From owner-sc22wg5@dkuug.dk Sat Jun 21 01:21:53 2003 Received: (from majordom@localhost) by dkuug.dk (8.12.8p1/8.9.2) id h5KNLrdF030948 for sc22wg5-domo; Sat, 21 Jun 2003 01:21:53 +0200 (CEST) (envelope-from owner-sc22wg5@dkuug.dk) X-Authentication-Warning: ptah.dkuug.dk: majordom set sender to owner-sc22wg5@dkuug.dk using -f Received: from math.jpl.nasa.gov (math.jpl.nasa.gov [137.79.7.57]) by dkuug.dk (8.12.8p1/8.9.2) with ESMTP id h5KNLlEc030943 for ; Sat, 21 Jun 2003 01:21:49 +0200 (CEST) (envelope-from vsnyder@math.jpl.nasa.gov) Received: from math.jpl.nasa.gov (localhost.localdomain [127.0.0.1]) by math.jpl.nasa.gov (8.12.8/8.12.8) with ESMTP id h5KNLgJe021260 for ; Fri, 20 Jun 2003 16:21:42 -0700 Received: from math.jpl.nasa.gov (vsnyder@localhost) by math.jpl.nasa.gov (8.12.8/8.12.8/Submit) with ESMTP id h5KNLgiK021256 for ; Fri, 20 Jun 2003 16:21:42 -0700 Message-Id: <200306202321.h5KNLgiK021256@math.jpl.nasa.gov> X-Mailer: exmh version 2.4 06/23/2000 with nmh-1.0.4 Reply-to: Van.Snyder@jpl.nasa.gov From: Van.Snyder@jpl.nasa.gov To: sc22wg5@dkuug.dk Subject: Directions for Modules TR Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 20 Jun 2003 16:21:42 -0700 Sender: owner-sc22wg5@dkuug.dk Precedence: bulk Early this week, I sent out a list of four changes I propose to make in the Modules TR, as compared to 03-123. Not counting Lawrie, who now has a competing proposal, I got four not-very-convincing responses. Aleksandar wants to put FORWARD on the interface block, because ABSTRACT is on the interface block. Lawrie and I have already explained why the situations are different. Richard suspected that putting a new word in the of a procedure header could lead to syntax problems, apparently because NOT putting BIND(C) in the caused syntax problems. I'm certain that there is no syntax problem in adding new bare words (without parentheses or other clutter) into the . Bill recollected that more people disliked the IMPLEMENTATION ... END IMPLEMENTATION bracket less than they disliked putting SEPARATE in procedure headers. I regard the decision to go in this direction as a hurried decision. I didn't like it at meeting 163, but I acquiesced. The more I think about it, the more I dislike it. Somebody -- I forgot who -- just didn't like SEPARATE because "so many things in Fortran are separate." I don't like FORWARD because the main use I see of separate interface bodies is when they and their corresponding procedure bodies are in fact in different program units, not "forward" in the same one. On the other hand, the procedure body is indeed SEPARATE from the interface, no matter whether they're in the same program unit or not. I like the economy of using one word in both the interface body and the procedure body, meaning exactly the same thing in both places. I haven't seen an argument yet that convinces me the extra complication of the alternatives so far proposed is necessary. Indeed, I have not yet been convinced it's desirable even if it's not necessary. I don't want to be sandbagged on this issue, like I was at the Oxford J3 meeting concerning a procedure to get error message text given an IOSTAT value. I'm leaving for Stuttgart on 17 July, and will spend ten days being a tourist in southern Germany, Austria and Czechia before arriving in Dresden. I'll be finishing a revision of 03-123, to be called N1537, this weekend. If you don't get something convincing to me before about next Tuesday, don't expect me to be sympathetic to entirely different directions in Dresden, or indeed to complaints during the next two weeks or so about the direction the draft of N1537 takes. -- Van Snyder | What fraction of Americans believe Van.Snyder@jpl.nasa.gov | Wrestling is real and NASA is fake? Any alleged opinions are my own and have not been approved or disapproved by JPL, CalTech, NASA, Sean O'Keefe, George Bush, the Pope, or anybody else.