Fortran


Fortran is a third-generation, compiled, imperative programming language designed for numeric computation and scientific computing.
Fortran was originally developed by IBM with a reference manual being released in 1956; however, the first compilers only began to produce accurate code two years later. Fortran computer programs have been written to support scientific and engineering applications, such as numerical weather prediction, finite element analysis, computational fluid dynamics, plasma physics, geophysics, computational physics, crystallography and computational chemistry. It is a popular language for high-performance computing and is used for programs that benchmark and rank the world's fastest supercomputers.
Fortran has evolved through numerous versions and dialects. In 1966, the American National Standards Institute developed a standard for Fortran to limit proliferation of compilers using slightly different syntax. Successive versions have added support for a character data type, structured programming, array programming, modular programming, generic programming, parallel computing, object-oriented programming, and concurrent programming.
Since April 2024, Fortran has ranked among the top ten languages in the TIOBE index, a measure of the popularity of programming languages.

Naming

The first manual for FORTRAN describes it as a Formula Translating System, and printed the name with small caps,. Other sources suggest the name stands for Formula Translator, or Formula Translation.
Early IBM computers did not support lowercase letters, and the names of versions of the language through FORTRAN 77 were usually spelled in all-uppercase. FORTRAN 77 was the last version in which the Fortran character set included only uppercase letters.
The official language standards for Fortran have referred to the language as "Fortran" with initial caps since Fortran 90.

Origins

In late 1953, John W. Backus submitted a proposal to his superiors at IBM to develop a more practical alternative to assembly language for programming their IBM 704 mainframe computer. Backus' historic FORTRAN team consisted of programmers Richard Goldberg, Sheldon F. Best, Harlan Herrick, Peter Sheridan, Roy Nutt, Robert Nelson, Irving Ziller, Harold Stern, Lois Haibt, and David Sayre. Its concepts included easier entry of equations into a computer, an idea developed by J. Halcombe Laning and demonstrated in the Laning and Zierler system of 1952.
A draft specification for The IBM Mathematical Formula Translating System was completed by November 1954. The first manual for FORTRAN appeared in October 1956, with the first FORTRAN compiler delivered in April 1957. Fortran produced efficient enough code for assembly language programmers to accept a high-level programming language replacement.
John Backus said during a 1979 interview with Think, the IBM employee magazine, "Much of my work has come from being lazy. I didn't like writing programs, and so, when I was working on the IBM 701, writing programs for computing missile trajectories, I started work on a programming system to make it easier to write programs."
The language was widely adopted by scientists for writing numerically intensive programs, which encouraged compiler writers to produce compilers that could generate faster and more efficient code. The inclusion of a complex number data type in the language made Fortran especially suited to technical applications such as electrical engineering.
By 1960, versions of FORTRAN were available for the IBM 709, 650, 1620, and 7090 computers. Significantly, the increasing popularity of FORTRAN spurred competing computer manufacturers to provide FORTRAN compilers for their machines, so that by 1963 over 40 FORTRAN compilers existed.
FORTRAN was provided for the IBM 1401 computer by an innovative 63-phase compiler that ran entirely in its core memory of only 8000 characters. The compiler could be run from tape, or from a 2200-card deck; it used no further tape or disk storage. It kept the program in memory and loaded overlays that gradually transformed it, in place, into executable form, as described by Haines.
This article was reprinted, edited, in both editions of Anatomy of a Compiler and in the IBM manual "Fortran Specifications and Operating Procedures, IBM 1401". The executable form was not entirely machine language; rather, floating-point arithmetic, sub-scripting, input/output, and function references were interpreted, preceding UCSD Pascal P-code by two decades. GOTRAN, a simplified, interpreted version of FORTRAN I for "load and go" operation was available. Modern Fortran, and almost all later versions, are fully compiled, as done for other high-performance languages.
The development of Fortran paralleled the early evolution of compiler technology, and many advances in the theory and design of compilers were specifically motivated by the need to generate efficient code for Fortran programs.

