Re: [xcbf] Ballot - Will changes be possible later on?

[James Bryce Clark <jamie.clark@oasis-open.org>]

Let me responsd to several TC list questions about OASIS 
process

    1.  Under the OASIS process, a TC determines that
a specification is satisfactory and complete before voting on it, and
confirms it by that vote.  Quality control is the primary
responsibility of members.  Generally there is no external review
for QA or technical accuracy.   When OASIS is notified of a
successful TC vote, we immediately post the spec (as approved) as a
completed TC product, not as a draft.  
    2.  Even at the OASIS Open Standard level, which
requires a membership vote, OASIS review is quite limited.  QA is
for the team and, broadly, for the voting members to
evaluate.    
    3.  To address a question raised by Sig. Triglia,
a standard may not be modified during a vote (at either the TC or
member-wide level).  This is to avoid ambiguity about which artifact
is the subject of the pending vote.
    4.   Taking an approved committee
specification to the membership for approval as an OASIS Open Standard
requires some specific additional steps, named in the TC Process
document.  OASIS does evaluate those submissions for compliance at
that stage.  If the package is approved it is submitted for a
membership vote.  As Mr. Griffin's noted, there are two ways that
objections might be addressed at that stage.   First, if
(and only if) a spec receives sufficient positive votes for
approval as well as 0>n>10% negative votes, there is a
30-day review loop for the TC, in which the TC may (i) vote to decline
the comments, (ii) withdraw or (ii) make modifications and resubmit the
modified spec to a new membership vote.  Second, as noted,
the TC also may vote to withdraw a submission during that
membership voting period, but that power to withdraw expires when
the voting period is completed.  
    These rules are set out clearly in the short normative
"TC Process" document at
[http://www.oasis-open.org/committees/process.shtml].  
I encourage all parties to read them to their own satisfaction.

    Beyond the rules, permit me to point out a few
practical matters from experience.
    --  There are two reasons why most TCs seek rough
unanimity before going to a vote, and attempt to address bugs, stability
problems or fundamental technical differences up front.
One is that the TC Process makes dissent fairly easy.  A small
number of opponents can have significant effect.  The other is that
some of the required elements for OASIS Open Standard submission, such as
certifications of actual implementation,  are easier to obtain if
you have a stable consensual product.  
    --  Please note that, in an OASIS membership
vote, only voting members (e.g., not individual members) cast a
ballot.  The TC list is publicly archived, so voting members may if
they wish review your prior discussions.
    --  Bear in mind that the voting membership of
TCs, and the decisions made by those voting members, can change over
time.   

    There also were some comments on the list, which I
haven't entirely followed, about 'taking the XBCF work' into some other
body.  Please note that on that path the TC would encounter a
considerable additional set of issues about IP rights, OASIS approvals
and the like.  Submitting works from OASIS TCs to other bodies is
not a foregone conclusion, due to our own institutional obligations to
keep the work open, preserve the IP, etc.

Best regards    Jamie Clark

> Alessandro Triglia wrote:
> > Phil,  Are the paragraphs
> > below a correct interpretation of the OASIS rules?
> > -----------------
> > Once a CS is approved by the XCBF TC and submitted to OASIS for standard
> > ballot, it will be impossible to make any changes to it - not even
> > editorial ones - until after the end of the OASIS ballot.   * *
> > *  Therefore, any wished changes that are not applied before TC
> > approval of the CS must wait until the end of the OASIS ballot, and then
> > they can only be applied if the ballot ends with a sufficient number of
> > YES votes and some NO votes.  This is the only way and the only
> > circumstance in which the document can be amended. * * * Of course, I may
> > be misunderstanding the OASIS rules.
> > Alessandro

~   James Bryce Clark
~   Manager Tech Stds Dev, OASIS Open 
~   +1 978 667 5155 x 203 office
~   +1 310 293 6739 cel