ISO/IEC JTC1/SC34
2006-05-22
Title:
Where to go after DIS 26300 – Cooperation?
Source:
Keld Simonsen
Status:
Expert contribution
Action:
For consideration by WG1 and SC34
Now that
the ODF standard has been approved as IS 26300, how do we proceed
with the next versions of the standard?
The DIS
ballot resulted in comments of about 20 pages, as contained in SC34
N0728. These comments needs to be addressed, and most likely the
final IS 26300 will be changed from the ballotted document. Thus
there will be two different standards for ODF, the OASIS standard and
the ISO/IEC standard. This is not at good situation.
What can
be done about this?
I see 3
scenarios:
Leave
it as is, and OASIS can go forward with the revision and submit
their standard as PAS for the next edition. This will result in a
number of different standards from OASIS and ISO, and make
implementers a bit confused on which standard to implement.
OASIS
can take the resulting ISO standard back in their process, and hope
that an OASIS ballot will not resolve in changes to the new OASIS
standard. This will establish one standard, but it is likely that
the changes to the 1.0 standard will be small, and that it will lead
to market confusion which version of the standard to implement. It
is also possible that the OASIS process introduces new changes.
Maintain
the standard jointly between OASIS and SC34. There is a model for it
in how the POSIX operating system standard is being maintained in
SC22, recorded in SC34 N0587. Here the participating standards
organizations are developing the standard together, so that there is
just one document maintained in a common group, called the ”Austin
Group” (from where the first meeting was held). The common
group consists of experts from each constituency and informal
decisions are taken a count of heads. If there are disputes that is
taken to the formal level in the common group, each constituency has
one vote, to be determined by the rules of the constituency. The
development and maintenance process is done in parallel in each of
the organisations, and according to the rules of the individual
organization. In ISO this is the normal 5 stage process. These
procedures ensures that the resulting final standard will be the
same in each organization, and the number of versions of the
standard will be fewer, probably only half the number that the PAS
procedures produce. It will also mean a quicker cycle for ISO, as
there will not be the delay from having the OASIS standard approved,
and then having a DIS ballot and a following period of comments
resolving. In the case of the first edition, the OASIS standard was
approved in May 2005, and the ISO 26300 will be published hopefully
some time late this year, which gives a delay of more than a year.
This would be much reduced if the balloting would happen in parallel
in OASIS and ISO.
I thus
recommend that we use the common maintenance model recorded in SC34
N0587.
