W3C home > Mailing lists > Public > whatwg@whatwg.org > January 2008

[whatwg] Some <video> questions

From: Oliver Hunt <oliver@apple.com>
Date: Tue, 29 Jan 2008 15:16:03 -0800
Message-ID: <F2C56CBF-3976-4A0C-B049-2957FA87943B@apple.com>
> The <video> element doesn't appear solve the problem of how to  
> embed video
> content in a player- and browser- agnostic fashion.
> HTML 5 provides an opportunity to normalize how video embedding is  
> done for
> most scenarios, making it as easy to embed video as it is an  
> image.  But as
> designed, <video> appears to only address an as-yet-undiscovered  
> combination
> of container and media formats.
> This seems like a missed opportunity at best.  Technically, it  
> doesn't seem
> like it'd be difficult to allow various media runtimes to register as
> first-class <video> clients.

The purpose of the <video> element is to provide native support for  
video content and
remove the need for external proprietary plugins, not to provide  
<object> mk 2.  The
<video> element is much more flexible in that it allows all controls,  
etc to be defined using
full css, rather than requiring the controls to be embedded in the  
plugin view.

Your assertion that <video> does not provide a browser agnostic  
mechanism for providing
video content appears to be driven by the idea that html5 is  
finalised -- it's not, we have yet
to make any decision on exactly which codecs and transport formats  
will be expected.

Finally, the method through which content is decoded (once you know  
the codec) in the
browser is an implementation detail --  WebKit uses QuickTime, WebKit/ 
Gtk uses gstreamer,
Firefox and Opera use builtin codecs (afaik) -- so it is not the  
place of the html5 spec to
define it.


> -- Charles
Received on Tuesday, 29 January 2008 15:16:03 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 16:59:00 UTC