From John.Nicholls@comlab Tue Dec 1 10:50:38 1998 Date: Thu, 2 Feb 95 21:09:16 GMT From: John.Nicholls@comlab To: Jim.Woodcock@comlab, Makoto.Someya@jp.co.unisys, alf@gb.hmg.dra.green, 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, 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, jbwords@com.ibm.vnet.winvmj, jgh@uk.ac.york.minster, Joan.Arnold@comlab, Steve.King@comlab, lupton@com.ibm.vnet.winvmj, maa@com.att.research, mark@ca.on.ora, mataga@com.att.research, mayers@com.tek.labs.poseidon, John.Nicholls@comlab, pjs@uk.co.ist, pyoung@uk.co.bt.srd, rbj@uk.co.icl.win, rda@uk.co.icl.win, rose@cs.uq.oz.au, sjd@uk.co.icl.win, srlm@br.ufpe.di, susan@uk.co.logcam, trevor@uk.co.praxis, uchi@jp.co.toshiba.ilab.ap, w869cp@uk.co.bae.mal.wa, will@uk.co.ist, wsinhee@nl.tue.win Subject: Minutes of Meeting 25 (ascii text version) Z STANDARDS PANEL BSI PANEL IST/5/-/52 DOCUMENT 170 ISO PANEL SC22 WG19 (Document Z-170.TXT printed 2 February 1995) ------------------------------------------------------------------- MINUTES OF 25TH MEETING, 24TH JANUARY 1995 [Note: The minutes incorporate revised versions of the proposals circulated to members of the Panel prior to the meeting, in the document CDPLAN] The 25th meeting of the Z Standards Panel was held on Tuesday 25th January 1995 in the Wolfson Building, Oxford Computing Laboratory. Present: Stephen Brien (morning), Rodger Collinson, Jon Hall, Steve King, Trevor King, Peter Lupton, Andrew Martin, John Nicholls (Convener), Jane Sinclair, Alf Smith, Pete Steggles, Susan Stepney, Jim Woodcock, John Wordsworth, Pete Young. Apologies: Rob Arthan, Will Harwood, Sam Valentine. The agenda was agreed. 25.1. ACTIONS ARISING FROM MINUTES OF MEETING 24 (DOCUMENT 168) Action 23.3 Investigation of ftpmail. Notes on the use of ftpmail access to the Oxford comlab server to be prepared. Action pending. Action 23.3 Information on other ftp sites. Action pending Action 24.2 Mode of working. No comments have been received on the proposal. Hence the proposed Mode of Working is agreed. The statement is: Subgroups meet, discuss issues, and prepare proposals for presentation to the Panel. Initial proposals may be made in the form of objectives, statements of principle, etc. When they have been prepared and circulated as proposed drafts of the current Working Draft, the Panel will review and comment, and may finally approve their acceptance in the Working Draft. Once proposals have been accepted and incorporated in the WD, they are subject to modification in the usual way, by the raising of issues and submission of proposals. Action 24.4.1 Syntax and tools subgroup. A marked up version of Annex B is to be prepared (see below). Action pending. Action 24.4.3 Toolkit Group. The meeting has been held. Action complete. Action 24.4 Revision of Sections 2,3, and 4. This work is continuing Action pending. Action 24.5 Plan for CD. This is described below. Action complete. Matters arising: John Nicholls reported that, following a message from Gordon Rose about the minutes of the previous meeting (item 24.4.5), he had agreed with Gordon that the use of the term "Object Z" is misleading. The Queensland group use the name Object-Z for their object-oriented extension to Z. The title of the paper by Stephen Brien (which does not refer to Object-Z) will be changed to: "Object-oriented extensions to Z and their relationship to the Z Standard". Action: JEN to change title of Document 166 in ftp server JBW to change title in list of documents. The minutes were approved. 25.2 REPORT FROM MATHEMATICAL TOOLKIT SUBGROUP John Wordsworth described the subgroup meeting held on 17th January, at which the latest revision of the paper by Sam Valentine (circulated to subgroup members) was discussed. The proposal involves an extension of the Toolkit, filling out the mathematical definitions (including the provision of a full set of arrows) and providing axioms for a general type, number, with nested subsets: Rational, Reals and Integers. The result is a much enlarged Toolkit, with 38 new functions/relations, 16 changed and 2 deleted. A difficulty has been found with the definition of decimal numbers, to allow for the positional significance of digits - for example, at the moment, the number 1234 is regarded as a string. How should the Toolkit be regarded? One view, held by some members of the Toolkit subgroup and others, is that "true" Z is the language sans Toolkit and that the Toolkit should not be regarded as an integral part of Z, but a definition, expressed in Z, of a useful set of relations, functions etc. However, many Toolkit operators are widely used, and need to be taught to new users of Z. The separatist view makes it difficult for newcomers to understand the notation. Why are some things in Z and others, apparently equally important, in a Toolkit? Historically, things may be in the Toolkit because they can