[Top] [All Lists]

Re: New type suggested: Multipart/Related

1993-05-20 11:05:53

On Thu, 20 May 1993 11:46:04 +0200, Harald Tveit Alvestrand

  alvestrand> Here is my entry to the content-disposition debate.
  alvestrand> This memo defines 2 things, which may or may not be a
  alvestrand> good idea:

  alvestrand> - A way to tell the recipient MIME reader which body
  alvestrand> parts contain only info that is "about" the other body
  alvestrand> parts

I think this makes the relationship between an organizer ( the
encolsing or containing entity) and an included component too
asymetrical. What if someone wants recursive references in their
document? The stuctured approach which you have used strongly implies
that documents are directed acyclic graphs. A quick look at
'multimedia' authoring tools will show you that this is not the case.

After the discussion a few months ago, it was my impression that
adding a new body-part header (content-disposition) was generally
acclaimed as being cleaner, easier to implement and administer, and
easier to parse at the MIME level than using the multipart syntax. I
lamentably have dropped the ball, but am picking it up, and will have
a draft out in a few days if it kills me. :)

Has the general consensus changed? I'd be happy to devote my time to
working on compound document formats, which is how I got drawn to this
in the first place.

  | J. Laurens Troost - UNIX Systems  | At Work: rens(_at_)imsi(_dot_)com    |
  | Investment Management Svcs, Inc.  | At Play: rens(_at_)century(_dot_)com |
  | 12 East 49th Street,  35th floor  |   Phone: (212) 339-2823   |
  | New York, New York         10017  |     Fax: (212) 444-1980   |
     -- IMS is unlikely to share any of the above opinions --

<Prev in Thread] Current Thread [Next in Thread>