From JKR@IBM-B.RUTHERFORD.AC.UK Mon Nov 4 17:26:20 1991 Received: from danpost2.uni-c.dk by dkuug.dk via EUnet with SMTP (5.64+/8+bit/IDA-1.2.8) id AA06290; Mon, 4 Nov 91 17:26:20 +0100 Received: from vm.uni-c.dk by danpost2.uni-c.dk (5.65/1.34) id AA09919; Mon, 4 Nov 91 16:26:22 GMT Received: from vm.uni-c.dk by vm.uni-c.dk (IBM VM SMTP V2R1) with BSMTP id 3958; Mon, 04 Nov 91 17:25:41 DNT Received: from UKACRL.BITNET by vm.uni-c.dk (Mailer R2.07) with BSMTP id 9567; Mon, 04 Nov 91 17:25:38 DNT Received: from RL.IB by UKACRL.BITNET (Mailer R2.07) with BSMTP id 0645; Mon, 04 Nov 91 16:24:24 GMT Received: from RL.IB by UK.AC.RL.IB (Mailer R2.07) with BSMTP id 0372; Mon, 04 Nov 91 16:24:24 GMT Via: UK.AC.RL.IB; 4 NOV 91 16:24:21 GMT Message-Id: <04 Nov 91 16:21:22 GMT JKR@UK.AC.RL.IB> Date: Mon, 04 Nov 91 16:21:22 GMT From: "JKR" (JKR at UKACRL) To: SC22WG5@dkuug.dk Subject: Vote on N743 X-Charset: ASCII X-Char-Esc: 29 Do you approve WG5 - N743 for submission to the SC22 Secretariat for registration as a committee draft? (CD)? My vote is NO. Reasons: 1. The operators and procedures are extended only for scalar arguments. This should be made clear throughout. I noticed the problem in the following places: 3/30, 4/40, 7/34, 8/13, 9/4, 9/22, 9/39. 2. 10/23-43. This description needs to be reworded. It talks about altering string and substring, which is not intended since the following sentence says that all arguments are unchanged. The same comment applies to 11/6-11. 3. 13/15 to 29/21. All the imitation interface bodies in the form of comments should be removed. They increase the overall length of the module by about 1000 lines (more than 25%) and are contary to one of the guiding principle of modules that information should be specied just once. When the comments are removed, one can see the structure of the module and one is left with a very useful index. 4. There are too many errors in the code. I think that it should be thoroughly checked on at least one Fortran 90 processor. I found the following, and am far from confident that there are no more: 21/29, 21/63, 51/55, 53/1. Delete ',OPTIONAL'. 32/25+. Add statements to define la and lb. 44/5, 44/55, 45/33, 46/38, 47/40, 48/45. Replace NULLIFY statement by "string=''". 54/43. Replace 'insert_ss' by 'insert_cs'. 55/46, 56/28. Add '('. 62/65, 64/2. Replace ',' by ':'. 67/71+ Add ENDDO 76/28. Replace 'ierr' by 'iostat=ierr'. 76/38. Replace 'CALL' by 'line='. I have not looked at the second example (pages 77-78). Minor comments: Title page. Replace 'Organisation' by 'Organization'. vi/12. Replace 'Griffith' by 'Griffiths'. 5/7. Delete 'the characters of' (seems to add nothing and confused me at first). 6/46. Change 'one' to '1' (for consistency with 'LEN(string)+1'). 9/13. Change 'transfered' to 'transferred'. 2.3.1. Say that if expr is a data object, it is not changed. 2.4, 2.5, 2.7. I would prefer a single statement in the lead paragraph of section 2.4 to the effect that all arguments are unchanged, rather than the repetition of this statement for every function. The same comment applies to sections 2.5 and 2.7. John Reid, 4 November 1991.