be defined in Z - putting them in the Toolkit allows us to have a simpler base language. There is a further difficulty with the separatist view, that the soundness of the logic, necessary for proofs in Z, needs to be established for the core notation. A much enlarged Toolkit may make this much more difficult. Jim Woodcock suggested that the subgroup should consider how to structure the Toolkit so that it would be easy to see how to state conformance of tools. For example, it might be divided into two parts, a "traditional" part and a new part. It should be possible to conform to the whole Toolkit or just the traditional part. Jim was also concerned about teaching Z to new users and explaining the difference between the Toolkit and other parts of the notation. There are also issues of character set support, not difficult for those working in some environments, which may become more complex when the use of UNICODE or ISO 10646 is considered. Further action: The Subgroup will meet again on 15th February in London. Comments, please to John Wordsworth (copy John Nicholls) on the issues mentioned above, and any others on the Toolkit, before the meeting. After the meeting, the revised paper will be distributed to all members. Action: JBW to call meeting on 15th February. 25.3 REPORT FROM SYNTAX SUBGROUP Syntax: The revision of the Syntax proposal by Peter Steggles and Will Harwood (document 169) was introduced by Pete Steggles. Document 169 contains a proposed new syntax, together with explanations and a rationale for decisions made in its development. The introduction of templates is a proposed addition to the language - one of its purposes is to simplify the syntax. The main part of Peter's presentation was a discussion of a related paper by Ian Toyn (University of York), discussing the template proposal. Ian has implemented this proposal and has pointed out improvements and corrections, which have been incorporated in the document. The Ian Toyn paper is available in compressed postscript format (9 pages of A4) from the FTP server ftp.cs.york.ac.uk as /hise-reports/cadiz/operators.ps.Z. Lexis: Susan Stepney reported that following reviews of the Lexis paper presented at the last meeting, a number of minor changes have been made. The revised document, 167 v2, is enclosed with this mailing. The chairman asked all members present at the meeting for their opinion on the proposals. There was general support for the directions indicated in the papers. The next stage is to circulate documents 169 and the revised Document 167 to all members, and request review and comment on technical direction. If a generally favourable response is received, the next stage is the construction of a replacement for Annex B. Copies of the syntax and lexis papers are enclosed with this mailing (and are on the papers subdirectory of the ftp server). Panel members are requested to send comments to Will Harwood (copy to John Nicholls) by the 27th February. Action: All members to comment on syntax and lexis papers by 27th February. 25.4 REPORT FROM SEMANTICS SUBGROUP No meeting has been held, but a meeting is planned for 9th February 1995 in Oxford, where a plan will be made for future work. Enquiries to Will Harwood, subgroup leader. Jim Woodcock will give an introductory talk on the approach being taken on the semantics at the X3J21 meeting in Pittsburgh in April. The material developed for this talk will form a basis for the text in the informative Annex, providing a tutorial on the semantic definition. 25.5 REPORT FROM CONFORMANCE SUBGROUP No meeting has been held. John Nicholls circulated a copy of the revised Conformity statement for VDM-SL, the result of a major revision. A version of this will be placed on the ftp server, in the papers subdirectory. John Nicholls asked all subgroup leaders to consider questions of conformance as they apply to the sections they are developing. 25.6 REPORT FROM OBJECT-ORIENTED SUBGROUP No meeting has been held. 25.7 REPORT FROM INTERCHANGE FORMAT SUBGROUP No meeting has been held. 25.8 PLAN FOR PREPARING A NEW VERSION AND A DRAFT CD Restructure for ISO/BSI requirements There are guidelines and requirements for the format and presentation of ISO (and BSI) standards. We have followed these fairly closely, but the current document (Version 1.0x) will need some revision to bring it into line. One source of information (for BSI standards - there are similar needs for ISO) is the following: BS 0 : Part 3 : 1991 A standard for standards: Part 3. Guide to drafting and presentation of British Standards. One change we will need to make is to make a clearer distinction between normative (i.e. defining) and informative definitions and text. I will propose some ways of indicating this distinction. Proposed CD structure The following is a proposed new structure, following the examples and requirements listed above. Sections