Re: [oiic-formation-discuss] My perspective. display perferct?

[robert_weir@us.ibm.com]

Simon Calderson <caldersons@yahoo.co.uk> wrote
on 07/01/2008 12:59:14 PM:

> 
> Dave Pawson <dave.pawson@gmail.com> wrote:
> > > Could you suggest some reasons why the proposed TC would
not 
> want to have a
> > > charter that takes into account the ODF 1.2 work which is
already nearing
> > > completion?  Is there a particular advantage to ignoring
this work?
> > 
> > Simply because it isn't an Oasis standard at this time of writing.
> > If, when the TC is constituted, 1.2 is the current Oasis standard,
> > then quote that.
> 
> Given that the charter cannot be redefined later, and it seems silly
> to constitute a new TC for interop with a standard that's likely 90%
> the same at worst, I don't see why versions need to be specified.
> 
> Thinking about it further, it also doesn't seem reasonable to 
> require that something be a standard before the TC can consider it.
> To be of most value to vendors and to promote best interop, it seems
> obvious to me to work on standards which are not yet finalized. 
> Surely the best point at which to raise interop concerns with a 
> standard is before it's finalized/ratified?
> 

That would be ideal, I think.  OASIS encourages
co-implementation during the development of a standard.  In fact,
a TC must point to successful uses of a draft before it can be eligible
for vote as an OASIS Standard.  So although it would be bad to release
a conformance test before the corresponding ODF standard is approved, I
see no reason why we could not start work on hypothetically on a 1.3 conformance
test while ODF 1.3 is still under development.  In fact,  goal
might be to co-release the test requirements document (and member implementations
of it) at the same time ODF 1.3 is approved.  

Wouldn't that be a good thing?

Much depends on the interest level of the actual TC
members.  If we have only 5 members, then we're probably going to
work single threaded, on one release at a time.  But with sufficient
interest the TC could create subcommittees, and have one work on ODF 1.1,
while another works on additions made to ODF 1.2, for example.

We found that to work well on the ODF TC.  While
the main TC was working on ODF 1.1 we had subcommittees working on specific
ODF 1.2 features, like metadata.

One way we can address Dave's concern of quoting an
OASIS version that does not exist is to not quote a version numbers at
all.  Leave it generic.

-Rob