Re: Concurrency, again

[John Wiegley]

>>>>> "EZ" == Eli Zaretskii <address@hidden> writes:

>> I very strongly disagree. I think "merge what we have and make it work" is 
>> the
>> wrong approach.

EZ> (Since then, John changed his mind.)

More accurately, Eli changed John's mind, by helping me see that (a) having
the support affects no one but modules that choose to use it, and (b) we can
call it experimental and subject to removal if the experiment fails. At the
very least we'd gain some experience by the attempt, rather than confining
ourselves to thought exercises.

EZ> IMO, we don't have Emacs-specific experience for such discussions. So
EZ> these discussions will most probably lead nowhere.

Yes, this.

EZ> My suggestion is to "learn by doing", collect experience, and _then_
EZ> discuss the design issues based on that experience.

I'm open to this. I'd *like* to see exploration of alternative paradigms for
expressing concurrency, so merging in the concurrency branch shouldn't be seen
as closing the door, or setting anything in stone.

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2