Quaestiones Informaticae Vol 1 No 3
Permanent URI for this collection
Browse
Recent Submissions
Item The Memory Organization of Future Large Processors(Computer Society of South Africa (on behalf of SAICSIT), 1980) Stein, David M.In this paper some specifics and some generalities on the subject of memory organization for future large processors are discussed. We briefly review the current hierarchy and discuss technological and economic changes which are expected to occur in the near future. Using simple models, we show how the impact of these changes will be felt and explain that the memory hierarchy will become an increasingly importarit part of future computers. We show that a need is developing for understanding the memory organization in a "Systems" sense, and we draw upon theories of organization and hierarchies to elaborate upon the issues and to gain insight.Item Thirty Years of Information Engines(Computer Society of South Africa (on behalf of SAICSIT), 1980) Scarrott, G.G.The era ofinformation engineering was initiated some thirty years' ago by the demonstration of the first stored program electronic computer. At that time there was intense innovative excitement among the pioneers which, however, became somewhat jaded in the subsequent decades as computers became a business and the business acquired its doctrinaire echelons of orthodox systems experts. · Mo~e re~ently, the.e~citement has begun to return as the triumphant progress of LSI technology has reopened the frontiers of systems engmeenng so that 1t 1s now appropnate to review the evolution of computers with the object of gaining an appreciation of where we are now by what route we arrived, and what can be expected to happen next. The paper includes brief reference to research into new system~ concepts, relevant to such a forecast that are now moving into the market place.Item High-level Programming Language for Interactive Lisp-like Languages(Computer Society of South Africa (on behalf of SAICSIT), 1980) Postma, Stef W.The language LISP 1.5 is neither easy to write nor to read. In the McCarthy etal. definitive manual on LISP 1.5 a higher level language, MLISP, is used From the early 1960's various people have proposed higher level languages for LISP. Two types oflanguages may be distinguished: those based on a LISP programming style, but more readable, and those based on an Algol style. The language PLISP is a language in which it is attempted to reconcile the two approaches to high level languages for LISP. The notation is based on the Algolstyle languages but is brief and adapted for terminal use. PLISP allows full access to all the LISP features in a consistent notation. In addition there are features that will appeal to the mathematically mature programmer: full infix notation for not only the assignment, arithmetic, and logical operators, but also for lists, sets, and chained relations. PLISP also has a consistent notation for the first order predicate logic and a variety of ways in which functions may be defined. Top-down program development is facilitated by the use ofan assertional language FEA which can be translated to PLISP for test-data generation. Furthermore a definite notation for function stubs is made available - the stubs yield values although the functions called have not been defined at that stage.