海角大神

You鈥檝e probably heard the big news by now: Giant retailers like Target know a lot about you. They know which part of town you live in, what credit cards you carry, what brands you prefer, whether or not you鈥檝e been divorced and 鈥� according to a recent cover story in The New York Times Magazine 鈥� when a woman is pregnant.

That last bit is the part of the story that gets everyone riled up 鈥撀� that Target can use predictive analysis to determine when a shopper is pregnant based on her buying habits possibly before she鈥檚 even told her coworkers. That seems creepy and Big Brother-ish to people. It seems way too personal.

But how different is it from Target predicting that in late July, I will buy lots of sunscreen and beach gear because my family goes on a beach vacation every August? That doesn鈥檛 sound creepy, does it?

Similarly, Target has simply figured out that when women become pregnant, they buy more of certain things: lotion, vitamin supplements, cotton balls. By analyzing shopping purchases, Target can assign a pregnancy prediction score to every shopper. If you score high, Target will send you special offers on things expectant mothers might need.

I don鈥檛 want to get too philosophical here, but is Target really sending a special offer to you? Or is it sending a special offer to a set of data points?

It seems to me, that in order to find the pregnancy example creepy, we have to believe that our personhood comes down to data. This is not a trivial point. As we plunge forward into the data-driven future, we will have to determine to what extent data define us as individuals so we can figure out when our privacy is being abused. Not an easy task.

I certainly don鈥檛 know where all the lines should be drawn, but it seems to me that predicting my life experiences and offering me relevant coupons is far from approaching the line of violating my civil rights or abusing my privacy. This seems purely transactional.

Of course, if the same predictive analysis were used by an insurance company to drop a woman from coverage before she showed up for prenatal care, that would cross the line. And it would be illegal.

But let鈥檚 get back to the Target example. If Target can predict what I might need based on figuring out things about me, I鈥檓 all for it. I want Target to analyze my purchases and figure out I鈥檓 a 50-year-old basketball player because I buy lots of ibuprofen and menthol-smelling rubs. And I want them to send me special coupons on athletic socks and basketball shoes. I buy a pair every year, and Target should know that. This knowledge can help both of us reach our goals: I want to drive to fewer stores to get my needs met, and Target wants me to buy more of the things I need at Target.

If being reduced to a set of data points is what it takes for the systems I interact with to predict my behavior, I鈥檓 OK with that.

When I google 鈥満＝谴笊� Science Monitor鈥� to find an article I鈥檝e recently read, I鈥檓 glad I only have to type the 鈥淐鈥� and the 鈥渉鈥� before Google fills in the rest. They know I type in that word a lot. Thank you, Google.

If a car company knows when I bought my last car and how long I typically keep a car and it starts sending me offers a few months ahead of when I鈥檓 thinking of buying a new car, I say 鈥淭hanks.鈥�

Speaking of cars, I wish I owned one that remembered my seat position like my brother鈥檚 does. He gets in his car after his wife has driven it, and the seat returns to the position (the set of data points) it knows he prefers. Thank you, Toyota.

The future will be full of interesting debates about privacy and data. There will be lawsuits, Supreme Court decisions, new legislation 鈥� the book on privacy will literally be rewritten.

I look forward to that, but in the meantime, I鈥檒l take any coupons Target wants to send me, preferably right to my smartphone. They should be able to figure out that I鈥檓 a bit of an environmentalist and opposed to traditional mailings.

Jim Sollisch is creative director at Marcus Thomas Advertising.