Re: Multiple "To:" and "Cc:" header lines in SMTP messages

[Pete Resnick <[e-mail]>]

This thread probably needs to be moved back to DRUMS.

I have some serious reservations about this whole discussion. I am not at
all convinced that we should in any way endorse in the new standard a
workaround for behavior that was broken in the old standard when it also
succeeds in invalidating behavior which is currently legal.

We have a situation now where some older versions of sendmail (numbers as
yet unspecified) bounce messages when they receive messages with large
numbers of addresses in single headers.  This situation shouldn't be
occurring with high frequency anyway since people tend to use either group
syntax or mailing list addresses. Unfortunately, Ned seems dealing with
some set of customers who require it out of his product.

Contrast that with a demonstrably large number of client mailers who (a)
generate messages with large headers (and hence would be causing these
bounces if they were prevalent) and (b) are unable to "properly" parse
multiple occurrences of headers. Given that except for Ned, we have only
heard complaints about (b) and not heard complaints of any magnitude about
(a), I don't see the impetus for now backing off of the old standard and
increasing the burden on mail clients. (I can go through why it's a pain in
the rear for us [and perhaps others] to parse and deal with multiple, and
perhaps discontiguous, recipient headers, but I'll leave that for another
time.)

We have had servers bounce messages for broken reasons in the past. We have
not in general rewritten standards to make all sorts of other behavior
non-conformant to solve the problems of a particular broken server.

And then on the issue of what the standard should say:

On 9/29/96 at 3:24 PM -0500, Ned Freed wrote:

>I disagree; the proper recommendation is "SHOULD NOT generate a single field
>for each separate address". The question of whether or not to generate
>multiple
>field for lots of addresses should be left open, as it is too
>environment-specific to recommend anything at this time.

I don't see any logic in this position at all, Ned. You are saying that we
should require people to handle multiple To: lines on incoming message. Why
not then make the standard for a single address on every To: line?
According to you, everyone's going to have to parse the blasted things
anyway. I see no justification in making this a "SHOULD NOT generate".

>I run into problems with sendmail's limits at a different site a couple of
>times every month on average so this is far from an academic point.

You are the only one I've heard speak in support of that position. Is
anyone else seeing this problem with their products? We certainly are not.

The better way to deal with this situation seems to me to (a) make the
standard say "MUST NOT generate multiple recipient lines"; (b) make
interpreting multiple occurrences as a single concatenated line a "SHOULD";
(c) if you need to generate ultra long recipient lists and have to deal
with a site that is broken, put something short (group syntax?) in your To:
line and put the address list in the body of the message.

pr

--
Pete Resnick <mailto:[e-mail]>
QUALCOMM Incorporated
Work: [phone] / Fax: [phone]