From John.Nicholls@comlab Tue Dec 1 10:51:20 1998 Date: Fri, 4 Aug 95 11:09:50 BST From: John.Nicholls@comlab To: A.Smith@gb.hmg.dra.green, Jim.Woodcock@comlab, Makoto.Someya@jp.co.unisys, S.H.Valentine@uk.ac.bton, an@jp.co.toshiba.rdc.isl, Andrew.Martin@comlab, baumann@ch.unizh.ifi, br@uk.ac.rl.inf, Stephen.Brien@comlab, cinzia@it.intecs.pisa, cjchamp@gb.hmg.dra.taz, cts@gb.hmg.dra.green, derek@uk.ac.le.mcs, drj@mil.ncsc.tycho, ecusack@uk.co.bt.fmg, gerhart@cl.uh.edu, ianh@cs.uq.oz.au, j.e.sinclair@uk.ac.open, jam@uk.ac.york.minster, jap@de.d400.uni-kiel.informatik, jarh@uk.co.praxis, jbwords@com.ibm.vnet, jgh@uk.ac.york.minster, Joan.Arnold@comlab, Steve.King@comlab, lupton@com.ibm.vnet.winvmd, maa@com.att.research, mark@ca.on.ora, mataga@com.att.research, mayers@com.tek.labs.poseidon, John.Nicholls@comlab, pjs%ist.co.uk@oxmail, pyoung@uk.co.bt.srd, rbj@uk.co.icl.win, rda@uk.co.icl.win, sjd@uk.co.icl.win, srlm@br.cnpq, susan@uk.co.logcam, trevor@uk.co.praxis, uchi@jp.co.toshiba.ilab.ap, w869cp@uk.co.bae.mal.wa, will%ist.co.uk@oxmail, wsinhee@nl.tue.win Subject: Minutes of Meeting 28 (email version) ----------------------------------------------------------------------- Z STANDARDS PANEL DOCUMENT Z-180.TXT IST/5/-/19/2 (Z Notation) ISO Panel JTC1/SC22/WG19 (Rapporteur Group for Z) MINUTES OF 28TH MEETING, 25TH JULY 1995 (email version) ----------------------------------------------------------------------- The 28th meeting of the Z Standards Panel was held on Tuesday 25th July 1995 at the Oxford University Computing Laboratory. Present: Rob Arthan, Stephen Brien, Jonathan Hammond, Andrew Martin, John Nicholls (Convener), Susan Stepney, Sam Valentine. Apologies: Jon Hall, Will Harwood, Randolph Johnson, Steve King, Peter Lupton, Jane Sinclair, Alf Smith, Pete Steggles, John Wordsworth, Pete Young. 28.1. OBJECTIVES OF MEETING John Nicholls outlined the purpose of this meeting: * to review and decide on actions to be taken regarding comments on Version 1.1, published on 30th June 1995. * to prepare final plans for preparation and submission of a CD draft. * to discuss activities at future meetings, including related events such as the Z User Meeting and FME'96. 28.2. ACTIONS ARISING FROM MINUTES OF MEETING 27 (DOCUMENT 179) Action 26.2.1 Investigation of ftpmail. The working of ftpmail has been confirmed. Action closed. Action 23.3 Information on other ftp sites. Two ftp sites are being set up by members of X3J21, at CMU Pittsburgh and ORA Ottowa. Details of these will be circulated when they are set up. Action closed. Actions 27.4 These actions have been completed as part of the work of the semantics subgroup. Actions complete. [Note: It was pointed out that some of the papers have not yet been circulated to members of the subgroup] Action 27.5 These actions are complete. Action 27.13 Discussions on Modules have been held. Action complete. See below for further actions on this topic. Action 27.15. Version 1.1 was published on 30th June and is now being revised. Action complete. 28.3. APPROVAL OF MINUTES The minutes were approved. 28.4. RETIREMENT OF GORDON ROSE John Nicholls reported receiving an email from Professor Gordon Rose, saying that, as he is retiring, he will no longer be able to remain as member of the Panel. The Panel unanimously agreed to send a message of thanks to Gordon for his work in the past years, and to send good wishes for his retirement. 28.5. INITIAL REPORT FROM NEW SUBGROUP: DISSEMINATION One objective of the dissemination group is to make people aware of the existence of the standard, and to allow current Z users and others to comment on the version now under consideration. Another aim is to explain the technical content of the WD, and to encourage existing and new users of Z to use the standard. Andrew Martin has drafted a WWW page with some of the headings that might be used for introducing the standard. This will include a FAQ and links to the sections of the WD. Mark Saaltink has proposed the use of a feedback mechanism similar to that used during development of the Ada standard, to allow general users to comment on the drafts. Senders of comment would be acknowledged but there would not necessarily be a commitment to answer or