ISO/IEC JTC1/SC22/WG5 N1590 Timetable for Fortran 2003 and the next revision John Reid Having a timetable for Fortran 2003 has been very helpful. I hope we can do the same for the next revision. Here is a first try, which I have developed with the help of Dan Nagle and David Muxworthy. 1. Fortran 2003 Stage Process ----- ------- 4 DIS registered 2004-05 DIS ballot initiated 2004-06 DIS ballot results available 2004-08 5 Standard published 2004-09 Corrigendum 1 formed (at WG5 meeting) 2005-07 Corrigendum 2 formed (by e-mail) 2006-07 2. Fortran 200x Stage Process ----- ------- 1 Repository started 2004-05 Preliminary choice of significant features 2005-04 Final choice of significant features 2006-02 Completion of separate edits 2006-05 2 New Work Item proposed to SC22 2006-08 First working draft available 2007-05 WG5 review of working draft 2007-07 WG5 approval of draft CD 2007-09 3 First CD submitted for registration and approval 2007-10 First CD ballot initiated 2007-10 First CD ballot comments available 2008-01 WG5 ballot resolution meeting 2008-02 WG5 review meeting 2008-07 Final CD submitted for approval 2008-09 Final CD ballot initiated 2008-10 Final CD ballot comments available 2009-02 WG5 approval of draft DIS 2009-04 4 DIS registered 2009-05 DIS ballot initiated 2009-05 DIS ballot results available 2009-07 5 Standard published 2009-08 Notes. A. This plan is based on the assumption that we aim for a minor revision in 5 years time, much as for Fortran 90/95. Reasons: 1. Vendors will need time to implement all of Fortran 2003, 2. Vendors will need time to optimize any of Fortran 2003, 3. Users will need time to learn much of Fortran 2003, 4. Users will need time to learn where Fortran 2003 needs improvement, 5. Authors will need time to prepare textbooks and other media supporting Fortran 2003, 6. Vendors will need time to keep the on-going costs of developing Fortran compilers within that which may be recouped from the on-going sales of Fortran processors, and 7. A minor update is indicated by the alternation between major and minor updates experienced so far. B. This will permit very few significant features, but several minor enhancements and editorial improvements. C. We need to allow time for outsiders to appreciate what is in Fortran 2003 and make informed suggestions for revisions. D. The preliminary choice of significant features (2005) will leave us with some candidates for which edits should be constructed, some decisions for rejection, and some open items. We would not preclude further items being suggested later in 2005. E. Any significant feature that is not well polished by the end of stage 1 should be rejected, as were exceptions and allocatable extensions for Fortran 95. F. Repository items should evolve to contain or point to papers that contain more detailed specification (ideally actual edits) and discussion, both for and against. G. Because of the size of Fortran 2003, we may reasonably anticipate significant interpretation requests.