[I'm not a member of the S/MIME working group. Please copy me the
messages you send.]
On Thu, Dec 18, 2008 at 11:23:03AM -0800,
Blake Ramsdell <blaker(_at_)gmail(_dot_)com> wrote
a message of 12 lines which said:
During AUTH48, a comment was sent by Stephane Bortzmeyer to the RFC
editor asking that we create and include an XML schema in
draft-ietf-smime-ibearch-09.
A small note to explain my point: it seemed strange to me that a RFC
defines a XML language, whatever its simplicity, without a formal
schema (whether in W3C Schema, in RelaxNG or in DTD).
BCP 70 "Guidelines for the Use of Extensible Markup Language (XML)
within IETF Protocol" says:
4.7 Validity and Extensibility
One important value of XML is that there are formal mechanisms for
defining structural and data content constraints; these constrain the
identity of elements or attributes or the values contained within
them. There is more than one such formalism:
...
This document makes the following recommendations for the definition
of protocols using XML:
o Protocols should use an appropriate formalism for defining
validity of XML protocol elements.
...
This is not just a matter of taste. It also has practical
consequences. In the version approved by IESG, there was no way
(natural language or formal schema) to say if <ibe:algorithm> is
mandatory or not inside a <ibe:request>.
Now, I understand it is AUTH48 and this is certainly a substantial
change, not just a spelling fix. Returning to the WG is certainly not
an appealing idea for most WG participants. But I am surprised that
the IESG did not catch this issue.