The collaboration book is mostly in the context of internet based and software development related collaboration. Here's a copy of some notes originally made on TiddlySpace.

_Because I am, according to @jrbl, supposed to "just fucking write a book already".

• Collaboration requires SharedGoals
• SharedGoals require SharedUnderstandings
• SharedUnderstandings require SharedLanguage
• Effective Collaboration comes from effective information sharing, access and comparison (so the participants can get new knowledge)
• Communication is not overhead
• Knowledge is Private, in a single head
• Knowledge becomes Information when uttered
• New knowledge (learning) comes from comparing information; placing it in context
• Computers are uniquely qualified tools for information comparison (c.f. Winograd & Flores)
• The Unix Way and I
• The best tools are open, in the sense that they make use of data which can be easily used by other tools.
• Information networks show not only links between things, they also show voids where there ought to information, but there is not: Networks can highlight ignorance and thus opportunities or needs for learning. See KnownUnknowns.
• Ideal information tools are ready to hand, keeping the task (learning) in the forefront. The information involved should be poised to inspire breakdowns (c.f. Heidegger and The Computer as Tool.
• Naming, in information networks, matters: We want to encourage discovery and collision as it is in the act of collision that the most learning happens.
• Identification is separate from naming. Identifiers must be truly unique and meaningless in order to be persistent and transportable in an otherwise transient network.
• Reuse by reference, not copy.
• The act of writing is fundamentally different from speaking or thinking. It requires a bit more follow through, and as a result writing a problem down, even if just for yourself, can lead to the solution. (Thanks to @FND for the reminder).
• When collaborating, group members can be more effective through stigmergy: leaving artifacts in the environment that others can be informed by.

Incomplete