RE: What is CONTENT?

["Iacovou, Danny" <danny.iacovou@ebenx.com>]

> -----Original Message-----
> From: Mark Day [mailto:markday@cisco.com]
> Sent: Tuesday, November 07, 2000 5:09 PM
> To: cdn@ops.ietf.org
> Subject: What is CONTENT?
> 
> 
> The current (-02) draft of the model document includes a 
> formulation of
> CONTENT in terms of CONTENT DATA UNITs.  No-one on the design 
> team seems to
> like this approach very much. I am happy to replace these 
> definitions or
> eliminate them entirely, but I find the leading proposed replacement
> somewhat unsatisfactory.
> 
> Barron's suggestion is that CONTENT is a "typed finite 
> sequence of octets."
> While this is correct, it seems incomplete.


  Maybe not the time to raise these questions, but we are talking about
  CONTENT. I have a video clip. This same video clip is in english,
  spanish, and french. Is CONTENT the video clip bits of each of these
  three cases - in which case I have 3 CONTENTS? Or is CONTENT the 
  "video clip" which comes in three different flavors?

  I would like to say "give me this video clip in spanish, if not then
  in english, if not then in french". Sort of like Gopher's "multiple
  views".

  A slightly different example along the same lines, which poses
  different problems.A have a speech, the text of it, a video of the
  speaker delivering the speech, and an audio only version of the speech.
  I ask for the speech. Is that the CONTENT? Or do I have to ask for a
  specific version of it? Is *that* the CONTENT? This is different from
  the previous example because now it takes different amounts of bits 
  to store each version of CONTENT.
  
  I like the notion that CONTENT consists of CONTENT DATA UNITS, because
  that leaves a space to associate forms of meta-data to the CONTENT DATA
  UNITS. I think the meta-data should be part of CONTENT.

  I've read the drafts and I am not sure where meta-data fits in the
  current architecture. In my opinion I think meta-data should be added
  in from the start, and not as an after thought as it is often done. I
  think today's Internet would be a far far more pleasent place to work
  if one could query based on meta-data attributes.

  BTW: I am not trying to imply that a canister ontop of CONTENT which
       holds multiple views, and other forms of meta-data couldn't be
       a solution.