are marked N (normative) and I (informative). Section headings in the main part of the definition, from Semantic Metalanguage onward, including Annexes, are based on the latest version of the Working Draft. Front and inside front cover I Contents I Foreword I Introduction I Scope N Normative references N (Definitions) N a possible section (Symbols and abbreviations)N a possible section Compliance criteria N Semantic metalanguage N Semantic universe N Language description I or N Name N a proposed new section Expression N Predicate N Schema N Paragraph N Specification N ANNEXES Abstract syntax N Representation syntax N Mathematical toolkit N Z Interchange format N Z character set N A deductive system for Z I or N status (I or N) for discussion Tutorial on Semantic method I a proposed new Annex Informative References I Glossary I Index I Revision and completion of normative definitions ------------------------------------------------ The completion and revision of these is the highest priority. We have subgroups discussing several of the syntaxes and the semantics, and once these revisions are complete, we will need to work on the main definitional sections: Expression, Predicate, Schema, Paragraph and Specification. The following have agreed to draft and edit specific sections of the text: Section Author/Editor Front part John Nicholls Introduction John Nicholls Scope John Nicholls Normative refs John Nicholls Compliance John Nicholls Semantic metalanguage Randolph Johnson Semantic universe Randolph Johnson Language description Randolph Johnson Name Expression Predicate Schema Paragraph Peter Lupton Specification Abstract syntax John Nicholls Representation syntax Will Harwood et al Mathematical toolkit John Wordsworth Z Interchange format Trevor King Z character set A deductive system Andrew Martin Tutorial on Semantic method Jim Woodcock Informative refs John Nicholls Glossary Steve King Index Steve King The following have offered to provide help in drafting and editing: Rob Arthan Rodger Collinson Jon Hall Jane Sinclair Responsibilities are as follows: Section Draft and update source text for editors the nominated section. Project Formatting and layout. editor(s) Control of source text. Final editing. Attribution The following is proposed: Section authors and editors will be listed on the title page. The following will be listed in the front part of the document: * members of the Panel (current members at the time of publication of a draft). * contributors to the work * a list of references to published work. An example of planned attribution is given in the enclosed sample pages. Copyright Copyright of the Standard will have to be assigned to ISO when a standard is agreed. I propose the continuation of the existing copyright notice. If any writer or editor has difficulty with this, please let John Nicholls know. IEC directives The following is an extract from IEC directives for the Technical Work of JTC1: 6.5.1 Preparation of committee drafts 6.5.1.1 The WD can be considered as having reached the stage of committee draft when: the main elements have been included in the document; it is presented in a form which is essentially that envisaged for the future International Standard; it has been dealt with at least once by JTC 1 or by a working body of JTC 1; JTC1 or one of its SCs has decided in a resolution during a meeting or by letter ballot that the WD be forwarded to the ITTF for registration as a CD. The JTC 1 or SC secretariat then forwards a copy of the WD in question to the ITTF which registers it as a CD and allocates a serial number to it (see Annex D). (The work has now reached stage 3). The number will remain the same throughout the reporting stages and for the published International Standard. Timetable for CD John Nicholls reported discussion of the timetable with the leader of the ISO secretariat for SC 22 who indicated that the following is needed: one month for preparation and distribution of the document. three months for the ballot period. During this period of four months, we can continue with completion of informative sections and completion or improvement of submitted text. ISO have indicated that it is not necessary to have such informative parts as an index for this submission. 25.9. TARGET DATE FOR COMPLETION OF CD INPUT In view of the activity needed and available resources, it was agreed that the following target be established: Completion of a draft suitable for submission to ISO for distribution and ballot: FRIDAY 30TH JUNE 1995 It is most important that all members consider the implications of this date. Those involved in drafting and editing sections of the document should give priority to completing those parts of the WD that need to be completed to meet the target. Those reviewing the document should consider what items are needed for their own support of the ballot for CD. This topic will be reviewed at the next meeting. Please send comments on the plan and target date before the next meeting. Action All members to comment on plan and schedule for CD development. 