Re: [dhcwg] I-D ACTION:draft-ietf-dhc-option-guidelines-00.txt

On Wed, Sep 12, 2007 at 12:15:02PM -0400, Internet-Drafts@ietf.org wrote:
> 	Filename	: draft-ietf-dhc-option-guidelines-00.txt

While working on draft-ietf-dhc-pxelinux, I had a random thought.

One of the areas in which the guidelines draft already discusses the
problem (a bit) is DHCP packet sizes.  That's more for guiding what to
do with large payloads (use TFTP or HTTP or etc) so far.

Several drafts have an almost 'boilerplate' description of RFC3396
behaviour, including the RFC3396 behaviour in their examples.  We've
gone back and forth on this with authors a few times, where the
re-description of RFC3396 is incomplete or misleading.

It ocurrs to me that anything beneath the new option's payload is
"DHCP Protocol".  So the thought I've recently had is our common
practice of including the Code and Length bytes (or words in
DHCPv6's case) is maybe the real source of the problem.  It seems
simpler, on the surface, to document only what code has been
assigned (external to any ASCII art), specify any fixed lengths of the
payload, and then the specific payload contents.

This creates a rather sudden and jarring inconsistency in existing
option specifications, however.

Or alternatively we could strive to make a single boilerplate.

-- 
Ash bugud-gul durbatuluk agh burzum-ishi krimpatul.
-- 
David W. Hankins	"If you don't do it right the first time,
Software Engineer		     you'll just have to do it again."
Internet Systems Consortium, Inc.		-- Jack T. Hankins