[Boost-users] Using boost::lamba with a vector of shared_ptr's