W3C home > Mailing lists > Public > public-html@w3.org > September 2008

Re: acceptable fallbacks [was: Re: Is longdesc a good solution? ...]

From: Dave Singer <singer@apple.com>
Date: Fri, 12 Sep 2008 10:37:46 -0700
Message-Id: <p06240824c4f0581f635a@[17.202.35.52]>
To: Henri Sivonen <hsivonen@iki.fi>
Cc: Justin James <j_james@mindspring.com>, 'David Poehlman' <david.poehlman@handsontechnologeyes.com>, 'Jim Jewett' <jimjjewett@gmail.com>, 'HTML WG' <public-html@w3.org>, wai-xtech@w3.org

At 12:52  +0300 12/09/08, Henri Sivonen wrote:
>On Sep 12, 2008, at 03:52, Dave Singer wrote:
>>>>In any case, full-text transcripts are useful for general audiences,
>>>>so once a full-text transcript has been written, it doesn't make sense
>>>>to hide it in a fallback chain only for deafblind users. Instead, it
>>>>  should be available to all users.
>
>(Quote was from me--not David Poehlman.)

I do apologize for any mis-attribution.

>>There is no 'fallback chain for deaf-blind users'.  There is 
>>material to enhance accessibility.  Why you might want it or need 
>>it is none of my business.  It *is* available to all users.
>
>
>If the user needs to go flip settings to indicate no visual content 
>and no audio content in order to discover a transcript, then it's 
>not available to all users *for practical purposes*.

Then choose another browser that makes it easier for you to get the 
content.  This aspect of the HTML spec. is describing a 'document' 
format, and what the elements in the document mean, in a sense.  How 
the user-agent determines what its user wants -- by keeping 
preferences, by drawing inferences based on other expressed 
preferences, by asking lots of questions all the time, by showing 
everything it possibly can -- is really rather out of scope.  For 
example, a document might say (somehow) "a transcript is available 
over <here>" and the user and UA are free to use that piece of 
information as they wish and why they wish.

>That's why an automatic alternative selection chain isn't the (only) 
>place where a transcript should be put.

I'm sorry if I implied that it should be.  I was wondering about (not 
really even proposing) longdesc.  I was also wondering about what 
should happen when an alternative selection chain fails to select 
anything;  at the moment, the answer is nothing and it's an authoring 
problem to avoid that case.  If the two got conflated, I also 
apologize!
-- 
David Singer
Apple/QuickTime
Received on Friday, 12 September 2008 17:40:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:23 GMT