Re: one more requirement

See comments below.

I think the bottom line is that I can see a good deal of utility in many of the
things being done with the "tv:" syntax, but in the case when the protocol
designator stands alone without a channel designation of some sort I think it
would be better viewed as an extension of HTML syntax, rather than a URI.


Dan Zigmond wrote:

> At the risk of degenerating into a purely philosophical debate, I'm not sure
> I agree that the "tv:" URL doesn't describe an "image."  The television
> broadcast is an image of sorts, and certain behaves like an image in terms
> of page layout.  (In other words, you can put it anywhere on an HTML page
> where you can put an image, and image attributes around position and size
> make perfect sense for television broadcasts.)  One of the big advantages of

Well, a television broadcast is an image of sorts, but there are some
fundamental differences. For example,
multiple images can appear on the same page without conflict. If multiple TV
broadcasts appear on the same page, what happens to the sound?There are many
different HTML tags with attributes for position and size which make perfect
sense for television broadcasts. This doesn't mean they should all be used for
TV broadcasts, any more than they should all be used interchangably for their
intended purposes.

> using a URI for TV broadcasts was that it can then integrate so well with
> the rest of HTML.  For example, the OBJECT tag provides a nice way of
> authoring content that can either include a TV broadcast or not, depending
> on the client:
>
>
> <OBJECT src="tv:">
>
> <OBJECT src="http://foo.com/still_image.png">
>
> </OBJECT>
>
> </OBJECT>

I can see the value of this, but it bothers me that there are a lot of
application level assumptions governing the interpretation of this "URI". For
example, how does a receiver with two tuners interpret it, since there are then
two default channels? Perhaps another way to accomplish the same thing
is:<OBJECT type="tv">
...
</OBJECT>
The application level assumption here is that if no src is specified, then use a
default source for the designated type.

>
>
>
> It also provides a natural URI for "pure TV"; when you want to get rid of
> all enhanced content and go back a full-screen television image, you simply
> go to the "page" described by "tv:" itself.

What would the syntax for this be? Is it just <img src="tv:"> as the only
content of the page body? If so, couldn't <tv> or <object type=tv> </object>work
just the same way? Or is the idea that you would have a hyperlink with
href="tv:"? Hmmm ... I admit I don't see a better way of doing this.

>
>
> At the very least, a television broadcast seems just like any other video.
> So a <TV> tag makes sense only if you also imagine a <VIDEO> tag.  If you
> believe ordinary video clips should be embedded in HTML using the OBJECT tag
> with URIs pointing to the source of the video, then using this technique for
> television as well seems to make sense.

Well, in this case the URI really does point to the source of the video. It
really tells you which video you want played. I would not have any problem
with:<OBJECT src="ntsc://wnyc">
</OBJECT>

>
>
>     Dan
>
>
> ---------------------------------------------------
> Dan Zigmond
> Senior Manager, Interactive Television Technology
> WebTV Networks, Inc.
> djz@corp.webtv.net
> ---------------------------------------------------
>
>
>
> -----Original Message-----
> From: Gomer Thomas [mailto:gomer@lgerca.com]
> Sent: Wednesday, January 06, 1999 12:22 PM
> To: WWW TV List
> Subject: Re: one more requirement
>
> Mike -
>
> To add these to the application scenario list we need enough detail to
> understand the application scenarios. How are these URIs used? Where do they
> come from? How is the application supposed to treat them?
>
> My understanding is that they typically appear in HTML pages. I think one
> example given in the WebTV internet draft is <img src=tv:>. Presumably the
> HTML engine rendering the HTML page is supposed to play the currently tuned
> channel in the space defined by the other parameters of the img tag, or use
> some default size if no size parameters are specified? Where would the HTML
> page typically come from in this scenario? A web site? VBI line 21 of an
> NTSC broadcast channel? I.e., what is the context within which a designation
> such as "7" or "abc" should be interpreted?
>
> I do not believe that we need to proserve legacy syntax, but I do believe
> that we need to ensure that the application requirements met by that syntax
> is met in some way within our framework.
>
> For example, my own personal view is that the syntax <img src=tv:> is
> unfortunate. It says to put in an image space something which is not an
> image in the usual sense, and it uses the syntax of a URI for something
> which is not identifying anything. If I had been part of the original design
> team, I would have recommended syntax more like the following for your three
> examples:
>
> *       <tv>
>
> *       <tv chan=7> or possibly <tv src=ntsc://7>
>
> *       <tv network=abc> or possibly <tv src=ntsc://abc>
>
> and for the other example sometimes given in this context:
>
> *       <tv station=wnyc> or possibly <tv src=ntsc:wnyc> (but how to
> distinguish network from station?)
>
> This proposed "tv" tag could of course have other appropriate attributes for
> specifying size, volume level, closed captioning, etc., or perhaps
> parameters like volume level and closed captioning could be specified by
> queries at the end of the URI.
>
>
> Michael A. Dolan wrote:
>
> I think we should add to the application scenario list (or requirements,
> whichever is more appropriate) the existing commercial practice for "tv"
> URI's in existing (NTSC) broadcast systems.  This list may not be complete,
> but the ones I know about are:
>
>         tv:                    [reference to the currently tuned channel]
>         tv:<channel_number>    tv:7
>         tv:<network_name>      tv:abc
>
> This syntax obviously presumes a specific transport and is very limited in
> scope, but one could envision a more complex scheme where this syntax falls
> out as valid *relative* URI's.
>
> In any event, I think this syntax should be addressed and accommodated if
> possible as we move forward with our designs.
>
>         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
>
> --
> Gomer Thomas
> LGERCA, Inc.
> 40 Washington Road
> Princeton Junction, NJ 08550
> phone: 609-716-3513
> fax: 609-716-3503



--
Gomer Thomas
LGERCA, Inc.
40 Washington Road
Princeton Junction, NJ 08550
phone: 609-716-3513
fax: 609-716-3503

Received on Wednesday, 6 January 1999 17:38:59 UTC