W3C home > Mailing lists > Public > public-html@w3.org > February 2012

Re: Encrypted Media proposal (was RE: ISSUE-179: av_param - Chairs Solicit Alternate Proposals or Counter-Proposals)

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Mon, 27 Feb 2012 23:17:41 -0500
Message-ID: <4F4C5565.4020406@mit.edu>
To: Mark Watson <watsonm@netflix.com>
CC: "<public-html@w3.org>" <public-html@w3.org>
On 2/27/12 11:01 PM, Mark Watson wrote:
> On the one hand, that could be an argument for defining that API more explicitly so that this topic can be subject to public review. On the other hand, browser developers might be more comfortable defining their own APIs that meet their own requirements in this respect.

One of the requirements for browser developers (either on their own or 
imposed by the CDM providers) might well be "make it illegal for any 
other browser developer to attempt to reverse-engineer the API we're using".

>> OK.  And CDMs could require running in a particular browser, yes?
>
> If there is no common API, this is the situation by default.

This obviously seems like a bad idea to me, given my preconceptions.  ;)

> If there is a common API, then I guess CDMs could place such restrictions, though I am not sure why they would ?

The history of the web is full of people placing single-browser 
restrictions for all sorts of good and not so good reasons.

We can't really keep people from doing such things if they want to hard 
enough, though we can certainly consider making it require extra work to 
do them.

> That just depends on whether you consider commercial video part of the "web".

Sure.  We're talking about a W3C standard for such a thing here, right? 
  If we were talking walled gardens, why is the discussion happening here?

> If you do, then this is the situation today: the 'binary blob' is Flash or Silverlight (or QuickTime, or a Netflix or YouTube native app, or ...).

Yep.  And it's being a problem right now, as I said.  This is not at all 
a hypothetical concern.

>> We have such barriers to entry to some extent now, with Flash (and Silverlight, and to some extent Java), but I think we should be reducing such barriers, not enshrining them in the web platform.
>
> Making the 'blobs' smaller with a well-defined and highly-constrained function should be an improvement.

Insofar as it makes them technically easier to port to new platforms, yes.

Insofar as it makes them into DMCA-protected measures that cannot be 
reverse-engineered, no....

> Making them go away altogether might be a bigger improvement by some measures but by others (range of content available) might be quite negative.

That's a key problem, yes.

-Boris
Received on Tuesday, 28 February 2012 04:18:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:46 GMT