From owner-sc22wg5@open-std.org Thu Nov 13 08:22:05 2008 Return-Path: X-Original-To: sc22wg5-dom7 Delivered-To: sc22wg5-dom7@www2.open-std.org Received: by www2.open-std.org (Postfix, from userid 521) id 69254CA343B; Thu, 13 Nov 2008 08:22:05 +0100 (CET) X-Original-To: sc22wg5@open-std.org Delivered-To: sc22wg5@open-std.org Received: from ns.nag-j.co.jp (218-42-159-107.cust.bit-drive.ne.jp [218.42.159.107]) by www2.open-std.org (Postfix) with ESMTP id 69B07C178D9 for ; Thu, 13 Nov 2008 08:22:03 +0100 (CET) Received: from 218-42-159-108.cust.bit-drive.ne.jp ([218.42.159.108] helo=[127.0.0.1]) by ns.nag-j.co.jp with esmtp (Exim 4.50) id 1L0WWV-0002hQ-4u for sc22wg5@open-std.org; Thu, 13 Nov 2008 16:21:51 +0900 Message-ID: <491BD5A7.8020104@nag-j.co.jp> Date: Thu, 13 Nov 2008 16:22:15 +0900 From: Malcolm Cohen User-Agent: Thunderbird 3.0a1pre (Windows/2008022014) MIME-Version: 1.0 To: WG5 Subject: Re: (j3.2006) (SC22WG5.3667) [ukfortran] N1755: Request for new features from MPI Forum References: <49137AD3.1070402@lrz.de> <20081111224927.8201CC178D9@www2.open-std.org> <20081111234923.517C5C178D6@www2.open-std.org> <20081112001017.26EB3C178D6@www2.open-std.org> <20081112091351.6581CC178D9@www2.open-std.org> <491BCB6F.8040406@sun.com> In-Reply-To: <491BCB6F.8040406@sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-sc22wg5@open-std.org Precedence: bulk J3 readers, please remember to change your reply to WG5 when it's a WG5 message, otherwise it won't get through... Robert Corbett wrote: > N.M. Maclaren wrote: > >> Aargh! No! C 'const'! It has NOT been a success - in fact, to describe >> it as a paving slab on the road to hell would be fair. [ Yes, I am referring >> to the saying that the road to hell is paved with good intentions. ] >> > > The best example of that in Fortran 2003 is the PROTECTED attribute. > A casual reader of the standard is likely to believe that a variable > with the PROTECTED attribute cannot be modified outside of the module > in which it is contained. A careful reading and the interps that > have been issued make it clear that PROTECTED is a promise made by > the programmer that he doesn't modifiy the variable outside of the > module This is not correct; to call it a "promise made by the programmer" is not even remotely accurate. PROTECTED is subject to almost precisely the same restrictions in use, and the same pitfalls in the guarantees, as INTENT(IN) - it's not quite a guarantee because we actually wanted INTENT(IN) to be easy to use in a world that includes things like old procedure libraries that lack INTENT specs. There are some subtle differences because one is a module variable and the other is a dummy variable, but those aside, almost exactly the same. Now it could be argued that we were trying to be "too nice" to the casual programmer back in Fortran 8X design days, and that is a reasonable argument: there is a tension between ease of use and guaranteed safety here. Perhaps we got that wrong (i.e. the guarantees are too weak). I might (not?) disagree, but it's nearly 20 years too late for that debate though. But if one accepts the INTENT(IN) facility with its minor warts, one cannot then argue against what we did for PROTECTED since that is not in practice any more open to abuse. Cheers, -- .....................Malcolm Cohen, Nihon NAG, Tokyo.