Computing Columnist: How Deep are the Mines?
We all know that our data is being mined.
I admit to being mildly amused and frankly admiring at the ads Gmail served me that related quite clearly to the topics of my emails. It was clear that my mail was being parsed for keywords, then then relevant ads were being served to me in the hopes that since I was already talking about the subjects, I was more likely to click through to the sponsors. In fact, I find such sponsorship less offensive than ads that have nothing to do with me or my interests popping up, interfering with my experience of a website, or worse yet, blocking my view until I click off of them or wait them out.
Then one day I was searching for information on a product. I wanted to find pros and cons. I discovered that the maker of the product had bought up all the key words - positive and negative. If you looked up the product name and "scam," you got a glowing review of the product. If you looked up "fake," "fraud," "rotten," "no good," whatever bad word you could think of, it led to a site - to all appearances a real site - that gave a positive review of the product. So much for research!
Then came the day when I was doing research for a client on a particular product in which I had no personal interest. From that point on, wherever I went on the internet, from YouTube to Blip to blogs to commercial sites, I was followed by ads for this product category! It was more than a little annoying, particularly as I had no interest whatsoever in the product. And besides, enough is enough.
Today, while posting a video to Facebook, I noted the sidebar ads, and realized that they had nothing to do with anything I had posted on Facebook, but did have something to do with searches I had performed on Google, and emails I had exchanged via Gmail. In other words, it appears my browsing and interactions are being mined for my interests in order to target me for advertising on a platform distinct from the one on which I expressed the interests. Yes, this does alarm me.
Most of the time, this is harmless, of course. If I send an email discussing an upcoming event I'm attending with friends, and I get served a related ad, that's just fine. My concern is, where does it stop, and who else gets to see the data?
Shared data in the aggregate is harmless. Even shared data confined to, say, Gmail and served on the basis of keyword matching is slightly concerning but generally of little threat.
But what if the government demands the information? The police? If I express a dissenting political opinion, there might come a day when this could be, um, troublesome. If I make a joke about not wanting to pay taxes, or engage in "hate speech," (people's comments are already being filtered on Facebook when they are deemed "hate speech," though it's clear that it's ok to hate some people and not others), or God forbid, admit to a crime, can this information be associated with me and used to prosecute me?
Of course, we already know that government agencies are targeting certain groups and individuals suspected of terrorism and plotting against the government and engaging in criminal activity, and monitoring their online activities. So certainly, we have ample evidence that the technology exists to see what we're doing, make sense of it (why pick this topic versus the dozens of others that I might mention in the literally hundreds of emails I might exchange in a week, and the hundreds of searches I might perform in that same time period?) and follow up on it. The technology even exists for those who place an ad targeting me to know whether I clicked on the ad, and whether I then actually took action once I reached the landing page associated with that ad. In other words, I can be followed and tracked all the way to the bank.
Harmless enough on its face, but the implications for our privacy - a right we Americans hold extremely close and dear - are worth consideration. But that horse, as they say, may be long since out the barn.