Re: proposed updated response to MPEG on codecs

On 23/10/2014 13:29, "David Singer" <singer@apple.com> wrote:

>
>On Oct 22, 2014, at 18:05 , Glenn Adams <glenn@skynav.com> wrote:
>
>> I suppose here you are referring to, e.g., specifying #color in a
>>profile definition document, but further restricting content to a subset
>>of the value space of #color, thus ending up with an effective
>>difference in what is required from a processor and what is permitted in
>>a document.
>> 
>
>I think that this is a weakened way to think of these profile tags.  I
>think a profile should be documented, with its restrictions and
>permissions, and they can be (in principle) as strange and arbitrary as
>they like.  
>
>As long as there is a one-one match between an identifier and its
>definition (the short identifier Fred identifies the Fred profile as
>documented in http://www.example.com/frozentests/ttml-fred), then we’re
>fine.  If Fred wants to say “the color red shall only be used when the
>TTML content is served from a server hosted in a country in which the
>communist party polls more than 12.5% in the most recent national
>election” then so be it.

I agree in principle, but it begs the question: what type of feature can
not be associated with an extension designation to apply either to a
content profile or a processor profile?

Seems like we can maintain the consensus by keeping the current 'should'
wording to permit external processor profiles to be defined by a
specification only, in case folk want to do that. The essence of the
combinatorial operators still works, but the same concept wouldn't be
expressible inside the document.

Nigel

Received on Thursday, 23 October 2014 13:53:39 UTC