Michael,
There is no specification that restricts what lists are allowed to do.
There will, however, be restrictions on who and how a domain's
name in origination addresses can be used soon enough. That is the basic
conflict.
I do not understand what this means.
And after you explain it further, please explain the basis for your certainty
about the future you are predicting.
Not surprisingly, predictions about the future of specific networking
activities/standards have had poor accuracy. Therefore, certitude about it is
always worth explaining carefully.
The danger of making design decisions based on a current batch is that
the next, innovative list will break the signatures. If we think
that's ok, then fine. If we think we are making something that will be
robust against all legitimate list behavior styles, then we are
definitely *not* fine.
That assumes that mailing list software's perogatives
trump all.
We are in a standards arena. Therefore, the question of what "trumps" what
needs to be based on standards, although a historical view of established
practice of course useful.
Here are some relevant facts:
1. History: The world of mailing lists has shown a tendency towards considerable
variation and has succumbed to standardization slowly and poorly.
2. The standards-based constraints on mailing list behavior are rather few,
really.
3. If the DKIM effort is going to make assertions or decisions that constrain
mailing lists beyond their current standards, then we a) must gain agreement
that that is within our charter, and b) must specify those constraints and gain
adoption of them.
I was under the impression that we must design something that lives within the
realities of existing standards, rather than assume that the current
implementations represent the complete range of choices. (And, by the way, I
suspect that existing range is rather wider than you believe.)
They don't. They need to be part of the larger
ecosystem here, and they certainly do not have a god-given
right to preserve the From: address and completely change
the content with complete impugnity.
Well, pretty much, they do.
Absent violations of an Internet standard, a mailing list's software may make
whatever changes the operator of the mailing wants or is willing to tolerate.
The object here is to reach an accommodation between these
two competing needs.
I think that you think we are in a negotiation with the mailing list community.
We aren't.
d/
--
Dave Crocker
Brandenburg InternetWorking
<http://bbiw.net>
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html