David,
Thanks for you posting.
To the specific concerns:
David MacQuigg wrote:
The only thing I might take issue with, if someone were to push, is the
terminology around Actors and ADMDs in section 2. To me, an actor is a
person or organization, not an SMTP relay.
The language for this area of the topic has gone through quite a few changes.
Generally, particular labels didn't jump out at us, demand that we use them, and
have everyone agree. So the challenge has been to find terms that have a
relatively intuitive meaning, for the way they are used. Given that email
touches the human "layer" of the network, it does invite some discomfort to use
a term that we typically use to refer to humans.
However some of that discomfort, for "actor" is helpful, I think. The section
seeks to get the reader thinking about different categories or spheres of
responsibility and control. Since the term 'actor' gets used in computer
science quite regularly, to identify discrete agents of actions, it seems to
work pretty well in this context.
For a relatively typical non-human example, see:
<http://en.wikipedia.org/wiki/Actor_model_theory>
> I don't like the emphasis that
Figure 4 seems to place on boundaries other than what I would call the Border
between two agents with no prior relationship. Again, as long as this is
just a general guide, I can build more detailed models, clarifying the
terminology as needed.
Hmmm. I'm not sure I understand what problem you are raising. But all will be
clear if you offer some wording (or diagram) changes.
Thanks!
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net