Return-Path: Delivery-Date: Received: from leopard.cs.byu.edu (no rfc931) by swan.cl.cam.ac.uk with SMTP (PP-6.5) outside ac.uk; Thu, 18 Aug 1994 16:06:33 +0100 Received: by leopard.cs.byu.edu (1.37.109.8/16.2) id AA29155; Thu, 18 Aug 1994 09:02:48 -0600 Sender: info-hol-request@lal.cs.byu.edu Errors-To: info-hol-request@lal.cs.byu.edu Precedence: bulk Received: from vanuata.dcs.gla.ac.uk by leopard.cs.byu.edu with SMTP (1.37.109.8/16.2) id AA29151; Thu, 18 Aug 1994 09:02:47 -0600 Received: from switha.dcs.gla.ac.uk by goggins.dcs.gla.ac.uk with LOCAL SMTP (PP) id <13554-0@goggins.dcs.gla.ac.uk>; Thu, 18 Aug 1994 15:58:16 +0100 Received: by switha.dcs.gla.ac.uk (4.1/Dumb) id AA13030; Thu, 18 Aug 94 15:58:06 BST From: tfm@dcs.gla.ac.uk Message-Id: <9408181458.AA13030@switha.dcs.gla.ac.uk> To: Matthew Morley Cc: info-hol@leopard.cs.byu.edu, tfm@dcs.gla.ac.uk Subject: Re: Where do type constructors come from? In-Reply-To: Your message of "Thu, 18 Aug 94 16:38:29 +0100." <199408181438.AA02884@borneo.gmd.de> Date: Thu, 18 Aug 94 15:58:06 +0100 > So HOL's type language is enriched quite a bit with > (t1,...tn)op constructors. HOL is a much more powerful type theory > than I hitherto imagined -- this I need to digest. Meanwhile, whence > came (t1,...tn)op? Historically, these came from LCF's logic PP\lambda. Type operators in this logic corresponded domain operations (separated sum, cartesian product, continuous function space). The earliest reference I have (on my shelf) is R. Milner, L. Morris, and M. Newey, A Logic for Computable Functions With Reflexive and Polymorphic Types, LCF Report No. 1, Department of Computer Science, University of Edinburgh (January 1975). Tom