respond to comments and questions. This proposal was welcomed and Mark is invited to develop the proposal further. Action: Mark Saaltink. FME'96. As part of the dissemination activities, John Nicholls said he intends to submit a proposal for a half-day tutorial, "Introducing the Z Standard" at FME'96 in March 1996, and will invite contributions from Panel members. Other activities of the dissemination subgroup will follow the completion of the CD submission. REPORTS FROM SUBGROUPS: ----------------------------- 28.6. Semantics subgroup Since the last meeting several meetings and exchanges of emails have taken place. These have been documented in messages circulated within the subgroup. Ideas for `type-value separation' have been extensively discussed, but at the time of reporting no final resolution has been made. Rob Arthan commented on the minutes of the meetings and said that in his view the separation is difficult and deep and would make a profound difference to parts of the language. There is an urgent need for resolution of the points made in recent meetings. It is planned that a further meeting of the semantics subgroup will take place as soon as possible to resolve this issue. Action: Will Harwood, members of group. 28.7. Logic subgroup The logic section is related to the semantics work in that a rules for generics are affected by the decisions. Andrew Martin is revising the version of Annex F in Version 1.1. Please send comments to him. 28.8. Syntax and Tools subgroup A copy of the Lexis report (X3J21/95-043) by Saaltink, Johnson and Milam has been received and is under consideration by Susan Stepney in revision of the Lexis. Susan will reply directly to the authors in considering this proposal, for which X3J21 is thanked. Considerable work remains to be done to complete and review the syntax and lexis proposals, and it seems likely that a further meeting of the subgroup will be necessary shortly. 28.9. MATHEMATICAL TOOLKIT SUBGROUP Report by John Wordsworth: ---- The Toolkit subcommittee has not met since the last meeting of the panel at the Open University in May. The subcommittee's proposal for a new toolkit was rejected by a poll of members of the full panel. Few comments have been received about how the toolkit presently in the draft could be extended in a way acceptable to a majority of the full panel. A note asking those who had voted against the proposal to explain how the toolkit could be acceptably extended provoked only two replies, both from panel members whose views had been made known in the meeting. The subcommittee's chairman supplied fixity paragraphs for the current toolkit in time for the draft of June 30. The views that have been expressed about how the toolkit might be extended, though they are few, advocate one or both of the following courses, but do not invite the subcommittee to proceed any further at this stage. add axioms for integers to Spivey 2. add number above integer, but not reals. Some find in infinite sequences and real numbers excesses that should not yet be part of the standard. ---- DISCUSSION: The Toolkit forum had been set up to exchange ideas for the Toolkit, but some of the proposed questions in the forum had not produced responses. Sam Valentine directly asked members some of the questions from his list. Several members stressed the importance of asking the right kinds of question, in the right order, when conducting such surveys. In discussions of the next stages, it was suggested that the Toolkit group should consider making a `conservative' proposal involving the addition of a more general concept of number. This should be sent for review before any further extensional work on the toolkit is considered. Susan Stepney agreed to consider this. Action: Susan Stepney 28.10. CONFORMANCE SUBGROUP Nothing to report at this meeting. 28.11. OBJECT-ORIENTED SUBGROUP Nothing to report at this meeting. 