Feeds

Amazon's Silk looks creepily Phorm-ulaic

Data-hoarding by proxy

Choosing a cloud hosting partner with confidence

Analysis A trick question for you...

What's the difference between Phorm's controversial WebWise system, and the kind of giant web proxy unveiled by Amazon yesterday? Technically, there isn't one. WebWise and Silk are doing exactly the same thing. Both intercept private web traffic – and massage it. Both also aggregate enormous amounts of private data and make behavioural inferences from this data hoard.

There are other differences, of course.

Phorm caused a furore because of the implementation policy. BT didn't inform its users they were guinea pigs in the Phorm trials – and planned to opt its users into the system.

Perhaps we have exceedingly short memories these days.

For 18 months, Phorm was the biggest UK technology story – and the bitter experience of the consumer backlash is branded into ISPs' memories. They'll be wondering this morning how on earth Amazon can get away with something so similar.

Privacy is like boiling the proverbial frog. We get accustomed to giving more personal information away, without objecting. But if the data aggregators turn up the heat a little too quickly, people notice. Phorm's stealth interception was brazen: a step too far for many users, as was Facebook's Beacon, Google's Buzz... it's a grim and growing list, is it not? But remarkably Gmail has continued to grow, despite reading your private exchanges and opening them for advertisements.

Amazon Silk has the potential to make WebWise look like a tiff in a teacup. As Sophos' Chester Wisniewski points out, the Silk privacy policy is complex and contradictory. It's far from clear whether a Silk user's HTTPS traffic is also intercepted, something Phorm declined to do. The device's MAC addresses will be discarded, along with other identifying information, after 30 days. That strikes me as a moot point: Amazon already knows who you are. After 30 days, Amazon will have all the behavioural data it needs.

"If you buy a Fire device, think carefully as to whether your privacy is worth trading for a few milliseconds faster web surfing experience," suggests Wisniewski.

Opera has such a system, an enormous one, that's growing faster than Google, that it calls (more honestly, I think) a "transaction cache".

Last year I quizzed them about how they planned to exploit it for behavioural information. Opera executives were quite clear: they weren't. And they don't need to. The Opera business strategy is a novel idea of performing arbitrage against competing ad networks, rather than data-mining the data itself for behavioural inferences. What makes Silk even creepier than Phorm, to me, is that it's already making those inferences – it anticipates the next page you'll visit by prefetching it. This makes the Kindle a data-collection tool for those "Customers who bought X also bought Y" nudges.

Amazon, like Phorm, is betting that you don't care enough about privacy to shop elsewhere. And from the gradual privacy ratchet, and the certain absence of opposition from rivals – nobody wants to poison the well – it may well succeed. ®

Top 5 reasons to deploy VMware with Tegile

Whitepapers

Choosing cloud Backup services
Demystify how you can address your data protection needs in your small- to medium-sized business and select the best online backup service to meet your needs.
A strategic approach to identity relationship management
ForgeRock commissioned Forrester to evaluate companies’ IAM practices and requirements when it comes to customer-facing scenarios versus employee-facing ones.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Website security in corporate America
Find out how you rank among other IT managers testing your website's vulnerabilities.
New hybrid storage solutions
Tackling data challenges through emerging hybrid storage solutions that enable optimum database performance whilst managing costs and increasingly large data stores.