[Numpy-discussion] future directions

[Alan G Isaac]

> Neil Martinsen-Burrell skrev:
>> The persistence of the idea that removing Numpy's legacy features will 
>> only be annoyance is inimical to the popularity of the whole Numpy 
>> project.  [...] Once scientists have working codes it is more than an 
>> annoyance to have to change those codes.  In some cases, it may be the 
>> motivation for people to use other software packages.


On 8/28/2009 2:28 AM Johan Grönqvist apparently wrote:
> I very much agree


I'm just a user but I've read the NumPy and SciPy lists
for years. Although this idea keeps resurfacing, I do
not have the sense that there is anything close to
developer unanimity that this is a good idea.  Note
that over the years proposals to *add* functionality
to NumPy keeps resurfacing too, with similar healthy
inertia.

I will speculate that the outcome will eventually be that
something like ndarray will become part of the
standard library, satisfying users who just want that,
and that NumPy will continue to provide its current
(and therefore legacy) functionality. That is just
speculation by a user, but the new buffer interface
does seem to lay the groundwork.

Alan Isaac