Possible subcommittee reorganization

[jon.bosak@sun.com]

Hello UBL TC,

In Tuesday morning's TC plenary in Santa Clara, we discussed our
experience so far with the suspension of most of our UBL
subcommittees following the May 2004 meeting in Hong Kong.  I
think it's fair to say that most of us feel that folding the SC
work back into the TC for the last 6 months has proven a net
benefit, but not without its downside.  Those present in the
Tuesday discussion came to the conclusion that we might be able to
secure the benefits of SCs while ameliorating their major
disadvantages if we structured their relationship with the TC in
light of lessons learned over the last three years.  I've
scheduled a further discussion of this for our Thursday afternoon
plenary call (Friday morning in Asia); please review the
outcomes of our Tuesday meeting below and be ready to contribute
comments by mail or join us for the discussion on Thursday.

Jon

==================================================================

Advantages of standing SCs

   Persistent mail archive and document archive

   Web page provides one-stop access to the work of the SC

   Tribal group dynamic leads to more investment in work,
   increasing focus and participation

   Possibly better fit with current OASIS direction with regard to
   form of organization

   Attractive working environment for people interested in a
   particular technical aspect but unwilling to participate in the
   policy and coordination work of the TC or unavailable for TC
   meetings (and the proposed revision to the OASIS TC process
   finally gives us officially recognized membership categories
   that allow us to have voting members of SCs who are not voting
   members of the TC)

   More direct and visible responsibility for the maintenance of
   works in progress

   Formal role for co-chairs, making consistent co-chair
   participation and travel easier to justify to member management

   Less dependence on TC for direct management of work teams

Disadvantages of standing SCs

   Much more overhead (though it's an open question how much of
   this would be required no matter how the group was structured
   if it were to be managed properly)

   Tribal group dynamic leads to territoriality and lack of
   communication with other SCs

   Possible failure of the TC to take responsibility for important
   recommendations of SCs, allowing decisions of the SCs to stand
   without sufficient review or buy-in by the larger group

Key question: Is it possible to structure the TC and its
subcommittees in a way that achieves the advantages of standing
SCs while overcoming the disadvantages?

==================================================================

Possible way to overcome SC disadvantages

 - The TC takes more responsibility for SC functioning, defining
   work lists and schedules for the SCs

 - The TC may form ad hoc work teams whose management is assigned
   to given SCs

 - SC schedules feature frequent milestones (monthly under the
   plan being considered)

 - SCs mail one-paragraph summaries to the TC each week, and the
   TC agenda features a standing slot for optional commentary upon
   or discussion of that week's SC work

 - Deliverables of SCs are recommendations to the TC, not
   decisions made independently of the TC

 - The ideal model is one in which each SC brings a question to
   the TC every month along with enough information and documented
   positions to enable the TC to make a decision

==================================================================

Possible SCs; note that these are intended to reflect the physical
division of participants in f2f meetings:

   DSC (Deployment SC)

      Scope: Education, documentation, interoperability, human
      interface (including input and output mappings)

      Document responsibility (continuing development, 1.0 issues
      list, 1.0 errata list, white papers, 1.1 issues list):
      Interface specs, FAQ, educational materials, implementation
      guides, programming aids

   BSC (Business SC)

      Scope: Data models and spreadsheets

      Document responsibility (continuing development, 1.0 issues
      list, 1.0 errata list, white papers, 1.1 issues list):
      Spreadsheets, EDIFIX data model

   XSC (XML SC)

      Scope: Schemas, code lists, extension methodology

      Document responsibility (continuing development, 1.0 issues
      list, 1.0 errata list, white papers, 1.1 issues list): NDR,
      Code List spec, Customization Guidelines

   LSCs

      Scope: Translation, regional deployment, regional
      requirements

      Document responsibility (continuing development, issues
      lists, errata): Translated documentation and data dictionary

==================================================================

Weekly TC meeting schedule under this plan

   Atlantic TC calls take place in the usual slot, shifted one
   hour to 16h00-18h00 UTC/GMT for the winter in order to maintain
   08h00-10h00 in San Francisco and 16h00-18h00 in London.

   Pacific TC calls take place in the usual slot, 00h30-02h30
   UTC/GMT, which in the winter will be 16h30-18h30 in San
   Francisco while staying at 08h30-10h30 in Singapore, Hong Kong,
   Perth, and Beijing and 09h30-11h30 in Tokyo and Seoul.

   Calls are structured as follows:

      First 0.5-1.0 hour: Standing TC items: Calendar, liaison
      reports, scheduling, status reports of teams managed
      directly by the TC (if any)

      Remainder of meeting: Devoted to that month's
      recommendations from the "subcommittee of the week" (one of
      DSC, BSC, XSC, or LSCs), conveyed by at least one co-chair
      of the subject SC(s) accompanied by other interested members
      of the subject SC(s) in that time slot

   Note that it will be necessary under this scheme for each SC to
   have an Atlantic co-chair and a Pacific co-chair, or better yet
   one chair (probably based on the U.S. west coast) who can
   participate in both time slots.

   The SCs would meet on their own different schedules once a week
   in the usual case, or split into Atlantic and Pacific calls if
   their members so desired.  SC timelines would be specified when
   possible in terms of monthly milestones representing decision
   points for the TC that align with the monthly cycle of formal
   SC reports.

==================================================================

Disposition of existing ad hoc teams

   Under the proposal floated above, we would make HISC a team
   under DSC and assign management of the Code List team to the
   XSC and management of the Tax XML and TBG17 teams to the BSC.

==================================================================

Gating items and risk factors

   Gating item: people willing to co-chair each of DSC, BSC, and
   XSC in such a way that each SC has at least one co-chair
   available to cover the monthly Atlantic call and at least one
   co-chair available to cover the monthly Pacific call.

   Risk factors: failure of the TC to prepare for a decision on
   the SC recommendation to be considered each week.

==================================================================

Here is what our meeting schedule for the next few weeks might
look like if this plan is adopted:

2004.11.10/11

   No meetings this week; JB to formalize SC plan with straw
   charters and worklists and submit it to the TC for discussion
   by mail

2004.11.17/18

   Mavis chairs Atlantic, Tim chairs Pacific; agenda: standing
   items; refine charters and worklists for new SCs

2004.11.24/25

   Mavis chairs Atlantic, Tim chairs Pacific; agenda: standing
   items; TC triage of proposed 1.1 items from worklist

2004.12.1/2

   Return to usual meeting management; agenda: standing items;
   finalization of detailed SC plan (including co-chair
   assignments, team leads, etc.); further discussion of proposed
   1.1 items from worklist; afterward, submission of SC plan for
   final TC approval by mail

2004.12.8/9

   Agenda: standing items; SC scheduling; further discussion of
   proposed 1.1 items from worklist; afterward, request for
   formation of new SCs to OASIS

2004.12.15/16

   Agenda: standing items; final discussion of proposed 1.1 items
   from worklist; final work schedule for 2005

2004.12.22/23

   No meeting

2004.12.29/30

   No meeting

2005.01.5/6

   First TC meetings under new structure

==================================================================