Printer Friendly

Perspective on recent software developments: a critical review of the present software situation with emphasis on contemporary programs and systems.

THE most obvious thought that springs from a scan of software releases in the last few months is a rather ungracious one: the absence in the U.K. of an RPG program for the Digital Equipment Co.'s range of PDP-8 and PDP-9 processors. In the U.S., Codon Corporation, Waltham, Mass., have an RPG that runs on the two small PDPs and a full blown Cobol for the PDP-10. In this matter the U.S. and U.K. roles seem to be inverted as in the past ten years the U.K. has made a speciality of putting machines with a scientific background to work in commercial computing. After burning their fingers so many times, the U.K. market seems to have become over-cautious and it now appears that the U.S. is ahead in cost cutting software for companies able to make unusual hardware decisions.

The mainstream of U.K. software releases has divided into two broad categories: the first aiming at packages and filling software gaps for ICL and IBM commonly available configurations, and the other aimed at extending remote computing facilities from computer service industry firms, such as SLA Ltd. and Time Sharing Ltd.

The coming months may well see a conjunction of the two categories quoted, as the 360 software monolith is now able to support new directions in remote computing services. Not only have the worst inadequacies of the 360 operating systems been removed in the last few months (for instance, the ASP--attached support processor system--is no longer clogged by being unaware of faults in input/output devices and is also able to link with i800-based data acquisition systems through shared files) but a combination of hardware and software releases enhances a number of marketing activities with potential in the service bureau area.

The hardware units cover a spectrum of direct access storage devices and terminal computing units. The software releases enable two IBM-based languages, PL/i and APL, to make inroads in the bureaux based on rival hardware. Language conversion programs are now available for Fortran, Algol and Cobol into PL/1. Programs of this type are definitely not for use where the applications program makes a big demand on machine facilities, or where the real benefits of PL/i (if any) could only be obtained by restructuring the application to exploit facilities exclusive to PL/1, but--and it is a big "but"--they provide a much-needed incentive for users heavily commited to existing languages to try out a scheme for a full-scale switch to PL/I.

APL has become in the U.S. one of the most favoured languages for developing interactive computer systems. Its abilities in this field were demonstrated in fair detail at Datafair 69 by four members of the University of Saskatchewan, P. Gross, A. Cropley, B. Hebb and R. Palmer, who showed how APL functions stored on disc can be accessed by other users in a time-shared APL system which can use inexpensive terminals.

APL has a double attraction for IBM 1130 users, in that it can be supported by the 1130 on a stand-alone basis, or by the 1130 as a terminal to a 360 complex. In the U.K. SIA Limited, among other bureaux, have arranged for their telecommunications facilities to be backed with software, which enables the 130s and other similar small computers to operate as remote terminals to the central CDC 6600. In addition, SIA are concerned with installing CDC-200 user terminals incorporating a card reader and line printer as well as a video display and keyboard in a number of remote situations. By itself, the SIA software which handles communications facilities so that remote terminals devices can be used for direct interfacing with the 6600 operating system would be of little value if the development of job control languages, such as CLEAT had not provided some foundation for SIA's claim to be developing 'user oriented software'.

CLEAT is a combination of a 'front-end' language to a complex suite of Fortran programs which combines job control aspects of program execution with the ability to interface a variety of data files as input and output to the running Fortran system. For reasons best known to SIA, CLEAT has been described as a language for engineers and technologists, whereas it is in fact much more of a language for computer applications programmers who do not wish to meddle over-much with the mechanics of a large Fortran system. The first of CLEAT's derivatives in this area is a subsystem for linear analysis and design of structures, acronymed LADS. An assessment of LADS should come from the structural engineering community, but the CLEAT'S approach has been given public prominence by the Ministry of Public Building and Works awarding a contract to Wilcock, Shearing and Partners (who created CLEAT) for GENESYS. GENESYS seems to be intended as a multi-Fortran operating system fora widely available set of machines with similar design objectives to CLEAT, (which is specific to the CDC-6600).

The biggest remaining weakness in putting the 360 software into the hands of remote computing users is the quite amazing incoherence and complexity of IBM's standard job control language. It is noteworthy that almost all of the IBM machine based US bureaux in this field have devised simplified job control software. It will be particularly interesting to see whether interactive PL/I (due next year) with its inbuilt data structures will have more customer appeal than interpretive CLEAT-type systems exploiting the computing capability of APL.

The original text from Vol. No.1 1969
COPYRIGHT 2009 A.P. Publications Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:OPINION
Publication:Software World
Date:Mar 1, 2009
Words:923
Previous Article:ICL/GPO link to demonstrate system 4 communications software.
Next Article:Everyone a computer specialist: a case for the development of user orientated autocode languages using the notation and terminology of the type of...
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |