I'd like to repeat my suggestion for the review manager to consider rescheduling this review without an acceptance decision now.
Usual fear, uncertainty and doubt from this individual. He is a consistently negative person - anything I ever propose he always agitates the same way, and I hope people will ignore his trolling. I will restate that apart from removal of a few member functions and types, and renaming outcome<T> and result<T>, the presented library is not going to change much because I have not been persuaded of any defects in it. I have been persuaded to add non-empty-capable varieties, and narrow observer varieties. But they're all typedefs of the same implementation, just as is currently the case in the presented library where there is a single implementation typedefed into varieties. The only major change in implementation I've agreed to is replacement of preprocessor-based stamping out of varieties with a template-based implementation which does exactly the same thing. Such limited changes are hardly cause to reschedule a review, but I'll leave that to the review manager to decide. Niall -- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/