Re: A Logical Model for Lists as Relations
From: Marshall Spight <marshall.spight_at_gmail.com>
Date: 12 May 2006 10:18:41 -0700
Message-ID: <1147454321.896930.245120_at_u72g2000cwu.googlegroups.com>
> If there's a point to all this, here it is: use sets where sets are
> useful, and use lists where lists are useful. Know the difference.
Date: 12 May 2006 10:18:41 -0700
Message-ID: <1147454321.896930.245120_at_u72g2000cwu.googlegroups.com>
David Cressey wrote:
>
> If there's a point to all this, here it is: use sets where sets are
> useful, and use lists where lists are useful. Know the difference.
I think this is an excellent point.
I note, in a kind of Sapir-Worf kind of way, that our programming languages shape the way we think about data structures. Given that programming languages don't do a good job of the above, it's not surprising that many programmers don't do a good job of the above.
I believe that a programming language with first class support for relations and lists would go a long way towards illuminating the distinction between ordered and unordered data. I would consider it idea if this could be done from a set-theoretic basis. (Hence the thread.)
Marshall Received on Fri May 12 2006 - 19:18:41 CEST