[uf-discuss] RFC: sHTML Video Thumbnailing

[Benjamin Hawkes-Lewis]

Charles Iliya Krempeaux wrote:

> It's the "cite" attribute that gives it value.
> 
> It lets me bind a set of thumbnails together (as being from the same
> "video") while allowing the thumbnails to be all over the place (and
> not necessarily in some container element, like a <span> or something,
> which binds them together).
> 
> (Did that make sense?  Did I explain that well?  Or would an example help?)

Ah, I see your problem. It's not pretty, but until HTML-next comes to 
town how about either of the following hacky solutions:

1. Using <cite> then trying to hide it

<span class="video-still"><img 
src="http://www.example.com/thumbnail.jpg" alt="Still: Dorothy 
encounters the Lion"><span class="full-video">(<cite 
class="full-video">(<a href="http://www.example.com/video">Wizard of 
Oz</a></cite>)</span>

with CSS:

@media all {
.full-video {display:none;speak:none;height:0;visibility:hidden;width:0}
}

If the citation is not hidden, it's not the end of the world: it still 
makes sense. Thanks to buggy treatment of display:none, I think 
mainstream screen readers will fail to read it even though they ignore 
speak:none.

http://css-discuss.incutio.com/?page=ScreenreaderVisibility

http://www.freedomscientific.com/fs_support/BulletinView.cfm?QC=1165

2. Using <object>

<object class="video-still" type="image/jpeg" 
data="http://www.example.com/thumbnail.jpg">
<param name="full-video" value="http://www.example.com/video" 
valuetype="ref"></param>
Still: Dorothy encounters the Lion.
</object>

Both would need some testing though.

--
Benjamin Hawkes-Lewis