RE: DOM-TV (was RE: ATVEF uri )

From: Adams, Glenn (gadams@spyglass.com)
Date: Wed, Feb 24 1999


Message-ID: <D181361D7C86D011925700805FFE898E01F38718@spybem01.nap.spyglass.com>
From: "Adams, Glenn" <gadams@spyglass.com>
To: "'Ted Wugofski'" <Ted.Wugofski@otmp.com>, "'www-tv@w3.org'" <www-tv@w3.org>
Date: Wed, 24 Feb 1999 11:35:41 -0600
Subject: RE: DOM-TV (was RE: ATVEF uri )


I agree entirely with your comments. I would suggest that ATVEF best be
served by (1) not referencing DOM0 or DOM1 behavior; but (2) explicitly
specifying the set of DOM classes/methods/properties/events and their
semantics that it will depend on in serving ATVEF functions. In doing (2),
there will be no ambiguity about the state of specification; there will be a
focus on TV related functions; and, there will still be an opportunity to do
this specification by using DOM1 as an example (retaining or abandoning
compatibility as desired - though, clearly, retaining compatibility would be
preferred, all things being equal).

		-----Original Message-----
		From:	Ted Wugofski [mailto:Ted.Wugofski@otmp.com]
		Sent:	Tuesday, February 23, 1999 9:58 PM
		To:	'www-tv@w3.org'
		Subject:	DOM-TV (was RE: ATVEF uri )

		<readying myself for slinging arrows>

		The current DOMs, whether we acknowledge that DOM-0 exists,
DOM-1, and
		the now-in-draft-stage DOM-2, are *not* sufficient for
enhanced
		television.

		If we assume that HTML will be used on television in the
same manner as
		HTML is used on computer screens, then these DOMs will
suffice.  But...

		I do not think HTML will be used the same way when creating
enhanced
		television content.  There will certainly be changes to the
event model,
		there is likely to be a stronger multimedia component
(SYMM-related
		stuff), there will likely be a different user agent paradigm
(i.e., no
		browser with a forward/backward/bookmark button).  In
addition, the
		content model might change in order to reduce
complexity/increase
		robustness of the receiver.

		Therefore, I strongly believe that there will be a DOM-TV.
What that
		DOM is, I don't know yet.  I do know, however, that it will
considerably
		overlap with DOM-1 and DOM-2.  

		In my recent review of DOM-0 (similar to Mr. Adams survey),
it became
		clear to me that starting from DOM-0 is not a good idea (to
be polite):
		its a non-standard, it overlaps with DOM-1 and DOM-2 which
are or will
		be standards, and it DOM-0 will still need to be extended.

		DOM-0 is undefined and poorly designed.  Mixing DOM-0 and
DOM-1 and
		DOM-2 is a disaster waiting to happen.  They have objects
and interfaces
		with similar names but different properties and methods.

		My recommendation is that the basis for a DOM-TV be DOM-1
and DOM-2.
		They are (or will be) standards with a clearly defined
specification.

		Addressing Mr. Dolan's comment that DOM-1 is too much, I beg
to differ.
		DOM-1 is considerably less than DOM-0.  DOM-1 does not
provide access to
		the Event model, User Agent, or CSS.  

		There is redundant functionality of the DOM-1 HTML
interfaces with the
		DOM-1 Core interfaces.  The HTML interfaces are
HTML-specific
		simplifications of features accessible through the XML-based
Core
		interfaces.  A receiver manufacturer could choose to only
support the
		HTML interfaces (which are certainly lighter weight) or they
could
		choose to only support the Core interfaces (which provide a
complete
		solution). The issue then becomes code-reuse and training.

