From owner-sc22wg5@dkuug.dk Fri Sep 5 16:04:33 2003 Received: (from majordom@localhost) by dkuug.dk (8.12.8p1/8.9.2) id h85E4XIe029943 for sc22wg5-domo; Fri, 5 Sep 2003 16:04:33 +0200 (CEST) (envelope-from owner-sc22wg5@dkuug.dk) X-Authentication-Warning: ptah.dkuug.dk: majordom set sender to owner-sc22wg5@dkuug.dk using -f Received: from inf.rl.ac.uk (nfs7.inf.rl.ac.uk [130.246.72.7]) by dkuug.dk (8.12.8p1/8.9.2) with ESMTP id h85E4MCp029936 for ; Fri, 5 Sep 2003 16:04:24 +0200 (CEST) (envelope-from j.k.reid@rl.ac.uk) Received: from numerical.cc.rl.ac.uk (numerical [130.246.8.23]) by inf.rl.ac.uk (8.11.6+Sun/8.8.8) with ESMTP id h85E4To07206 for ; Fri, 5 Sep 2003 15:04:29 +0100 (BST) Received: from rl.ac.uk (jkr.cse.rl.ac.uk [130.246.9.202]) by numerical.cc.rl.ac.uk (8.8.8+Sun/8.8.8) with ESMTP id PAA10346 for ; Fri, 5 Sep 2003 15:13:26 +0100 (BST) Message-ID: <3F589AAA.7030902@rl.ac.uk> Date: Fri, 05 Sep 2003 15:16:10 +0100 From: John Reid Reply-To: j.k.reid@rl.ac.uk Organization: Rutherford Appleton Laboratory User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en MIME-Version: 1.0 CC: sc22wg5@dkuug.dk Subject: Re: (SC22WG5.2970) Paragraph numbers References: <200309041843.h84IhXMm018889@dkuug.dk> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-sc22wg5@dkuug.dk Precedence: bulk WG5, I should tell you that it was me that started this. I was in London on Tuesday for a meeting to prepare the UK position for the SC22 meeting and discussed our line number problem with people from other programming languages. They explained the reason for the ban, which is that page and line numbers change if the standard is translated to another natural language. The C solution is to number the paragraphs within each clause. These numbers have been allowed by ISO. I began by asking Richard and Van what they thought of the idea and whether it was technically possible with Latex. Dan joined in the discussion. It looks as if it is technically possible, but we were not unanimous on its desirability; it was clearly time to ask the rest of you. My idea is that we should discard the line numbers so that each page of the published standard is the same as the final draft, apart from the header and footer. See the C standard for an example of what it looks like - Van's specimen is not quite right because he includes line numbers too and does not start the paragraph numbering at 1. We would switch from referencing page and line numbers to referencing clause and paragraph numbers. Not quite so easy to find, but they suffer less change when a draft standard is edited and would allow references to the standard to be valid for both the English and Japanese versions (Makki: please comment on whether this would be helpful). It will make it easier to prepare corrigenda (bear in mind that I have edited all our corrigenda) and TRs. > I understand we're not allowed to have more than one numbered list per > subclause, because a reference like "Item (3) in the list in 16.2" is > ambiguous. > > Assuming we can legally use paragraph numbers, and assuming a list is > part of a paragraph, not that each item is a separate paragraph, does > that mean we can legally have more than one numbered list in a > subclause? I remember this from the ANSI rules that we used for F90, but I cannot find it in the ISO rules. It is not a problem once you restrict yourself to a paragraph. > If the numbers are really going to be permanent there's not going to > be much call for a new Fortran, because we won't be able to fit very > much extra into it before it collapses under the weight of paragraph > numbering anomalies. At the moment we don't even have permanent > subclause numbering! I am not suggesting this, but I think they should be permanent for corrigenda. This would facilitate national bodies printing revisions with corrigenda incorporated, which is certainly what readers (including us) want. Cheers, John.