(j3.2006) enhanced ASYNCHRONOUS attribute

[Robert Corbett]

Does Clause 2.4 of DPTR 29113 change the language
specified by the Fortran 2008 standard in any way?
AFAICS, it imposes no new requirements on Fortran
processors.  Therefore, any processor that conforms
to the Fortran 2008 standard should also conform to
the Fortran 2008 standard plus Clause 2.4.

There might be programs that conform to the
Fortran 2008 standard alone that will not be
conforming under Fortran 2008 plus Clause 2.4.
A program that invokes a procedure written in a
language other than Fortran could always be subject
to restrictions beyond those imposed on programs
written entirely in Fortran.  I presume that
Clause 2.4 allows a processor to designate
procedures written in Fortran as asynchronous
communication initiation or completion procedures.
In that case, some programs written entirely in
Fortran might cease to be conforming when
Clause 2.4 is in effect.  Which programs will
cease to be conforming will be processor
dependent.

I looked up the word "affector" and found that it
does not seem appropriate.  I know the word is
already used in a similar manner in the standard.
The most obvious definition I found was "an entity
that affects."  What does a variable "affect"
during asynchronous communications?  An alternate
definition was "one who feigns or pretends," but
that seems only slightly more appropriate.  Is
there a meaning of the word "affector" that is
more appropriate?

Robert Corbett