On 2/24/18 2:41 PM, Stefan Seefeld via Boost wrote:
As one of the more vocal opponents to the announced "move", allow me to comment.
I think there is a real danger that two distinct questions get conflated:
a) a technical review of infrastructure to build Boost libraries with CMake.
b) whether and how individual library developers / maintainers can be mandated to use any particular tooling (be it to build, to document, to bug-track, etc., etc.)
It's true, a successful completion of a) makes it more likely for the code to be accepted by the community. But neither should a positive review result in a mandatory use by all libraries, nor should a review be biased based on the perceived danger of having to migrate if the review is completed successfully.
Right now my only focus is to get a fair and complete evaluation the only serious candidate for a CMake build system. Other considerations/decisions can be deferred until we past this hurdle. Robert Ramey