So, at the risk of being accused of poisoning the well, let me
ask, other than the "we don't want to change because PEM will be
published as an RFC tomorrow" argument, is there some other
argument against this proposal?
Everything you said with respect to "merging" PEM and MIME is correct.
However, there are two technical points that you do not mention.
1. PEM is, by design, intended only for Internet text messages. In the
general case, PEM will fail miserably if any attempts are made to
use it to send any sort of binary data.
2. PEM includes a step to encode its "output" in a printable form.
In order for PEM to make use of MIME, there would have to be changes to
the specification. PEM and MIME specifications are both on the
fast-track right now. Given how long PEM has already been delayed,
there is a very strong commitment to not slow down the process to change
it to be compatible with MIME.
During informal discussions in San Diego it was agreed that the change
could easily take place during the transition from Proposed Standard to
Draft Standard.
On the other hand, MIME could make use of PEM today. The only thing
that is required is that a complete MIME message be rendered in a text
form (ie perform any necessary printable encodings, eg base64) and then
fed to a PEM filter. Our implementation of PEM has always operated like
a filter and will continue to do so. In fact, I am pretty sure I can
fold PEM into Marshall's implementation of MIME in a few hours.
I believe those are the only two issues of significance.
Jim