December 21, 2010

Descartes famously turned "divide and conquer" from a strategy dear to generals into a general problem solving method. Unattributed, its converse is no less true and practiced by all. To stall a resolution, make sure to confuse the issues concerned as, together, they will prove intractable.

Take net neutrality for instance. If one offers a desirable commodity for free, the day will come when the demand for it will exhaust your capacity to deliver. There can only be two answers to this simple issue. Ration the commodity or price it to market.

In the United States, one would imagine the latter solution to be a no brainer. Not so for cable and wireless communications. The likes of Comcast and Verizon would rather ration service availability at will than charge their customers in proportion for their bandwidth usage.

This illustrates the difference between ideology and practice, between capitalism and the capitalists. As the adoption of a "net neutrality" principle would severely limit the leverage of bandwidth sellers over content providers, the former are against it. And so they claim they should be free to manage their property, i.e. their network, as they see fit.

Monopolists and oligopolists are great defenders of property rights when regulators try to free the markets from distorsions. Whether the operators have confused these two issues enough to stop a resolution will be tested by how "the Federal Communication Commission [...] go along with proposals put forward by its chairman, Julius Genachowski", Richard Waters alerts us (*). "Serious lobbying [is] going on behind closed doors".

Issue confusion may not always be so obvious. Take the case of WikiLeaks and his founder's personal troubles. As Ravi Somaiya reports, "the Swedish government['s representative] urged the [English] court to separate the contentions over the WikiLeaks releases from the sexual accusations" (**). Julian Assange would rather not, as it makes the Swedish prosecution look like a persecution. But, truly, is this the real issue?

WikiLeaks has not yet shared its stash of US military and diplomatic corps secrets with the whole media. Yet even without this, Julian Assange's tribulations have provided an attractive story for all to cover. "After nine days in Wandsworth Prison, [he] emerged into an explosion of photographers' flashbulbs and spotlights". Add torrents of ink poured by editors pondering over First Amendment rights and Freedom of the Press.

The US government itself helps develop the hysteria. "The Air Force is barring its personnel from using work computers to view the Web sites of The New York Times and more than 25 other news organizations and blogs", writes Eric Schmitt (***). Understandable but hilarious nonetheless. "Some Air Force officials acknowledged that the steps taken might be in vain". These at least must have heard about King Midas' donkey ears.

Still how could so many secret documents be leaked in the first place? This is the real issue all these adventitious developments have confused.

The blame has been placed on Private Bradley Manning (1). If he is but a convenient fall guy, it should be worrisome in the extreme. If he is guilty as charged, shouldn't the media dig further on how the US government made itself so vulnerable to a very junior employee, especially to one said to have been reprimanded and demoted prior to the leaks. Isn't inappropriate data appropriation a classic revenge of disgruntled staff?

More often than not, lack of sharing suppresses significant links between data held in separate silos. And the titillation of reading unvarnished paintings of country heads encourages sharing. But isn't the danger of oversharing with fragile personnel a worthy issue when cyberwars loom?

As far as eprivacy is concerned, the confusion of the issues is at its maximum. According to Tanzina Vega, "a Commerce Department task force [...] recommend[s] the creation of voluntary codes of conduct that would address emerging technologies" (****). I do not want to prejudge the entire report, but asking the wolves for new sheep clothing is a sure way not to clear the confusion.

Indeed the only way forward is to go back to Descartes' program and start to unravel this noxious nexus.

Justifiably, as they contemplate how to corral companies which collect and exploit user information online, both the FTC and the US Commerce Department worry about the economic impact. But they should focus on the legitimate stakeholders which, besides the consumers, are the advertisers and the advertising space providers.

To include the current advertising networks in their concerns is like trying to consult insider trading networks on how to dam insider trading. If you let human behavior dictate the law, admit might makes right and stop demanding respect for the law.

Can advertisers benefit from targeting specific consumer profiles without hurting consumers' rights to privacy? That is the question which innovation must address and ePrio has shown the answer to be positive (2). Current advertising networks should be free to adopt this or any other competing solution but why protect last year's bold innovators who behave as today's sleepy, although still greedy, incumbents?

Having taking care of the economic issue, one can more easily face the issue of how to craft the most efficient regulation. When budgets are squeezed and knowing one faces "a list of 500 ad networks and 600 analytics companies", according to "Jules Polonetsky, the co-chairman and director of the Future of Privacy Forum" as quoted by Tanzina Vega (*****), ease of enforcement must be the starting point.

My past advice to the FTC (3) is still valid today. To be successful, enforcement must involve the consumers and give them a visible target they know and care about, i.e. the site which they are willingly visiting when it strikes them with a targeted ad. While such sites are countless, they are also the most likely to play by the rules for two obvious reasons. They are the first ones to benefit from targeted advertising, as they sell their ad space, and they cannot run away, as they need consumers to come visit again.

Notice that empowering the consumers does not prevent what Jules Polonetsky calls "privacy watch dogs" from playing a positive role. But instead of watching over a teeming crowd of obscure intermediaries, they would check content providers' practices.

Ignore the intermediaries all you want, my alert reader will wonder, but isn't their compliance indispensable? Well, remember that, to a company like Yahoo "that earn[s] a large percentage of [its] revenue from advertising that is tailored to users based on the sites they have visited", intermediaries are tame suppliers. What Yahoo demands, they will execute and Yahoo will audit the resulting process lest its deep pockets be found responsible.

Lacking Yahoo's scale, small content providers cannot hope to audit the ad networks which buy their space. Yet who doubts unfettered competition will provide them with bonded suppliers they can trust? Ask not what Society can do for ad networks, but what ad networks should do for Society.

Economic and enforcement issues solved, wouldn't a rule to mandate online personal profile management better than a "do-not-track" system?

Philippe Coueignoux

• (*) ......... Operators must tread carefully with the data explosion, by Richard Waters (Financial Times) - December 16, 2010
• (**) ....... Freed on Bail, Wikileaks Founder Offers Defiant Note, by Ravi Somaiya (New York Times) - December 17, 2010
• (***) ..... Air Force Says Leaked Cables Are Off Limits On Its Network, by Eric Schmitt (New York Times) - December 15, 2010
• (****) ... A Call for a Federal Office to Guide Online Privacy, by Tanzina Vega (New York Times) - December 17, 2010
• (*****) . Microsoft, Spurred by Privacy Concerns, Introduces Tracking Protection to Its Browser, by Tanzina Vega (New York Times) - December 8, 2010
• (1) see the entry for Bradley Manning in the Wikipedia
• (2) see ePrio's proposal, backed by the forbidding prose of its patent portfolio
  ..... see Patent Number 6,092,197 and Patent Application Numbers 20060053279 and 20090076914 at the US Patent and Trademark Office
• (3) look up my plain language Comments to the FTC, November 2007