Using Processes in CLIM

• To: moon@brazil.cambridge.apple.com
• Subject: Using Processes in CLIM
• From: Jon L White <jonl%kuwait@lucid.com>
• Date: Tue, 2 Apr 91 13:53:50 PST
• Cc: clim@BBN.COM
• In-Reply-To: David A. Moon's message of Tue, 02 Apr 91 11:44:12 <9104021656.AA04762@brazil.cambridge.apple.com>

re: It would also be bad if CLIM got bogged down in trying again to 
    standardize all the things that X3J13 subcommittees were unsuccessful
    in standardizing. . . .
    There are very sound reasons why the standardization process is slow 
    and unwieldy and lags behind what vendors can do on their own without 
    thinking about other platforms and the needs of other classes of users.

You have come full circle in your reasoning.  This is precisely the
reason that I've recommended that CLIM make a minimal part of some
innards freely available and NOT seek "standardization".  The latter 
can only bog down for years what could be just a simple useful utility.

For example, my original suggestion involved only the QUEUE and EVENT
datatypes (with a few primitives), since not only do mulltiprocessing 
implementations have some form of these hidden deep down in their bowels,
but many non-multitasking applications do so also.  Especially when any 
sort of asynchrony (such as interrupts) is available, _regardless_ of the
question of a multitasking model.

DCPL had it right -- put these useful things in a non CLIM package
and avoid the bog.  If they're really useful the bog is irrelevant;
and if they aren't useful, they'll die a quiet death.


-- JonL --



0,,

• Re: Using Processes in CLIM
• From: "David A. Moon" <moon@brazil.cambridge.apple.com>

• Prev: Using Processes in CLIM
• Next: Using Processes in CLIM
• Index: CLIM Mailing list (5-Oct-90 to 31-Dec-91)
• Thread: CLIM Mailing list Index (5-Oct-90 to 31-Dec-91)