Painful as this is, I am convinced that we need to re-evaluate and re-argue the interval delimiter.
I have gone through the exercise of reviewing the entire archive to attempt to find all of the discussion on this. What I find is that there have been only two members of this forum who argue against the slash character as the interval delimiter. Most everyone else is comfortable with it, to varying degrees. And while those opposed make a convincing case that ISO should never have used it, I'm not convinced that there is a compelling argument that ISO's mistake was so aggregious as to justify violating it in our spec.
I am concerned that having two different interval delimiters serving the same purpose could significantly weaken this spec. And this seems more serious to me than the similar problem (still under evaluation) of having both basic and extended forms (hypens and colons) because at least in that case both are legal in 8601. In the case at hand we are talking about prescribing an alternative that is not 8601 compatible, for a feature that is adequately (if not ideally) handled by 8601.
So I am hereby re-opening this for debate. If you still feel strongly about this then please offer your arguments again.