Re: [Evolution-hackers] Bug #50046.

[David Woodhouse <dwmw2 infradead org>]

On Tue, 2004-01-06 at 12:20 +1030, Not Zed wrote:
> I still don't see what your issue is here.  We just don't display it, it
> isn't lost.  Just like we don't display comments in addresses either.

There are two separate problems with the address lists, and one problem
with X-header formatting. All are caused by the same fundamental design
problem w.r.t. the way we think about user interaction.

The first problem is that when I _send_ a mail with a list in the
headers, that list really is _lost_. Not present in the mail which gets
submitted to the SMTP server. Gone. Absent.

Look for the 'To: Some people : ;' header which I just painstakingly
typed into this mail. It was pure person-to-person communication as
documented by RFC2822, and Evo ate it.

The second problem is that even if I'd sent this mail with pine to
ensure the To: header isn't removed, you'd not be able to see it. Yes,
you could argue that it's not 'lost' in that case -- it's still present
in the underlying mail spool. But to be honest, if a mail reader strips
something from a mail it's supposed to be displaying, with no hint that
information is being removed, and the user has no idea that perhaps they
should try a different mail reader on this mail because there's more
information there than Evolution likes to display, that information
really does get lost to the _user_. 

And the third is the reformatting of X-headers by considering various
lengths and types of whitespace to be interchangeable for the purpose of
displaying text to the user, rather than only for the purpose of
_interpreting_ the content of mails. Go talk to Anna about how
whitespace (and borders and margins) are all syntactic sugar and we
should consider them with contempt when the sender has explicitly laid
something out in a way that will be readable by the recipient. Hint:
don't stand too close.

These are three technical issues which I'm happy enough to look at
fixing. They highlight a severe underlying user-interface misconception
which seems to be causing resistance to the fixes which I'm suggesting.

-- 
dwmw2