Re: Formal objection to Encrypted Media Extensions progressing to Proposed Recommendation without greater user protection

On Tue, Aug 16, 2016 at 10:49 AM, Harry Halpin <hhalpin@w3.org> wrote:

>
>
> On 08/16/2016 07:46 PM, Mark Watson wrote:
>
>
>
> On Tue, Aug 16, 2016 at 10:36 AM, Harry Halpin <hhalpin@w3.org> wrote:
>
>>
>>
>> On 08/16/2016 07:28 PM, David Singer wrote:
>> >> On Aug 16, 2016, at 10:22 , Harry Halpin <hhalpin@w3.org> wrote:
>> >>
>> >>
>> >>
>> >> On 08/16/2016 06:35 PM, David Singer wrote:
>> >>> I think you’ll need to explain why the choice of how much, and how,
>> the user is warned, needs to be made by you/us, and not by the browser
>> maker.  You also haven’t given details of the ‘user harm’ you talk about
>> for EME itself.
>> >> The details of 'user harm' are adequately explained by the concerns
>> over
>> >> the DMCA and the use of DRM that have been well-documented elsewhere
>> >> [1].
>> > No, those criticisms were almost exclusively focused on the risk to
>> security researchers. Not one of the sub-headings you link to appears to
>> apply to users.  Current politics notwithstanding, repeatedly stating
>> something doesn’t intrinsically make it true.
>>
>> I don't agree.  Insofar as security researchers cannot audit the
>> security of DRM systems, then it is *users* who will face any harm due
>> to the lack of security audits. That is a unique feature of DRM that EME
>> enables. EME can also, due to clearkey, be considered a DMCA-compliant
>> system itself (i.e. clearKey's Key System is equivalent to a CDM).
>>
>> >
>> >> Any installation or use of software that is compliant with the DMCA
>> >> can be considered a risk to users and security researchers as is noted
>> >> on WIkipedia [1].
>> > Users and security researchers are not the same people.
>>
>> See above.
>>
>> >
>> >> Although the FO is filed as an individual, one task of W3C is to assure
>> >> there is representation of the interest of users and to ensure the Web
>> >> is secure.
>> > Sure, but you need to say what the *user* harm is.
>>
>> If there is a part of a browser that makes security research difficult
>> and possibly illegal,  then that part of the browser is rather
>> self-evidently dangerous to end-users.
>
>
> ​Unless the browser implementor takes additional measures to compensate
> for that difficulty / possible illegality.​
>
>
> I really do appreciate the work by this WG to compensate, but I do not
> think they are sufficient without 'off by default.' We can agree to
> disagree here, and let the WG make a decision and pass it to the Director.
> In that case, the FO still stands.
>
>
>
>> While I appreciate the effort,
>> sandboxing may help but there is no such thing as a perfect sandbox.
>>
>
> ​There is no such thing as a perfect security audit either. We're talking
> about different kinds of risk reduction.​
>
>
>
> Agreed. However, the DMCA provides a 'chilling effect' on security audits
> and research on DRM systems in general, regardless of the particular
> details of an audit or research.
>

​But there are things that can be done to compensate for the risk that any
such effects cause. Sandboxing and strict input / output validation is one
thing. Funding additional security audits could be another. Actively
encouraging independent security research into that component is another.



>
>
>> >
>> >> So, I think the decision should be made by the Working Group
>> >> with the best interests of users in mind, not just the browser makers.
>> >> While you can consider 'off by default' to be unreasonable, I think if
>> >> one takes that this same approach has been adopted by similarly
>> >> controversial APIs (Geolocation API), I think it's quite reasonable.
>> >>
>> >>   cheers,
>> >>        harry
>> >>
>> >> [1]
>> >> https://en.wikipedia.org/wiki/Digital_Millennium_Copyright_A
>> ct#Criticisms
>> >>
>> >>>
>> >>>> On Aug 16, 2016, at 4:28 , Harry Halpin <hhalpin@w3.org> wrote:
>> >>>>
>> >>>>
>> >>>>
>> >>>> On 08/13/2016 05:05 PM, Paul Cotton wrote:
>> >>>>> Note that I am responding to an email and proposal from Harry from
>> earlier in this thread.
