Tuesday, 19 January 2010

PETs - Stephan Engberg's response

Stephan Engberg, one of the speakers at the European Commission's December 2009 workshop on PETs (privacy enhancing technologies), has responded to my blog post on the recently published PETs workshop papers, and has kindly given his permission to post his emailed comments here:

"Interesting comments.

As speaker at the workshop I cannot help noticing that key elements are not reported accurately (amendment ongoing but perhaps not in conclusions).

I therefore suggest to have a look at my presentation to some of the key ommissions.
http://ec.europa.eu/justice_home/news/events/workshop_pets_2009/presentations/ENGBERG_Stephan.pdf

Especially my definition (page 12 below) of PET eID (or "National Id 2.0" as I prefer to label it ) - this is critical because it is about empowering the citizen in free choice negotiations and ensure that markets can selfadjust EVEN if consumers prior made a mistake by providing identifiable data elsewhere.

The interim report was severaly biased towards mere internal data protection (does NOT provide privacy) and rally bad applications claiming PET status for no justifiable reason while ignoring many of the most obvious and successfull PETs (e.g. Elections and CAR GPS Navigation)

We need much more precise defintions and economic framework understanding which was exactly what I was trying to provide.

E.g. definitions on PET (Enabling value and data sharing without transferring control), Privacy (Security from the point of view of an Individual forcussing on Rrisk management and minimisation in generel) and PET eID (Enable Context Isolation to ensure the DEMAND control value chains) while also demonstrating the economic fundamentals and how PETs are critical for innovation, security and efficiency.

PETs are not about what service providers do themselves (internal security meassures does not provide privacy), but what infrastructure and governments do (e.g. preventing the citrizens/consumers from "having to identify" in the first place).

Consumers are always in a conflict of interest negotation with Service providers about price and control (power in later transactions) . Unless we ensure the tools for consumers to MAINTAIN data control (i.e. remain un-identifiable which is not necesarily the same as anonymous), service providers will try to take control and create lock-in.

So when Government make eID which assume identifikationc it is truely anti-privacy because bureaucrats WANT CONTROL for the sake of power. When infrastrure cartel standards prevent PET because THEY want gatekeeper control for (power and thereby) profits.

Here we should notice that Identity Gatekeepers are about a huge problem in themselves as they present privacy invasive and severly market distoring elements - Microsoft Passport was an Identity Gatekeeper trying to get into the role of being the "trusted party" to all transactions, and now we see the same with SAML and eGovernment gateways.

PET are prevented by lack of authoriy attention to the critical requirements for markeds processes to work.

BAD GOVERNMENT or rather bureaucrat polcies are preventing PETs through lack of interoperability in communiocation protocols and technical interfaces while also preventing PETs from maturing by killing the marked demand to pay for the industry to evolve.

PETs are a huge success - ask the criminals, look to Car GPS Navigation, Democratic Elections and Broadcase communication. Problem is that governents (incl. EU) amd infrastructure cartels (through non-interoperable technical standards) are preventing the PET providers from making legitimate PETs for consumers in the private sector and citizen in eGovernment.

POINT: If we want prosperity through market innovation and efficient eGovernment, we CRITICALLY NEED PETs to empower the citizens.

Even though these are not simple issues, when bureaucrats claim Security through Surveilance, they are in reality undermining security and severely damaging the economy from evolving.  They prevent PETs themselves for the
sake of their interest while trying to claim PETs is a failure because nobody wants them.  And most critical processes have no justification for this whatsoever - when shiopping or getting eGovernment services you are NOT talking to terrorists or committing crimes justifying surveillance.

Legal Conclusions - we no longer can suffice with "Data Protection" and trying to regulate what is allowed to do with Identifiable Data -the problems are simply scaling out of control with digital integration and especially cloud. Instead we need to ensure Technilogy Design prevent abuse of data simply to prevents markets (and not only PET markets) from failling.

It goes so far as to realise that this blog is selling Personal Data in direct violation with the ePrivcacy Directive (look in the thirdparty links yourself) and that the costs are much bigger than the benefit of what looks like a "free service". This is not intuitively obivous.

Regards
Stephan Engberg
(I consent to this being digitally linkable to my name meaqning exactly zero privacy in this context risking abuse out of context)"

I am grateful for the clarification as to what was said at the presentation.

I agree that we need PETs to empower citizens. And I broadly agree about the conflict of interest and why PETs are not being adopted (see my blog post on PETs and compliance & enforcement - moving to PETs costs hard cash, and deprives businesses & governments of data which they think (probably rightly) would give them an edge, whether in crime policing or commercial competitiveness terms, so why should they do it unless it would benefit them e.g. because it's compelled by law with large penalties for breach, or because people will pay more for it?).

Many people do think eID could and should be implemented without full identification, i.e. more granular disclosure with pseudonymity - see e.g. Dave Birch's brilliant and very readable paper "Psychic ID: A blueprint for a modern national identity scheme" (PDF).

It's interesting that the Article 29 Working Party also want current laws to be beefed up to require PETs etc (see my report of the Article 29 Working Party's Future of Privacy paper), although as mentioned in that blog post I'm taking a "wait and see" attitude as it may be over-optimistic to think their recommendations will be taken up fully by the Commission.

However, I don't follow the references to car GPS etc as demonstrating the success of PETs - surely they underline the need for PETs? And I don't quite see how SAML and eGovernment gateways necessarily constitute identity gatekeepers?

Finally, I don't believe this blog is selling personal data in breach of the ePrivacy Directive in view of the third party links, and would be interested to know in more detail why that has been suggested?

In terms of selling data, this blog doesn't even have ads. And merely linking to a third party site surely wouldn't reveal to the third party site the IP of a visitor to this blog, at least in the case of a basic simple link. The Javascript for Delicious, tweets, Digg and AddThis sharing might, I suppose, tell Delicious etc that a particular IP address has visited my blog, but how does that violate the ePrivacy Directive? I'd be interested in any further views on this, or indeed any other points.

©WH. This work is licensed under a Creative Commons Attribution Non-Commercial Share-Alike England 2.0 Licence. Please attribute to WH, Tech and Law, and link to the original blog post page. Moral rights asserted.