Re: Wiki requirements and implementation alternatives update

From: Giuseppe Pascale <giuseppep@opera.com<mailto:giuseppep@opera.com>>
Date: Friday, November 15, 2013 9:51 AM
To: Bob Lund <b.lund@cablelabs.com<mailto:b.lund@cablelabs.com>>
Cc: public-inbandtracks <public-inbandtracks@w3.org<mailto:public-inbandtracks@w3.org>>
Subject: Re: Wiki requirements and implementation alternatives update

A bit off-topic, but the text right below the "implementation alternatives" also need some more generalization.

In particular where you say "a kind from the HTML5 kind table if the UA can determine the kind" as I'm not completely sure if there is always an obvious mapping between the various format and the "kind"s available in HTML5.

<Bob> You are right; there isn't always a mapping which is why the metadata is needed in the first place. I thought the sentence said "UA should set kind ..." I'll double check. </Bob>


This is also true but what I meant is different, that is: there may be a mapping but may not be obvious or unambiguous or even unique.

The last column in  the table here http://www.w3.org/TR/html5/embedded-content-0.html#audiotracklist-and-videotracklist-objects describes for various formats how the different kind values should be generated. Enumerating rules in this way would make the mapping obvious and unambiguous.

If you see the OIPF contribution sent earlier on this list you will notice how the value of kind there is related to the existence of certain other fields/values in various places.

Those rules could be added to the table for MPEG-2 TS – notice the 'main-desc' kind in the referenced table.


(regardless if the OIPF proposal is correct or not, the problem still exist)

/g