On Sun, 2 Mar 2003 nolan@celery.tssi.com wrote:
> There is something bad to say about every major ISP on the issue of
> RFC3464 compliance.
And that doesn't suggest problems with RFC 3464 to you?
Not necessarily.
There are some syntax issues that could reasonably be "blamed" on lack
of clarity or the fact that you need several RFCs in order to get the
complete picture. But then again, that is the nature of IETF standards
and anyone who is serious about such standards should know this and
should do the right thing. Which brings me to my second point.
Frankly, in my opinion, a large ISP or an email service provider should
not be permitted to use as an excuse the fact that IETF standards may
require study to get right. They should be setting an example. Which
is the basis for my comment.
But then again, there is Microsoft....
For example, my issue with AOL and 3464 is that they do not use it all
the time. It seems that the type of DSN you get depends on which part
of their system rejects the message. They also don't report the
Original-Recipient.
With MSN/Hotmail, it is their seemingly random use of "HMOBHMOE" (from
memory that may not be exactly right) as the Original-Recipient, which
is otherwise always reported as the same as the final. Like AOL, they
do not report the Original-Recipient in general.
Now, to be fair, the Original-Recipient is an optional component of the
report. However, as list managers, I'm sure we all appreciate its
intrinsic value. How else are you to track down a subscriber who has
had their email forwarded 1, 2, 3, etc., times before it finally fails.
(Okay, yes, I know there are other infrastructure issues that are in
play here, but the point is still valid.) Yes, there are always VERPs,
but then the whole point is with 3464 you don't need them and you get a
HUGE performance win for large lists (at the list server and the $BIGISP
receiving server). And, if they tell me who the actual subscriber is I
will actually remove them from the list; a win for both of us!
As most of us have never worked for a major ISP, and are unlikely to
do so, perhaps someone could analyze what concerns or problems large ISP's
may have with RFC 3464 compliance.
I really doubt they're all being non-compliant just out of spite.
I don't think they do it out of spite. I think they do it for the money
out of ignorance. We wouldn't be a capitalist society if this were not
so.
Jim
References:
|
|
