[Numpy-discussion] future directions

[Neil Martinsen-Burrell]

On 2009-08-27 19:56 , David Goldsmith wrote:
> --- On Thu, 8/27/09, Fons Adriaensen<fons@kokkinizita.net>  wrote:

[...]

>> 3. Finally remove all the redundancy and legacy stuff from the
>> world of numerical Python. It is *very* confusing to a new user.
>
> I like this also (but I also know that actually trying to achieve it
> would ruffle a lot of feathers).

I think that feather ruffling is *not* the problem with this change. 
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.  Numpy enjoys some of its ongoing popularity among active 
scientists because of its stability and the ease of transition forward 
from Numeric.  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.

I think that as we go forward it is important to balance not confusing 
new users (a problem that can be addressed with better documentation and 
pointing people to modern ways of doing things) with not alienating 
existing users (who are in some cases influential in recruiting those 
new users in the first place).  For software developers, 
compatibility-breaking changes seem like they call for just a few small 
tweaks to the code.  For scientists who work with software, those same 
changes may call for never choosing Numpy again in the future.

I think that this is a balance that we should be aware of when 
introducing changes.  It makes sense that we will all see this balance 
differently, but I think that we need to acknowledge that this is the 
essential tension in removing cruft incompatibly.

-Neil