On Mon, Apr 14, 2003 at 11:19:42PM -0400, Andrew Brown wrote:
> >> now numbering is clear - numbers with letters are development versions,
> >> without - release versions. it's clear
> >
> >Wasn't there a 1.4.3A release? I don't quite recall.
> 
> there's mention of something called 1.4.3A in the cvs logs, but i
> think that was just something that led up to the actual 1.4.3 release.
> 1.4.3A was, in all likelihood, just a "beta" of 1.4.3.
  
Check the RCS log for "sys/include/param.h,v".

----------------------------
revision 1.56.2.6
date: 2000/12/13 23:49:07;  author: he;  state: Exp;  lines: +2 -2
Apply patch (requested by he):
  Bump visible version number to 1.4.3A, to indicate that we're
  no longer exactly at 1.4.3.
  Here: bump to 1.4.4, since format doesn't allow 1.4.3A.
----------------------------

> >> after that change it won't be IMHO better but will look like linux
> >
> >C'mon. If an idea is a good idea then who cares where it came from?
> >I think it is a perfectly fine idea. It may even nicely fit into
> >__NetBSD_Version__ in sys/param.h. The current system would have
> >broken if we had a 1.4.4 release.
> 
> how so?  that would 104000400, no?

Same file. Here's the diff:

1c1
< /*    $NetBSD: param.h,v 1.56.2.4 1999/12/20 21:35:51 he Exp $        */
---
> /*    $NetBSD: param.h,v 1.56.2.5 2000/07/01 18:02:32 he Exp $        */
68c68
< #define __NetBSD_Version__  104000300 /* NetBSD 1.4.3 */
---
> #define __NetBSD_Version__  104000400 /* NetBSD 1.4.3 */

So if 1.4.3 was 104000300 and 1.4.3A was 104000400, what would 1.4.4
have been?
-- 
Kevin P. Neal                                http://www.pobox.com/~kpn/

Seen on bottom of IBM part number 1887724:
DO NOT EXPOSE MOUSE PAD TO DIRECT SUNLIGHT FOR EXTENDED PERIODS OF TIME.