> To: jhawk@MIT.EDU
> Cc: gnats-admin@netbsd.org, netbsd-bugs@sun-lamp.cs.berkeley.edu
> From: Chris G Demetriou <Chris_G_Demetriou@LAGAVULIN.PDL.CS.CMU.EDU>

> i've applied your patch, but have a comment:
> 
> > ***************
> > *** 1,4 ****
> > ! #	$NetBSD: bsd.README,v 1.7 1994/06/30 05:31:09 cgd Exp $
> >   #	@(#)bsd.README	5.1 (Berkeley) 5/11/90
> >   
> >   This is the README file for the new make "include" files for the BSD
> > --- 1,4 ----
> > ! #	$Id: bsd.README,v 1.2 1995/01/02 23:43:13 jhawk Exp $
> >   #	@(#)bsd.README	5.1 (Berkeley) 5/11/90
> >   
> >   This is the README file for the new make "include" files for the BSD
> 
> please don't include patches that include hunks like this.
> 
> We've been working (slowly) to get the entire tree to be converted
> over to the $NetBSD$ id's, and we'd _really_ rather that normal $Id$'s
> creep back in.

Noted, sorry about that...

> It's not clear what a good (completely automated) way to keep a local
> version number, keep the NetBSD version number (for ID purposes), and
> keep the two from mixing in the master source tree.  But i think it's
> a bad idea to change the NetBSD version id in diffs you send back.

I'm halfway reluctant to just add another line (with $Id$) following
the $NetBSD$, because then line numbers on diff's will be off
(also, it will mean I cannot just strip out the RCS part of the diff
and have everything be the same), yet I'm also reluctant to forgo
version control within my own source tree... 

Am I missing a clear solution here?

Thanx.

--
John Hawkinson
jhawk@panix.com