reply-all should be discouraged

cognitive.libertarian+ml at gmail.com cognitive.libertarian+ml at gmail.com
Tue Feb 23 17:00:31 EST 2010


* Jeff Kletsky <jmk at wagsky.com> [2010-02-22 18:21]:
> Not RFC2822
>
> Non-standard header not beginning with X-
>
> http://tools.ietf.org/id/draft-ietf-drums-mail-followup-to-00.txt expired 
> in 1998 and contains the comment
>
> This proposal does not represent any consensus opinion in the drums
> working group at this time (21 November 1997).

You're making the same mistake Fred Bone did.  While the RFCs make a
good guide for what's proper, you're trying to reverse that and expect
everything proper to appear in an RFS.  There are copious good
practices to message composition (like trimming quotes and not
top-replying to a message) that do not find their way into an RFS.

> Insisting everyone support your non-standard demands sounds like
> Microsoft's approach to the Internet.

Couple problems with this interpretation.  I'm certainly not demanding
that everyone (or anyone) compose messages using the mail-followup-to
header.  I'm simply advocating it as a proper means for someone else
with a "demand" to have their request met automatically - enabling
tools (as opposed to humans) to do the job.  If someone chooses not to
use it to request copies, I really could not care less as it's for
their own benefit.

Moreover, this is not the way of microsoft.  Quite the contrary.
Microsoft would rather endorse your stance of having humans do the
work (manually state reply preferences in the body), so they could
then sell an extra premium accessory that would find such requests and
compose accordingly.  Then it could be patented so others couldn't use
the algorithm.  Then their next version would embed a flag in an
MSTNEF attachment, and certainly not make use of an openly used header
already used by the GNU community.  

> Why don't you insist that all mailers natively support
> Exchange-isms?

You've obviously completely missed the thesis of the discussion.  It's
actually users of poorly developed MUAs (like Outlook) that have a
harmful effect on list policies.  Catering to unsuitable MUAs is
precisely what downgrades the quality of the forum.  You have to
realize that some users have lousy clients, and some have advanced
clients.

> I agree, for those that know about the header and have agents that
> support it, it is convenient. 

Catering to the deficient clients drags everyone down, and it's a poor
philosophy to support.  The way forward is to, at a minimum, ensure
that quality open tools play well with the list settings.  Some of
those using nonstandard proprietary garbage will be compelled to get a
better client, and rightly so.

> However, it is not appropriate to demand that a non-standard feature
> be supported by all agents. 

What's not appropriate is to use a straw man to convert someones
advice for a good practice, and represent this point as some kind of
demand.  Again, no one on this list demanded that anyone use the
mail-followup-to header.  And no one has to.  But it remains an
effective option for those that use it.

> That is exactly what the standards-track RFCs are for.

RFCs do not even begin to try to cover all good practices and
conventions.  They're good at what they cover, but to take this rigid
stance that something cannot be proper unless it's in an RFC is
absurd.

> I find that there is plenty to filter on in the headers emitted by
> the GNUCash mailing lists.

Precisely my point.  Yet not a single one of these list headers appear
in public replies masquerading as personal replies.



More information about the gnucash-user mailing list