[SciPy-User] Contributing to SciPy was Re: Least-squares fittings with bounds: why is scipy not up to the task?

denis denis-bz-gg@t-online...
Wed Mar 14 11:30:17 CDT 2012


On Mar 14, 3:17 pm, David Warde-Farley <warde...@iro.umontreal.ca>
wrote:
> On 2012-03-14, at 6:12 AM, Scott Sinclair wrote:
> I agree, certain sorts of recipes are a better fit than others. However, it would be nice if we had some clear and simple guidelines as to what belongs and what doesn't rather than making it a matter of subjective judgment; otherwise the only fair way forward seems to be accepting almost everything.

"Has anyone used this recipe in living memory ?"
would be a clear guideline.
(SO etc. track that with member voting, up / down and when.
Is there a simple off-the-shelf voting package that we could use for
recipes ?)
You're right, the tradeoff isn't easy:
accept everything -- hodepodge -- or cut through the jungle.

OT / beyond-topic, I think that each major area
(cluster fft integrate interpolate io ...) should have an owner;
all recipes with no owner go into "old/..." aka "nobodyknows/..."
I'm sure that's been discussed, no volunteers ...

cheers
  -- denis


More information about the SciPy-User mailing list