ietf-mxcomp
[Top] [All Lists]

Re: Ruminations

2004-03-27 14:55:15

Mark C. Langston wrote:
On Sat, Mar 27, 2004 at 01:50:11PM -0600, Gordon Fecyk wrote:

Some of those changes
are going to break some functionality that a minority of users rely on

[...]

As I see it, the answers as to what should be broken and what should not are
already being decided by the users.

Gordon, could you clarify this apparent contradiction?  One the one
hand, it appears to me that you're okay with breaking user-desired
functionality.  On the other, it seems to me that you think the users
should be the arbiter of what's broken.

Is it that you're using two different senses of the term "user"?  In
your first statement, the "users" to which you refer seem to be
end-users -- those people sitting at a computer, typing email into an
MUA.  In your second statement, the "users" to which you refer seem to
be people responsible for running MXes and MTAs -- those people who
control the filtering that occurs at the points prior to incoming email
being queued.

Do you consider both groups to be the same?

It looks like he's referring to 2 sets of users, which probably have some overlap. The first set of users, under just about any proposal, are those that do direct-to-MX delivery in the MUA or on a roving MTA. Obviously, others could fit into this group of users as well, I'm just presenting the most apparent. The second set of users are anyone who runs a filter at any level. That filter could be at an edge MTA, or it could be in an MUA.

I think the clearest way to resolve this is to reword what Gordon said as follows: (Gordon, please correct me if I'm wrong) 1. Some of those changes are going to break some functionality that a minority of *senders* rely on 2. As I see it, the answers as to what should be broken and what should not are already being decided by the [intermediate and final] *receivers*.

Philip Miller


<Prev in Thread] Current Thread [Next in Thread>