My 1992 view of the problems of computer programming in 1992

4 hours ago

In 1992, the author reflected on the history of IBM's FORTRAN compilers, noting the significant investment in the 1970s to develop the advanced FORTRAN H compiler.
By 1992, compiler technology had advanced so much that undergraduate students could write passable compilers in a semester, making them much easier to develop than in the past.
The author questioned why IBM hadn't invested in creating a 'super FORTRAN I' compiler, given the improvements in compiler-writing ability.
The answer highlighted that compiler quality is no longer the main bottleneck; instead, fundamental problems in programming methods, languages, and thinking persist.
The author concluded that programming remains a 'black art' because we lack effective ways to express and manage programs, even though tools have improved beyond our ability to use them optimally.

Hasty Briefsbeta