[Python Edinburgh] Hack Session

[Dougal Matthews]

I've got a few idea's that I'd like to throw out there for consideration.

1. Benchmarking
There is a project called djangobench[1] that was started by Jacob
Kaplan-moss. Basically the idea is to monitor the performance of Django as
its versions progress. I can see a couple of ways we could extend this; The
obvious one, write more benchmarks and the other is to create a website that
automatically runs the benchmarks and keeps track overtime to see how the
current trunk is compared to the latest release. This would ideally
automatically run when new commits are made to SVN.

This could be extended further and create a generic benchmark system for
python that would then have a front end etc. but that maybe is a bit too far
for the first pass.

2. Django 1.3
Django 1.3 is coming up soon(ish)[2] - there are loads of bug fixes etc.
that need to be done.

3. Django Bursar
Bursar[3] is a generic payment gateway, its got much of the code there but
its far from a finished project. It's a spin-off from satchmo[4] and thus
implements a number of gateways - they are not all confirmed as working and
mostly untested. It's a great idea and I would like to see it as a more
generic Python library with then a set of django helpers. I've used it in a
project fairly recently so I'm quite familiar with the internals as I had to
dive in and do some bug fixing.



It may well be that we are best to work on a number of projects, for example
we have a set of mentors and then other's that choose to go with one of the
projects or don't really mind. This would help with the diversity as for
example I realise lots of my idea's are very Django or web centric.

Dougal

[1]: http://github.com/jacobian/djangobench
[2]:
http://www.djangoproject.com/weblog/2010/sep/30/django-1_3-release-schedule/
[3]: http://bitbucket.org/bkroeze/django-bursar/overview
[4]: http://www.satchmoproject.com/

On 30 September 2010 23:09, Dougal Matthews <dougal85 at gmail.com> wrote:

> On 30 September 2010 20:00, John Sutherland <john at sneeu.com> wrote:
>
>> I think I suggested PyPy [1] on Tuesday, although I’m not sure how
>> much use that would be to anyone (I certainly haven’t used PyPy).
>>
>
> I haven't used PyPy either but I should really learn more about it. I
> wonder if there is quite a bit of overhead and understanding before we can
> get into it? We may need some expert knowledge.
>
> Another possibility might by a port of Ruby’s chronic library.
>
>
> Chronic looks really nice. I'd like that to have that thanks!
>
> Cheers,
> Dougal
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/edinburgh/attachments/20101005/a7ea49f7/attachment.html>