Re: [Evolution] Dealing with broken mailers

[Not Zed <notzed ximian com>]

On 12 Mar 2001 22:05:36 -0800, Richard Zach wrote:
> I've recently suggested a couple of enhancements for Evolution to better
> deal with messages sent by "broken" (ie, non-standards-compliant) mail
> user agents. To wit, in
> <http://bugzilla.ximian.com/show_bug.cgi?id=1751>  I pointed out the
> nasty habit of some mailers (in particular, Eudora!) to generate From
> headers of the form "From: <address> (by way of <resender-address>)" for


If you want to see such details (which aren't even important) you can
use view-source.

> resent messages, which Evolution does not display in their entirety, and
> in <http://bugzilla.ximian.com/show_bug.cgi?id=1688> suggested that
> Evolution should deal gracefully with non-standard back-reference
> prefixes (e.g., "RE:", "Re(2):", "AW:").


Well the fact that is deals with "re:" at all is merely a cosmetic
thing, if you wanted it to deal with everything in existance it would
never be complete.

> It seems neither of these will be addressed. I don't care too much about
> these particular issues, however, the rationale given seems to me to be
> wrongheaded. The rationale is this: If it's fixed in Evolution, it will
> encourage other mailers to keep on being broken.  I'd agree that working
> around or accomodating messages generated by broken mailers should not
> be a priority (at least right now, pre-1.0).  To suggest, however, that
> intentionally making Evolution not deal in an optimal way with messages
> generated by Eudora (on the suggestion that to do so would encourage
> Eudora to keep on generating broken headers) is just to shoot yourself
> in the foot. Even if Evolution could have that kind of effect of making
> other software standards-compliant--which I don't think it will--it
> leaves the problem of older versions of broken mailers, which certainly
> won't be retroactively fixed.


However, neither of these issues actually affect interoperation with the
given mailers in any practical way.  It is not like you cannot actually
send or receive mail from these mailers, for instance, or that any
(particularly) useful information is lost whilst doing so.

> In my opinion, optimal interoperability and compatibility with existing,
> if broken, MUAs should be a priority, especially if the MUAs in question
> are used by (I suspect) millions of people.


Well IMHO there are other reasons not to waste time on 'fixing' these
interoperability issues, including the time wasted!