W3C home > Mailing lists > Public > public-html@w3.org > June 2009

Re: comments on draft-barth-mime-sniffing

From: David Singer <singer@apple.com>
Date: Thu, 18 Jun 2009 17:08:26 -0700
Message-Id: <p0624087bc660873b7a65@[17.202.35.52]>
To: robert@ocallahan.org
Cc: Maciej Stachowiak <mjs@apple.com>, Adam Barth <w3c@adambarth.com>, Shane McCarron <shane@aptest.com>, Ian Hickson <ian@hixie.ch>, Larry Masinter <masinter@adobe.com>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>, public-html@w3.org
At 12:01  +1200 19/06/09, Robert O'Callahan wrote:
>On Fri, Jun 19, 2009 at 7:05 AM, David Singer 
><<mailto:singer@apple.com>singer@apple.com> wrote:
>
>At 16:32  +1200 18/06/09, Robert O'Callahan wrote:
>
>>On Wed, Jun 17, 2009 at 7:15 PM, Maciej Stachowiak 
>><<mailto:mjs@apple.com>mjs@apple.com> wrote:
>>
>>On Jun 16, 2009, at 11:12 PM, Adam Barth wrote:
>>
>>I think roc was specifically referring to content for the <video> tag.
>>Because no such content currently exists, the legacy compatibility
>>requirements are much less onerous than, say, for HTML.
>>
>>
>>I think Dave's point was that if you support codecs and containers 
>>in <video> that are currently supported by widely used plugins, 
>>then you are more likely to face these legacy issues with broken 
>>MIME types when deploying <video>. Content authors might well 
>>expect that already published MP4 files which work in the QuickTime 
>>plugin or a Flash-based player, should continue to work if embedded 
>>via <video>.
>>
>
>
>They may expect that, but breaking that expectation is less of a 
>problem than breaking actual existing content.
>
>
>sorry, you've lost me.  breaking that expectation does break 
>existing content;  and I can't see how doing content sniffing breaks 
>more content than not.  can you explain?
>
>
>Currently there is almost no existing content that uses the <video> 
>element to play media files that are being served with incorrect 
>Content-Types. (Please correct me if I'm wrong about this.) 
>Therefore, not sniffing for <video> should break almost no existing 
>content.
>
>Maciej suggests there is a significant amount of content using 
>plugins (Flash or Quicktime, I suppose) to play media files served 
>with incorrect Content-Types, and therefore authors will be 
>surprised if they attempt to use <video> to play those files and 
>they don't work.
>
>I say that "author surprise" is much less of a problem than 
>"existing Web pages stop working".

But I may well be able to re-author my HTML but completely unable to 
influence my hosting provider to provide the correct MIME types.  I 
don't want users trapped into using the old because we've made the 
new inaccessible to them.
-- 
David Singer
Multimedia Standards, Apple Inc.
Received on Friday, 19 June 2009 00:11:31 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:04 UTC