[SciPy-dev] numpy.random.randint

Alan G Isaac aisaac at american.edu
Fri Mar 10 13:05:02 CST 2006

On Fri, 10 Mar 2006, Robert Kern apparently wrote: 
> The keyword in the functions does not necessarily match 
> with the attribute names on the objects. I don't find this 
> particularly shocking. Or bad. But that's just my tastes. 

> So I'm -1 on changing numpy.random and all of scipy.stats. 

I'll make a last comment and then let this go.
The point you make above seems fine as an abstract point.
But I think in this concrete case, where *all* we are doing 
is generating arrays with core attributes that are well 
known, the current size keyword and lack of shape keyword 
are quite "surprising" and bad for that reason.

I.e., will the current design produce questions from and 
confusions for new users?


More information about the Scipy-dev mailing list