Tue, 11 May 2010 00:23:52 -0700, T J wrote: [clip]
It seems reasonable that we might have to follow the deprecation route, but I'd be happier with a "faster" fix.
1.5 - Provide numpy.random.lomax. Make numpy.random.pareto raise a DeprecationWarning and then call lomax.
2.0 (if there is no 1.6) - Make numpy.random.pareto behave as Pareto distribution of 1st kind.
I think the next Numpy release will be 2.0. How things were done with changes in the histogram function were: 1) Add a "new=False" keyword argument, and raise a DeprecationWarning if new==False. The user then must call it with "pareto(..., new=True)" to get the correct behaviour. 2) In the next release, change the default to "new=True". Another option would be to add a correct implementation with a different name, e.g. `pareto1` to signal it's the first kind, and deprecate the old function altogether. A third option would be just to silently fix the bug. In any case the change should be mentioned noticeably in the release notes. -- Pauli Virtanen