On Mon, Dec 30, 2013 at 1:04 PM, William Stein <[email protected]> wrote:

> On Mon, Dec 30, 2013 at 12:47 PM, Nils Bruin <[email protected]> wrote:
> > On Monday, December 30, 2013 12:35:37 PM UTC-8, R. Andrew Ohana wrote:
> >>
> >> Moreover, I don't think we have to force the switch, it should be
> >> perfectly possible to support both python 2 and python 3 for a period of
> >> time.
> >
> >
> > What's the benefit of that? The sage process itself will be running on
> one
> > python process. We can't be hybrid with that. So either you make the
> whole
> > sage library Py3-capable or not. We could ship both Py2 and Py3 so that
> we
> > can run some scripts under Py3 and the sage process under Py2 (or vice
> > versa), but having two pythons seems a lot of admin for very little
> benefit.
> > I could see a development version that for a while has both to aid
> debugging
> > adapting the sage library, but outside of that I don't see a benefit of
> > having both.
>
> The benefit is that we could switch to python3 by default; however, if
> somebody runs into a show-stopper issue, we can tell them: "heh, just
> do 'sage --py2' to start your copy of Sage up using python2 instead".
>  It directly addresses the main argument that I made against switching
> above.  It sounds technically difficult, but it is certainly possible.
>

This would certainly be very nice, but

Alternatively, we could have a build flag that determines whether
> everything gets built with py3 or py2.
>

this would probably be much easier.

It all sounds like a lot of extra pain and work, but it is possible
> and addresses the main argument against switching.
>

I don't think it is that much work to maintain, as for most python code
there isn't that large of a difference between python 2 and 3. There will
of course be some trickier bits, but how we deal with most of these issues
should occur during the initial porting process.


> Your main argument against this is "for very little benefit" --
> however, given that only 2% of download from PyPi are for py3, and in
> light of the discussions I linked to, evidence suggests there's
> potentially substantial real benefit to continuing to support the
> python2 ecosystem, at least for now.
>
>  -- William
>
>
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "sage-devel" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to [email protected].
> > To post to this group, send email to [email protected].
> > Visit this group at http://groups.google.com/group/sage-devel.
> > For more options, visit https://groups.google.com/groups/opt_out.
>
>
>
> --
> William Stein
> Professor of Mathematics
> University of Washington
> http://wstein.org
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to [email protected].
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/groups/opt_out.
>



-- 
Andrew

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to