> While we do this, can I check your view on the possible document status?
> - If we need RFC2119 keywords to make the guidelines normative, would we
> make the document BCP - or are you happy with standards keywords in an
> informational document?
Actually, I was thinking Proposed Standard, as an Applicability
Statement (see RFC 2026, Section 3.2). BCP would be an alternative
also.
I think that, *if this is the way the WG decides to go*, it would not
require extensive document changes. I see it as something like this:
----
a. Make udpzero a Proposed Standard and call it an Applicability
Statement, or make it a BCP.
b. Put the 2119 language into udpzero Section 5.1, instead of in
udpchecksums Section 5.
c. Change the intro paragraph in udpzero Section 5:
OLD
This section identifies requirements for the protocols that are
transported over a transport connection that does not perform a UDP
checksum calculation to verify the integrity at the transport
endpoints.
NEW (for AS)
This section is an Applicability Statement that identifies REQUIRED
restrictions on the use of techniques that involve not performing UDP
checksum calculations to verify the integrity at the transport endpoints.
NEW (for BCP)
This section specifies best current practice for REQUIRED restrictions
on the use of techniques that involve not performing UDP checksum
calculations to verify the integrity at the transport endpoints.
(Or some similar sort of language to indicate that the restrictions
must be followed.)
d. Change udpchecksums Section 5 to remove the numbered list and the
RFC Editor note, and to make this other change:
OLD
However, some protocols, such as tunneling protocols that
use UDP as a tunnel encapsulation, MAY omit computing the UDP
checksum of the encapsulating UDP header and set it to zero,
subject to the constraints described in RFCXXXX.
NEW
However, some protocols, such as tunneling protocols that
use UDP as a tunnel encapsulation, MAY omit computing the UDP
checksum of the encapsulating UDP header and set it to zero,
subject to the constraints described in [I-D.ietf-6man-udpzero],
Section 5.1.
e. Make udpzero a normative reference.
----
With those changes, most of udpzero remains as informational
exposition leading up to Section 5, and Section 5 is the normative
part.
Barry
--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
