ARGON

[Sean 'Captain Napalm' Conner]

On some network somewhere in cyberspace, Alaric B. Williams transmitted:
> 
> On 20 Dec 96 at 1:01, Sean 'Captain Napalm' Conner wrote:
> >   In the January 1986 issue of "Computer Language" there's an article on
> > KISS, which, as far as I know, only exists in this article.  The declaration
> > of variables (traditional, numbers and strings) is as follows:
>
> Eurgh - it's a COBOL picture! :-)
> 
  Or Ada.

> > > "Alaric's *important* work"
> > > "Alaric's /secret/ letters"
> > > "Alaric's _wierd_ jokes"
> > > 
> > > etc...
> > 
> >   What's wrong with having a metacharacter followed by an open set of
> > commands, much like, oh, HTML?  I mean, it doesn't have to follow HTML
> 
> That's an implementation detail. Conceptually, the style is still 
> "with" the character, if you see what I mean. Indeed, for storage, 
> the styles are best stored in a a seperate RLE stream.
>  
  But not EVERY character needs a style byte (unless you're writing ransom
notes).  You basically have a consecutive number of characters with the same
style, therefore it seems wasteful to include syle information for each
character.

> > A program can register (to the system) new sytles, or
> > such styles can be included in some meta-data for that text file.
> 
> Complex! The same text may appear wrongly in other contexts... text 
> is a very lightweight concept, it can get copied about etc. into 
> places where the styles used may not apply. That's why I wanted 
> uniform "style space".
>  
  Okay, fair enough.

> >   -spc (I find it odd that you came up with a cooperative system as 
> > 	well, even though I've to be convinced it's a win ... )
> 
> Odd in what way???
> 
  Strange, or coincidental.  Fare and I have gone head to head several times
over cooperative vs. preemptive multitasking (check both the tunes and the
tunes-lll mailing list archive).  I don't see how cooperative multitasking,
in a general purpose system, can be faster than a preemptive multitasking
system.

  True, a cooperative system can save less "state" than a preemptive system,
but in all the preemptive systems I've seen, the only "state" that's been
preserved have been registers.  And on those systems with virtual memory,
it's always been another "register" that needs to be saved to save the
memory space of the process.

  I don't know.  I've given up arguing with Fare about it (and I'm sure Fare
is relieved) but I've to be convinced that, in a general purpose system,
cooperative will be faster, as as responsive to, a preemptive system.

  -spc (Not intending to start another coop/preemptive fight here)