From Keith.Bierman@eng.sun.com Thu Mar  2 09:19:14 1995
Received: from Sun.COM (koriel.Sun.COM) by dkuug.dk with SMTP id AA20063
  (5.65c8/IDA-1.4.4j for <SC22WG5@dkuug.dk>); Fri, 3 Mar 1995 02:20:32 +0100
Received: from Eng.Sun.COM (engmail2.Eng.Sun.COM) by Sun.COM (sun-barr.Sun.COM)
	id AA23778; Thu, 2 Mar 95 17:20:13 PST
Received: from chiba.Eng.Sun.COM by Eng.Sun.COM (5.x/SMI-5.3)
	id AA14345; Thu, 2 Mar 1995 17:20:01 -0800
Received: from localhost by chiba.Eng.Sun.COM (4.1/SMI-4.1)
	id AA07840; Thu, 2 Mar 95 17:19:15 PST
Message-Id: <9503030119.AA07840@chiba.Eng.Sun.COM>
To: ljm@slac.stanford.edu (Leonard J. Moss)
Cc: SC22WG5@dkuug.dk
Subject: Re: (SC22WG5.745) Re: Re: Interoperability 
In-Reply-To: Your message of "Thu, 02 Mar 95 15:03:55 PST."
             <199503022329.AA17393@dkuug.dk> 
Date: Thu, 02 Mar 95 17:19:14 PST
From: Keith.Bierman@eng.sun.com
X-Charset: ASCII
X-Char-Esc: 29

> 
> >I neglected to mention in my previous missive that some common (GUIish
> >stuff) has names longer than 31 characters, another reason for some
> >sort of external to Fortran mechanism.
> 
> Perhaps we need a quick mod to F95 to accomodate longer names?  I would
> think that's a small enough edit it could be agreed at the WG5 meeting and
> changed in the document in time for the CD ballot (that is, providing
> people can agree on how long is long enough ;-).

Whether or not the names are longer, it is often helpful to have some
way to map to foreign names. It would be, for example, a crude side
door to C++ (the user would have to know the mangling algorithm for
that particular release of that particular C++ compiler, but it would
be better than nothing). Systems that keep Fortran and C namespaces
separate via underscores or other special symbols, etc.

I am not against longer names being allowed, merely against
considering them a solution to the existing situation.