Vijay is now loading an assembly and all the raw data from a given file into datastructures (PE Headers, CLI Headers, sections, metadata streams, metadata tables, etc) ... basically all the stuff I won't have to worry about with the Zsolt - since it is going to use the unmanaged metadata api.  This weekend I am refactoring things a little and starting to design the next layer -> my implementation of the functionality that the unmanaged metadata api provides.

Anyone have any requests for things that you just can't do with reflection?