Brown, J. S. and P. Duguid (2000). The Social Life of Information. Boston
, MA, Harvard Business School
The central theme of this book, never overtly spelled out by the authors, for better or worse, is that Human
interaction revolves around issues of trust, and trust in the anonymous computer realm is hard (but not impossible) to
come by. Reputation systems are an important component of that, but in reality we judge the trustworthiness of a
person on a million different factors, and it is hard to code that many different variables. A firm handshake, a shared
joke, social capital, and a legion more of these nearly imperceptible cues allows us to work together. We're an
overblown troop of monkeys in some ways, and would be foolish to deny that we're hardwired for these kinds of
I would like to buy a few dozen copies and pass them out in airports while
I wear saffron robes. Or leave them in hotel rooms Gideon style. It's a vindication for a small, yet vocal, community
of people who have addressed these issues is the past, while not blaming or talking down to the people who have
refused to include the human in their design. It also gives some practical advice for people who would like to examine
information from a more holistic point of view, including how to introduce a new technology into an already existing
social system (Alexander Graham Bell did this). The Social Life of Information is one of those rare books that
informs without preaching, advocates without subjecting, and entertains without pandering. It is a smart attempt at
stepping away from the technological roller coaster (without getting out of line) and seeing how the social systems
enveloping the technology batter it about. This is an important read for any person involved in information technology
TSLOI argues that there needs to be a focus on social networks in designing information systems.
The authors advocate paying attention to where there is
The idea is that these are the effects of overreliance on information is some reductionist sense.
These all amount to a "Myth of Information" that the authors try to paint as having definite exceptions.
autonomous agents (bots)
The authors start with the premise that AI agents are becoming increasingly sophisticated, and will continue to do so. The second premise is that agents are not objective, that they have political missions either through programmer design or constraining architecture. (clampe note: this relates well to Lessig's point on the use of convenience as a means to pull an end run around choice and freedom).
An important term here is "sympathetic disobedience", or the ability tobreak rules to achieve overall goals. Without this, agents have "autonomy without accountability" while their owners get "accountability without control" Finally, the importance of trust in agent systems, as typified with the relationship between banks and ATM's, and the failure of digicash systems.
Best use of a German term
gemeinschaft- the sort small, community-based way of life broken down by industrialization (see communitarians)
People have not taken to telecommuting as was expected by futurists like Toffler. Mainly focuses on the loss of incidental learning in co-presence. (clampe Q: What are the implications of Olson and Teasley's work on "war rooms" at Ford?)
"Technology design has not taken adequate account of work and its demands but instead has aimed at an idealized image of individuals and information." This seems to be a driving thesis for the whole book, and arguably for the existence of SI.
Practice vs. Process. TSLOI relies on the Orr studies of Xerox technical reps. "Orr's study suggests that a process view of work can result in similar displacement, cutting off the lateral ties in the name of cross-functional efficiency." The book argues that these lateral ties are important for problem solving, and are often re-established in informal ways like breakfasts.
(clampe Q: How does this concept of lateral ties work into Weick's work on sensemaking? Lateral ties as fostering sensemaking?)
Information vs. knowledge vs. IT
The authors see information as some sort of more objective entity, while knowledge implies a knower. A focus on knowledge, rather than information would more clearly be person-centered.
communities of practice
Shout out to Lave. Relates this to Bruner's work on "learning to about vs. learning to be". In other words, collocated practice is very important.
networks of practice
aka "occupationals groups" or "social worlds". Are larger than communities of practice, and the agents in the network are often unknown to each other.
A question of the risks and benefits of uniting identities in the firm. How much to uncouple a group from an organizational process in order to innovate? How to integrate innovations once they are created?
Firms tend to cluster together for a variety of reasons. For instance, the authors propose the term "ecologies of knowledge" to describe tech hotbeds like Silicon Valley and Boston.
Basically, forming organizations and innovating is easier when there is an infrastructure to support it. (clampe Q: How does this relate to complexity? For instance, some of the reason clustering takes place is because of adaptation of agents from failed firms.)
This is an excellent example of the "stubborn technologies" the authors are interested in. Despite increasing amounts of information in electronic forms, the amount of paper is increasing. Why?
Discusses in a shallow way (understandably) the role of documents. Gets into the "document as darts" paradigm, and mentions some sympathy for McLuhan's "the medium is the message." However, they say "documents do not merely carry information, the help make it, structure it, and validate it."
This is more fully treated in other readings for this week, but briefly, shared documentation creates virtual "social worlds" of collaborators.
fixity vs. fluidity
Takes several forms here. Mentions LaTour's "immutable mobiles", but then deconstructs how information is immutable in the electronic world. One aspect of this is product vs. service (clampe Note: this is seen as a key distinction in the Open Source movement.) The authors argue that a certain amount of fixity is very efficient.
Mentions the British Open University and the University of Phoenix. The basic crux of this chapter is the concern that the university system is dying out.
We depend on third party organizations to place a stamp of approval on people to make our sorting process more efficient, i.e. college degree, the law bar, etc.
"The ability of the degree to shelter these activities from close scrutiny, immediate justification, and micromanagement helps provide society with more diverse and versatile candidates than it knows to ask for."
The university system enables more exploration of the design space, while drawing together people with similar intellectual interests.