Discussion: new operators for numerical computation

[Warren B. Focke]

(Huaiyu Zhu) said:
>On Thu, 20 Jul 2000 00:58:06 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:
>>
>>>    (2) Which is matrixwise or elementwise
>>
>>I can't see a good answer here. NumPy has been around for a long time
>>and I believe it has a fairly large user base which is a strong
>>argument in favor of the ".x" operators being matrixwise. However,
>
>We may need also to consider future userbase.

If we are considering future user base, it seems to be a question of
what will produce the least surprise.  Since whatever we come up with
can't mimic Matlab exactly, I think that preserving compatibility
should be deemphasized, and we should try to make it good rather than
compatible.  I do agree that having .* and * and having them mean the
exact opposite of what they do in Matlab is bad, though.

But I think also that we should consider not just user base, but code
base - how much code is already out there that will break under the
various suggestions.  My impression is that MatPy is still in a
development phase, such that people shouldn't expect everything to be
written in stone yet, and that people are, at the moment, more
experimenting with it that using it; while there is a good deal of
NumPy code out there in production use which would break under a
change in the meaning of the current operators.  Although, at least,
it should be possible to convert it all with sed or something similar.
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University