Re: [arch-d] <draft-lazanski-consolidation-00>

[Christian Huitema <[e-mail]>]

On 11/12/2020 6:46 AM, Vittorio Bertola wrote:
>
>> Il 12/11/2020 04:29 Christian Huitema <[e-mail]> ha scritto:
>>
>>
>> On 11/11/2020 4:37 PM, Martin Thomson wrote:
>>
>>>> I am more concerned about the loss of privacy that seems to be 
>>>> occurring at the application layer via extensive tracking of users and 
>>>> sharing of user information, rather than my perception of what is now 
>>>> occurring at the network layer.
>>> A totally legitimate concern, but that line of argumentation sounds like whataboutism. Improvements in privacy with respect to the network do not drive privacy violations at the application layer.
>>
>> What Martin says. The draft is supposedly about consolidation. There
>> is definitely a relation between concentration and privacy: if
>> traffic gets consolidated on a small number of big platforms, the
>> owners of these platforms see a lot of data and metadata, which
>> affects privacy. But the relation between that and end-to-end
>> encryption is tenuous. For example, Google was getting just as many
>> cookies before web pages started being sent over HTTPS. The
>> differences is that before encryption other actors were also able to
>> read these cookies. That was hardly a better privacy posture.
>>
> I note one big risk to privacy that is generated by the introduction
> of end-to-end encryption, together with the IoT and cloudification
> trends: end-users - not just network administrators - lose the ability
> to scrutinize what their applications and devices actually send to
> their makers, or - in case of self-masquerading technologies like DoH
> - even to detect that a communication is happening and to block it if
> they want so. There was a talk about this at the IETF 105 plenary:
> https://datatracker.ietf.org/meeting/105/materials/slides-105-ietf-sesse-lessons-from-privacy-measurement-arvind-narayanan-00

Device inspection is an interesting issue, but it is far to being as
clear-cut as you present it.

Take the example of a doorbell camera. It sends motion activated video
streams to a cloud server, from which they can be accessed by the home
owners using their cell phone. If the traffic was in the clear,
inspection would reveal exactly that: motion activated video streams to
a cloud server. It will not reveal whether the cloud servers provide a
copy of the video streams to the police. It will not tell whether the
servers perform face recognition on the incoming images, and use that
for tracking the movement of people walking nearby. Encryption will
prevent that inspection, but it will also prevent third parties from
tapping into the video stream and performing their own parasitic
surveillance. From a privacy point of view, encryption there is a net
gain -- but the presence of the devices themselves may well be a huge loss.

The inspection performed by researchers revealed that some devices where
performing extra tasks, not described in the public specification of the
device. Maybe a child's toy also sending data about the children to its
manufacturer. Inspection by researchers has revealed such shenanigans in
the past, and the inspection was typically done by breaking the device
encryption using spoofed server certificates. Should we then recommend
that all devices should accept spoofed server certificates? That would
be opening the gates to a free-for-all of hacking and banditry. It would
also not solve the "cloud connection" problem shown with the doorbell
cameras.

The device problem is real, but the solution certainly is not to make
the devices easy to break by third parties. Instead I would rather see
consumer bureaus stepping in and providing "clean device" certification
labels, after inspecting devices and services in which the protections
have been lifted by the manufacturer. Or, law enforcement could step in
and severely punish device manufacturers engaged in deceptive practices,
either on the device or in the cloud servers.

>
> Indeed, one of the side effects of DoH for some (more technical) users

> is to disrupt their Pi-holes and other DNS-based tracker blockers.
>
> Consolidation underpins all of this, as in the "surveillance
> capitalism" business model whoever has more opportunities to acquire
> information wins.

There is definitely a relation between surveillance capitalism and
consolidation, but you don't need to discuss the encryption of
communication to make that point.

-- Christian Huitema





> -- 
>
> Vittorio Bertola | Head of Policy & Innovation, Open-Xchange
> [e-mail] <mailto:[e-mail]> 
> Office @ Via Treviso 12, 10144 Torino, Italy