Historic Documents in Computer Science
collected by Karl Kleine
FORTRAN II General Information Manual and
IBM 7090/7094 Programming Systems: FORTRAN II Programming
are two IBM manuals of 1963 describing the FORTRAN II language.
IBM 7090/7094 Programming Systems: FORTRAN IV Language, 1963, and
IBM System 360 and System 370 FORTRAN IV Language, 1974.
FORTRAN IV was next to the ANSI standard of 1966 for long time the reference language for Fortran as used by legions of scientists and engineers.
But what you should really study is the Revised Report on the Algorithmic Language Algol 60, here in a rendition of the Algol Manual der ALCOR Gruppe, together with a report for a subset of the language and the describtion of basic input/output procedures. There is also an online HTML version of the revised report.
Note: The IFIP working group WG2.1 on Algol decided early on that they will keep the copyright and that the Algol 60 report shall be free to copy and distribute, independent of some publisher. That was in 1960, long before any word of open source was voiced. It was just proper scientific procedure for such a document. If only standard bodies of today would show such an attitude!
The first usable Algol 68 system was a subset by the british RRE (Royal Radar Establishment, Malvern), called Algol68-R for the ICL1900 and ICL2900 series machines.
Algol 68 Revised Report reformatted and crossreferenced by W. B. Kloke. There is also an official sublanguage of the revised Algol 68 by Peter Hibbard, and an official report on the hardware representation, which makes good reading for the problems of that time with available peripheral equipment and the use of stropping conventions to indicate the kind of symbols in the program text.
This was further helped by Pascal-S, a subset for running student jobs in a quick turn-around batch system. This report contains the whole compiler, written in Pascal, of course.
Whereas Wirth's Pascal was a practical compromise for teaching structured programming of reasonably small programs in a batch environment, a number of derivates appeared rather soon, often very specific to the needs of the various operating environment, and loaded with features, the best known being Turbo-Pascal by Borland. As a reaction to that, there are two ISO standards on Pascal, ISO/IEC 7185:1990 Pascal, and ISO/IEC 10206:1990 Extended Pascal.
Any discussion of Pascal would however be incomplete without Brian Kernighan's critique Why Pascal is not my favorite programming language.
A team at Digital and Olivetti defined a worthy successor: Modula-3 Report (Revised), DEC SRC, november 89, but as usually the case, the actual use and feel of a language is largely determined by available libraries: Some Useful Modula-3 Interfaces, DEC SRC, december 93. The language evolved further and is described on a web page and a pdf snapshot thereof. More recent stuff is found on www.m3.org.
After PL360 there have been a number of similar languages, mostly forgotten today.
Programming and Software Systems
Metcalfe's famous sketch may not be missing here. And then there is of course the joint ethernet specification by Digital, Intel, and Xerox (V1 of Sept. 1980, and V2 of Nov. 1982). Later ethernet specs are IEEE standards in the 802 series.