Web Anatomy was a surprisingly poor book given the authors pedigrees. The authors try to build the case for "design frameworks" a not altogether distinguished resource apart from design patterns and components.
The distinction is somewhat arbitrary. For example in many implementations such as the MailChimp pattern library, design patterns and components (reusable code), live together.
Ultimately the ambiguity forces the authors to spend too much text justifying itself.
The book effectively is broken down into an intro that lays the foundation of what frameworks are and why they are important, and next a section that covers examples of frameworks and finally a conclusion which purports to demonstrate how to build your own framework libraries, but instead attempts to justify itself once again. Here its unclear how segments made it past the editor, specifically how the authors claim frameworks are an integral solution to solving Amtrak's usability issues are odd.
Another oddity is the choice of case studies, the authors choose easy and at times arbitrary (movie sites) frameworks to document. Admitting in some cases that the findings were not their own but simply the outcome of a workshop.
Ultimately frameworks have the most value when viewed as a research deliverable owned by the research team, this allows it to be liberated from the communicative work pattern libraries have to deliver and allows it to have a more natural update process that integrates design research into the development process.
I like the concept but feel this book is a little lacking. I feel the authors should have avoided specifics and just focussed on developing and consuming interaction frameworks or gone for the full reference guide like Christian Crumlish and Erin Malone’s Designing Social Interfaces. Read more on purecaffeine.com.