In article <[EMAIL PROTECTED]>, Steve Holden <[EMAIL PROTECTED]> wrote:
> Aaron Watters wrote: > > The cost paid for these minor improvements is too high in my > > book. But I suppose if it is going to happen do it sooner > > rather than later. Just *please* *please* don't > > systematically break the pre-existing code base again for a > > very long time, preferable ever. > > I'm pretty sure the 3.0 compatibility breakage is a one-shot deal. If > it's not I won't be the only one looking for Guido with a bog stick in > my hand ... Depending on what you mean, that appears to be either a truism or an absurdity. If you mean, 3.1 won't break code like 3.0 did ... well, of course. If you mean, there won't be a 4.0 that means the same thing for compatibility that 3.0 means, then I can't imagine how you could be convinced of this. Changes to Python in 3.0 won't satisfy the continuing "need" for change thereafter. Donn Cave, [EMAIL PROTECTED] -- http://mail.python.org/mailman/listinfo/python-list
