00:00:00 --- log: started forth/01.10.10
00:27:06 --- nick: MrBRB -> MrReach
00:27:09 --- part: MrReach left #forth
08:12:36 --- join: MrReach (mrreach@209.181.43.190) joined #forth
09:30:56 --- quit: MrReach ()
10:37:41 --- join: futhin (thin@h24-66-209-114.cg.shawcable.net) joined #forth
11:51:11 --- quit: futhin ()
17:31:12 --- join: MrReach (mrreach@209.181.43.190) joined #forth
20:52:25 --- join: futhin (thin@h24-66-209-114.cg.shawcable.net) joined #forth
20:52:32 <futhin> hey mrreach
20:53:42 <futhin> forth needs to be revitalized among the programming community
20:54:03 <MrReach> bbiab
20:54:06 <futhin> in order to do this, it needs more support and a stronger community
20:55:41 <futhin> forth needs more accessbile information:  easy to read documents are needed:  forth philosophy, intro/tutorial of forth for new coders, intro to forth for experienced coders (deal with their prejudices), proper forth coding methodology
20:57:02 <futhin> forth also needs more presence.  more source code and more programs coded in forth are needed online.  It is helpful to people to know that others actually code in forth, as well as being able to see well-written forth code.
20:57:25 <futhin> and to be able to take forth code and improve on it.
20:59:00 <futhin> it might even be helpful to have a new standard..  chuck moore wanted a publication standard, not an execution standard..  and i assume that he knows what he's talking about since he is god.   i've also heard people say that the ansi standard isn't the greatest
21:02:59 <futhin> what little source code i've seen online isn't exactly well-written and well-commented..  this isn't friendly to newcomers to forth.  it is important to be friendly to newcomers because we want to enable people with a powerful programming language that eliminates artificial layers between the computer and user.  we also want more people to contribute to the forth community and to come up with bright ideas that affect us all.
21:57:48 <futhin> agh.. i'm gonna go to bed
21:57:49 <futhin> good night
21:57:53 <MrReach> ok
21:58:03 <futhin> no response to my revitalization talk?
21:58:22 <MrReach> just a sec
21:58:28 <futhin> what do you think of it, etc :)
22:00:14 <MrReach> I'm talking to my sis in Montana
22:02:43 <MrReach> ok, I'm with you now
22:02:48 <futhin> ah ok
22:02:49 <MrReach> lemme read the backscroll
22:04:29 <MrReach> ok, what you prob don't know ...
22:04:56 <MrReach> Is that Mr. Moore was vehemently opposed to the ANS standardisation of Forth in the first place
22:05:13 <MrReach> he considered it an unnecessary burden
22:05:38 <MrReach> and you're right, the example code out there is not well commented at all
22:05:56 <MrReach> that is why forth becomes known as a "write only" language
22:06:29 <MrReach> because the code is indescipherable to anyone but the author, and even to the author after a couple of months
22:06:52 <MrReach> he isn't god, btw, he's a geek in a closet
22:07:10 <MrReach> but he does remarkable things in very wierd ways
22:07:50 <MrReach> you there?
22:14:31 <futhin> yes
22:14:35 <futhin> sorat
22:14:49 <futhin> got distracted by the extremeprogramming site
22:15:33 <futhin> mr. moore didn't want an execution standard, he wanted a publication standard.  ansi was an execution standard
22:16:12 <futhin> "I had reservations about ANSI. I worried that it would be a disaster and not merely a dubious advantage. All of my fears of the
22:16:12 <futhin> standard and none of the advantages of the standard have come to pass. Any spirit of innovation has been thoroughly
22:16:12 <futhin> quenched. Underground Forths are still needed. I said I thought the standard should be a publication standard but they wanted
22:16:12 <futhin> an execution standard. ---
22:17:13 <MrReach> heh
22:17:36 <futhin> that's mr.moore
22:17:40 <MrReach> yes
22:17:52 <MrReach> you must be reading Jeff Fox's wesite
22:17:56 <futhin> yup
22:18:04 <MrReach> he's the only one who has anything to do with chuck anymore
22:18:19 <futhin> there's barely any other content on the web for forth philosophy
22:18:28 <futhin> and it's pretty fragmented and all over the place... annoying
22:21:28 <MrReach> there's some REALLY good forth coders out there, though
22:21:35 <futhin> the ansi standard wouldn't have been screwed up if they had listened to moore :)
22:21:45 <MrReach> Speuler, for instance, has far more experience than I do
22:22:08 <MrReach> I know lots of comp sci, but Speuler can code rings around me
22:22:27 <futhin> even if there are really good forth coders out there, where's all the code online?  where are all the well-written easy to read well-commented code, where's the community?   forth needs to be revitalized! :)
22:22:41 <futhin> forth coders are selfish bastards ;)
22:22:48 <MrReach> busy bastards
22:23:10 <MrReach> the Forth Interest Group doesnt have much presence, though
22:23:14 <MrReach> you're right about that
22:23:20 <futhin> bah, there are a lot of bastards in other languages, and they still put up a good deal more code and things coded in their particular language
22:23:31 <futhin> bah, FIG is dead
22:23:40 <futhin> and completely useless
22:23:50 <MrReach> not useless, good library
22:24:09 <futhin> i heard that their mandate was to promote forth, but they seem to be failing that ridiculously
22:24:24 <MrReach> yes, I agree ... I wonder why?
