[This is a draft introduction to the term social protocol and may be expanded to (1) a slightly fuller exposition of social protocols, (2) the theme of a collection of related papers, or (3) a history of policy related work at the World Wide Web Consortium. Please send any comments to me.]
Joseph M. Reagle Jr.
Resident Fellow, Harvard Law School
Research Engineer, W3C/LCS, Massachusetts Institute of Technology
"Social protocol" is a term I use to discuss protocols (or their applications) that enable individuals and communities to express social capabilities. This would include tools necessary for creating rich content, managing trust relationships, making verifiable assertions or recommendations, and enabling agent assisted (or automatic) decision making. The ability to make verifiable assertions, to build reputation, to solicit advice, and defer to a trusted source are all real world capabilities. This is what meat space relationships are built upon. If we want sophisticated cyberspace relationships, we will need similar mechanisms.
Social protocols have substantive policy issues related to their development -- as do many other protocols, such as DNS. However, the key difference between a social protocol and something like DNS is that little social behavior, cues, or relationships are carried through DNS. The key function of social protocols is to express and operate upon such information for users and their computer agents. I would not consider email to be a social protocol, rather it is the encoding and transport mechanism for the social protocol of natural language. A good example of a social protocol is the warn function in the online chat application AOL Instant Messenger:
A warning is a form of electronic vilification. It allows a user who has been affected by the online behavior of another user to express an opinion about that behavior.
[http://www.aim.aol.com/faq/warnfaq.html]
While I do not consider email content itself to be a social protocol, I would the filtering of email on the basis of email headers. What about filtering on the basis of the content of the email? Probably. However, it is not my intent to amass of a list of applications and features which I would consider to be social protocols. Rather, I wish to present a general concept and state that the determining factor for analysis is the degree to which the semantics and operation of social behavior/structure are captured within the data structures and protocol of the application. This is critical because it is at this point that social/legal concerns and methods have an impact on technical design.
I first used the term social protocol in April 1997 to describe to describe a type of technical work that I found compelling at the World Wide Consortium (W3C). This work was part of the W3C's Technology and Society (T&S) domain to which I'd recently joined as a Policy Analyst. The work of T&S was strongly reflected in the lead project of the time, the Platform for Internet Content Selection. (PICS). I used the term in an introduction to the work of the W3C and T&S at a IMA conference, hosted by Brian Kahin, on Internet Controls. My concluding slide stated that:
A social protocol is not so much an "Internet Control," but a way of using meta-data and negotiation to control the interactions one has with others on the Internet/Web.
[http://www.w3.org/Talks/9704-IMA/slide9.htm]
Interestingly, it has been this tension between the ability to control ones own interactions and the possibility of this ability being co-opted by centralized authority that characterized much of the debate [1] [2] regarding the nature of PICS in the course of that year. [Res97]
My use of the term in April 1997 was strongly influenced by three factors:
Subsequently, I've used the term in a number of presentations and expositions, most substantively in a paper [CR97a] with Lorrie Cranor regarding the design of the Platform for Privacy Preferences (P3P). The term was also used in the course title of the Harvard Law School's (Fall '98) course "The Law of Cyberspace -- Social Protocols."
There are also a couple of variant terms that describe loosely related concepts. Social Informatics was a term that evidently preceded my usage to describe a broader body of research and study "that examines social aspects of computerization -- including the roles of information technology in social and organizational change and the ways that the social organization of information technologies are influenced by social forces and social practices." Tim Berners-Lee has used the term "social machines" in 1999 presentations at LCS's 35 Anniversary and WWW8 conferences.
Others work that is related to social protocols include:
[CR97a] Cranor and Reagle. "Designing a Social Protocol: Lessons Learned from the Platform for Privacy Preferences Project" Telecommunications Policy Research Conference. Sept. 97.
[CR97b] Cranor and Resnick. Protocols for Automated Negotiations with Buyer Anonymity and Seller Reputations. Proceedings of the 1997 Telecommunications Policy Research Conference.
[Cra98] Cranor. P3P Guiding Principles. W3C:NOTE-P3P10-principles-19980721.
[LR98] Lessig and Resnick. The Architectures of Mandated Access Controls. Proceedings of the 1998 Telecommunications Policy Research Conference.
[Les97a] Lessig. What Things Regulate Speech (requires Adobe Acrobat reader)
[Les97b] Lessig. The Law of the Horse: What Cyberlaw Might Teach. Stanford Technology Law Review. Working Paper. 1997.
[RW98] Reagle and Weitzner. Statement on the Intent and Use of PICS: Using PICS Well. W3C:NOTE-PICS-Statement-19980601.
[Rea96] Reagle. "Trust in a Cryptographic Economy and Digital Security Deposits: Protocols and Policies" Thesis E.E. MIT 1996 M.S.
[Res97] Resnick. PICS and Intellectual Freedom FAQ.