>> >>>>> https://lists.w3.org/Archives/Public/public-html-media/2016A
>> ug/0005.html
>> >>>>>
>> >>>>>> The text I would add to the EME spec would be similar:
>> >>>>>> "A conforming implementation of this specification MUST ensure
>> that this
>> >>>>>> API cannot be used without the user's express permission due to the
>> >>>>>> inherent risks in the activation of a CDM in a user agent. The API
>> MUST
>> >>>>>> be disabled by default, and should only be activated when the user
>> gives
>> >>>>>> express consent and is fully informed on a per-origin basis."
>> >>>>> This request would appear to be (at least partially) covered by the
>> existing text in the EME CR:
>> >>>>> http://www.w3.org/TR/encrypted-media/#navigator-extension-re
>> questmediakeysystemaccess
>> >>>>>
>> >>>>> 3.1.1 Methods
>> >>>>> requestMediaKeySystemAccess
>> >>>>> Note
>> >>>>>
>> >>>>> Calling this method may have user-visible effects, including
>> requests for user consent. This method should only be called when the
>> author intends to create and use a MediaKeys object with the provided
>> configuration.
>> >>>>> Requests access to the specified Key System. When
>> supportedConfigurations is specified, the configuration specified by at
>> least one of its elements must be supported. The resulting
>> MediaKeySystemAccess will correspond to the first such element.
>> >>>>> Any permission checks or user interaction, such as a prompt, MUST
>> be performed before resolving the promise.
>> >>>>> Were you aware of this text in the EME specification?  Can you live
>> with the current text since changing this text to be normative and changing
>> it to a MUST (I believe) would be a “breaking change” and would require
>> that we re-publish another EME CR?
>> >>>>>
>> >>>>> /paulc
>> >>>>> HME WG Chair
>> >>>> Paul,
>> >>>>
>> >>>> Thanks for noting 'use-visible effects' but it does not either
>> explicitly 1) require a user-visible effect for fully informing the user"
>> and "gaining their consent" as well "turning off EME by default". Simply
>> put, the text you are noting says MAY have user-visible effects and as so
>> is too weak to include in a test-suite or support EME being off by default.
>> So, it does not cover my objection, which normatively requires much
>> stronger text, including "off by default". So, no I can't live with that
>> text and require the change (or a semantically equivalent one) that I
>> suggest in the github repo and e-mail list:
>> >>>>
>> >>>> https://github.com/w3c/encrypted-media/issues/304
>> >>>>
>> >>>> Thus, my formal objection still stands.
>> >>>>
>> >>>>  cheers,
>> >>>>      harry
>> >>>>
>> >>>>>
>> >>>>> From: Harry Halpin [mailto:hhalpin@w3.org]
>> >>>>> Sent: Tuesday, August 2, 2016 5:00 AM
>> >>>>> To: public-html-media@w3.org
>> >>>>> Subject: Re: Formal objection to Encrypted Media Extensions
>> progressing to Proposed Recommendation without greater user protection
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>> On 08/02/2016 03:00 AM, Paul Cotton wrote:
>> >>>>>> An individual who registers a Formal Objection should cite
>> technical arguments and propose changes that would remove the Formal
>> Objection;
>> >>>>> http://www.w3.org/2015/Process-20150901/#WGArchiveMinorityViews
>> >>>>>
>> >>>>> Before I record your formal objection at [1] I would like to ensure
>> that your “proposed change” for your objection is understood.
>> >>>>>> There is prior art in HTML for similarly powerful and
>> privacy-invasive features such as the Geolocation API [4]. Thus, there
>> should be no procedural problem in requiring user consent and disabling EME
>> by default in all browsers.