28.12. INTERCHANGE FORMAT SUBGROUP A new version of Annex E is being written by Jonathan Hammond. Please send comments to him. 28.13. REPORT FROM X3J21 [The following is not an official report of the meeting.] The X3J21 group had a meeting in Montreal on 8-9 July 1995. Stephen Brien and John Nicholls had been invited to attend the meeting to discuss progress on the Z standard. The following Z-related topics were discussed at the meeting: * Tutorial on semantics Stephen Brien gave a talk on Z semantics, a continuation of the talk given by Jim Woodcock at the previous meeting. He outlined some of the issues arising in the `type-value' separation. Using sample semantic equations, he showed how the semantic metalanguage is used to build up types and values. During the meeting he prepared a short written report based on the talk (a copy is with Haim Kilov) and a revised version will be distributed later. * Lexis proposal The report by Mark Saaltink, Randolph Johnson and William Milam is in response to the proposed Lexis Annex in Version 1.1. In general, it seems that the proposal for a human-oriented representation is well accepted, though there may be slight differences of emphasis on whether this form is, or should be, suitable for tool support. * UCS support X3J21 is preparing a statement requesting UCS support for Z (and VDM-SL). Bill Harvey will send a draft copy of the proposed submission. Highest priority will be given to symbols in the Z language although it is also intended to seek support for the metalanguage symbols. * Plan for preparing a Committee Draft for Z Current plans for submitting a CD for registration at the SC22 plenary in September were discussed and it was generally agreed that this would be appropriate. * The use of Z in ODP standards An extensive discussion of the recent exchange of proposals and reviews of the ODP proposals in IS 10746 (Parts 2 and 4), was held. The US delegation had proposed the rejection of Part 4 and made a number of counter-proposals. At the conclusion of the discussion, John Nicholls offered to get in touch with the author of the Part 4 proposal, Richard Sinnott of Stirling University, and to offer the possibility of help from members of the Z Panel to bring the proposal more in line with the Z Standard. Richard Sinnott has now accepted an invitation to attend the next meeting of the Z Panel in September to consider the next stages. [Note: Panel members who would like to read the standards papers can obtain them from the ftp server: ftp.gmd.de in the subdirectory: documents/iso/RM-ODP 28.14. Plan for developing a Committee Draft Report by John Nicholls The preparation of a CD for registration is our next important step. The version just prepared (Version 1.1, available on the ftp server ) meets the following criteria for submission as a CD, as required by JTC1 procedures: * the main elements are included in the draft; * it is presented in a form which is essentially that envisaged for the future IS. However, we do not propose to submit this version, as it contains a number of errors we wish to correct before submitting as a CD. The proposed plan is as follows: * correct errors in Version 1.1, and annotate the draft to indicate known areas for immediate future work. * prepare and distribute a revision, Version 1.2, by approximately 24th August 1995. * discuss this at Meeting 29 of the Z Panel in Limerick on 6th September 1995. * if agreement is reached, take Version 1.2 to the SC22 meeting in Annapolis 18-22 September 1995. I believe it is a appropriate action to be taking after the substantial work that has taken place on the development of the proposed standard. The meeting endorsed this plan, which represents the current working plan for development of the CD. Further actions on the proposal to register a CD will be reported later. THINGS TO DO FOR VERSION 1.2 Immediately after the meeting, a small ad hoc group developed the following list of things to do for the next version. It is of course not complete, nor does it represent a commitment to do them, but it gives the idea of what was being discussed. 1. Correct typos throughout document 2. Resolve differences between concrete and abstract syntaxes (e.g. where they don't agree on schema) 3. Resolve abstract syntax problems with substitution - we need two kinds of \odot 4. Remove pred expr from concrete syntax. 5. Provide semantics for generics. 6. Free types (only in concrete syntax for now). 7. Toolkit: add omitted definitions. 8. Clarify fixity definitions. 