🚀 go-pugleaf

abell@netcom.com (Steven T. Abell) writes (about the P5):
>
>Unfortunately, you can't just brush of an obnoxious architecture by saying
>"The compiler people will worry about this." Let's stay real, folks.
>Underlying models matter, and the Intel model is about as bad as they come.
>If this is what we have to live with, I suppose we'll manage somehow,
>but if you think it doesn't affect you, you're wrong.
>

For Next, beyond all others, this is not quite right. Next and Next's
OS does the best job of isolating the user from the hardware and internals
of any OS I've seen.

Consider NextStep 486: reports from developers are that porting to it
is a very easy affair, usually just recompiling assuming the developer
has not circumvented Nextstep in some way.

I suppose my question is: if you are contending there will be impact beyond
the compiler and low level OS stuff... what are the effects you forsee and 
why haven't they shown up in Nextstep 486 as major problems? Generalizations 
are fine, but since Nextstep even takes care of byte swapping, I don't see 
where the "underlying model" is going to effect the user or the Nextstep 
developer.  (assuming that they don't write assembly -- a pretty good 
assumption on Nextstep.)

Thanks!
-Chris