On 5/25/07, Robert Dodier <robert.dodier at gmail.com> wrote:
> On 5/24/07, Andrej Vodopivec <andrej.vodopivec at gmail.com> wrote:
>
> > Function names should be suggestive.
>
> Yes, nintegrate certainly is suggestive, isn't it?
> It suggests all kinds of things. You've already said that
> you don't want some of the features other people have mentioned:
>
> > I didn't want to write a function nintegrate which would make
> > whatever it can to get the best numerical result.
>
> which is why I'm asking you to come up with a less-suggestive name
> (or accept other people's ideas about the scope of applicability
> of the function).
The name stands for numeric integratation. It is taken from
Mathematica. When I said that I didn't want it to do whatever it can
to get the best result I meant that it shouldn't use symbolic
integration because it is not a numeric method.
So if nintegrate(sin(x), x, 0, 2*%pi); fails with an error is OK by me
because quad_qags fails for this problem. Other numerical methods
available can be added later, so nintegrate(sin(x), x, 0, 2*%pi),
nint_method=romberg would produce the correct answer (whatever the
name and syntax is).
If we want to first try symbolic integration and if that fails try
numeric integration, I think this should be done in some other way.
This is consistent with Mathematica. NIntegrate[...] only does numeric
methods, N[Integrate[...]] will try symbolic methods and if that fails
it will try numeric methods.
So I still think nintegrate is a good name. Maybe I came out with a
suggestion too soon and it is not clear exactly what I want (maybe I
don't know what exactly I want too), but I did get some good
suggestions.
--
Andrej