Re: Survey ready on Media Text Associations proposal

On Mar 11, 2010, at 12:54 PM, Silvia Pfeiffer wrote:

> On Fri, Mar 12, 2010 at 2:36 AM, Eric Carlson  
> <eric.carlson@apple.com> wrote:
>>
>> On Mar 10, 2010, at 8:34 PM, Maciej Stachowiak wrote:
>>
>>>
>>> On Mar 10, 2010, at 8:25 PM, Philip Jägenstedt wrote:
>>>
>>>> On Thu, 11 Mar 2010 12:10:01 +0800, Silvia Pfeiffer <silviapfeiffer1@gmail.com 
>>>> > wrote:
>>>>
>>>>> On Thu, Mar 11, 2010 at 2:56 PM, Philip Jägenstedt <philipj@opera.com 
>>>>> > wrote:
>>>>>> Is the intention that we specify a track selection algorithm  
>>>>>> with exact
>>>>>> rules for which track to enable based on settings, or should  
>>>>>> this be left to
>>>>>> UAs to override ad-hoc?
>>>>>
>>>>> I don't think anything is prescribed to UAs in the HTML spec -  
>>>>> if at
>>>>> all we can only make recommendations on what override rules a UA  
>>>>> uses.
>>>>> No?
>>>>
>>>> There are very clear rules for resource selection for <source>  
>>>> and no provision for the UA to override this. I'm asking if the  
>>>> intention is to specify with the same level of detail which  
>>>> <track> to select and not allow UAs to override this, or if we  
>>>> should simply say that the UA can do whatever it wants. I don't  
>>>> like the latter because it will certainly lead to poor  
>>>> interoperability.
>>>
>>> I would prefer if the rules for <track> are as precise as the ones  
>>> for <source>, but I also think that this level of detail can wait  
>>> until after the proposal is submitted to the HTML WG.
>>>
>>  I agree with Maciej and Philip, we should have precise rules for  
>> <track> selection.
>
> Does HTML on other things require user preference settings to be
> implemented in browsers?

I don't think the spec has to require that the browser exposes  
particular preference settings. It just has to define what the  
algorithm is based on some input values (such as captioning being  
enabled, user's language settings, high contrast being enabled, etc).  
Some of those input values might be exposed via media query  
properties. The mapping from particular settings in the UI to these  
input values can be up to the browser. (For example, Safari on Mac  
would likely pull many of these settings from OS-wide preferences  
rather than having its own separate settings).

Regards,
Maciej

Received on Thursday, 11 March 2010 21:55:26 UTC