JTC1/SC22 N2531

Date: Mon, 28 Jul 1997 16:12:54 -0400 (EDT)
From: "william c. rinehuls" <[email protected]>
To: [email protected]
Subject: SC22 N2531 - Minutes of JSG June 30, 1997 Meeting

_______________ beginning of title page ________________________________
ISO/IEC JTC 1/SC22
Programming languages, their environments and system software interfaces
Secretariat:  U.S.A.  (ANSI)



ISO/IEC JTC 1/SC22
N2531



August 1997



TITLE:
Minutes of SC22/Java* Study Group (JSG) Meeting on June 30, 1997 in
London, United Kingdom



SOURCE:
Secretariat, ISO/IEC JTC 1/SC22



WORK ITEM:
N/A



STATUS:
N/A



CROSS REFERENCE:
N/A



DOCUMENT TYPE:
JSG Meeting Minutes



ACTION:
To SC22 Member Bodies for information.


* Java is a trademark of Sun Microsystems


Address reply to:
ISO/IEC JTC 1/SC22 Secretariat
William C. Rinehuls
8457 Rushing Creek Court
Springfield, VA 22153 USA
Tel:  +1 (703) 912-9680
Fax:  +1 (703) 912-2973
email:  [email protected]

_________________ end of title page; beginning of minutes _______________

              SC22 Java(TM) Study Group (JSG) Meeting Minutes

                         BSI, London, U.K.

                           June 30, 1997


Java is a trademark of Sun Microsystems


Monday June 30

1. Commencement

The meeting commenced at 9:30 with the convener, Bob Mathis, as chair. 

2. Local Arrangements

Bob introduced Jean Stride, BSI meeting coordinator, who explained the
local facilities.
 
7. Introduction of Attendees

The following people were in attendance:

Convener SC22 JSG
        Bob Mathis
Austria
        Martin Schoenhacker (HoD)
Canada
        Tim Boreham
        Simon Tooke (HoD)
Finland
        Petri Mahonen (HoD)
France
        Pierre Parquier (HoD)
Japan
        Toshiaki Kurokawa (HoD)
        Shinichi Niho
Norway
        Magnus Alvestad (HoD)
U.K.
        Ian Brackenbury
        Mike Curtis
        Alex Fiennes
        Francis Glassborow
        Lois Goldthwaite
        Kevlin Henney
        Derek Jones (HoD)
        Neil Martin
U.S.
        John Benito
        Frank Farance
        Gary Fisher
        Rob Gingel
        Roger Golliver
        Rex Jaeschke
        Scott Jameson (HoD)
        Dmitry Lenkov
        Randy Meyers
        Kevin Miller
        Tom Plum
        George Willingmyre

        Other U.S.

        Pete Smith
        Fred Tydeman
        Philip White
SC24 Liaison
        David Duke
        Ivan Herman
        Jean Stride
CEC Liaison
        Ken Thompson

3. Approval of Agenda

We will need a drafting committee to work on any resolutions.

4. Appointment of meeting secretary

Rex Jaeschke volunteered to take the minutes.
 
5. Review of previous minutes

The minutes from the Cupertino, California, meeting held in January 1997
(mail message 273) were accepted without change.

6. Review of Action Items

All were completed or withdrawn as follows:

Rex Jaeschke
  * Distribute the meeting minutes - done
  * Refine the technical issues list procedures - withdrawn
Bob Mathis
  * Forward our statement to JavaSoft - done
  * Establish a date and host for an April meeting - withdrawn
  * Respond to MHEG - done
  * Respond to ECMA TC39 - done
  * Write a short summary of the meeting for general release - done
Derek Jones IST5-53, UK Java Panel
  * Investigate BSI's hosting of a June meeting in the UK - done

8. Reports of National Member Bodies

8.1. Heads of Delegation present

        Austria         Martin Schoenhacker
        Canada          Simon Tooke
        Finland         Petri Mahonen
        France          Pierre Parquier
        Japan           Toshiaki Kurokawa
        Norway          Magnus Alvestad
        U.K.            Derek Jones
        U.S.            Scott Jameson

8.2. Austria - Voted YES. We voted Yes on accepting them as PAS submitters,
since we are willing to look at any proposed standards they wish to
submit.

8.3. France - Voted NO with comments. Prefer that the various parts of
Java standardization be done in the same group, however, if done in
separate groups they should meet together. Want the name `Java' in the
name of the standard(s). Having Sun do the maintenance is unacceptable.
Prefer a more traditional approach to standardization such as the
NWI-FCD process.  

8.4. USA - A new technical committee, J22, was formed in May as the U.S.
TAG. (see email 439) The US voted NO with comments as reported on the
reflector.
 
8.5. Canada - Abstain with comments.

8.6. U.K. - no vote has been taken yet. No indication given on what that
vote might be.

8.7. Japan - The vote will be taken today.  Recommendation has been YES
with comments.

8.8. Finland - Probably NO vote.

8.9. Netherlands - Unofficially reported that a NO vote is expected.

8.10. Norway - Voted NO with comments.

9. Liaisons

9.1. SC24

Ivan Herman reported that if the core technology makes it through the PAS
phase, it is presumed that Sun might want to use that process for other
APIs including those of interest to SC24.

9.2. SC29

SC29/WG12 deals with multimedia and wants to incorporate references to
Sun's Java specification (as reported at the January 1997 meeting). They
want rapid action, so they want the PAS process to succeed.
        
9.3. European Community

Has no policy on Java specifically. They are very much in favor of an
international standards process that can respond quickly.
 
9.4. ECMA TC-39

See item 10 below.
  
9.5. IEEE Internet Practices Study Group

Nothing to report.

