Re: [dhcwg] draft-ietf-dhc-dhcpv6-netboot-05

[Stephen Jacob <[e-mail]>]

On Tue, Oct 06, 2009 at 04:41:04PM -0400, Thomas Narten wrote:
> >    precedence        A single unsigned octet indicating the order in
> >                      which this URL should be processed, if more than
> >                      one URL appears in the message.
> 
> Later, we learn that lower numbered precedence actually means "use
> first".  It would be good to add text here saying lower precedences
> indicate higher precedence. Also, a lower numbered precedence is
> treated higher precedence? Why? this seems counter-intuitive to
> me. (Not a big deal, but let's pick an ordering that is consistent
> with what people would expect and what other "precedence" fields
> mean.)

I agree that it would be good to say in that paragraph what the
order is.

I don't agree with your objection to the chosen ordering, however.

To me, it does not mean "low precendence is high precedence". Rather, 
"a low numeric value for precendence indicates high precedence",
which is subtly different -- and not contradictory.

In my experience it is quite common to have "low values indicate
higher priority/precedence" because that way everybody knows what
the highest possible priority is (0, or 1). If it goes the other
way, how do you choose a value for "really high priority"? 100?
1,000? 10,000? So, I think this is a reasonable choice.

> >    The value of the precedence field MUST NOT be zero.
> 
> Again, this is not good for interoperability. What happens when it is?
> Is the value effectively reserved?  Why and for what?
> 
> And what does an implementation do if it receives an option with such
> a value?

I tend to agree that unless there is a compelling reason to forbid zero,
zero should simply be the highest precedence.

Regards,
Stephen
-- 
 Stephen Jacob | [e-mail] | [phone]
 Nominum, Inc. |  http://www.nominum.com/  | [phone]