Re: Proposed version number change for 21.2
Eric Eide wrote:
sb> There wouldn't be a 21.2. That's part of the point.
There would
sb> be a 21.2.XX where XX is the number where we finally release.
This seems unnecessarily confusing from a user's perspective, unless 21.2.XX is
going to be publicly released for more than one value of XX.
This could work with an odd/even version approach, i.e. betas would be
21.1.xx, the initial release would be 21.2.0, updates would be 21.2.1
etc.
Additionally, I think there is a risk that users will think that the
software
is a for-gurus-only version if the XX is anything other than zero.
So gcc-2.8.0 is the `for-users' version? :)
--
Glynn Clements <glynn(a)sensei.co.uk>