22:24:33 <futhin> stupidity? :)
22:24:38 <futhin> nobody there?
22:24:43 <MrReach> I wonder if they sit down once a year and ask, "How many forth coders are there now?"
22:24:55 <futhin> i doubt that they do that
22:25:03 <MrReach> and use that as a benchmark for measuring success
22:25:32 <futhin> i doubt that they are very cohesive and organized..  their site is kinda lame
22:25:39 <MrReach> well, to do that successfully, they need to track downloads of free systems ... and how many commercial systems are sold, and in which countries
22:26:06 <MrReach> I actually hear more from the Dutch Forth Workshop than I do of FIG
22:26:15 <futhin> yeah
22:26:47 <futhin> the russia fig group has a better page too, sorta  (their older version is better than their newer version)
22:28:29 <futhin> i am interested in eventually writing up documents on forth, putting up a site, ensuring there's lots of quality code, etc..  i want to promote forth 
22:28:59 <futhin> maybe develop a new standard? :)
22:29:07 <MrReach> I'd like to see Forth uniform across all the PC platforms ...
22:29:18 <MrReach> Linux, Windows, MacOS
22:29:43 <MrReach> no ... not a new standard ... a de-facto library
22:29:44 <futhin> yes exactly
22:29:50 <futhin> hmm
22:29:55 <futhin> yeah
22:30:00 <futhin> that'd might be a better idea
22:30:13 <MrReach> one of the major probs with forth is that it doesn'thave the services that most programmers have come to expect
22:30:19 <futhin> yes
22:30:37 <futhin> most coders these days expect oo
22:31:01 <MrReach> there's no reason at all that Forth should come out of the box with bindings to glibc and OpenGL
22:31:10 <futhin> and what little oo i've seen seems to have awkward syntax, but i don't really know
22:31:12 <MrReach> shouldn't come, rather
22:31:35 <futhin> there's no reason OpenGL shouldn't be coded in forth ;)
22:31:50 <MrReach> it's huge
22:31:58 <MrReach> it's proven ...
22:31:58 <futhin> probably be smaller in forth? :)
22:32:04 <MrReach> it's hardware specific ...
22:32:25 <MrReach> and, most importantly, it's already written and ported
22:32:38 <futhin> yeah whatever! :P
22:33:10 <MrReach> heh, one does not rewrite a large library unless one's life depends on it
22:33:12 <futhin> real coders have chest hair and can code the opengl in 2 days :P
22:33:38 <MrReach> libc and OpenGL are excellent tools as they are
22:33:55 <futhin> yeah..  but can they be improved? :)
22:33:58 <MrReach> don't mess with them. Instead ...
22:34:12 <MrReach> let forth shine in writing unique solutions on the cutting edge
22:35:03 <futhin> what are the true advantages of forth?  how does one convince a hardcore c/c++ coder on the advantages of forth over c/c++ ??
22:36:07 <futhin> i talked about words and extensibility to my friend but he responds "that's what libraries are"
22:36:27 <futhin> i'm having trouble convincing my friends on the advantages of forth.. :(
22:37:36 <futhin> hmm?
