From Miles.Ellis@etrc.ox.ac.uk Mon Sep 22 19:18:05 1997 Received: from oxmail4.ox.ac.uk (oxmail4.ox.ac.uk [163.1.2.33]) by dkuug.dk (8.6.12/8.6.12) with SMTP id TAA19156 for ; Mon, 22 Sep 1997 19:18:04 +0200 Received: from ermine.ox.ac.uk by oxmail4 with SMTP (PP) with ESMTP; Mon, 22 Sep 1997 18:14:37 +0100 Received: from [163.1.85.1] (com1.etrc.ox.ac.uk [163.1.85.1]) by ermine.ox.ac.uk (1.1/8.8.3) with ESMTP id SAA15900 for ; Mon, 22 Sep 1997 18:14:36 +0100 (BST) X-Sender: mellis@ermine.ox.ac.uk Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 22 Sep 1997 18:15:03 +0100 To: sc22wg5 From: Miles Ellis Subject: Interoperability with C - ballot results I have just received the results of the SC22 ballots on (a) the registration of the PDTR, and (b) its approval. In both cases the voting was 14 countries in favour, 2 against (UK and US), 2 abstaining, and 5 not voting. The PDTR has therefore been registered as PDTR 15815. However, there were over 200 comments attached to approval ballots, of which around 160 were attached to the two negative votes, and the remainder to approval votes. WG5 has been requested to prepare a Disposition of Comments Report and a recommendation on the further processing of the PDTR. I shall put the complete document on the server tomorrow. However, this result does leave us in something of a quandry. In Vienna, we passed a resolution (with only one vote against) requesting the Primary Development Body to accept responsibility for the completion of this project. They have agreed to do so, BUT ONLY IF IT IS CARRIED OUT AS PART OF THE DEVELOPMENT OF FORTRAN 2000 AND NOT AS A TECHNICAL REPORT. It appears to me, therefore, that we now have three choices: 1. Continue with the work as at present, with Michael Hennecke as Editor, assisted by other members of WG5 as required. [Remember that in Vienna he indicated that he could not hope to complete the work in a reasonable timescale unless he had assistance from J3 and other members of WG5.] In this case there are 200 or so comments to be individually addressed by the Development Body before a modified document is submitted to the full WG5 membership for approval - and possibly more changes to be made as a result of that ballot. 2. Notwithstanding the approval in the SC22 ballot, proceed as resolved in Vienna, and forward the document and comments (as useful input) to J3, with a request that they take over the work as a priority item for Fortran 2000. In this case I shall have to request SC22 for approval to adopt this approach - although I do not anticipate any serious problem if that is what WG5 wishes to do as we would not be abandoning the work, simply progressing it in a different, and potentially faster, way. 3. Abandon the whole project on the grounds that there are insufficient resources to continue the work in a timely fashion. If we take option 1, then it is imperative that Michael Hennecke can depend on a substantial amount of committed support (something that has been notable by its absence so far) if the project is to be ready for submission as a DTR by the time of our meeting in Sweden next June. It may be too late for integration into Fortran 2000 even then, but it will certainly be too late if it does not become a fully approved TR by the end of next year. My personal feeling, although I would love to be proved wrong, is that it is probably already too late for the TR to be integrated into Fortran 2000 and that this option will mean that the TR will co-exist with the Fortran 2000 standard and will then be integrated into the next revision (Fortran 2007?). The delay in integration would be unfortunate, but would at least allow the details to be thoroughly tested (assuming that the TR's proposals are integrated into F2000 compilers) before they were cast in stone. If we take option 2 then, by definition, the interoperability feature WILL be in Fortran 2000. On the other hand, it may be in a rather different form from that currently envisaged, and it will not have had the advantage of "field testing" which the TR approach is intended to provide. I very much hope that we don't take option 3, but the somewhat entrenched positions that were appearing in Vienna must make it an outside possibility. I am away from the office for the rest of the week after tomorrow, so I shall give the whole issue some more thought during that time. I suggest that everyone else does so as well. We shall need to decide what to do quite quickly if we are not to drift towards option 3. I had thought that we had made a (difficult) decision in Vienna, but two of the six countries (Finland and Japan) who voted for the resolution to ask J3 to take over the work (as part of Fortran 2000) voted to approve the PDTR - in both cases without any reservations - while two others either abstained (Austria) or did not vote (Sweden). So we clearly need to re-examine our Vienna decision, since it is directly contrary to the result of the SC22 ballot. Miles Ellis Convenor: ISO/IEC JTC1/SC22/WG5 Email: Miles.Ellis@etrc.ox.ac.uk Phone: +44 1865 270528 Fax: +44 1865 270527