Add a section to discuss techniques to quantify privacy

[bslassey]

There are several different ways to quantify privacy. Entropy has been particularly prevalent in discussions around web-based fingerprinting. Other measurement techniques may be better or worse for evaluating the impact on user privacy for a given API such as k-anonymity, differential privacy and probably others. I think we should add a section discussing the merits of each and when each is applicable to a given issue.

[EricMwobobia]

@bslassey This is a great idea. I had suggested in the first meeting we highlight the severity of the respective risks - that way we know the privacy exposure level. I, however, noticed the team did a great job in the Self-review questionnaire which helps assess the exposure - https://www.w3.org/TR/security-privacy-questionnaire/#threats . A section on tested techniques should be a good addon to help users get more from the threat model