Category Privacy

Describe CategoryPrivacy here.

Data is inherently frictionless, and do not exhibit properties of wear and degradability. However, in the interest of privacy, data needs to have some friction, and wear out with use, and degrade.

Some patterns that have been implemented to deal with these issues are:

• Restrict the size of the data set returned. Telstra Whitepages limit the number of results returned, and require users to be more specific in their queries.

• Adding wrong data to results. Cartographers often add fictional landmarks and features to prevent copying from rivals.

• Make data difficult to scrape. A lot of web services are starting to impose tests that it is viewed by a human, and not an automaton. Another example is through the use of Flash interfaces, data is more difficult to scrape. Google will probably not cache it either. ;) See also CaptchaTest

• Liberal use of no-cache directives, and expiry headers.

• Data scrubbing to removing identifying information. Court cases involving children do not identify the children in the court transcripts. Now that a lot of more court documents are available on-line, perhaps it is time that all personally identifiable information be removed from the judgements and transcripts.

• Throwaway identities, handles, Nome de'plume

• Explicitly marking documents as being confidential or personal. For instance, Microsoft documents, emails.

• Narrowcast. Information which should be presented to qualified audience. Password protected RSS. Email instead of the web.

• Using DRM (Digital Rights Management) technologies as well as license servers which must be consulted at every document access, where authority servers may also expire documents. A temporary key which expires is issued by the server to the intended audience in order to view the document.