"Richard Hennessy" <rich.hennessy at verizon.net> writes:
> I have been thinking about renaming pwdelta() and diff_pwdelta() in
> pw.mac for a while. I was considering diracdelta() and
> diff_diracdelta() but now I think it would be better to use delta()
> and diff_delta(). If no one objects I will make the name change
> shortly.
>
> Rich
I don't actually use (and, therefore, load) pw.mac myself, so this is
not an issue for me, but:
"delta" is a pretty generic name. As far as I can tell, this change
would mean that any code using your library (or a library that used your
library) would be precluded from using "delta" to mean a function. I
suspect that delta is the most used greek letter in maths, so you may
wish to be cautious about such a change.
There's also the question of separating out the mathematical concept
(distribution on some domain which, when integrated against a test
function, samples at a point) with your specific implementation. For
example, it makes sense to talk about Dirac distributions on C, R^n or
even more exotic measure spaces. I suspect that specifically flagging
the functions as "pwdelta()" and "diff_pwdelta()" makes it clear to a
user exactly what the function is able to do.
Rupert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 315 bytes
Desc: not available
URL: <http://www.math.utexas.edu/pipermail/maxima/attachments/20110930/1a7b990e/attachment.pgp>;