The reading notes record thoughts from things I read. 這網誌是我的一些閱讀後的思考和摘要記錄。My website 我的網頁:

Privacy in the world of big data

Personal data privacy is everyone’s concern.  But our understanding of such may be limited.  Scientific American has an article on this in the November 2013 issue.  It was written by a privacy advocate on privacy in the world of big data.  We may read it bearing in mind his bias, but there is some insight in his observation.

We think privacy of personal data mainly concerns our identity card number and data which may easily identify us.  For the privacy advocates, privacy is about naked information available to some but unavailable to others.  Privacy is the arbiter of who gets to be in more control.  At present, most of the debate on privacy is the type of personal data which can be collected, used, disclosed.  The most talked about point is the trade-off.  In the modern environment where we need the society network for goods and services, there is a need to provide our personal data to other providers.  The question is framed to be how much privacy are people willing to give up for certain benefits.  The underlying notion of this question is that if people would only share more, they would enjoy more convenience or create more value in online networks.  The disadvantages of giving too much personal data away have been limited to unwanted advertisements.

In fact, this form of privacy is only an illusion covering the real situation.  Information about a person include anything that a person did, whether online or got into the network from other means.  Anything we interact with another entity, say the government, banks, purchases, services are all recorded.  Anything we did in any kind of social media are all recorded.  This is done with our tacit consent.  The debate on privacy is lopsided as no one has complete information on what kind of personal data is stored and processed by others.  What we can see are the unwanted advertisements, unwanted friends in social network, and some secrets disclosed by Snowden.  But that is only the tip of the iceberg.

You may encounter the scenario where a Google search gives you the desired results; or Amazon recommends the books your like.  They are able to do that because they maintain a profile on you.  The information in your profile is built up from your previous connections with them.  In fact, anything you did with them are recorded, analyzed, and used to refine your profile so that they can provide the goods and services most likely suit your needs.  This kind of refined and targeted service is a two-edged sword.  On the one hand, people may get information on what they like, and at the same time, information limited to what they like.  It is an information filter which may keep one-half of the real world away from you.  The big data also work for the commercial world.  With personal information on the population obtainable, medical insurance companies may be able to devise sales strategy excluding high-risk customers, thus depriving those who need insurance most.

A more frightening scenario, which was predicted by novelists long time ago, and which has gradually becoming a reality, is the behaviour changing tactics based on big data.  A profile on anybody can be set up using any personal data which can be collected from the network.  The profile can be refined as more information is gathered and analyzed, to the extent which could reflect the preference and intention of such person.  A harmless proposal is to locate the position of the person at all time.  If this person likes to have a cup of coffee whenever he leaves the MTR station, a text message could be sent to his phone indicating the best coffee shop nearby.  This can be used as a business opportunity in selling advertisement slots to coffee shops.  This could also be used in electioneering where a person may be steered to vote for a particular candidate.

With personal information so widespread, there is no way we could keep it all from others.  The author suggests that people should be given control on the level of privacy they could keep.  One of the ways is to attach a monetary value to personal information.  All information identified with a person are given a code so that a small amount is credited to the person whenever it is used.  In this way, businesses and the government may be more careful in using others’ personal information, in stead of freely mining the big data.  But this charging system may be very complicated, and I think it may itself become a source of personal information.

Incidentally, I just saw this video on YouTube on a social media experiment.  This guy used his smartphone to check social media on other users near his location.  Then he checked their status, profile, posts and photo which were open to public, and then approached them by name.  People were surprised that a complete stranger could know so much of their personal details just by checking on a smartphone.