| 1 |
On Mon, Jul 20, 2015 at 6:03 AM, Ben de Groot <yngwin@g.o> wrote: |
| 2 |
>> I don't see any strong technical reason to switch from python2 + |
| 3 |
>> python3 to python2-only enabled. Some people don't like having two |
| 4 |
>> versions of python installed -- that's about the gist of it. |
| 5 |
> |
| 6 |
> Indeed, there is no strong technical reason, except that some people |
| 7 |
> like to keep their systems more lean. But I think having a smaller |
| 8 |
> stage3 tarball is a more important reason. The python team has |
| 9 |
> historically left that up to the RelEng team, which has been averse to |
| 10 |
> handling that themselves. |
| 11 |
> |
| 12 |
>> So, I'm personally not going to make that change without some kind of |
| 13 |
>> vote on it. I can arrange a vote within the python team if you like. |
| 14 |
> |
| 15 |
> I would like to hear from the other team members, yes. |
| 16 |
|
| 17 |
I have sympathy towards those who are asking for only one Python in |
| 18 |
stages (as in, I would be fine with that), but I very much think we |
| 19 |
should not leave Python 3 out of generally installed systems by |
| 20 |
default. We need to move through the transition, and increasing the |
| 21 |
barriers to Python 3 adoption will only make that process slower. |
| 22 |
|
| 23 |
I also feel like a voting process for this is probably not a solution. |
| 24 |
|
| 25 |
Cheers, |
| 26 |
|
| 27 |
Dirkjan |
Archives