Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site pur-ee.UUCP
Path: utzoo!linus!security!genrad!grkermit!masscomp!clyde!floyd!harpo!eagle!mhuxl!ihnp4!inuxc!pur-ee!iuvax!cjl
From: cjl@iuvax.UUCP
Newsgroups: net.lang
Subject: Re: Strong Typing and Ignorance - (nf)
Message-ID: <1228@pur-ee.UUCP>
Date: Thu, 8-Dec-83 18:28:01 EST
Article-I.D.: pur-ee.1228
Posted: Thu Dec  8 18:28:01 1983
Date-Received: Sat, 10-Dec-83 22:15:20 EST
Sender: notes@pur-ee.UUCP
Organization: Electrical Engineering Department , Purdue University
Lines: 23

#R:utcsrgv:-289200:iuvax:11800005:000:933
iuvax!cjl    Dec  8 14:51:00 1983


     I agree it is very hard to  find  an  untyped  language,  so  let
assume  every  existing language is typed. What we concern now is WHEN
and HOW DECENT the type checking is done.

     The definition of static type checking and dynamic type  checking
given  by  Condict describes WHEN the type checking is done, but prob-
ably should not include HOW DECENT it is done.

     We can reserve the strong type checking and weak type checking to
describe  HOW DECENT the type checking is enforced by the language. Of
course, the degree of adequacy is ambiguous. I usually think  the  set
of  complete type checking rules (name compatibility or structure com-
patibility) where the Pascal family enforces is decent. But  the  non-
enforced  type  checking of LISP and the incomplete type checking of C
are examples of weak type checking.

C. Jim Lo, Dept. of Computer and  Info  Sci.,  IU-PU  at  Indianapolis
cjl.Indiana@UDEL-relay