Normative references to stable non-standards track documents
are indeed allowed when those documents are of the nature of
algorithm descriptions, mathematical theorems, and the like -
they don't even have to be RFCs. See RFC 2026 section 3.1,
first paragraph. MD5 and HMAC are the archetypical
examples of this type.
RJ Atkinson wrote:
At 15:04 30/08/00, Kurt D. Zeilenga wrote:
I post to ietf-discuss because two of the referenced
Informational RFCs, RFC 1321 (MD5) and RFC 2104 (HMAC),
are also normative references for numerous other
Standard Track protocols.
Neither of those are protocols. Instead, they specify
mathematical algorithms. As such, I do not believe
they need to be standards-track to be referenced NOR
are they in the set of things IETF would normally
If not, would it be appropriate to submit a request
to the IESG to consider reissuing the specifications
detailed by RFC 1321 and RFC 2104 on the Standard Track?
IMHO, that is not needed and not particularly desirable.