[Mailman-Users] A modest proposal: Reply-To munging considered *carefully*

[Barry Warsaw]

[Your original message probably didn't make it to mailman-developers  
since it was spelled "mailman-devel at python.org" -BAW]

On Oct 10, 2009, at 4:19 AM, Stephen J. Turnbull wrote:

> 2.  A formal proposal of a new header field, Responses-To, whose sole
>    purpose is to allow mailing lists conforming to RFC 2369 to hint
>    to MUAs whether a reply or followup is usually appropriate on the
>    list.
>
> I've cross-posted to Mailman Developers because I'd like to solicit
> the opinions of those folks about whether Mailman should implement
> this draft RFC early in the process.  Concrete proposals are a ways
> off, though.  Followup to Mailman Users, please.

So a couple of things.

I'm not opposed to implementing draft RFCs or even defacto proposals  
in Mailman as long as we label them experimental.  IOW, should the RFC  
change, we'll obviously need to change Mailman to conform.  A lot of  
email lore is just defacto standards and we already support many of  
them.  We'll take them on a case-by-case basis, but Mailman should not  
blindly rule out supporting them.  For the ones we like, Mailman  
support can go a long way to proving the feasibility and perhaps  
nudging others to support the proposals as well (see my Archived-At  
contents proposal).

In general, I like Stephen's proposal as a way to help reduce the  
ambiguity in this very common workflow.  I'll bikeshed on this detail:  
I think the header should be called Response-Precedence or maybe List- 
Response-Precedence.

The 'List-*' prefix is evocative of RFC 2369 and since this is a list  
specific header, it should probably be prefixed as such.  A counter  
argument would be that because it's the user and not the list that's  
making the value choice, it should not be a List-* header.

Response-Precedence is evocative of the defacto Precedence header, and  
it makes some sense to me because you're specifying the precedence of  
where responses should go, based on the conditional you posted.  *-To:  
headers feel more like they should contain an address but this has a  
limited vocabulary, as does Precedence, so I like it not being a *-To  
header.

As for going the RFC route, I think it's not a bad idea, but I have no  
idea how much work is involved.  ;)

-Barry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 832 bytes
Desc: This is a digitally signed message part
URL: <http://mail.python.org/pipermail/mailman-users/attachments/20091012/697a8d01/attachment.pgp>