Jeff,
I think we have two different issues here: 1) the format for a
self-signed data item to be used in the certificate request
processing, and 2) what message type to be used in that and analogous
processing.
If one uses the existing certificate format as the data item
to address the first need, then existing code for encoding and
decoding certificates could be employed. The newly proposed data item
can be constructed (perhaps reusing existing code fragments), and is
more compact, but does it warrent whatever extra work is involved?
The second issue can be examined in light of the first, even
though the two are separable. If we used self signed certificates in
conjunction with MIC-ONLY message formats, is the result less secure,
significantly harder to process, etc. (vs. newly defined message
types). Again, the question is whether the effort required to support
this functionality at every PEM UA is significantly different between
the two alterantives, given existing code capabilities.
Steve