AMDG Emil Dotchevski wrote:
I haven't followed the discussion and I apologize if I'm repeating something, but in my mind if three is a useful library A, and if we could provide a layer (wrapping?) which makes library A work better with Boost, the only question should be how popular library A is, and how many of library A's users would benefit from an easier Boost integration.
That's perfectly true if the question is whether such a wrapper should be created, but...
Specifically, what platforms that library runs on is not important.
Portability is important for deciding whether to include such a wrapper in Boost. From the requirements: "A library's interface must portable.... If a portable implementation is not possible, non-portable constructions are acceptable /if reasonably easy to port to other environments/..." [emphasis mine] In Christ, Steven Watanabe