Information Flow: The Logic of Distributed Systems (Cambridge Tracts in Theoretical Computer Science, Series Number 44)

Impressive formalization of information flow via mathematics and logic, but in the end it's not clear if this subject holds any practical value. For, in practice, the model of information flow always requires a priori working models of the phenomena under consideration before the principles of information flow can be put to use. For example, consider currency fluctuations as information about some nation's economy. You can model this using Barwise's scheme, but doing so doesn't bring anything to one's understanding that we didn't already have from economics. No real clarification on what information is, either (although there's a lengthy discussion of this at the outset). Requires a fairly good working knowledge of set theory and basic formal logic. Poorly edited.

There's a lot of informal discussion at the beginning about what information "really is", but very little in terms of innovative content. The book contains a good formal approach that expands on earlier work in the same field, but it hangs on too many ill-fitting concepts as the "Xerox principle" and the notion that knowledge can be quantified. The book also completely ignores the ontological perspective which other researchers have addressed for over a decade.