Re: history of Fortran, good post on LinkedIn

Feb 01, 2026 1 Replies


According to Nioclás Pól Caileán de Ghloucester snipped-for-privacy@Taf.com:


>>I want to quote an old comp.compilers post by its moderator about how
>>FORTRAN programmers are not bothered to consult the FORTRAN standard
>>(circa FORTRAN-66) so they insist that they know FORTRAN when they do
>>not, so a new FORTRAN standard (circa FORTRAN-77) made a
>>backwards-incompatible change to accept this wrong belief of what
>>FORTRAN really is. Alas searching for it takes too long (the 3 search
>>options offered by
>
>I'm pretty sure I didn't say that. Possibly someone else did but I don't
>recall that either. In fact F77 tried hard to stay compatible with F66 and
>the few incompatibilities were well documented and had good rationales.

In 1990 I led a project with Shell. All calculations were still required to use FORTRAN IV. Because there was substantial graphics involved we got dispensation to use c on VMS. (Using transputers, also occam was allowed.) Compatibility was a priority.


John Levine, snipped-for-privacy@taugh.com, Primary Perpetrator of "The Internet for Dummies",

Groetjes Albert

In software, compatibility is the bane of good design. Just because something made sense for a particular project, doesn't mean it makes sense for YOUR project.

Or, if it is even "best practices" any longer.

I chuckle when I think of how many devices using Linux kernels are carrying the cost of file system support when they likely never would have put "support for a filesystem" on their list of requirements.

And, when the whole notion of a global namespace actually makes their design LESS secure/robust!

(And ACLs? etc. Really?? All that extra, UNNEEDED code that just makes your product buggier and more bloated...)

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required