Re: [css3-writing-modes] Re-Summary of Tr in UTR#50 and text-orientation discussions

Asmus Freytag wrote:

> Seems to me that a measure of "how bad" and "how probable" the
> occurrences of such "undefined" behavior are in realistic scenarios
> should figure into the discussion. John points out correctly that
> fallbacks do have various kinds of costs in this instance, so one
> would like to see them at the minimum offset, if not dwarfed by the
> expected benefits.

Exactly!