(j3.2006) Liaison to IFIP WG 2.5

Aleksandar Donev donev1
Wed Aug 22 16:44:11 EDT 2007


On Wednesday 22 August 2007 11:54, Lawrie Schonfelder wrote:

> WE were
> constrained in this by having to carry about with us all of the baggage of
> F66/77 where a single processor, linear memory model of a computer was
> heavily embedded 
OK, then, I propose you work on designing a brand new language that does not 
actually map onto real machines and see how far you get in the real world. It 
has been tried before, and is still being tried. Someday it may work out. It 
is not Fortran, and it never will be. You are guarding some nonexistent tower 
of Pure Fortran from the very people that are trying to renovate the existent 
tower of Practical Fortran so it keeps standing and helps real programmers.

> Co-arrays are a step back in the other direction.
Which may just be the *right* direction. Different from what you want, sure, 
but that doesn't make it wrong.

Best,
Aleks



More information about the J3 mailing list