011. Why individuals?

[Jon Bosak <bosak@boethius.eng.sun.com>]

I am convinced that the voting members of OASIS TCs should be
individuals, not organizations.  The logic is pretty simple.

   1. You can't prevent individual members from having votes
      in TCs, because then individual membership would be
      meaningless.

   2. You can't prevent organizations from stacking committees
      by sending individuals.  All you can do is to try as hard
      as you can to make sure that the individuals aren't
      useless as participants.

The simplest and strongest form of committee organization that I can
think of given these two realities is an organization that consists
entirely of individuals, regardless of how they managed to get their
dues paid and who cuts their paychecks.

It may be objected that this allows organizations to stack committees
by sending more members.  The only reply I can make to this is to
repeat point number 2, only more loudly: YOU CAN'T PREVENT
ORGANIZATIONS FROM STACKING COMMITTEES BY SENDING INDIVIDUALS.
Because if you say to me, BigShotCo, that I can only have so many
representatives, then I can set up a sponsorship fund through seven
layers of offshore holding companies that will send dozens of
promising young grad students to represent my interests as
individuals, and there's nothing you can do about it except write
increasingly more convoluted rules that I will continue to ignore.
Personally, I would rather have multiple company representatives out
where everyone can see them rather than attempt to implement toothless
but onerous rules that will only encourage the nefarious to go
undercover.

Note that there is nothing at all to prevent people working for OASIS
member organizations from having their own individual memberships.  In
fact, you have one such person already on this list: me.  I work for
OASIS member Sun Microsystems, but I have my own individual membership
(which happens to have been granted me in perpetuity in consideration
for the xml.org domain name, but which under other circumstances I
could have bought for $250 a year along with all the other individual
members).  So I'm not actually on this committee as a representative
of Sun Microsystems, although Eduardo Gutentag and Eve Maler are.  Do
you believe that the three of us are here by coincidence?  Does it
matter?  Given the experience that we bring to this exercise, wouldn't
it be strange if we could exercise only one vote among the three of
us?  And given that I've got my own membership, wouldn't it be two
votes among the three of us rather than one?  I don't know where
trying to anticipate every wrinkle in this would lead the rest of you,
but it would lead me right off the edge.

So how are we to prevent organizations from sending whomever they
want, whenever they want?  By recognizing that a stacked committee is
generally an acute rather than a chronic condition.  No sane
profit-oriented organization is going to fund participation by more
than one or two highly skilled employees on the off chance that the
extras might be needed for a close vote; that's not how it works.
Stacking is generally something that's done on special occasions, when
something of particular importance to an organization is on the
agenda.  The way to stop occasional stacking is by putting in place
rules that make occasional attendance impossible.  As it happens, such
rules are just the sort of rules we would want to have in order to
discourage dilettantes and troublemakers anyway, so two purposes are
served at once here.

It is very important to understand that when I say "participation by
individuals" or "criteria for voting membership of individuals" I mean
the word "individual" in an absolute sense.  I mean that
organizational affiliation really doesn't matter for purposes of
meeting the membership criteria in a committee.  In particular, I mean
that individuals cannot discharge their obligation to attend and
participate regularly by sending substitutes.  Individuals is
individuals.  Otherwise what I'm suggesting won't work.  It all hinges
on continuity of individual membership (including four face-to-face
meetings a year) and absolutely clear and mechanical criteria under
which members are given voting status and under which they will
automatically lose that status if they allow it to lapse.  The kind of
criteria I have in mind for voting status are discussed separately
under the heading "012. Criteria for voting membership."

I admit that this does not prevent an organization with the means and
determination to continuously fund a large number of active XML
experts from exerting disproportional influence in a TC.  But this
prospect is not going to keep me awake nights, for several reasons:

1. I don't think we can reasonably prevent it anyway.

2. The process is set up from the beginning to accommodate
   multiple solutions in the same space.  If company Y wants to
   create a TC to ram its DTD through, company Z can do the same
   thing with its own committee.  Big deal.

3. There is absolutely nothing to stop companies Y and Z from
   doing the same thing outside of OASIS, in which case they
   wouldn't have to put up with the annoyance of holding votes at
   all, let alone subjecting themselves to harangues from
   individual members riding theoretical hobby horses (which will
   certainly be a feature of the structure I have proposed), so I
   find the idea of a company bent upon world domination using
   this process a bit unlikely anyway.

4. A committee specification is NOT an OASIS standard (or
   whatever we decide we can call the thing that is approved by a
   vote of the OASIS membership).  This distinction is basic to
   the proposed process.

Jon