FORTRAN

The initial release of FORTRAN for the IBM 704 contained 32 types of statements, including:
  • and statements
  • Assignment statements
  • Three-way arithmetic statement, which passed control to one of three locations in the program depending on whether the result of the arithmetic expression was negative, zero, or positive
  • Control statements for checking exceptions ; and control statements for manipulating sense switches and sense lights
  • goto|, computed,, and assigned
  • loops
  • Formatted I/O:,,,,, and
  • Unformatted I/O:,,, and
  • Other I/O:,, and
  • ,, and
  • statement.
The arithmetic statement was reminiscent of a three-way comparison instruction available on the 704. The statement provided the only way to compare numbers—by testing their difference, with an attendant risk of overflow. This deficiency was later overcome by "logical" facilities introduced in FORTRAN IV.
The statement was used originally to give branch probabilities for the three branch cases of the arithmetic statement. It could also be used to suggest how many iterations a loop might run. The first FORTRAN compiler used this weighting to perform at compile time a Monte Carlo simulation of the generated code, the results of which were used to optimize the placement of basic blocks in memory—a very sophisticated optimization for its time. The Monte Carlo technique is documented in Backus et al.'s paper on this original implementation, The FORTRAN Automatic Coding System:
The fundamental unit of program is the basic block; a basic block is a stretch of program which has one entry point and one exit point. The purpose of section 4 is to prepare for section 5 a table of predecessors which enumerates the basic blocks and lists for every basic block each of the basic blocks which can be its immediate predecessor in flow, together with the absolute frequency of each such basic block link. This table is obtained by running the program once in Monte-Carlo fashion, in which the outcome of conditional transfers arising out of IF-type statements and computed GO TO's is determined by a random number generator suitably weighted according to whatever FREQUENCY statements have been provided.

The first FORTRAN compiler reported diagnostic information by halting the program when an error was found and outputting an error code on its console. That code could be looked up by the programmer in an error messages table in the operator's manual, providing them with a brief description of the problem. Later, an error-handling subroutine to handle user errors such as division by zero, developed by NASA, was incorporated, informing users of which line of code contained the error.

Fixed layout and punched cards

Before the development of disk files, text editors and terminals, programs were most often entered on a keypunch keyboard onto 80-column punched cards, one line to a card. The resulting deck of cards would be fed into a card reader to be compiled. Punched card codes included no lower-case letters or many special characters, and special versions of the IBM 026 keypunch were offered that would correctly print the re-purposed special characters used in FORTRAN.
Reflecting punched card input practice, Fortran programs were originally written in a fixed-column format. A letter "C" in column 1 caused the entire card to be treated as a comment and ignored by the compiler. Otherwise, the columns of the card were divided into four fields:
  • 1 to 5 were the label field: a sequence of digits here was taken as a label for use in DO or control statements such as GO TO and IF, or to identify a FORMAT statement referred to in a WRITE or READ statement. Leading zeros are ignored and 0 is not a valid label number.
  • 6 was a continuation field: a character other than a blank or a zero here caused the card to be taken as a continuation of the statement on the prior card. The continuation cards were usually numbered 1, 2, etc. and the starting card might therefore have zero in its continuation column—which is not a continuation of its preceding card.
  • 7 to 72 contained the statement field.
  • 73 to 80 were ignored.
Columns 73 to 80 could therefore be used for identification information, such as punching a sequence number or text, which could be used to re-order cards if a stack of cards was dropped; though in practice this was reserved for stable, production programs. An IBM 519 could be used to copy a program deck and add sequence numbers. Some early compilers, e.g., the IBM 650's, had additional restrictions due to limitations on their card readers. Keypunches could be programmed to tab to column 7 and skip out after column 72. Later compilers relaxed most fixed-format restrictions, and the requirement was eliminated in the Fortran 90 standard.
Within the statement field, whitespace characters were ignored outside a text literal. This allowed omitting spaces between tokens for brevity or including spaces within identifiers for clarity. For example, was a valid identifier, equivalent to, and 101010DO101I=1,101 was a valid statement, equivalent to 10101 DO 101 I = 1, 101 because the zero in column 6 is treated as if it were a space, while 101010DO101I=1.101 was instead 10101 DO101I = 1.101, the assignment of 1.101 to a variable called DO101I. Note the slight visual difference between a comma and a period.
Hollerith strings, originally allowed only in FORMAT and DATA statements, were prefixed by a character count and the letter H, allowing blanks to be retained within the character string. Miscounts were a problem.