<- People Behind Informatics


All 0-9 A B C D E F G H I J K L M N O P Q R S T U V W XY Z




 
Algol 60
The driving concern in the development of the programming language Algol was to establish a notation for programs that would be a suitable carrier of algorithms and programs among computers of different types and capabilities. Algol was the result of a collaboration of American and European committees. When the work was initiated in 1956, the computing scene in Europe was still dominated by one-of-a-kind computers, while in the USA the most commonly used computers were from manufacturers' standard series. The problems of designing adequate programs were becoming acute everywhere. One approach to overcoming these problems was to replace the cumbersome machine languages by more convenient notations. The selection of such notations was guided by the applications of computers in science and engineering that dominated the early years of their invention, namely, solving the problems of mathematical analysis by techniques developed in numerical analysis. Thus programming notations came to be inspired by the algebraic notations that had a long tradition in mathematical analysis. Experiments with using such notations for programs were being made in many places. The idea of establishing a programming notation that would be equally suitable in programming computers of different makes and types and for exchange of programs was first proposed in Europe in 1955, and became the subject of a working group of GAMM (Gesellschaft für angewandte Mathematik und Mechanik). In 1957 this group joined a similar effort conducted within ACM. In June 1958 a GAMM-ACM working conference in Zurich worked out a proposal, later known as the Zürich Report or Algol 58, for an algebraic programming language designed for use in programming a variety of computers, as well as for describing algorithms in publications. The Zürich Report gave rise to intense public discussion. As a result, a 13-person committee met in Paris in January 1960, where a new version of the language, Algol 60, was discussed and agreed upon. Some minor corrections to Algol 60 were published in the Revised Report on the Algorithmic Language Algol 60 in 1962. The most prominent characteristics of Algol-60 is the compact and precise definition of the language in the Algol report. The most important features of the language are: • General expressions • A notation similar to the usual mathematical notation • Block structure • Scoping rules • Modes of parameter passing (by value and by name) • Recursion • The type Boolean • A formally specified syntax An interesting battle is going on up to the current days in some notational issues. In Fortran the equal sign (=) is used to express assignment. This has the awkward consequence that the mathematically absurd expression x=x+1 is a valid statement in Fortran, meaning: “take the current value of x, increment it by 1 and assign the result to x”. This problem was cured by Algol introducing the special := double-symbol for assignment. The above statement is written in Algol as x:=x+1, which is unusual – intentionally –, but not absurd. Algol uses the = symbol to check whether two values are equal (e.g. if x=y then …). Irony of the history is that in the programming language C the old Fortran convention was used (for whatever reasons) and thus, all the currently broadly used C-based languages, such as C++, Java, C# etc. still apply the old, awkward usage of =, and use the strange double symbol == for equality. [Source: Encyclopedia of Computer Science, Fourth Edition, Anthony Ralston, Edwin D. Reilly, David Hemmendinger; Laszlo Böszörmenyi: Notes to the Virtual Exhibition "People behind Informatics"]
 

 

<- People Behind Informatics


Home  |  Top  |  Search  |  Gallery  | Glossary  | Sitemap  |  Making Of  |  Help