22:37:51 <MrReach> down to the metal
22:38:04 <MrReach> and SIMPLE
22:38:35 <MrReach> how many languages can you keep the operation of the entire compiler in your head?
22:39:21 <futhin> that wouldn't convince my friends heh :(
22:39:52 <MrReach> ok, then ...
22:40:13 <MrReach> point out that the interactive nature of forth allows one to debug words immediately from the console
22:40:29 <MrReach> which allows a 4x faster development cycle
22:40:40 <MrReach> can't beat C at what C is good at
22:40:48 <futhin> which is what?
22:41:13 <MrReach> C is good at writing fast, compact code in a regular manner
22:41:44 <MrReach> it was designed to write operating systems in, and it's very good at that
22:41:53 <futhin> forth isn't good at writing fast, compact code in a regulard manner??
22:42:00 <MrReach> nope
22:42:22 <MrReach> certainly not "regular" ... because you alter the language itself to describe your problem domain
22:42:58 <MrReach> certainly not "fast" ... until recently, all forths were threaded interpreters, which mean 1/6 the speed of C, at best
22:43:23 <futhin> what?!  the words are compiled
22:43:23 <MrReach> compact? can't much more compact than forth, I take that back
22:43:44 <MrReach> yes, but not compiled to machine code
22:43:49 <futhin> yes they are
22:43:58 <MrReach> usually compiled to tokens
22:43:58 <futhin> when the words are put into the dictionary??
22:44:04 <MrReach> not usually
22:44:34 <futhin> what the hell are people doing creating forths that compile to tokens rather than machine lang!?
22:44:35 <MrReach> it hasn't been until the last 4 years or so that machine code comiling forths have been in the picture at all
22:45:00 <MrReach> futhin: because it makes the compiler 100x more complicated
22:45:00 <futhin> i've got books older than 4 years and they all say compiled to machine language
22:45:14 <MrReach> no, primitives are machine code
22:45:27 <MrReach> high-level use the "inner interpreter"
22:46:13 <futhin> um..  the high-level words are compiled into the dictionary, in that it's all machine code that CALL/JMP to the primitives
22:46:16 <MrReach> but, you know, base-line machines now are well over a TerraMIP ... speed isn't an issue anymore
22:46:27 <MrReach> no, it isn't
22:47:23 <futhin> speed is an issue when you've got a bunch of brain dead coders adding more layers and layers between the computer and the user.   i dislike bloat-ware like windows..  
22:47:42 <futhin> my computer is a p133 with 32 megs of ram and i expect it to be able to handle everything :)
22:48:11 <MrReach> heh, my server is a P133/48
22:48:54 <MrReach> my desktop is a K6-2/350MHz/128MB ... and it's getting a bit doggy with Winblows
22:48:54 <futhin> i've always been behind on the computer curve..  i was using a 386 when everyone was on 486 at 66mhz
22:49:04 <MrReach> So I'll be upgrading soon
22:49:06 <futhin> so i've been sensitive to crappy software
22:49:18 <MrReach> so was I
22:49:28 <futhin> i really think that a new os needs to be developed
22:49:39 <futhin> and i'm thinking possibly in forth..
22:49:40 <MrReach> I don't
22:49:49 <MrReach> I think Linux is a suitable OS
22:49:52 <futhin> windows and linux are both crap :/
22:50:13 <futhin> linux isn't exactly userfriendly to non-technical people
22:50:21 <MrReach> the prob w/ linux is that Windows how clean setup can be, and linux/x doesn't match up
22:50:39 <MrReach> is that Windows SHOWED how clean ...
22:50:49 <MrReach> you are correct
22:51:02 <MrReach> however, the linux situation is being incrementally improved
22:51:09 <futhin> i kind of think the man pages are horrid too :)
22:51:30 <futhin> i think there should be a help system that comes with the os, but it should be wellwritten and easy to understand
22:51:33 <MrReach> heh, use man2html
22:51:45 <futhin> man2html just changes the format.. ?