9. Make typeface conventions more consistent and document the typographic conventions. 10. Review Editor's notes (i.e. remove redundant ones, change some to Informative Notes) 28.15. MODULES PROPOSAL A short discussion on the Modules proposal (document mod.ps) was held and it was decided to extend this discussion at the next meeting, with Rob Arthan chairing the discussion. It would be possible to discuss the current proposal, a simple extension to existing Z, with Ian Hayes, co-author of a paper on modules at the 1992 Z User Meeting. [Note added in minutes: Since this meeting, Rob has indicated that he will bring forward a full language proposal for discussion at the Limerick meeting. This will be made available on the ftp server.] Action: Rob Arthan. 28.16. Subgroup organizers The following are organisers of subgroups and should be consulted about proposals and meetings: Semantics and logic: Will Harwood Syntax and tools: Will Harwood Toolkit: John Wordsworth Conformance: John Nicholls Object-oriented: Pete Young. Interchange format: Jonathan Hammond Dissemination: John Nicholls 28.17. SCHEDULE OF FUTURE MEETINGS Z Panel meetings (and related events) are scheduled as follows. The mark * denotes newly listed dates. Wednesday 6 September 1995 University of Limerick Ireland (9th International Z User Meeting, 7-8 September 95,Limerick Ireland) (SC22 Plenary: 18-22 September 95, Annapolis USA) (Z Twenty Years On, IRIN, 10-12 October 95, Nantes France) Tuesday 21 November 1995 BSI Headquarters, Chiswick London Tuesday 23 January 1996 Logica Cambridge Friday 15 March 1996 * Oxford (FME'96 - Formal Methods Europe, 17-22 March 96, Oxford) Tuesday 7 May 1996 * location to be decided Tuesday 9 July 1996 * location to be decided 28.18. NEXT MEETING Wednesday 6th September 1995 at 10:30 am. Location: University of Limerick, Ireland 28.19. CLOSE OF MEETING The meeting closed at 2:35 p.m. John Nicholls --------------------- end of z-180.txt ----------------------------- From John.Nicholls@comlab Tue Dec 1 10:51:52 1998 Date: Fri, 11 Aug 95 10:43:25 BST From: John.Nicholls@comlab To: A.Smith@gb.hmg.dra.green, Jim.Woodcock@comlab, Makoto.Someya@jp.co.unisys, S.H.Valentine@uk.ac.bton, an@jp.co.toshiba.rdc.isl, Andrew.Martin@comlab, baumann@ch.unizh.ifi, br@uk.ac.rl.inf, Stephen.Brien@comlab, cinzia@it.intecs.pisa, cjchamp@gb.hmg.dra.taz, cts@gb.hmg.dra.green, derek@uk.ac.le.mcs, drj@mil.ncsc.tycho, ecusack@uk.co.bt.fmg, gerhart@cl.uh.edu, ianh@cs.uq.oz.au, j.e.sinclair@uk.ac.open, jam@uk.ac.york.minster, jap@de.d400.uni-kiel.informatik, jarh@uk.co.praxis, jbwords@com.ibm.vnet, jgh@uk.ac.york.minster, Joan.Arnold@comlab, Steve.King@comlab, lupton@com.ibm.vnet.winvmd, maa@com.att.research, mark@ca.on.ora, mataga@com.att.research, mayers@com.tek.labs.poseidon, John.Nicholls@comlab, pjs%ist.co.uk@oxmail, pyoung@uk.co.bt.srd, rbj@uk.co.icl.win, rda@uk.co.icl.win, sjd@uk.co.icl.win, srlm@br.cnpq, susan@uk.co.logcam, trevor@uk.co.praxis, uchi@jp.co.toshiba.ilab.ap, w869cp@uk.co.bae.mal.wa, will%ist.co.uk@oxmail, wsinhee@nl.tue.win Subject: Correction to Minutes of Meeting 28 From: John Nicholls To: Members of Z Standards Panel Date: 11th August 1995 CORRECTIONS TO MINUTES OF MEETING 28 Rob Arthan has sent me the following note about the minutes of Meeting 28 (document 180). I aplogise for misunderstanding and misquoting him, and am glad to send this clarification. ------------------- From: Rob Arthan John, Quotation from Minutes: "Rob Arthan commented on the minutes of the meetings and said that in his view the separation is difficult and deep and would make a profound difference to parts of the language. There is an urgent need for resolution of the points made in recent meetings." I'm sure I didn't say this. None of the adjectives "difficult", "deep" or "profound" reflect my views about the problem in question. An outline of my current view (and I have held it for some time) is that there are that the semantics doesn't quite hang together and we need to think over some technical issues to fix this. I don't see this as theoretically problematic, nor do I think the decisions will make any significant difference to Z practitioners. Regards, Rob. ------------------ end of message ----------------------------