Re: comments on draft-barth-mime-sniffing

Adam Barth wrote:
> On Mon, Jun 1, 2009 at 2:39 AM, Robert O'Callahan <robert@ocallahan.org> wrote:
> > On Mon, Jun 1, 2009 at 8:18 PM, Ian Hickson <ian@hixie.ch> wrote:
> >>
> >> The current HTML5 spec's position on this issue is that the Content-Type
> >> header is completely ignored in the processing of <video>, I believe.
> >
> > Hmm, really? In our media element implementation, we honor Content-Type
> > completely and do no sniffing whatever.
> 
> I've removed all mention of audio or video formats from the I-D.  I
> think it's wise to wait and see how <video> and <audio> get deployed.
> If we can follow Content-Types strictly, that would be great.

Video formats are incredibly diverse, and typically consist of:

    - The container file format, or sometimes no container.

    - The container (if there is one) has 1 or more streams of video
      (and audio).

    - Each video stream may be any common video encoding (codec).

E.g. you can have MPEG video inside an AVI, MPEG video inside an MPEG
container, WMV video inside a WMV container, or WMV inside an AVI
container, etc.  Although it's a hideous mess, there's a certain
amount of orthogonality between the container formats and the video
encodings they can contain.

Am I right that the Content-Type merely says the type of the container
format, and says nothing about the format of encoded video(s) within?

-- Jamie