ISO/IEC JTC1/SC22/WG5/N1576
                                              8 September 2003


To:      WG5
From:    Van Snyder
Subject: Changes proposed to TR 19767 at J3 meeting 165


The purpose of technical report 19767 is to address the deficiencies of
Fortran to support large programming projects.  In a large project, it is
likely that conflicts arise between names of global entities, including
submodules.

To reduce the possibility of name conflicts, it was proposed that the
name of a submodule ought to be local to its ancestor module.  This
proposal was approved by a straw vote.

This requires that both the ancestor module and the parent submodule be
mentioned in the case of a submodule that is not a child of the module.
To accomodate this, the following modification of the SUBMODULE statement
was proposed:

SUBMODULE ( <module-name> [ : <parent-submodule-name> ] ) <submodule-name>

Beyond this syntax change, which affects [7:21] in 03-199r1, edits (with
respect to 03-007r1) to implement this change are not extensive:

[405:19,22] Insert "non-submodule" before "program unit", twice.
{Submodule names are not global identifiers.}

[406:9] Delete "and".
[406:10] Append ", and".
[406:10+] insert
  "(4) If the scoping unit is that of a module, its submodules".


Another proposal was to require a submodule to be a child of a module,
but to allow one submodule to access another by use association, provided
they have the same parent module.  This allows a more general (DAG
instead of tree) relation between submodules of a module.  It was
observed that the present proposal allows one to put a module procedure
interface in a submodule and its separate module procedure body in one of
its descendants.  Replacing the submodule hierarchy with inter-submodule
use association would not allow to put a separate module procedure body
in one submodule and access its corresponding module procedure interface
body from a sibling submodule.