Re: Scheme is too complicated

[Ian Bicking <ian@bickiia.earlhall.earlham.edu>]

Harvey J. Stein writes:
> You're talking about the common lisp sequence functions now, except
> that you're mixing sequences & dictionaries.
> 
> Sequences    - lists, vectors, strings.
> Dictionaries - hash tables, alists, trees.

Common Lisp thought of everything, didn't it?  If only they knew when
to stop...

I think it is possible to unify sequences and dictionaries.  There's
two ways I can think of that you can view dictionaries as sequences:
(1) associative arrays, (2) unordered lists of (key . value) pairs.

(1) would imply that the sequence functions would act on the values in
the dictionary, possibly recreating the key/new-value association
after the transformation.

(2) would imply that sequence functions would act on the (key . value) 
pair itself.  Just like association lists.

Unfortunately, both those concepts are valid and will give different
results.  I'd vote for (1) being the way
sequence-functions-on-dictionaries are interpreted, with special
functions to give (2).

I don't see any actual reason for association lists except
compatability.  A hash-table should mimic all the association list
behavior.  Unfortunately, if you make sequence functions act like (1)
you won't get this alist like behavior, you'll need seperate functions 
that act on the associations.  Not a big deal, but then hash-tables
wouldn't be drop-in replacements for alists.

The other place where alists are different is being able to construct
them manually, e.g.:

'(('key1 . "value1")
  ('key2 . "value2"))

I think allowing this is a bad thing anyways.  An alist is only
distinguished from a list in concept, not in type.  My recent work
with Tcl (which does this for everything) makes me think this is a bad
idea because it cripples the ability for the language implementation
to be intelligent and undoes some of the advantages of dynamic typing.


<------------------------------------------------------------------->
< Ian Bicking                 |  bickiia@earlham.edu                >
< drawer #419 Earlham College |  http://www.cs.earlham.edu/~bickiia >
< Richmond, IN 47374          |  (765) 973-2824                     >
<------------------------------------------------------------------->