It is the title of an interesting paper by Roger Dingledine and Nick Mathewson. They are members of the Free Haven project. This project studies topics such us onion routing (technology used by TOR), or Mixminion an anonymous email network.

The paper presents two challenges: usability and network effect.

• Usability is a typical challenge of security solutions. The authors show that often privacy setting requires technological skills that are opposed to ease of use for everybody. The easy solution is often to delegate security decision to the user, who is not necessarily the best person to decide. This reminds me the security model of Android, where you have to decide (too) many parameters.
• Network effect; efficient anonymity requires to have a lot of traffic to hide within. This rises the problem of bootstrapping. And here is a nice tradeoff. If your system is extremely secure, it will most probably be difficult to use, thus attract fewer people, thus reducing the strength of anonymity. On the other hand, if the system is easy to use, thus less secure, it may attract more users, thus strengthening anonymity.
  For instance, in the design of Mixminion, they had to answer the following tradeoff:

  Since fewer users mean less anonymity, we must
  ask whether users would be better off in a larger network where their messages
  are likelier to be distinguishable based on email client, or in a smaller network
  where everyone’s email formats look the same.

The three described use cases, Mixminion, TOR, and JAP, are excellent illustrations of the issues. An excellent paper.

Citation: N. Mathewson and R. Dingledine, “Anonymity Loves Company: Usability and the Network Effect,” Proceedings of the Fifth Workshop on the Economics of Information Security WEIS 2006, pp. 547-559.

In July 2010’s issue of IEEE Security & Privacy, Bruce Schneier in a one-page paper presented his taxonomy. It is extremely interesting. My comments are in italics.

• 1. Service data is the data used to manage the service such as your name.
  You have control on the creation, although you may be obliged to give sometimes real data.
• 2. Disclosed data is what you post on your own pages.
  You normally have full control on it.
• 3. Entrusted data is what you post on other people’s pages.
  You have control on the creation, but lose control on its life.
• 4. Incidental data is what other people post about you.
  You do not have control on the creation, nor on its life. Of course, your entrusted data are incidental data for other people.
• 5. Behavioral data is data the site collects about your habits by recording what you do and who you do it with.
  This is the “raison d’être” of many social networks. Never forget that there is no free lunch. Most of the business models are based on “selling/using” your profile. You have no control, excepted that you may try to control your behavior.
• 6. Derived data is data about you that is derived from all the other data.
  This is where the social networks are polishing your profile and thus increasing its value. The more they know you, the more valuable ads/personalized services they will be able to offer. You have definitively no control.

Category 5 and 6 are the most interesting ones from the privacy point of view. How can you control what the social network provider will infer from your activity on the social network.

The reference of the paper is
B. Schneier, “A Taxonomy of Social Networking Data,” IEEE Security and Privacy, vol. 8, 2010, p. 88.