Privacy and Trust, Revisited
The friendly folks from Opinity apparently disagree with me (and possibly also with Ben Laurie too, since we’re now apparently in complete agreement nirvana or something?) on that whole privacy and trust thing. I did like how I might possibly be in collusion with Orwellian NSA forces working to subvert the minds of the digerati with my “toxic assertions.” Just doing my part to usher in 1984, I guess.
In any case, before I descend into sillyness completely, I thought I should point out that “reputation” does not create trust as Tom states. Obviously, reputations can be good or bad, so I’ll give Tom the benefit of the doubt and assume that he meant “good reputations create trust.” But I don’t think even that is always really a truism.
I certainly don’t trust J. Random Ebay Seller because he has 98.4% positive feedback. Under no circumstances would I trust Jay to function as a babysitter or as my senator simply because he had a feedback score of 98.4%. I’m willing to take a calculated risk with Jay when it comes to purchasing stuff, but that doesn’t mean there is a real bone fide trust relationship going on. If I came back to Jay and did more business with him, ok, now maybe we have a case for trust going on. If I emailed Jay with a few questions before buying, ok, maybe we might have trust starting to form. But in the absence of a past relationship, most transactions that are done on the Internet are done as the result of simple calculated risk, not real trust (at least, not trust as it’s defined in my handy dictionary, YMMV). Feedback scores give more information with which to make an informed decision, but they do not, in and of themselves, create trust. Especially because most feedback systems cannot deal with the issue of building up a high score and then blowing it all on a single big con. However, in most cases, there is something closer to genuine trust going on between the purchaser and the feedback scoring system itself. I tend to trust the EBay feedback system to be a mostly accurate measurement of the reliability of a particular seller.
Right now, I have a little less trust in the Opinity system than I might have had before, purely because my judgement is clouded by having one of their head guys claim that “I don’t seem to get it.” Trust is a funny thing.
And before I let the strange analogy with the NSA go, I should also point out that, concerning the issue of privacy, there are a whole lot more statistically dangerous things than the NSA. Especially if you happen to not be a real terrorist. My original post was mainly about whether a service could trust its users. I’m not sure why everyone keeps bringing up spies. Deep-seated geek fascination/paranoia perhaps?
Anyways. I’m amazed it took this long for someone to substantially disagree with my argument. When I originally wrote the post, I climbed into a full-body asbestos suit in anticipation of the flames. But then the flames never came, Ben pretty much completely agreed with me, and the suit got all itchy and scratchy and gave me a nasty case of mesolithioma. (Long pause, then in Monty Python voice…) I got better!
I’m looking forward to Tom’s next post wherein he actually explains why I’m wrong. In my next post, I’ll give you the secrets to true bliss and you will long for that post like you’ve never longed for a post before!
Update: Kevin Kelly seems to agree with me as well. Or rather, I agree with him.
Update: I probably don’t need to tell you that I mostly disagree with Bruce Schneier on this one. A similar entry with very interesting comments can be found on his blog.
Update: Many moons ago, Penny-Arcade summed up the problem perfectly.