Le 10/02/2013 17:07, Rob Weir a écrit :
But it is not question of mathematics. It is a question about how we make controversial changes and how we keep the ecosystem informed about breaking changes. Hiding behind one (but not the only) mathematical convention for this calculation avoids dealing with the fact that a broken changes was checked in without any review on this list and without any attempt to inform users and application developers about it. I hope we agree that this should not be how we do things.
Agreed. But for the 0^0 issue, you were very verbose against such a change, when it affects very few users whereas for the broken extensions issue, that would affect a good deal of users (perhaps millions), you were just asking about how to communicate. Nothing said about the comments from Bernard and Hans. The difference in the handling of both issues is what I find rather strange. Hagar
