wow, lots of e-mails I need to read and answer. Thank you!!!

On Tue, Nov 10, 2009 at 9:12 AM, David Mertens <[email protected]> wrote:
> So you want a feature comparison?  Obviously it would be nice if we had a
> chart comparing Matlab and PDL features.  However, it becomes a murky
> discussion, because at some point Matlab starts tackling problems that are
> clearly outside of PDL's problem-domain, like symbolic math (but see Steffen
> Müller's Math::Symbolic package).  On the other hand, Perl is a general
> systems programming language and has WAY more features than Matlab, and at
> which point do you stop?  We could go down CPAN and compile a never-ending
> feature list that Matlab couldn't match.

Yes and the rest of your mail can be a good base for such comparison.
When evaluating an alternative to Matlab companies will primarily look at the
feature set they are familiar with from Matlab an want to know how X or Y
are matching those features.

So if we would like to convince companies to use Perl and PDL we need
to convince
the people that we have a solution for their problems and that we can match the
capabilities of the competition. Not only that of Matlab but also of Numpy.

One of the ways people do this is that they write down a set of requirements
and then create a table trying to figure out how each solution can fill the
various requirements. For example this is what I got from one of the companies
when they were comparing Subversion with Clearcase:
http://www.pti.co.il/clearcase_vs_subversion.html

We don't know what are the requirements of each potential user of Matlab vs PDL
so the best we can do is to create a table comparing the feature of our solution
with that of the competition. Here I am writing "our solution" as I
already understand
it is not PDL vs Matlab but
Perl+PDL+CPAN+Padre+other things vs
Matlab vs
Python + Numpy + other things

It is also especially good to have a mapping of Matlab features to
"our features" as
it will make it easier for people who are familiar with the Matlab way
of thinking and the
terms Matlab uses.

We can of course include items that are not easy to do with Matlab and
point out that how
easy they can be done with our solution but better restrict it to
fields that are likely interesting
to the potential users.


Gabor

_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to