I couldn't help with the qmail bit (it wasn't a dig at qmail)
23:41:40 <JasonWoof> but I don't know any convenient way of deleting those messages from the queue
23:41:53 <JasonWoof> there'd probably be a lot of them
23:41:57 <fission> yeah in postfix it's a one-liner
23:42:05 <JasonWoof> one-liner to do what?
23:42:14 <fission> delete messages from a queue
23:42:17 <JasonWoof> delete all messages in the queue from a particular address?
23:42:33 <fission> well that's a matter of grep-and-cut, but yeah, it would be one line
23:43:03 <JasonWoof> I made a horrid one-liner to do that for qmail
23:43:05 <JasonWoof> didn't save it though
23:43:07 <fission> heh
23:43:29 <fission> I recently had some assface send spam from a local account, so I had to cook one up for that special occasion
23:43:43 <fission> of course, I also redirected said user's site to lemonparty, so I think my revenge was appropriately exacted
23:44:26 <JasonWoof> grep -r ... | sed ... | xargs find -name | xargs rm
23:44:32 <JasonWoof> something nasty like that
23:44:56 <fission> oh heh rm
23:45:08 <JasonWoof> um... how do you delete files?
23:45:20 <fission> well, in postfix you can delete by messageid
23:45:34 <JasonWoof> might be able to with qmail too
23:45:44 <fission> could be
23:45:45 <JasonWoof> I didn't read any documentation or anything
23:45:54 <fission> I just mean that you don't have to mess with rm directly on queuefiles
23:45:57 <JasonWoof> I know there's software to clean the qmail queue
23:46:06 <JasonWoof> yeah
23:46:09 <fission> normally I just parse the output of mqueue, grab messageids and then use postsuper
23:46:10 <JasonWoof> I was in a huge rush
23:46:13 <JasonWoof> wanted to get the spam out of the queue
23:46:32 <fission> yeah, that's a total pain in the ass, spam in the queue
23:47:05 <JasonWoof> I only had to do it once, and didn't want to stop qmail while I read all about it
23:47:33 <JasonWoof> I've been enjoying xargs though
23:47:41 --- quit: Quartus_ (Read error: 104 (Connection reset by peer))
23:47:44 <JasonWoof> I didn't know about it for the longest time
23:47:45 <fission> yeah xargs is sweet ass sweet
23:47:51 <JasonWoof> and wrote my own script that does pretty much the same thing
23:47:52 <fission> especially find -print0 ... | xargs -0
23:48:29 <JasonWoof> heh
23:48:48 <JasonWoof> xargs is more polished though
23:48:51 <JasonWoof> I like it
23:49:14 <fission> eh?
23:49:16 <JasonWoof> only trouble I've had with xargs is that it works differently on FreeBSD than on Linux
23:50:00 <JasonWoof> cmdline switches are different
23:50:09 <fission> no, I mean 
23:50:11 <fission> 23:48 <@JasonWoof> xargs is more polished though
23:50:15 <fission> I don't understand
23:50:35 <JasonWoof> xargs is more polished than the crappy bash script I wrote for the same purpose
23:51:24 <JasonWoof> and on FreeBSD you have to pass a switch if you want the program it runs to be able to take input from the terminal
23:52:09 <fission> oh heh
23:59:59 --- log: ended forth/06.11.10