Despite the scuff up over Charter's sale of user browsing history to behavioral advertising firm NebuAD, the discussion of the privacy implications of such systems have been muted. As I mentioned last week, there's some dissent within the halls of ISPs between engineers and marketing departments over the use of such systems, but for the most part the public here remains unaware that their albino antelope fetish has become a revenue stream.

In the UK, the debate over such systems has been much more heated. That's in part because the largest behavioral advertising firm in the UK, Phorm, began its money-making life as a spyware and rootkit developer named121 Media. And whereas NebuAD is fairly forthcoming about their systems, Phorm tries to warm consumers to the idea by pretending that their system can protect users from phishing scams.

The company also found itself under fire when it was revealed that they conducted a secret trial with British Telecom, without anyone bothering to tell consumers. A former rootkit developer secretly buying your browsing history under the guise of an anti-phishing solution? What could go wrong?

Wikileaks today, um, Wikileaked the confidential results of Phorm's trial with British Telecom. The leaked memo (pdf) shows, among other things, that personal IP addresses were involved in the tests, contrary to Phorm's claims that personal identifiers would be made anonymous. Wikileaks also claims that charity ads were replaced by more profitable endeavors:Given the trial was conducted secretly (potentially in violation of British law), users obviously didn't know about it. The report concludes that just 15-20 users (0.1% of the 10,000 tested) actually noticed the trial and had a "negative reaction." Of course none of the users who noticed glitches caused by the system understood they were caused by Phorm. The test concludes that to sell users on the idea, PR is obviously essentialYou think? Might I suggest that carriers interested in insulting their users' intelligence could do what Charter did and suggest that such systems are so beneficial, they're no different than users getting faster speeds. The memo goes on to note that (just like NebuAD's system) opting out of Phorm's system does not opt users out entirely. The memo also notes that the system they tested (which may have evolved since) was a bit of a pig:There's plenty more technical specifics in the report for those interested in studying these systems more closely. I'd particularly be interested in reading thoughts from ISP network administrators. If you're unfamiliar with NebuAD, see my interview with NebuAD CEO Bob Dykes.