Re: Don't put descriptive info in a URI, only identifying info

From: Michael A. Dolan (miked@tbt.com)
Date: Mon, Dec 21 1998


Message-Id: <3.0.5.32.19981221141521.00976650@cts.com>
Date: Mon, 21 Dec 1998 14:15:21 -0800
To: fin@finseth.com
From: "Michael A. Dolan" <miked@tbt.com>
Cc: gomer@lgerca.com, connolly@w3.org, ietf-url@imc.org, www-tv@w3.org
Subject: Re: Don't put descriptive info in a URI, only identifying info

At 02:26 PM 12/21/98 -0600, Craig A. Finseth wrote:
>	...
>   >I believe that in most, if not all cases, the time dimension will be
>   >specified by identifying the "event" (TV program, in ordinary
>   >terminology), with perhaps in some cases an offset relative to the start
>   >of the event, rather than by explicitly specifying a clock time.
>
>   This is one way to do it that is more object-oriented and fits the object
>   model much better.  However, there is not agreement on the list that this
>   is the (only) possible or useful solution.  There are other possible URI
>   designs that do not directly reference an EPG object, and thus might need
>   absolute (GMT) time.
>
>Umm, you just proved the opposite of your point.
>
>We all agree (I believe) that there is a requirement to designate
>time-related content on a channel

yes.

>(e.g., an event).

Not if you literally mean an event in EPG terms.  This implies a solution
limited to existing EPG event data only, which I do not agree with.

>The open question is whether that requirement may _only_ be met
>by putting time-related information in the URI itself.

It is not that simple.  Specifying time (or not) in a scheme is not the
only design criteria.

>You agree that there are valid schemes that meet the requirement
>without such encoding.

Syntactically "valid", yes.  The best design - not necessarily.

>That indicates that the encoding is not itself
>part of the requirement.  (The fact that one can devise schemes that use
>the encoding is not relevant to the question.)

There are always multiple solutions to a problem.  I am not advocating
multiple concurrent schemes to do the same thing, nor am I in any way
advocating that eliminating time encoding is the utmost design goal in
meeting the URI requirements.  There may be other reasons for needing a
time-based approach that outweigh a general desire to avoid it.  In the
end, we must consider all reasonable solutions, of which there will be
compromises, including possibly desires to eliminate time in the URI(s).

	Mike

---------------------------------------------------------------------------
Michael A. Dolan, Representing DIRECTV,  (619)445-9070   FAX: (619)445-6122
PO Box 1673 Alpine, CA 91903, Overnight: 20239 Japatul Rd, Alpine, CA 91901