		Ted



		> -----Original Message-----
		> From: Michael A. Dolan [mailto:miked@tbt.com]
		> Sent: Monday, February 22, 1999 8:55 AM
		> To: Rob Glidden; Adams, Glenn
		> Cc: Ted Wugofski; www-tv@w3.org; Philipp Hoschka
		> Subject: Re: ATVEF uri 
		> 
		> 
		> Rob/Glenn-
		> 
		> Noone in TV-land is trying to characterize IE and Netscape

		> behavior.  The
		> issue is that DOM1 is too much, and there is a need for 
		> something less.
		> 
		> We can call it DOM0v2, or DOM0.5, or DOM-TV-0
		> 
		> DOM-TV-0 may never be used on the Internet, but that
doesn't 
		> make it any
		> less fruitful to pursue...
		> 
		> 	Mike
		> 
		> At 08:39 AM 2/23/99 -0800, Rob Glidden wrote:
		> >
		> >-----Original Message-----
		> >From: Michael A. Dolan <miked@tbt.com>
		> >To: Philipp Hoschka <ph@w3.org>
		> >Cc: www-tv@w3.org <www-tv@w3.org>; Ted Wugofski 
		> <Ted.Wugofski@OTMP.com>
		> >Date: Sunday, February 21, 1999 3:07 PM
		> >Subject: Re: ATVEF uri
		> >
		> >
		> >>Philipp-
		> >>
		> >>Excellent question.  I am pushing ATVEF that this be 
		> addressed as part of
		> >>an overall standardization effort of all the items in
the spec.
		> >>
		> >>Any help W3C would like to offer on this topic (pointer
to 
		> the old DOM0
		> >>document that was previously there at W3C, or other
legacy 
		> DOM0 work) to
		> >>help us define it would be greatly appreciated.
		> >
		> >My interpretation of "DOM0" was that it was simply a
reference to the
		> >"unspecified situation before DOM", much like "DHTML" was
a 
		> reference to an
		> >unspecified collection of various features from various
vendors.
		> >
		> >So "DOM0, version 2" seems like an unfruitful pursuit.
		> >
		> >Rob
		> >
		> >>
		> >>Thanks,
		> >> Mike
		> >>
		> >>At 06:22 PM 2/21/99 +0100, Philipp Hoschka wrote:
		> >>>
		> >>>the new ATVEF 1.1 spec (dated 2 Feb) is at
		> >>>
		> >>>http://www.atvef.com/atvef_spec/TVE-public-1-1r26.htm
		> >>>
		> >>>one question: the ATVEF spec says:
		> >>>
		> >>>Mandatory support is required for the following
standard
		> >>>specifications:
		> >>>
		> >>>...
		> >>>- DOM 0
		> >>>...
		> >>>
		> >>>
		> >>>The W3C DOM rec states
		> >>>
		> >>>"The term "DOM Level 0" refers to a mix (not formally
specified)
		> >>>of HTML document functionalities offered by Netscape
		> >>>Navigator version 3.0 and Microsoft Internet Explorer
version 3.0.
		> >>>In some cases, attributes or methods have been
		> >>>included for reasons of backward compatibility with
"DOM Level 0"."
		> >>>
		> >>>Given that DOM level 0 is not formally specified, how
can one
		> >>>test whether it is supported in ATVEF ?
		> >>>
		> >>>On 15/02/1999, Ted Wugofski <Ted.Wugofski@OTMP.com>
wrote:
		> >>>>You might want to look at the latest ATVEF
specification 
		> (1.1), which
		> >>>>provides a new and improved URI system.  Off the top
of 
		> my head, the URL
		> >>>>is http://www.atvef.com
		> >>>>
		> >>>>Ted
		> >>>>
		> >>>>-------------------------------------------
		> >>>>Ted Wugofski         voice: +1 817 285 1853
		> >>>>Gateway              fax:   +1 817 285 9567
		> >>>>
		> >>>>mailto:ted.wugofski@otmp.com
		> >>>>http://www.gateway.com
		> >>>>
		> >>>>
		> >>>
		> >>>
		> >>>
		> >>------------------------------------------------------
		> >>Michael A. Dolan, Representing DIRECTV,  (619)445-9070
		> >>PO Box 1673 Alpine, CA 91903        FAX: (619)445-6122
		> >>
		> >>
		> >
		> >
		> >
		> ------------------------------------------------------
		> Michael A. Dolan, Representing DIRECTV,  (619)445-9070   
		> PO Box 1673 Alpine, CA 91903        FAX: (619)445-6122
		> 
		>