>> >>>>> The reference you provide as a “prior art” precedent in [4] is in a
>> non-normative section of Geolocation API and it does NOT disable the
>> Geolocation API by default.  Can you explain why you think this reference
>> is useful here?  Are you suggesting similar text be added to EME?  If so
>> could you suggest exact text that would remove your formal objection?
>> >>>>>
>> >>>>> The exact phrasing is " A conforming implementation of this
>> specification must provide a mechanism that protects the user's privacy and
>> this mechanism should ensure that no location information is made available
>> through this API without the user's express permission"
>> >>>>>
>> >>>>> To clarify, I agree that the spec should have used capitalization:
>> " A conforming implementation of this specification MUST provide a
>> mechanism that protects the user's privacy and this mechanism should ensure
>> that no location information is made available through this API without the
>> user's express permission."
>> >>>>>
>> >>>>> However, I do believe that all browsers *do* indeed ask for user
>> consent before using the Geolocation API and that it is disabled by
>> default, which is clear prior art for EME. This is indeed the case on the
>> browsers I use, but if others activate Geolocation API without user
>> consent, please do inform me, as I don't use Microsoft products.
>> >>>>>
>> >>>>> The text I would add to the EME spec would be similar:
>> >>>>>
>> >>>>> "A conforming implementation of this specification MUST ensure that
>> this API cannot be used without the user's express permission due to the
>> inherent risks in the activation of a CDM in a user agent. The API MUST be
>> disabled by default, and should only be activated when the user gives
>> express consent and is fully informed on a per-origin basis."
>> >>>>>
>> >>>>> Given there is already a place where user consent MUST be asked in
>> the EME spec (" User agents must ensure that users are fully informed
>> and/or give explicit consent before Distinctive Identifier(s) are exposed,
>> such as in messages from the Key System implementation"), there is no
>> reason why this MUST can't be broadened. I would also remove the "or." The
>> user must be fully informed AND give explicit consent.
>> >>>>>
>> >>>>> Furthermore, given this normative, testing for express user consent
>> on a per-origin basis should be part of the test-suite.
>> >>>>>
>> >>>>>       cheers,
>> >>>>>               harry
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>> /paulc
>> >>>>> HME WG Chair
>> >>>>>
>> >>>>> [1] https://dev.w3.org/html5/status/formal-objection-status.html
>> >>>>>
>> >>>>> From: Harry Halpin [mailto:hhalpin@w3.org]
>> >>>>> Sent: Monday, August 1, 2016 6:42 PM
>> >>>>> To: public-html-media@w3.org
>> >>>>> Subject: Formal objection to Encrypted Media Extensions progressing
>> to Proposed Recommendation without greater user protection
>> >>>>>
>> >>>>> [Note this is a formal objection as an individual in a private
>> capacity, not on behalf of my organization]
>> >>>>>
>> >>>>> I'd like to fill a formal objection against Encrypted Media
>> Extensions progressing to Proposed Recommendation status without adequate
>> protection for users.
>> >>>>>
>> >>>>> I believe that this work is so problematic given the well-known and
>> well-documented problems with DRM,  it should not happen as a standard at
>> all at W3C. That being said, for reasons which I do not agree with and hope
>> he reconsiders, the Director has approved both the scope of the charter and
>> the move of the Working Draft to Candidate Recommendation.
>> >>>>>
>> >>>>> If it does happen, then it seems given the well-documented
>> problems, some harm mitigation should be pursued. The security research
>> community has broad support for the EFF covenant being a normative
>> requirement [1]. However, it appears consensus is not forthcoming on either
>> EME itself or the EFF covenant, with the Technology and Policy IG also
>> failing to gain any consensus for even further discussion (as it failed to
>> even get chartered).
