Re: [dhcwg] WG last call on draft-ietf-dhc-mipadvert-opt-01.txt (SECOND REQUEST)

[Henrik Levkowetz <[e-mail]>]

Tuesday 28 October 2003, [e-mail] wrote:
>     TL> On Wednesday, October 22, 2003, at 09:17 AM, Kim Kinnear wrote:
>     >> I don't see the value in having zero length suboptions.  I don't
>     >> think we generally allow them (though someone will doubtless
>     >> correct me if I'm wrong).  If there is some valid, semantically
>     >> interesting reason for allowing zero length suboptions, then I
>     >> think the meaning should be spelled out, otherwise I think they
>     >> should be disallowed.
> 
>     TL> I agree with this.  There is a precedent, but it's a big hassle
>     TL> to support this - it's better for the suboption to just have a
>     TL> boolean value.
> 
> I have been following this and was feeling uncomfortable about zero
> length options. Yes, we have to support it for the nwip sub-options but
> I really do not like how it continues to break what was a nice 
> paradigm on dhcp options: code, length, data.
> 
> I know dhcp packets are small and there are times when we want to be as
> space efficient as possible. However I am not sure this tradeoff is
> worth the cost of saving two bytes.  I agree with Ted. I think it is 
> better for this suboption to just have a boolean value.

Ok, I've taken this to heart and will do the draft update accordingly.

Thanks,

	Henrik