From D.Cannon@cen.ex.ac.uk Sat Apr 3 23:31:48 1993 Received: from danpost4.uni-c.dk by dkuug.dk with SMTP id AA28557 (5.65c8/IDA-1.4.4j for ); Sat, 3 Apr 1993 21:31:56 +0200 X400-Received: by mta danpost4.uni-c.dk in /PRMD=minerva/ADMD=dk400/C=dk/; Relayed; Sat, 3 Apr 1993 21:31:24 +0200 X400-Received: by /PRMD=uk.ac/ADMD= /C=gb/; Relayed; Sat, 3 Apr 1993 21:32:01 +0200 X400-Received: by /PRMD=UK.AC/ADMD= /C=GB/; Relayed; Sat, 3 Apr 1993 21:31:48 +0200 Date: Sat, 3 Apr 1993 21:31:48 +0200 X400-Originator: D.Cannon@cen.exeter.ac.uk X400-Recipients: non-disclosure:; X400-Mts-Identifier: [/PRMD=UK.AC/ADMD= /C=GB/;<10504.9304031931@cen.ex.ac.uk>] X400-Content-Type: P2-1984 (2) Content-Identifier: IEEE LIS and ... From: D.Cannon@cen.ex.ac.uk Sender: D.Cannon@cen.ex.ac.uk Message-Id: <10504.9304031931@cen.ex.ac.uk> To: sc22wg15@dkuug.dk Cc: " (Jim Isaak)" , "stephe (Stephen Walli)" Subject: IEEE LIS and TM discussions Acknowledge-To: D.Cannon@Exeter.ac.uk X-Organisation: University of Exeter, Computer Unit X-Mailer: ELM [version 2.3 PL0] X-Charset: ASCII X-Char-Esc: 29 To all WG15 member bodies: The intention of this note is to highlight a problem which has begun to emerge from the IEEE Posix groups - the people who produce the drafts which we (as an ISO WG) consider for standardisation - over the last few months. The problem affects us (as an ISO WG) because it is causing a fundamental reassessment within IEEE Posix of the type and number of Posix drafts which they produce: it will have a direct effect on what documents are available to be offered to ISO. The essence of the problem, running from group to group and from committee to committee, is one of diminishing resources - less manpower, and less money to fund attendees at the IEEE Posix meetings - and of increasing demand on the resources that remain - more work in terms of drafts to be produced together with higher fees to fund the standards-making activity. This mis-match between supply and demand was highlighted in a paper by Stephe Walli: "POSIX - Caving In Under Its Own Weight", which had wide circulation prior to the January meeting of the IEEE Posix groups. The debate on the content of that paper lasted the whole week of the meeting, and raged both within and outside the formal meetings of groups and committees. The discussion came to a climax at the Sponsor Executive Committee (SEC) meeting on the final evening of the week... The proposals to SEC were, briefly: . Test Methods (TMs) to become non-normative annexes in the drafts. . Language Independent Specifications (LIS) to be suspended, with a proposition to WG15 that when WG11 completes its work, IEEE Posix will consider drafting its work in WG11 terms. Meanwhile drafts will be submitted to ISO WG15 in 'c', Ada, Fortran or Modula-2 form, as appropriate. The TM Proposal: After a well-focussed debate the proposal to suspend the requirement for TMs in Posix standards was passed without objection. An action was placed (on SCCT, which sponsored the original proposal) requiring the SCCT chair to bring a proposal on continued suspension or removal of the TM requirement before the April meeting of the IEEE Posix groups. The LIS Proposal: Stephe Walli presented the Proposal, which was to suspend the requirement for LIS forms of the Posix drafts, allowing documents to be forwarded to ISO WG15 in base-language form, pending a revue in the interim by an Ad-Hoc group of the SEC. The revue was to assess the value of the LIS work, or other potential alternatives, and the Ad-Hoc was charged with reporting back to the April SEC meeting. One major concern was that LIS is only good for another language binding writer - the standards-making exercise. Implementors of code conforming to the standard will possibly regard the LIS as unnecessary clutter or a potential source of ambiguity in the standard. SEC went into committee for a long, well-informed debate on the exact wording of the proposal, together with a minimum set of issues to be addressed by the Ad-Hoc. It was pointed out that relationships with ISO were on the whole good up to now, and that this had involved a certain amount of repair work following disruptions with the Fortran (and other) standards. The committee were concerned to maintain this relationship, and wrote into the Ad-Hoc's terms of reference the requirement to liaise closely with the ISO community, specifically the WG15 member bodies, to ensure they are aware of the manpower situation and also to take into account any input provided. It was noted that P1003.1 was in ballot at the moment in LIS form: one of the charges on the Ad-Hoc was to decide how this should progress. The committee, formed solely to frame the LIS resolution then adjourned, handing the meeting back to the SEC plenary with the completed resolution for SEC to consider. - The LIS resolution, framed by the committee phase, was considered. The clause imposing suspension of the LIS requirement was dropped, pending the report of the Ad-Hoc group, which was to be established and chaired by Stephe Walli, so for the moment the IEEE Posix groups are still committed to producing drafts in LIS form. This may change after the April meeting. What the IEEE Posix groups are looking for is more support: if the member countries of ISO believe that Language Independent Specification (LIS) drafts of the Posix standards are of vital importance then the IEEE ultimately needs ISO to come forward with manpower to help them produce the drafts. In the meantime the IEEE Posix groups are looking for guidance to help them find a way forward. As a WG15 member body you are urged to consider the problem outlined above, and to offer your comments to the chair of the IEEE Ad-Hoc group: Stephen Walli stephe@mks.com before the next meeting of the IEEE Posix groups, 19-23 April in Irvine, USA. There is an IEEE PASC-LIS mailing list, established specifically to air this discussion. We can request to get onto it by emailing: pasc-lis-request@ui.org with your email address, name and affiliation. and requesting to be added to the pasc-lis mailing list. Stephe Walli is happy to send the pasc-lis mailings to date so you can catch up on the discussion. David Cannon, Convenor, UK IST/5/-/15 (Posix) panel. (D.Cannon@exeter.ac.uk)  -- ____________________________________________________________________________ Dave Cannon University of Exeter, Computer Unit Laver Building, North Park Road Systems/Network Programmer Exeter, EX4 4QE, Devon, UK Phone: +44 (0)392 263956 Fax: +44 (0)392 211630 ____________________________________________________________________________