In article
<M12392(_dot_)001(_dot_)6e950(_dot_)1(_dot_)941221193152Z(_dot_)CC-MAIL*_OU=SECCG_OU=AZBH_PRMD=MOT_ADMD=MOT_C=US_(_at_)MHS>
Paul_Lambert-P15452(_at_)email(_dot_)mot(_dot_)com writes:
As a start, on documenting the requirements:
Why does MIME-PEM need 5 identifiers and 3 name forms?
What problems (requirements) are these mechanisms attempting to solve?
There is a desire to allow people to use alternatives to the
currrently defined (and not very successful) PEM hierarchy. The
mechanisms as defined are flexible and work well. I do not think they
are excessive. They have been discussed to death for very long periods
of time. I, for one, am quite happy with the specification and would
like to see it remain exactly as it is now, modulo grammar and other
similar fixes. It is far too late in the process for us to consider
revising the whole way of thinking that went into this spec unless
someone can demonstrate a real concrete problem with it.
The spec is also perfectly implementable -- you can get
implementations NOW. Its thus fairly obvious that there is no extant
problem in implementation caused by its flexibility.
To summarize, I think we have a spec that more than covers everything
we want to be able to do, does not cause us any harm with the
provisions it makes for things beyond what we might immediately want
to do, and has been implemented. It is time to move on.
--
Perry Metzger perry(_at_)imsi(_dot_)com
--