22:51:48 <futhin> that wouldn't help
22:51:54 <MrReach> I like the organisation of the man pages, I don't appreciate their presentation
22:51:57 <futhin> man pages tend to be pretty cryptic etc
22:52:22 <MrReach> some are better than others
22:52:32 <futhin> i think linux is a little too complex.. needs to be simplified
22:52:49 <MrReach> M$ learned the hard way in the early 80s that you *NEVER* let the coders write the public documentation
22:52:54 <MrReach> agreed
22:53:06 <MrReach> linux configuration is a bitch
22:53:07 <futhin> but i want to develop a new os.. for fun and profit ;)
22:53:15 <MrReach> ok, help yourself
22:53:37 <MrReach> for technical merit, I don't think you can beat the linux kernel, though
22:53:38 <futhin> i think that windows and linux can't really be fixed,  the os needs to be started from scratch
22:53:50 <futhin> with more thought to design
22:54:14 <futhin> i like QNX a lot.. seems to be very well designed & very modular
22:54:21 <futhin> and very tight code
22:54:35 <futhin> but it's proprietory :)
22:58:48 <MrReach> well ...
22:58:49 <futhin> i think QNX is an _excellent_ os..  they could do a better job of packaging everything up and then could probably market it as a  desktop os  unless microsoft was paying them off ;)
22:58:56 <MrReach> what is WRONG with the linux kernel?
22:59:10 <MrReach> I can't point out LOTS wrong with the various bundles
22:59:28 <MrReach> can point out, rather
22:59:36 <futhin> what's wrong with the linux kernel is that qnx kernel seems to be better ;)
22:59:41 <MrReach> heh
23:00:10 <futhin> the design for qnx is very modular.. the kernel is something like 20k or so..
23:00:32 <futhin> last i heard it was something like 15 but it's probably  bigger now
23:00:48 <futhin> it fosters a lot of the other stuff in the linux kernel to the outside, to be run separately..
23:02:15 * MrReach nods
23:02:38 <MrReach> ummm ...
23:03:18 <MrReach> a 20k (or even 200k) kernel in an age when you can't buy a computer with lesst than 64MB of memory is rediculous
23:04:26 <futhin> qnx is mostly for embedded applications
23:04:48 <MrReach> oh, you and I are talking about different animals, then
23:04:54 <futhin> however, is wasting 100k of memory smart?  or 1 megabyte?  or 20 megabytes?  i don't really think so.. 
23:04:57 <MrReach> linux is miserable for emedded apps
23:05:14 <MrReach> who says it's a waste?
23:05:39 <MrReach> a serial driver takes roughly 10k of machine code, not matter what the language
23:05:54 <MrReach> 20-30k is a reasonable expectation
23:06:25 <MrReach> but unloading functionality into modules doesn't improve effiency one lick
23:06:41 <MrReach> in fact, it adds a bit of overhead becuase of the required indirection
23:07:00 <futhin> yeah, true
23:07:27 <MrReach> the modules themselves may or my not be more efficient than what you might find in a monolithic kernel
23:08:51 <MrReach> you know what I think is amusing?
23:09:09 <futhin> what? :)
23:09:12 <MrReach> a system that spins down the dirves after 15 mins or an hour, then wakes on LAN
23:09:26 <futhin> heh
23:09:54 <MrReach> my desktop used to spin down the drives, until I installed ICQ ... that ended that
23:10:10 <MrReach> talk about bloatware
23:11:25 <futhin> heheh
23:27:40 <MrReach> I'm trying to figure out how to implement asyncronous times in win32forth "properly"
23:27:48 <futhin> heh
23:28:03 <MrReach> it looks like I'm gonna have to drop to assembler ... what a bother
23:28:50 <futhin> i think i saw a library somewhere with some asynchronous something?
23:29:03 <MrReach> heh
23:29:26 <futhin> actually nevermind
23:29:30 <futhin> i'm thinking swiftforth
23:29:33 <futhin> and it's all blurry
23:29:55 <MrReach> did you download it?
23:30:04 <futhin> swiftforth.. yeah
23:30:13 <MrReach> pretty nice system
23:30:20 <MrReach> love the OO syntax
23:30:30 <MrReach> the implementation is a nightmare, though
23:30:44 <futhin> heh yeah
23:32:45 <futhin> gotta go to bed, gotta get up early
23:32:47 <futhin> good night
23:32:50 <MrReach> me too
23:32:54 <MrReach> nighty night
23:33:06 --- quit: MrReach ()
23:33:29 --- quit: futhin (byebye)
23:59:59 --- log: ended forth/01.10.10