On Sun, 11 Apr 1999, Curt Sampson wrote:

> Because at one time this code had more than just this; it also
> included the username in the prompt if the username was not `cjs,'
> among other things. Since we don't need this, obviously we can
> simplify the code as per your message.
> 
> And again, Charles, NOTE THAT FOLLOWUPS ARE REDIRECTED AWAY FROM
> PORT-VAX. Or do you have some reason to think that this is some
> sort of VAX-specific issue?

Well, I started with the cc to port-vax.  The current-users list has
always seemed a VERY SCAREY place to me where people often get in heated
arguments about stupid things, so I was a little reluctant to leave the
comfortable fold of the port-vax list. But then I'm patient and
even-tempered so I'll probably manage alright here.  I am afterall typing
this from my MicroVAX II.

In defense of Charles, though I'm guessing he wasn't thinking of this as a
VAX specific thing, so I guess that means that I'm not really defending
him?  Hmmm... go figure.  Wandering off onto an unrelated path...  I would
argue that pretty much all generic NetBSD issues are in fact VAX specific.  
Well at least in one regard, that one thing being performance.  Given that
the older VAX based systems are the slowest ones supported by NetBSD, it
only follows that they're the systems most likely to be impacted by coding
decisions.

I'd never suggest compromising the generality of code for the sake of
performance unless it was absolutely needed.  But everyone must think of
the wee little diskless VAXstation 2000 with 6Meg of RAM trundling along
at its sub-MIP (sub-VUP) speed, reading files, writing files, and swapping
over 10Mbps ethernet on day 8 of its build of NetBSD's world.

Everybody needs a VAX.  VAXen are great for getting one to code more
optimally, that or they'll drive you insane.

-brian.
---
Brian "JARAI" Chase | http://world.std.com/~bdc/ | VAXZilla LIVES!!!