[Mailman-Users] Digest message format?

Gregory Leblanc GLeblanc at cu-portland.edu
Tue Apr 25 21:59:09 CEST 2000

> -----Original Message-----
> From: J C Lawrence [mailto:claw at cp.net]
> Sent: Tuesday, April 25, 2000 12:53 PM
> To: Gregory Leblanc
> Cc: Mailman-Users list (E-mail)
> Subject: Re: [Mailman-Users] Digest message format? 
> On Tue, 25 Apr 2000 11:31:14 -0700 
> Gregory Leblanc <GLeblanc at cu-portland.edu> wrote:
> > I realize that HTML email is evil, but I think this is one place
> > where it would be REALLY useful.  I'm subscribed to a couple of
> > digested mailman lists, and I've tried both plain text and MIME
> > encodings, neither is satisfactory.  It seems that all of the
> > pieces are there, just not put together in any useful fasion.
> > Here's what I'd like to propose as a "option": HTML messages.  The
> > digest functions already parse the messages, remove the subjects,
> > and chop it up into pieces for the MIME parts, and stick
> > everything together in a semi-coherent fasion for the plain text
> > version.  To get useful HTML, all it would take would be to have
> > the digest function put in a name anchor (you know, the little
> > page.html#gohere thing) where it puts the messages together, and
> > put links into where it puts all of the subjects that point to
> > those name anchors.  Plus, if you make (leave?)  the default
> > digested mode as plain text, you won't get people who can't handle
> > HTML email getting it, the same way that you avoid that with MIME.
> Reliably and correctly parsing and reconstructing MIME messages is a
> non-trivial problem.  What you are proposing is even less trivial.
> That said, I'd argue that the major problem with digests is that
> people read them rather than bursting them back into individual
> messages and then reading those messages as per normal.  Digests are
> very useful in helping handling transport related problems (eg
> message count, mailbox quotas, work interruptions, etc), but are an
> absolutely lousy form for people to actually use, read, or
> manipulate.  

HTML digests solves that problem.  Besides, how would you break the digest
up into useful parts again, it's already been smeared together.  I think
HTML is a much better solution to this problem.  I don't know python, so I'm
not sure how well suited to this it would be, but it's probably less than 20
hours coding to put together an alpha quality release of HTML digests for
somebody who does know it, or something similar using Perl.  

More information about the Mailman-Users mailing list