From: Kevlin Henney (kevlin_at_[hidden])
Date: 2000-11-29 12:23:46
In message <9035rl+sp42_at_[hidden]>, William Kempf <sirwillard_at_my-
deja.com> writes
>> I suspect that somewhere in here we might be in agreement ;-)
>
>I'm not so sure. You're still in favor of COW, and frankly I don't
>think that's valid, much less the/a preferred approach here.
In which case I think we are in agreement: I am definitely _not_ in
favour of COW. The reason I raised COW is because it causes so many
problems, and hence why I would not be in favour of that as an approach.
COW is the reason that a hidden, reference-counting optimisation is
difficult, and that any reference-counting implementation, to be
workable in this case, would not be an optimisation but a semantic
requirement. This is perhaps where we part company, because I think such
a semantic requirement is unreasonable as it embodies a different
purpose.
For the record I also think that there is too much focus on optimisation
here: That is why am I have been trying to debunk the optimisation-
focused approach :->
Kevlin
____________________________________________________________
Kevlin Henney phone: +44 117 942 2990
Curbralan Limited mobile: +44 7801 073 508
mailto:kevlin_at_[hidden] fax: +44 870 052 2289
http://www.curbralan.com
____________________________________________________________