10. ECMA-262 (previously known as JavaScript/ECMA Script/LiveScript)

10.1. Report from ECMA General Assembly meeting

George Willingmyre reported on the General assembly held last week.

Currently called ECMA Script with a final name yet to be determined,
hopefully at the July '97 meeting. The lack of a name is holding up the
formal process. The document won't be forwarded to JTC1 until after
September '97.

The document is done and available for download, from www.ecma.ch.

10.2. JSG Reaction/Discussion/Action

The issue of conformance arose and there was some discussion.

10.3. Technical issues

None

10.4. Recommendations to SC22 & JTC1

The drafting committee was asked to work on wording for JSG's continued
existence, the handling of ECMA-262, and the handling of Java.

10.5. Future JSG activities in this area

The future of JSG will be determined at the August 1997 plenary. JSG or
its successor should decide whether ECMA-262 is of interest to it and if
not, where responsibility for it should lie.

The U.S. can support that any ballot resolution be handled by JSG.
However, it does not support that the on-going work be assigned to
SC22/JSG.

Canada supports this and expects that JSG (or its successor) will be more
focused and that ECMA-262 may well be a distraction to JSG's primary
interests.

There were no objections to the positions put forth by the U.S.

The convener appointed Scott Jameson to head a drafting committee to draft
related resolutions.

11. Java-Related Issues

11.1. Should all Java-related standards activities be in a single group?
What are the implications of JTC1 reorganization, Sun's PAS application,
and activities of other groups?

There ensued a discussion of how best to carve up the tasks:

Keep language, core libraries, and JVM tightly coupled.  

Don't want separate standards for these unless they come from the same
committee; otherwise, problems there are problems with coordination (e.g.,
impact on JVM with respect to enhancements of the language).

JSG should state firmly that it wants to have these three.

We could debate this for a long time, however, the bottom line is that
it will depend on what Sun ultimately submits.

Can we afford to wait for Sun? In fact, might we inform Sun of what we
perceive the industry wants/needs?

The JVM needs to be targetable by languages other than Java, so perhaps
it should be dealt with separately from the language specification.

Considering the JVM as a general-purpose VM makes the problem more
complicated; perhaps we should start out thinking of it from a Java-only
perspective without making it more difficult for other languages to
target.

Yes, we should make a statement about possible organization.

11.2. Sun's proposal to be recognized as a PAS submitter

Should we make an official statement re this?

No, we've already indicated our respective positions via the National Body
votes. What more could we add?

If we all agree, it won't hurt to issue a statement; however, there seems
to be no benefit in doing so.

Maintenance must be handled by a standards group.

The name `Java' needs to be available to the community for use in
conformance testing.

11.3. Scope (from our perspective) of the Java standards effort

There was no discussion on this.

11.4. Liaisons with other JTC1 subcommittees and working groups

As in January, 1997, we are open to liaise with interested groups.
There's nothing more to do right now.

11.5. Immediate reactions to SC22, and thus JTC1, about Sun' application
to become a PAS submitter (ballot closes July 14)

We had no value-added contributions to make in this regard.

11.6. Contingency planning and discussion

A lengthy discussion ensued regarding as to what plans, if any, we can
make for future meetings. Clearly, our future relies heavily on the
outcome of the PAS submission vote and the content and completeness of any
subsequent technical submissions.

12. Other potential activities

12.1. Collection of technical contributions, error reports and/or
interpretations

There was no support for doing work along these lines.

12.2. Roly Perera's work (530)

Nothing to do here.

12.3. Conformance testing
        
The U.S. National Institute of Science and Technology (NIST) will host a 
Conformance Testing Workshop on September 11-12 (Thu-Fri), in
Gaithersburg, Maryland, USA.

12.4. JVM error reporting

There was a brief mention of the efforts by the University of Washington
in the U.S. with respect to their `High Integrity Java' efforts in
security testing.

13. Discussion of Report to the SC22 Plenary

The Convener's Draft Report (see email 516) was discussed with several
suggested improvements being made. It was decided to make this a report of
JSG as a whole rather than simply a convener's report.

14. Other Items

The following issues were mentioned in passing with no suggestions that we
address them further:  ANDF, ActiveX, AT&T Inferno, IBM's Universal
Virtual Machine, and Microsoft/Colusa OmniVM.
 
15. Next Meeting Planning

With so many variables, it was determined we could not make definite plans
for another meeting.
 
16. Resolutions

R-1: ECMA-262

JSG recommends that, if SC22 is assigned ballot resolution for the fast
track of ECMA-262, JSG serve as the ballot resolution group for the fast
track ballot.

R-2: ECMA-262

JSG recommends that, if SC22 is assigned responsibility for ECMA-262, SC22
rely on ECMA's TC39 for maintenance and revision, and not establish an
SC22 Working Group for this standard.

R-3: JTC1/SC22

Whereas JSG believes that the Java language, Java Virtual Machine, 
and core APIs are very interrelated, JSG recommends to SC22:

* That SC22 seek the responsibility for the initial PAS submissions based
on Sun's pending application.

* That JSG should be the ballot resolution group for Sun's initial Java
submission(s).

Whereas some National Bodies have requested that future maintenance of Java
be done within JTC1, JSG recommends to SC22 that JSG be assigned future
maintenance of any standard(s) assigned to SC22 based on these initial
submission(s).

R-4: JTC1/SC22

JSG recommends that SC22 continue the Java Study Group for another year.

R-5: Thanks to the host

JSG thanks BSI for their hospitality and meeting support.

17. Adjournment

The meeting was adjourned at 17:30.

__________________________ end of SC22 N2531 _________________________