[tei-council] some questions about non-textual components and media

[Sebastian Rahtz]

On 18 Dec 2012, at 19:24, Martin Holmes <mholmes at uvic.ca>
wrote:
> 
> In the case of born-digital documents, there's nothing to capture: 
> you're authoring directly. Imagine you're writing for a journal like 
> DHQ. You (as author) know you want to include a media file, which you do 
> in a simple way; DHQ has its own rendering pipelines, and it handles the 
> display of video or audio in ways conforming to its own styleguide, 
> depending on the output format. It's not the author's or encoder's job 
> to worry about that.

true. but the person encoding the archives of DHQ in TEI wants to preserve
the final look of the journal.
...
> I think if the author wants to do such complex stuff, they can import 
> the HTML5 elements and attributes.

yeah, ok, i guess thats always the fallback

> 
....
>> what the heck. i have the code ready for audio and video anyway, cos I just
>> abuse <graphic> :-}
> 
> That worries me more than the implications of a new relatively basic 
> <media> element. Especially if the content is audio; it's flat-out 
> misleading, surely.


oh, its Christmas, you'll forgive me, wont you? must I switch to <ptr target="foo.mp4" rend="transclude"/>?

Do remember that if I meet
 	<media target="foo.mp4"/>
and
	<media target="foo.tiff"/>
I really really really don't want to work out that I should use <video> for one, and <img>
for the other, by looking at file suffixes. Either let's have <video> and <audio>,
or mandate mime types.

I suspect actually that I could decide to pull the audio from a video file
using HTML <audio>, so thats another reason not to use <media>, because
it is not precise enough  - what aspect of the media did the author include?

--
Sebastian Rahtz      
http://www.justgiving.com/SebastianRahtz
Director (Research Support) of Academic IT Services 
University of Oxford IT Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431