From ljm@slac.stanford.edu Mon Mar 13 04:12:11 1995 Received: from SCSW6.SLAC.Stanford.EDU by dkuug.dk with SMTP id AA29176 (5.65c8/IDA-1.4.4j for ); Mon, 13 Mar 1995 21:10:09 +0100 Received: from [134.79.128.74] (MOZART.SLAC.Stanford.EDU) by SCSW6.SLAC.STANFORD.EDU (PMDF V4.3-10 #6987) id <01HO34Z72WGW002EGE@SCSW6.SLAC.STANFORD.EDU>; Mon, 13 Mar 1995 12:11:00 -0800 (PST) Date: Mon, 13 Mar 1995 12:12:11 -0800 From: ljm@slac.stanford.edu (Leonard J. Moss) Subject: Re: (SC22WG5.757) strategic miscue X-Sender: ljm@popserv To: SC22WG5@dkuug.dk Message-Id: X-Envelope-To: SC22WG5@dkuug.dk Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7BIT X-Charset: ASCII X-Char-Esc: 29 At 02:37 3/13/95, Lawrie Schonfelder wrote: >[...] > >... WE should also >review our working mechanisms and procedures. It should be possible to get >requirements set much more firmly and earlier than we have done this time. I strongly agree with Lawrie on this. The new procedures, with a clear separation between the tasks of setting requirements and developing the draft, are a definite improvement over the old (F8x) free-for-all. However, I think the process suffered a great deal this time from a vague statement of requirements. Both X3J3 and WG5 developed good procedures for managing requirements databases, yet in the end, the actual requirements document (the B9 resolution) made no direct reference to items in these databases, but instead attempted to paraphrase various ideas from them. The result was considerable confusion on the part of X3J3 as to what WG5 actually meant -- for example, "minimal and exact field width editing", "allow some classes of user-defined functions in declarations", "exception handling" and (my favorite ;-) "object oriented programming". One approach that I believe would help considerably is to make better use of the requirements databases: when it is time for WG5 to pass a resolution establishing the F2K requirements, that resolution should consist entirely of a list of approved items from the WG5 Repository. This is not a cure all, of course -- it's still possible to approve vague requirements. Nevertheless, limiting the choice to those items that have been previously discussed and approved on their individual merits will go a long way, I believe, to firming up the process. -- Leonard J. Moss | My views don't necessarily Stanford Linear Accelerator Center | reflect those of SLAC, MS 97; P.O. Box 4349; Stanford, CA 94309 | Stanford or the DOE