J3 / 97-215 Date: 11 Aug 1997 To: J3 From: R. Maine Subject: Correspondence about f90 opengl bindings The following are excerpts from my recent correspondence with William F. Mitchell on this subject. ----------------------------------------------------------------- From: mitchell@cam.nist.gov (William F Mitchell) To: maine@altair.dfrc.nasa.gov Subject: Re: Fortran 90 OpenGL bindings Date: Wed, 30 Jul 97 14:07:26 EDT Richard, Thank you for responding for my request for feedback on the proposed Fortran 90 bindings for OpenGL last month. I am sorry I took so long to get back to you, but I have been on travel and other assignments. I only received four replies to my request, and truly appreciate your taking the time to be one of them. I don't think the OpenGL ARB will adopt my proposal unless I provide supporting comments from Fortran experts. Most of them are not familiar with Fortran and do not feel qualified to assess the proposal. So I must request help from the Fortran experts, and it would be very much appreciated if you could read through the proposal. If you think it sounds reasonable, please send some supportive comments that I could forward to the OpenGL ARB. You can obtain a copy of the technical report from the f90gl web page http://math.nist.gov/f90gl or from my web page (URL in .sig). The common comment in the responses I received was to examine the ISO Technical Report on interoperability between Fortran and C. I have read this report. I thought that it might make the Fortran 90 binding for OpenGL obsolete, but I do not believe this is the case. Instead, I think it will provide the means for a portable _implementation_ of the binding. The binding itself is defined entirely from the Fortran side, so the interoperability of Fortran and C is not an issue in the proposed bindings, just in an implementation of them. This is another major difference from the fortran 77 bindings for OpenGL which I should have included in the list that you have seen. I will be adding: The Fortran 90 interface to OpenGL is defined entirely on the Fortran side of the Fortran/C interface, so the issues associated with interoperability between Fortran and C are hidden from the user. Thank you, -- Bill William F. Mitchell | william.mitchell@nist.gov Mathematical and Computational Sciences Division | na.wmitchell@na-net.ornl.gov National Institute of Standards and Technology | Voice: (301) 975-3808 Gaithersburg, MD 20899 | Fax: (301) 990-4127 http://math.nist.gov/acmd/Staff/WMitchell/ ------------------------------------------------------ maine> Do you mind me submitting your email and the tech report to J3 as a maine> paper at its August meeting? I'm not sure what you mean by submitting it as a paper, but anything that can drum up interest is welcome. This may be a good opportunity for some of the Fortran experts to discuss it and provide some feedback.... maine> Mostly submitting it as a paper means that maine> it is copied for everyone there and becomes part of the public record maine> of the meeting. I can't guarantee that any time will be allocated to maine> discussing it (almost certainly not a lot of time), but submitting it maine> as a paper would be necessary to have much chance of that. Yes, that's fine. As a NIST IR it's already public. I wasn't really expecting much discussion in the meetings proper -- I'm sure a pretty full agenda already exists -- but was thinking it might come up at coffee breaks, etc. Whatever works! Thanks, -- Bill