Actually I ask the same question in reverse. There really
needs to be some pretty good reasons why we would even
consider S/MIME.
S/MIME is common building block in many systems today. Its structures
have proven flexible enough for use in e-mail messaging (mua-mua;
mta-mta in communities of interest). It also provides security
components for SIP and I have found it in web-based form signing, off
the shelf file encryption and disk encryption products. S/MIME has loads
of running code, is present in existing product libraries and is proven
to robust survivability through an array of e-mail transmission
scenerios. Much more than a "niche" player I wouldn't you think? Given
this proven track record of interoperability and adaptability I would
need to see some very good reasons to __need__ to create "something
else".
If the best solution absolutely requires "something else" then as the
best solution it should succeed and be supported by participants of this
group. I would require quite some convincing that we do need to create
something else in regard to a signature format.
Craig.