Re: What is Orthogonal (Exactly)?

>> In itself, a programming language is orthogonal if its features can
be
used without thinking about how that usage will affect other features. <<

In the old days, there were only 5 or 7 valid ways to write an array subscript in Fortran II and I am damned if I can remember them now. I think it was something like constant, (constant * variable), variable, (variable + constant) , (constant * variable + constant). This had to do with the register hardware on the IBM machine which generated the displacment.

Later programming langauges and versions of Fortran were not so attached to the hardware, so their rule was that any expression which returned an integer was a valid subscript. Simpler rule, wider set of possibilities. In short the subscript syntax was orthogonal -- context free. In 25 words or less, anywhere in the language that a scalar value makes sense, then an expression of the same datatype makes senses. Received on Mon Jun 17 2002 - 05:34:49 CEST