| Marcin 'Qrczak' Kowalczyk 2004-10-23, 8:56 am |
| mikel <mikel@evins.net> writes:
> The normal mode of operation is to build things the easiest (and
> perhaps slowest) way, then profile the code and substitute different
> (faster) data structures. Common Lisp makes it especially easy to
> substitute different data structures without perturbing the structure
> of your program (or, in fact, even restarting it!).
I would disagree. Operations would have to be more polymorphic.
For example when you change between alist, plist, hash table,
(in case keys are integers from a small range) vector, or your
own representation (e.g. using a hierarchy of vectors) - you have
to update operations in the code.
Unless you introduced your own abstractions in the first place,
but this you can do in any language.
--
__("< Marcin Kowalczyk
\__/ qrczak@knm.org.pl
^^ http://qrnik.knm.org.pl/~qrczak/
|