FORDESK(ID:5832/)Interactive FORTRAN IIInteractive FORTRAN II developed at the University of Western Australia, featuring extensions to permit graphic tablet input (also refer to FORDESK I) Related languages
References: […]two interpretive conversational languages for the PDP-6 which fulfill the conversational on-line desk calculator aims of FORDESK. These are a version of advanced BASIC, the language developed by Dartmouth College and used by the G.E. time-sharing systems, and a language called AID which is based on the JOSS language developed by the RAND Corporation (J. C. Shaw, 1964). COMPARISONS WITH BASIC AND AID BASIC is designed for business and educational problem solving applications as well as for scientific work, while AID is designed chiefly for scientific and engineering problem solving By comparison FORDESK falls between the two, being better suited for most scientific work than BASIC and better suited for business type applications than AID. Also, as it is hoped eventually to implement all of FORTRAN IV, features such as double precision or complex arithmetic that exist in neither BASIC nor AID will become available in FORDESK. FORTRAN is a harder language to learn than either BASIC or AID as it is further removed from normal English syntax It is interesting to note though, that current FORDESK I users do not seem inclined to spend the time to learn either BASIC or AID, preferring to use FORDESK, with which they are familiar. In basic design all three languages are much the same, having arithmetic, jump, test and loop instructions but each has some features not found in the others. BASIC has a powerful set of matrix operations which allow matrices to be manipulated in a similar fashion to scalar variables and also some powerful character manipulation instructions AID has a set of functions PROD, SUM, MAX and MIN whose arguments can be specified in an iterative clause similar to mathematical notation but which a FORTRAN programmer would have to translate into a DO loop. AID and BASIC have a syntax better suited to conversational operation. The "up arrow" symbol for exponentiation in BASIC and AID is preferable to the " " used in FORTRAN while AID also has "' !" for absolute value and "[ ]" interchangeable with "( )" for greater legibility. These features cannot be incorporated into FORDESK as they are not compatible with FORTRAN IV. Variable names are restricted to one letter in AID and to one letter or one letter followed by a digit in BASIC, restrictions which limit the number of variables that can be used and prevent a user making his names mnemonic. FORDESK I users found the restriction to one letter variable names inconvenient because of these reasons. However, FORDESK IV follows the FORTRAN convention of variable names being up to six alphanumeric characters long with characters after the sixth ignored. FORDESK has a more extensive set of arithmetic functions than either BASIC or AID and it goes one step further in its debugging facilities by allowing breakpoints to be set in a program. The current version of FORDESK occupies 5K of core as does BASIC, but AID needs 11K of core, thus the use of AID results in increased overheads in the swapping system. FORDESK is designed to operate re-entrantly to further reduce its swapping overheads FORTRAN statements typed into FORDESK can start immediately after the line number so it is not necessary to space or tab across to represent card image. Rather, statement numbers are delimited by a colon or a tab and similarly comments are indicated by a "C" followed by a colon or tab. The line numbers provided by FORDESK are for reference in FORDESK commands only and are not used as FORTRAN statement numbers. Thus if resequencing of FORTRAN statements is requested, the actual program does not have to be modified. In addition, this will mean that it will be possible to input normal FORTRAN programs from sources other than the teletype. Having received a statement, FORDESK interprets it immediately and if possible executes it. If an arithmetic statement is executed at this stage, the result is typed out immediately as a guide to the correctness of the user's program. Execution is suspended if a jump instruction is typed which would transfer control to a currently undefined statement number. Execution resumes when the statement number is defined. If an error is detected then FORDESK types a diagnostic message and asks for the same line to be retyped. The user can return to command mode from text mode by pressing the ALTMODE key on the teletype instead of entering a FORTRAN statement. in Proceedings of the Fourth Australian Computer Conference Adelaide, South Australia, 1969 view details Moore et al. (1966) discuss their experiences with a remote console time-shared system at the University of Western Australia. In particular they discuss FORDESK, a Fortran compatible, on-line system that enables simultaneous editing, translation, execution, and debugging from a user's console. FORDESK was first released toward the end of 1965. Moore and Erickson (1966) describe the use of a CRT with a lightpen in a timeshared environment. Of particular interest is an application involving polynomial curve-fitting. The curve-fitting program allows dynamic location of the axis, adjustment of scale, and choice of the degree of polynomial all by lightpen picking of displayed "light buttons." in [ACM] ACM Computing Surveys 2(4) Dec1970 view details British Petroleum (BP) in the 1960s ran a refinery just south of Perth in a coastal town called Kwinana. Being a huge industrial concern they had access to some expensive technology, one item of which was a light pen. The computing centre at UWA got wind of this and struck a deal. In exchange for use of the pen the Computing Centre agreed to write the code for the BP maintenance system. A light pen had applications in graphics where it would benefit crystallography and medical research. Procurement of the light pen was a further example of the close network of personal links within the state. External link: Online copy at Alex Reid's site in [ACM] ACM Computing Surveys 2(4) Dec1970 view details |