I think it would be a good idea to provide some technical notes for various boost class implementations. For example, I have recently been studying smart pointers - I've ready Andrei Alexandrescu's discussion of the topic in his book, 'Modern C++ Design'. It's good, and fairly comprehensive, but doesn't cover all of the nuances of some of the boost smart pointer implementations. For those of us interested in learning more about the language, and more about boost classes and the rationale behind the implementation - and for those of us who simply don't have the technical background to "just know" the answer at a glance, as I'm sure some of you contributors do - why not provide a simple rationale for adding a particular member function to a particular class? I'm all for sharing, but I would like to know the reason for half the implementation. I believe this would give something new to the C++ community in a simpler way than has been done in the past - information generally gathered individually and painfully over years of study. Thanks, John Calcote Novell, Inc.