On Tue, Aug 16, 2005 at 06:45:57PM -0700, bear wrote:
> SRFI-44 is, plainly and simply, a Bad Idea.  Abstraction over
> collection types invites the misuse, or grossly inefficient
> use, of those collection types by inviting people to consider
> them interchangeable when, in effective use, they are not.  To

Actually these approaches (concrete+efficient and abstract+inefficient)
can be easily reconciled by first defining the concrete datatypes and
then collecting (pun?) them under an abstraction layer.  And if what you
state was the only problem with SRFI 44, it would be quite in line with
a language where even strings don't guarantee their most effective usage
patterns :)

The idea for SRFI 44 was probably taken from Common Lisp, the users of
which can undoubtedly tell a lot about its merits...

> I could go on about other reasons why SRFI-44 is a Bad Idea;
> but why?

Well, if I have to write some kind of rationale about why SRFI 44 is
ignored, I would be well advised to write something more elaborate :)
That said, I think SRFI 44 has already taken more than its share of
bashing.

However, it seems to me that the SRFI way of disagreeing is to write a
new SRFI to replace the former one.  Would you consider doing that?
(And would it have any content besides "don't"?)

Panu

--
personal contact: xxxxxx@iki.fi, +35841 5323835, +3589 85619369
work contact: xxxxxx@helsinki.fi, +35850 3678003
kotisivu (henkkoht):	http://www.iki.fi/atehwa/
homepage (technical):	http://sange.fi/~atehwa/