[whatwg] <video> application/octet-stream

Benjamin M. Schwartz bmschwar at fas.harvard.edu
Wed Jul 21 08:25:49 PDT 2010

On 07/21/2010 10:24 AM, Chris Double wrote:
> Or the developers of said browser could obey the mime type that the
> server sent, not have to write or maintain error prone content
> sniffing code that could behave differently across browsers ("Chrome
> content sniffs this as Ogg but you dont!!", etc), and solve even more
> pressing problems!

I agree. Consider direct URL links (i.e. a URL entered directly into the
browser's address bar).  If you sniff content types there, you have to
sniff for _all possible_ types, which creates a major risk of
misidentification, followed by displaying garbage (or worse).  If you
don't sniff for direct access, but do sniff for <video>, then you create a
situation in which people's videos will play in a webpage, but won't play
when you link to the video file directly.

Sniffing is bad.  Make them fix their servers.

As for slippery slopes ... as long as a large fraction of <video> viewers
(currently a majority) use browsers with strict type checking, I expect
sysadmins to fix their servers PDQ.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: OpenPGP digital signature
URL: <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/attachments/20100721/f7bac0c7/attachment-0002.pgp>

More information about the whatwg mailing list