From maine@altair.dfrc.nasa.gov Sat Feb 3 18:31:16 1996 Received: from altair.dfrc.nasa.gov ([130.134.34.72]) by dkuug.dk (8.6.12/8.6.12) with SMTP id SAA23967 for ; Sat, 3 Feb 1996 18:31:11 +0100 Received: by altair.dfrc.nasa.gov (5.0/SMI-SVR4) id AA20225; Sat, 3 Feb 1996 09:31:34 +0800 Date: Sat, 3 Feb 1996 09:31:34 +0800 Message-Id: <9602031731.AA20225@altair.dfrc.nasa.gov> From: Richard Maine To: Craig.Dedo@mixcom.com Cc: sc22wg5@dkuug.dk In-Reply-To: <199602031448.PAA21357@dkuug.dk> (message from Craig Dedo on Sat, 03 Feb 96 08:47:05 -0600) Subject: Re: (SC22WG5.1027) (x3j3.1996-50) Status of Fortran 95 CD (FOR X3J3/136) content-length: 1764 Craig said (reformatted according to email standards so that people with terminals less than about 300 characters wide can read it) > From some of the discussion it is apparent that this discrepancy > [i/o specifier kinds] is unintentional. No. Exactly the opposite. It is evident that it is intentional. There have been several comments noting that this was done intentionally in the first place in writing f90. Published interpretation 85 explicitly addresses the issue. Jerry re-raised the question about a year ago, bringing up some of the same points being asked now (that's obviously the discussion that I alluded to recalling). That sounds like 3 different times that the question was considered. > It also creates portability problems. As Keith mentioned before, any program that uses files with more than 2 billion records has severe portability problems in any case. You are, of course, aware that no standard unix interface in a 32-bit system can support such files for pretty much the same kind of reason as comes up here (and if it is a full 64-bit system, then the existing Fortran inquire is ok). The millions of systems that must be at issue here must be something like MS-DOS and VAX systems I suppose . > What is your recommendation? I thought he made his recommendation pretty explicit. It was pretty clear to me that he was addressing exactly this discussion when he said that we were not to consider new technical issues for f95. The process of making proposals for f2k is pretty well established - follow it. This is a plausible subject for a new requirement, which we can then duely consider along with other requirements. It is not something that was overlooked. -- Richard Maine maine@altair.dfrc.nasa.gov