>> >>>>>
>> >>>>> To myself, the danger of EME is that over the last year suddenly
>> over millions of people had a content decryption module installed without
>> their explicit consent on their computer. For many users, such as those of
>> Firefox, the DCM was installed via a silent update they had no control
>> over. Such a content decryption module can serve as both a technical
>> security risk, as it puts a highly privileged process in the user's
>> computer outside their direct control by design, and as a legal risk
>> subjects any inspection of the DRM-enabled system on their computer due to
>> the anti-circumvention provision of the DMCA or local equivalent. Thus,
>> regardless of whether or not one agrees with EME, it seems the *least* the
>> W3C should do is warn the user about the installation and activation of a
>> CDM on their machine - and that the CDM should not be installed and EME
>> should not be activated without explicit user consent. Thus, in all
>> configurations, EME should be *de-activated by default.*
>> >>>>>
>> >>>>> There is prior art in HTML for similarly powerful and
>> privacy-invasive features such as the Geolocation API [4]. Thus, there
>> should be no procedural problem in requiring user consent and disabling EME
>> by default in all browsers.  While it can be argued many users will want to
>> watch protected videos and will turn them on, just as many users will want
>> to use Google Maps with Geolocation APIs, there does not seem to be a
>> reasonable case for having such
>> >>>>> powerful and possibly dangerous features enabled by default.
>> >>>>>
>> >>>>> Current language in the spec is so weak as it may not be enforced
>> and so EME does not have to be disabled by default: " User Agents have some
>> flexibility to determine whether consent is required for a specific
>> configuration and whether such consent may also apply to other
>> configurations. For example, consent to one configuration may also imply
>> consent for less powerful, more restricted configurations. Equally, a
>> denial of consent for one configuration may imply denial of consent for
>> more powerful, less restricted configurations."
>> >>>>>
>> >>>>> The only place where the spec [5] makes normative statements around
>> consent, but due to how DRM (i.e. Key Systems) are designed, it is unclear
>> how the user agent should interpret these statements
>> >>>>>
>> >>>>> "If a user agent chooses to support a Key System implementation
>> that cannot be sufficiently sandboxed or otherwise secured, the user agent
>> should ensure that users are fully informed and/or give explicit consent
>> before loading or invoking it."
>> >>>>>
>> >>>>> "User Agents should ensure that users are fully informed and/or
>> give explicit consent before a Key System that presents security concerns
>> that are greater than other user agent features (e.g. DOM content) may be
>> accessed by an origin..."
>> >>>>>
>> >>>>> " User agents must ensure that users are fully informed and/or give
>> explicit consent before Distinctive Identifier(s) are exposed, such as in
>> messages from the Key System implementation."
>> >>>>>
>> >>>>> Although these statements show some progress towards trying to
>> mitigate the dangers of DRM, given that DRM systems work in conjunction
>> with anti-circumvention legislation that prevents a developer from knowing
>> whether or not they can sufficiently sandboxed (or 'otherwise secured') and
>> whether a Distinctive Identifier is used by the DRM system, so the use of
>> EME will *always* present security concerns that are greater than the rest
>> of the user agent features involving the Open Web Platform.
>> >>>>>
>> >>>>> Is the Working Group amendable to having EME and associated DRM
>> systems (i.e. "Key Systems") *normatively* disabled by default in order to
>> protect users?
>> >>>>>
>> >>>>>   cheers,
>> >>>>>       harry
>> >>>>>
>> >>>>> [1] https://www.eff.org/deeplinks/2016/03/security-researchers-t
>> ell-w3c-protect-researchers-who-investigate-browsers
>> >>>>> [2] https://en.wikipedia.org/wiki/Digital_rights_management#Oppo
>> sition_to_DRM
>> >>>>> [3] https://en.wikipedia.org/wiki/Digital_rights_management#Shor
>> tcomings
>> >>>>> [4] https://www.w3.org/TR/geolocation-API/#implementation_
>> considerations
>> >>>>> [5] https://w3c.github.io/encrypted-media/
>> >>>>>
>> >>> David Singer
>> >>> Manager, Software Standards, Apple Inc.
>> >>>
>> >>
>> >>
>> > David Singer
>> > Manager, Software Standards, Apple Inc.
>> >
>>
>>
>>
>
>