[ruby-talk:9710] Re: 101 Misconceptions About Dynamic Languages

Christoph Rippel <[e-mail]> wrote:
>math-ruby ml - the lag of generics is almost staring at you. In C++ you would
>implement something simple like ZZ/(n) (sorry for the math lingo)
>
>class ComRing {
>// maybe create a top abstract class
>};
>
>template <unsigned int n>
>class ZZ_mod : public virtual ComRing {
>static const unsigned int N = n;
>// implement constructors and arithmetic operations
>// which depend on the class constant N.
>};
>
>try to express this in Ruby - it is going to be a nightmare. There are a lot of
>competent people who consider a lack of support for generics as a language
>deficiency - for example support for generics in java

Here  I  must make  a  comment;  I know  a  lot  more about  programming
mathematics than you do;  it has been one of my  main activities for the
past 25 years.

 The functionality  you want  is to  describe a  mathematical structure,
such as a  ring; the basic idea  is that, as soon  as certain operations
and fields are defined, you can define certain other ones. A template in
C++ expresses perfectly  this, but also a mixin in  Ruby. The only thing
missing in Ruby here is the  ability to overload on argument types; this
was  circumvented by  a hack  around 'coerce'.  Otherwise Ruby  is fully
equivalent in expressiveness to C++.

 Now, there are problems in the actual library and class design in Ruby:
I  think too  many things  are classes  were most  of the  functionality
should have  been put  in mixins:  most operations  on arrays  or hashes
should have been  put in mixins, and  class IO should have  been a mixin
(this came up  recently in a discussion about doing  the perl equivalent
of 'tie'  a hash: if most  of the hash methods  has been in a  mixin the
implementation would have been very easy,  where in contrast now one has
to  copy a  lot of  methods whih  are trivially  implemented in  term of
others). Actually  I think it  is still not too  late to reform  Ruby by
pushing the  functionality in  mixins. As  far as I  can see  this would
change the language in a perfectly upward compatible way.

Best regards,
   Jean MICHEL