In boost.graph what would be better : a) Manipulate the original graph - it would save space(crucial for big graph) -if the user want he can copy the graph object to a new object and then feed into functions b) Generate a new graph - although space will be allocated but user has the opportunity to directly input the graph.

I'm not sure I understand the context of this question. In general, the choice to copy or modify is a design decision that depends on the application. Andrew Sutton andrew.n.sutton@gmail.com