25.10. PROPOSAL FOR A HUMAN-ORIENTED ASCII VERSION. The proposal by Mark Saaltink was welcomed in principle. More detail is needed and Mark is invited to submit a more detailed proposal, with perhaps a detailed statement of design principles and a sample set of proposed. One source of ideas may be the similar approach taken in the B-Tool. Action: Mark Saaltink to be invited to prepare a more detailed proposal. 25. REPORTS FROM OTHER GROUPS Report from Randolph Johnson: On January 8-9 the X3J21 committee met in San Diego and accomplished substantive work on four topics. We discussed and approved, subject to minor adjustments in the wording, a proposal that the Standard should include a standard human oriented ASCII representation for Z which conforming tools would be expected to support for input and output. We also finished our work on the content of a proposal to add to UNICODE some 14 characters need for standard Z or the example mathematical toolkit which are not currently in UNICODE. (We did not include additional characters used in the metalanguage of the Standard.) We discussed the comments Mark Saaltink sent to you about the standard. We made plans for an extended meeting at the Software Engineering Institute in Pittsburg from 30 March through 3 April. This will include 1.5 days for a technical presentation by Peter Gorm Larsen on the VDM-SL standard, 1.5 days for our actual committee meeting, and 1.5 or 2 days for a technical presentation by Jim Woodcock on the Z Standard. Following meetings are in Winnipeg in July, Austin in October, Hawaii or Florida in January '96, and Dearborn, Michigan in April '96. 25.12. NEW MEMBERS The following new members are proposed: Andrew Martin (Oxford University PRG) is taking over a project looking at the logic underlying Z, and is developing a proof tool for the Z logic in the standard. Peter Mataga is proposed to replace Mark Ardis, representing AT and T Bell Laboratories. Peter will be known to members for his work with Pamela Zave on the use of Z for a telephone system, presented at the last Z User Meeting in Cambridge. Mark Saaltink is with ORA in Ottowa Canada and is known for his work with Z and other formal methods. Mark presented a paper at the 1991 Z User meeting on the adaptation of the Z Toolkit for use with Eves. The new members were proposed as follows: Andrew Martin: Proposed by Jon Hall, seconded by Jane Sinclair; Peter Mataga: Proposed by John Wordsworth, seconded by Pete Young; Mark Saaltink: Proposed by Alf Smith, seconded by Rodger Collinson. All were unanimously accepted and the chairman formerly welcomed them as new members. The thanks of the Panel are extended to Mark Ardis for his work on the Panel. 25.13. TITLE OF DOCUMENT The current title, Z Base Standard, was taken over from the ZIP project. It does not make much sense in the present context. When it becomes a standard, the word standard is superfluous, and the word Base does not mean much. That leaves Z, Z Notation, . . . Ideas please, for the next meeting. Action: All members 25.14. SUBGROUP MEMBERSHIP The following is the current list of subgroup members. Conveners are indicated in upper case. Those willing to work within subgroups should notify the relevant convener. Members who cannot attend meetings may take part by email, phone or other means. Semantics and logic: Rob Arthan, Peter Baumann, Randolph Johnson, WILL HARWOOD, Jon Hall, Peter Lupton, John Nicholls, Sam Valentine, Jim Woodcock. Syntax and tools: WILL HARWOOD, Colin Parker, Sam Valentine, Susan Stepney, Pete Young. Toolkit: Alf Smith, Susan Stepney, Sam Valentine, JOHN WORDSWORTH. Conformance: JOHN NICHOLLS, Colin Parker, Sam Valentine. Object-oriented: Jane Sinclair, Susan Stepney, PETE YOUNG. Interchange format: TREVOR KING, John Nicholls. 25.15. SCHEDULE OF FUTURE MEETINGS Z Panel meetings are scheduled as follows: 21st March 1995 IBM South Bank, London (2-3 April 1994, X3J21 meeting Pittsburgh USA) 23rd May 1995 Open University, Milton Keynes 25th July 1995 Oxford 6th September 995 Limerick Ireland (SC22 Plenary: 18-22 Annapolis USA) September 1995 21 November 1995 location to be decided 23 January 1996 location to be decided 25. NEXT MEETING Tuesday 21st March 1995. Location: IBM South Bank Building, London. 25. CLOSE OF MEETING The meeting closed at 3:30 p.m. John Nicholls Enclosed: Contact list dated 2 January 1995 by Joan Arnold Z Standard - Proposed Lexis by Susan Stepney, Document 167 (v2) dated 25 January 1995. Revised Syntax document by Pete Steggels and Will Harwood. Sample of front part of proposed new document, showing title, attributions, etc. ---------- end of Z-170.txt -------------------------------------