IBM 610 workstation computer 3468

RJ

On the contrary, it is about Fortran. It's about the deficiency of languages that follow the design of the von Neumann-style computer, of which Fortran is a prime example. Backus defines, in that paper, a language that can be implemented on conventional hardware and has the attributes he had come to believe were desirable.

As to whether it could have been produced in the early '50s, it's not really so different from Lisp, which was also very early. It looks different, but that's mostly syntactic sugar. It uses a "point-free" style, which is currently fashionable in some parts of the functional language community, but which can be written in languages that were not specifically designed with that in mind.

Of course, a lot of the basic work in language theory and compiler technology was done in the 20-odd years between the first Fortran and Backus giving the Turing Award Lecture, so he would not have understood when doing Fortran originally that there were possibilities for doing things better. But in the 25+ years since he publicised these ideas, the fact that people persist with, and defend, Fortran, ignoring his insights, is a sad reflection on the state of the industry.

--brian

-- Wellington, New Zealand

"What's life? Life's easy. A quirk of matter. Nature's way of keeping meat fresh."