UBL Liaison SC wind-up meeting 29 March 2004

[jon.bosak@sun.com]

Hello UBL Liaisons and LSC observers,

This message is intended to serve two purposes: first, to remind
you that we have a regularly scheduled UBL Liaison Subcommittee
call this coming Monday morning 29 March at 7 a.m. California
time, and second, to follow up concretely on a suggestion made
during the LSC meeting 25 February that we wind up the LSC as
currently constituted and restart it after the TC meeting in Hong
Kong with a charter better suited to what we intend for UBL 1.1.

1. Our next UBL Liaison SC meeting will take place 07h00 (standard
   time in San Francisco) on Monday 29 March 2004.  This is 16h00
   (daylight time) in London, 17h00 (daylight time) in Copenhagen,
   and 23h00 (standard time) in Singapore, Hong Kong, and Perth.

   That's right -- London and Copenhagen switch to daylight time
   28 March, but San Francisco and other major cities in the
   U.S. make the change on 4 April.

   We will use this number for the call:

      #############################################
      U.S. domestic toll-free number: (866)545-5198
      Int. access/caller paid number: (865)521-8904
      Access code: 3470093
      #############################################

   Preliminary agenda:

    - UBL 1.0 progress

    - Liaison reports

    - Calendar review and speaker planning

    - Restarting the LSC

    - Other business

2. Here's what I intend to cover Under the agenda item "restarting
   the LSC."

   The UBL Liaison SC was set up in order to get input from
   various organizations interested in helping us to create a
   workable set of schemas for UBL 1.0.  I think that it has
   served this purpose quite well, particularly in the early
   stages of the effort, though there remain some incompletely
   digested items that will have to be followed up properly in the
   release later this year that we are calling UBL 1.1.

   It was understood from the beginning that organizations
   appointing liaisons to UBL were not committing to UBL adoption.
   But it seems to me that continuing to list all the current
   liaisons as active following the announcement of UBL 1.0 may be
   misunderstood as a formal endorsement of the work as finally
   released.  I would of course very much like to see such
   endorsements, but I don't think that this is the way to do it;
   I think it would be better to use this address list as a way to
   invite quotes for a press release later this spring after UBL
   1.0 passes through OASIS standardization.

   What I would like to suggest instead is that we wind up the
   current LSC, referring to it historically in the future as "the
   UBL 1.0 Liaison SC," and then figure out how to incorporate the
   liaison functions moving forward.  I can see several different
   ways to do this.

   The most straightforward continuation of the current
   arrangement would be to start a new subcommittee whose
   membership would consist of liaisons appointed by organizations
   with an active interest in using UBL 1.0 or in helping us
   advance toward UBL 1.1.  (UBL 1.1 is the name for a release
   planned later this year that would incorporate refinements to
   the data definitions based on the work of the UBL localization
   SCs and backward-compatible schema modifications that would
   incorporate further work on a number of deferred XML schema
   technical issues.)

   A reconstituted liaison subcommittee would have two main
   functions:

    - Represent industry data standards organizations intending to
      create industry-specific profiles of UBL in the 1.1 time
      frame.

    - Maintain relationships with organizations sharing a desire
      to coordinate as closely as possible in the future evolution
      of UBL.  I'm hoping that this would include our liaisons
      with as many of the organizations currently participating as
      possible, but I don't want to make any assumptions about
      this.  Such liaisons would serve principally to convey
      scheduling information to UBL and to report on UBL progress
      back to the participating organizations.

   It's an open question in my mind as to whether these two
   purposes are best served by one subcommittee or two.  My
   inclination is to prefer to see both sets of interests
   represented in one set of meetings, so I would prefer at this
   point to see a "UBL 1.1 liaison SC" not split along functional
   lines.

   But there is another axis to consider here: time zones.  We
   have demonstrated that it is virtually impossible to hold one
   set of meetings that can be attended by everyone whose
   organization might be interested in UBL.  This consideration
   gains force if (as I expect) the groups interested in defining
   industry profiles of UBL are organized regionally.  This
   thought leads to the notion of two sets of liaisons that would
   meet by phone as "the UBL Atlantic Liaisons" and "the UBL
   Pacific Liaisons."

   This is in line with what we have attempted with some success
   to accomplish with "Atlantic coordination calls" and "Pacific
   coordination calls."  A similar "hemispherical" organization is
   beginning to emerge with regard to the UBL localization SCs,
   which until now have all been centered in Asia but with the
   recent addition of the ESLSC are going to extend across Europe
   and the Americas as well.

   The daunting prospect of actually trying to organize and run
   all these calls leads me to a rather radical solution that I
   would like the current LSC to consider before I take it to the
   TC.

   There appears to be general support for the idea that we will
   have to reduce the number of UBL SCs following the release of
   UBL 1.0.  The ideal would be to do all our business in UBL
   plenary TC calls; if we did that, we wouldn't have all the
   communication problems that we have been experiencing as we try
   to reach agreements that go beyond individual SCs.  There are
   two things that prevent the adoption of this obvious solution:
   first, OASIS TC rules that we cannot change require a majority
   of the entire TC in order to constitute a quorum; and second,
   we cannot get everyone into the same time slot.

   Traditionally, the quorum problem has been solved in
   organizations like ours by lowering the percentage of voting
   members needed to constitute a quorum.  It is quite common
   among societies with a distributed membership to set quorums as
   low as 10 percent of the members.  We cannot do this for the
   TC, but we certainly could do it for specific SCs.  So here is
   a plan that I'd like us to talk about in the call on Monday:

    - Do away with the idea of liaisons as a separate subcommittee
      and just appoint liaisons to the UBL TC itself.

    - Form two "coordination subcommittees": an Atlantic
      Coordination Subcommittee (ACSC) and a Pacific Coordination
      Subcommittee (PCSC).

    - Make the voting membership of both the ACSC and the PCSC
      identical to the voting membership of the whole TC (so any
      voting TC member can participate in either SC), but lower
      the quorum requirement to 10 percent.

    - Hold weekly ACSC and PCSC meetings on the same day in
      California, for example the ACSC at 8 a.m. on Monday in San
      Francisco and the PCSC that same day at 4:30 p.m. in San
      Francisco (which is at reasonable times Tuesday morning in
      most of Asia and Australia).  These times have worked out
      well when we've tried this in the past.

    - Make sure that any important issues are properly noticed in
      meeting announcements so that everyone with an opinion will
      attend one of the two coordination meetings.

    - Perform all management business in these two weekly
      meetings: liaison reports, calendar planning, plenary
      meeting schedules, cross-SC queries and issue tracking,
      translation issues, everything.

    - Attempt to achieve a consensus on all major issues in both
      *CSC meetings and secure formal TC approval by posting the
      outcomes by mail and relying on common consent to move
      forward.  Then the only issues needing resolution by the TC
      would be issues in which consensus proved impossible to
      achieve in the weekly *CSC meetings.

   Please think about this and be ready to discuss on Monday.
   Other ideas are most welcome.

Jon