Subject: Re: take 2; which way should we go for /etc/rc...
To: John Nemeth <jnemeth@victoria.tc.ca>
From: Charles M. Hannum <root@ihack.net>
List: tech-userlevel
Date: 12/08/1999 08:58:11
jnemeth@victoria.tc.ca (John Nemeth) writes:

> On Dec 2, 10:01am, Jason Thorpe wrote:
> } On Thu, 2 Dec 1999 10:54:12 -0500 (EST) 
> }  der Mouse <mouse@Rodents.Montreal.QC.CA> wrote:
> } 
> }  > >> f)	Full SYSV style run levels.
> }  > > IMHO, this is the best solution.
> }  > 
> }  > And IMO, it's one of the worst.
> }  > 
> }  > We're BSD, dammit.  If you want to go start NetSysV and do this, I'll
> }  > fully support your project.
> } 
> } Unfortunately, I don't think "We're BSD, dammit" is a compelling argument.
> 
>      I do.

So does that mean we should go back to pcc (or gcc 1.39+++), old
config and static interrupt configuration, monolithic non-configurable
/etc/rc, completely separate hardware support for different machines,
etc, etc?  That's the `technology' that `BSD' shipped with.

As Frank said, if you want BSD, go get a VAX and put 4.3 on it.  Or a
HP 9000/300 and run 4.4-Lite.

It's precisely that sort of `no, we must do what BSD did' attitude
that puts the `legacy' in `carrying on the BSD legacy'.