Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!mcvax!cernvax!ethz!heiser From: heiser@ethz.UUCP (Gernot Heiser) Newsgroups: comp.lang.modula2 Subject: Re: Sets Message-ID: <213@bernina.UUCP> Date: Wed, 14-Oct-87 10:41:44 EDT Article-I.D.: bernina.213 Posted: Wed Oct 14 10:41:44 1987 Date-Received: Sun, 18-Oct-87 21:36:32 EDT References: Reply-To: heiser@bernina.UUCP (Gernot Heiser) Organization: ETH Zuerich, Switzerland Lines: 16 In article Info-Modula2 Distribution List writes: >Wirth created the idea of a bitset because it would be easy to implement >a very efficient way to do sets the size of the computer's word. This >means sets are likely to be 8, 16 or 32 bits, possibly other sizes, >depending on the implementation. I rather believe that this was done to support bit-fiddeling in Modula-2, since BITSETs aren't really good for anything else. None of Wirth's compilers support sets bigger than BITSETs. In this sense, a Modula-2 set is not an abstract data type but an image of a machine word. The question is whether they ought to be called "sets". -- Gernot Heiser Phone: +41 1/256 23 48 Institute for Integrated Systems CSNET/ARPA: heiser%ifi.ethz.ch@relay.cs.net ETH Zuerich EARN/BITNET: GRIDFILE@CZHETH5A CH-8092 Zuerich, Switzerland EUNET/UUCP: {uunet,...}!mcvax!ethz!heiser