non-text documents

• Mail folder: WWW Talk 1992 Archives
• Next message: Jim Davis: "document caching"
• Previous message: Tim Berners-Lee: "Re: page break in HTML"
• Reply: Edward Vielmetti: "Re: non-text documents "
• Reply: Dan Connolly: "Re: non-text documents"

Date: Fri, 16 Oct 1992 12:11:46 -0400
From: Jim Davis <davis@dri.cornell.edu>
Message-id: <199210161611.AA06348@willow.tc.cornell.edu>
To: www-talk@nxoc01.cern.ch
Subject: non-text documents

Can you tell us when WWW and Viola will support non-text
documents?  I would very much like to be able to return
pictures including Postscript.  I know this has been discussed
in the past.

One issue is how will the client know what kind of data
is being returned.  It seems there are several possibilities:

1) Make the client figure it out itself by examining the data (e.g.
for "magic numbers").  Simplest, but makes the client do too much
work.

2) Use typed links.  A server which wishes to return non-text data
first returns a constructed document containing anchors where the link
type somehow encodes the data type.  Benefit here is that it allows
the user to choose the kind of data if there is a choice.

3) the HTTP protocol to allow an HTTP server to inform the client of
the kind of data to be returned, maybe even negotiate about it.

4) Extend HTML to allow embedded non-text data, perhaps by using MIME.
Servers will probabaly construct a new document for each query, using
MIME's external reference feature where possible to avoid copying
large data files.

best wishes