Jeff,
Hmmm... Ok, I see your point, Certainly Steve Kent has made a point of
running a more orderly mailing list and formally disposing of all issues on
the mailing list. He's been silent recently, and I guess it's not well
defined whether, as chair, it's his responsbility to should manage the
dialog or whether the responsbility now falls on the authors of the
proposed specification. I guess I'll accept the challenge and see what we
can do to be responsive. (I'm no longer at tIS and hence no longer in
charge of the implementation, but I'm still one of the authors of the
document and am still active in the PEM WG.)
You raised two issues:
o Addition (and removal) of the PGP identifier
o Kazu's note about the syntax of <keysel>.
With respect to the PGP identifier, there was indeed some discussion not
visible on the mailing list aimed at accommodating the PGP community, and
the result showed up in the spec. There was then reaction against it, and
apparently the decision is to remove it. The criteria for openness does
not preclude side conversations, "design teams" and other communications;
it only requires that the specifications be openly available and that all
viewpoints be discussed before adoption. That's the state we're in now.
(I think the reaction against the PGP identifier was during the meeting;
that was open and reported in the minutes. I'm not aware of any offline
discussion on this aspect of the matter, but perhaps there was.)
With respect to the Kazu's point, that definitely needs a response. I'm
expecting Jim Galvin to respond to that point. This is probably just a
matter of Jim getting back from travel and digging through piles of mail.
Please do raise your points. You've been a valuable contributor and your
views are most welcome. The only challenge is how to reach closure when
viewpoints vary. Thsi WG has had a hard time reaching consensus on quite a
few issues.
I've discovered that even very small subsets of this WG have a difficult
time reaching agreement on specific details.
Steve
At 9:14 AM 12/13/94, Jeff Thompson wrote:
Steve Crocker writes:
o No debate has taken place on the pem-dev mailing list.
Most of the details in the current specification have actually been
discussed in the open in one form or another. Even so, nothing
precludes further discussion.
You are claiming that MIME/PEM issues have been addressed in the open.
This is exactly what I didn't want to hear. They have not been
addressed openly, and hearing you say they have only underscores my
sense that the authors are out of touch with the rest of the pem-dev
voices. Let me clarify this with two examples, one from the past and
one that is currently active.
The decision to add the PGP identifier happened offline from pem-dev.
Shortly after, Steve Dusse posted a message asking why MIME/PEM needs
to support an identifier from another standard which has incompatible
encoding protocols. He suggested that the better approach is for them
to define their own MIME body part. This suggestion enjoyed *zero*
response from the authors. And now recently, the decision was made
offline to remove the PGP identifier, not due to discussion on
pem-dev, but rather some other mysterious process. The same outcome
of Steve's original suggestion has happened but not by addressing
concerns openly. This is extremely disquieting.
For a current example, Kazuhiko Yamamoto writes:
On page 19 and 20, an example of Originator-ID is given as follows;
---
Originator-ID:PK,MHkwCgYEVQgBAQICAwADawAwaAJhAMAHQ45ywA357G4fqQ61aoC1fO6B=
ekJmG4475mJkwGIUxvDkwuxe/EFdPkXDGBxzdGrW1iuh5K8kl8KRGJ9wh1HU4TrghGdhn0Lw8g=
G67Dmb5cBhY9DGwq0CDnrpKZV3cQIDAQAB,EN,2,galvin(_at_)tis(_dot_)com
---
This shows us that <keysel> is '2'.
BNF of <keysel> is defined on page 28 and 32 as follows;
<keysel> ::= <encbin>
<encbin> ::= 1*<encbingrp>
<encbingrp> ::= 4*4<encbinchar>
<encbinchar> ::= ALPHA / DIGIT / "+" / "/" / "="
It seems me that <keysel> should consist of at least 4 characters with
this spec. But <keysel> example up above is only one character.
Please tell me whether my understanding is wrong or the I-D is
inconsistent.
I haven't seen a response to his question.
I have my own questions I'd like to post, but I feel they will be
ignored as Steve Dusse's was before and Kazuhiko Yamamoto's is now.
If responses are being given, they should happen on pem-dev.
I'd like the name of someone personally accountable for addressing
questions, as we enjoyed with Steve Kent in the original PEM effort.
It appears Steve Crocker is no longer directly involved in the TIS
effort (what is CyberCash anyway?), so it has to be someone else.
- Jeff
--------------------
Steve Crocker
CyberCash, Inc. Work: +1 703 620 1222
2086 Hunters Crest Way Fax: +1 703 391 2651
Vienna, VA 22181
crocker(_at_)cybercash(_dot_)com