Honestly, I don't need to serialize std::tr1::shared_ptr - why would any of us ever use it as long as we have boost::shared_ptr which (with serialization is a superset of std::tr1::shared_ptr.)
That's a completely bogus argument.
I understand your objection to boost::serialization depending upon the implementation details of boost::shared_ptr. I just don't see any way to address this other than making big changes to either the serialization libary or shared_ptr. Even then, its not obvious to me that its even possible.
Doesn't Peter's own serialization library succeed in not requiring special access to boost::shared_ptr? If so, it is an existence proof that there's a way to do it, which I think means that big changes may need to be made to Boost.Serialization. It will be a reasonable expectation for a Boost.Serialization user to serialize std::tr1::shared_ptrs in the future. Has Peter described the algorithm he used? Dave