Am 07.01.2014 10:59, schrieb Paul Moore:
On 7 January 2014 09:40, Georg Brandl <g.brandl@gmx.net> wrote:
Very nice, thanks. If I was to make a blasphemous suggestion I would even target it for Python 3.4. (No, seriously, this is a big issue - see the recent discussion by Armin - and the big names involved show that it is a major holdup of 3.x uptake.) It would of course depend a lot on how much code from unicode formatting can be retained or adapted as opposed to a rewrite from scratch.
Will the relevant projects actually support only 2.X and 3.4/5+? If they expect to or have to support 3.2 or 3.3, then this change isn't actually going to help them much. If they will only support versions of Python 3 containing this change, then it may well be worth considering the impact of delaying it till 3.5.
Yes, exactly. Another, and probably better, proposal would be to make 3.5 the "ultimate" viable porting target: we now know pretty well what the major remaining roadblocks (real and perceived) are for our developers and users. The proposal would be to focus entirely on addressing these roadblocks in the 3.5 version, and no other new features -- the release cycle needn't be 18 months for this one. This is similar to the moratorium for 3.2, but that one came too early for 3.x porting to really profit. In short, I am increasingly concerned that although we are going a pretty good way (and Nick's FAQ list makes that much clearer than anything else I've read), but it is not perceived as such, and could be better. We have brought Python 3 on the community, and as such we need to make it very very clear that we are working with them, not against them. A minor release dedicated to that end should be a very direct representation of that. I know about the "release everything to PyPI" strategy, but it just doesn't have the same impact. It would be very cool to have multiple projects working together with us for this, and at the release of 3.5 final, present (say) a Mercurial that works on 2.5 and 3.5. Mostly pipe-dreams though... Georg