W3C home > Mailing lists > Public > www-tag@w3.org > February 2014

Re: Comments on w3ctag/eme/

From: Henri Sivonen <hsivonen@hsivonen.fi>
Date: Thu, 20 Feb 2014 12:50:25 +0200
Message-ID: <CANXqsRLytZU12p4_X8fOonwyeJMPywft9wtRezhibrD5vGh4Hg@mail.gmail.com>
To: www-tag <www-tag@w3.org>
On Wed, Feb 19, 2014 at 3:50 PM, Henri Sivonen <hsivonen@hsivonen.fi> wrote:
> Therefore, I don't think non-linkability is a major plug-in-specific
> disadvantage.

To elaborate on this: Sometimes EME-integrated DRM gets portrayed as
more open or Webby by claiming linkability advantages. The linkability
point does have Webbiness (but that much openness) merit when
presented in the context of a Web vs. iOS app dichotomy. However, when
EME is used, direct links to video files don't work and you have to
link to the HTML page that contains a <video> tag anyway. (The use of
the MSE companion API also has this effect even without EME.) The
linkability of EME video is in practice the same as the linkability of
<object> video. Therefore, the linkability point has no merit in the
EME vs. NPAPI/ActiveX plug-ins case, but bringing up a pro-EME talking
point from another context in a context where the point has no merit
looks like TAG is after-the-fact rationalizing EME, which might not be
the intent but doesn't look good. (The before-the-fact rationale for
EME clearly being arranging a substitute for PlayReady-in-Silverlight
in the absence of Silverlight but in the presence of PlayReady or a
PlayReady substitute—regardless of the cause of the absence of

> As I have said earlier on this list, it is technically correct that
> EME involves "encrypted" content,  but talking about encryption evokes
> the wrong connotations about who the adversary is.
> Calling DRM implementations "encryption technologies" amounts to weasel words.

To elaborate on this: The essence of DRM is restriction. Encryption is
incidental. Encryption is just a mechanism that makes data useless by
default (i.e. unless the key is provided on the condition of
restrictions being in place). The name of the Restricted Media
Community Group gets this right. The title of the spec is a trap
despite being technically correct, since the title focuses on the
incidental rather than the essential. Focusing on encryption is
falling into the trap.

Henri Sivonen
Received on Thursday, 20 February 2014 10:50:53 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:33:24 UTC