<< 1 >>
Rating: 
Summary: Excellent book on applied graph theory with the BGL
Review: I really like the BGL book! What strikes me the most is the accessibility - the introductions hit the right key immediately, so there's really no need to keep reference material at hand. The samples are equally good, and they're the right size to be easily transformed into solutions to related problems. The book currently rests on my bedside table, and only my favorites make it there...
The Boost Graph Library is powerful, and the BGL book is an excellent companion for graph-related studies or real-world graph problem solving. Room for improvements:
* The index
* Even more samples (maybe not so elaborate in the text, could be on the CD) Finally: I like the book a lot, and even if you're not "a graph person", chances are that you will be after reading the BGL book. I highly recommend it!
Rating: 
Summary: A fine addition for any STL/C++ programmer
Review: The boost graph library extends the STL tool box of containers into graph containers. Wished for a general purpose graph? This is the place. Never studied graph's in a CS course, but know basic data structures like lists, hashes and trees? This is a fine place to start. The book goes from basic, what is a graph good for, (solving search, and routing problems) to what are the various drawbacks to the different approaches. If you are used to using the C++ STL library you will feel right at home using this. (BTW this library works with MSVC 6.5)
Rating:
Summary: Solid and entertaining
Review: The first part of this book applies the algorithms of the Boost Graph Library (BGL) in fun ways. For example, it solves the problem of Make-style file dependency checking using BGL. And it explores everybody's favorite pastime, "Six Degrees of Kevin Bacon" and similar problems. These real-life examples greatly spice up an otherwise very boring topic. BGL uses concept checking and property maps and a few other Boost approaches that aren't very well explored in the text. Granted, it's not a book about C++ programming in general (like "Modern C++ Design") but a bit more coverage of these key ideas would have been nice. Also, unlike the STL, several of the alogirithms are named by their canonical names instead of simply specifying their behavior and performance (e.g. names like dijkstra_visitor and bellman_visitor vs. list and map) . This means that library users may need to refer back to the text to figure out which algorithm they desire. The choice to typeset all code in bold italic text was very unfortunate, but the use of C-Web to explain larger routines more than makes up for this. The latter part of the book is a dry reference to BGL. More of the exploratory material would have been preferrable. Overall, a great addition to my C++ library.
<< 1 >>
| 
