Re: nanosecond-resolution time stamps

[Paul Eggert]

>> For example, if you'd prefer the format (19981 28208 421496 053)
> 
> That would reduce compatibility problems, most likely.
> Tho we might as well add 6 more digits rather than just 3 (just so that
> all the sub-second integers encode the same number of digits).

OK, that sounds good, and I'll revise the patch to do that.

> I think I'd rather wait for Emacs-24.2 for this change, tho maybe
> I could be convinced otherwise.

Obviously there's no rush.  To be honest, though, I'd like to do it now,
since it's all in my head and I'd rather get it done while it's
relatively fresh in my mind.

What would it take to convince you?  (Could I buy you a beer maybe?  :-)

Failing that, how about if we put in the underlying machinery to
do the nanosecond time stamps, *except* that we don't expose it to
Lisp?  That is, the Lisp format is (19981 28208 421496) just as
before.  This would let us shake out all the porting glitches
at the C level, without causing any problems to Lisp users.  (On the
other hand one could argue that this approach would cost us all the
porting pain without giving us any feature gain....)