From martin@anduin.ocf.llnl.gov Fri Jul 12 23:16:19 1996 Received: from anduin.ocf.llnl.gov (anduin.ocf.llnl.gov [134.9.49.1]) by dkuug.dk (8.6.12/8.6.12) with SMTP id XAA17740 for ; Fri, 12 Jul 1996 23:16:18 +0200 Received: by anduin.ocf.llnl.gov (4.1/SMI-4.0) id AA00477; Fri, 12 Jul 96 14:16:15 PDT Date: Fri, 12 Jul 96 14:16:15 PDT From: martin@anduin.ocf.llnl.gov (Jeanne T Martin) Message-Id: <9607122116.AA00477@anduin.ocf.llnl.gov> To: sc22wg5@dkuug.dk Subject: Absent Vote To: Those who will be in Dresden I will not be joining you even though there are several important matters to be decided that affect the future of Fortran. For most of them, I would like to be present for the discussions, particularly the content of Fortran 2000, before making any decisions about voting. But there is one decision I have already made, and I would like to take this opportunity to express my opinion on that issue. It would be nice if a portable program could be written without the need for conditional compilation, but that is not reality. Every portable program I have seen at LLNL uses some form of conditional compilation even though they all run on "UNIX" systems. CRI (UNICOS) seems to things slightly differently from most others, but even between SOLARIS and SUN OS there are differences, primarily in library routines that provide functionality outside the scope of any programming language. Conditional compilation is also heavily used for different versions of a code so that only one source need be maintained. It is a crime that there is no standard Fortran-knowledgeable means of selecting portions of a Fortran code to compile. Therefore I would cast my vote (if I had one) for the CoCo functionality as described in N1192. For those who will be at the Dresden meeting, I wish you well. Good Luck, Jeanne Martin email: jtm@llnl.gov phone: (510) 422-3753 postal code: L-561 fax: (510) 423-8704 Lawrence Livermore National Laboratory Livermore, CA 94550 USA