ODF 1.2 review part staging

[robert_weir@us.ibm.com]

On Monday's TC call I mentioned that I was focusing over the next couple 
of weeks on JIRA defects related to ODF 1.2 Part I and III,   Obviously, 
this would mean less emphasis on Part II (OpenFormula) in this time frame. 
 

Let me explain the logic behind this.    Since we have 3 different parts 
to ODF 1.2, we have two possible approaches to stages the work:

A) Wait for all parts of ODF 1.2 to be ready for review and send them all 
out for review at once.  This would mean that there would be no public 
review for anything until everything was ready.  When we send a 
specification out for public review, we are required to do it for a 
minimum of 60 days, and during that time we are not permitted to change 
it.  So this approach would mean that for 60 days the TC did nothing.  So 
the net result is that those who participate in the review process, both 
the public and SC34, would get hit by a big chunk of material to review at 
once, then the TC would sit on its hands for 60 days, and then the TC 
would get hit by a large number of review comments.  This does not appear 
very efficient to me. The two resources (reviewers and editors) are 
alternately at 0% and 100% utilization.

Or

B) Send the individual parts of ODF 1.2 out for public review when each 
part is ready.  Since the parts vary in length and degree of maturity, I 
think that they may be made available for public review in a staggered 
fashion over the next few months.  Part III Packing is the closest to 
being ready for public review, followed by Part I and then Part II.  As 
one part is out for public review, the TC can finish the editing of the 
parts that are not yet ready.  The net is that we're fully utilizing both 
the capacity of those who participate in the public review, as well as the 
TC's capacity to fix defects.  No one sits around waiting.

Of course, some defects may cut across two or more of the three parts, so 
I think that we want to have some substantial period at the end, where all 
three parts are under simultaneous public review.   Remember, a 60-day 
public review is the minimum required by OASIS.  We can have a longer 
public review if the TC desires this.

As for defect processing in JIRA, my recommendation would be to push 
forward the "most mature" parts first, to get them ready for public 
review.  So in my case, I have defects assigned to me in Parts I and II. 
Since Part I is closer to being ready for public review, I'm going to work 
on those now.  This doesn't mean that OpenFormula is less important.  This 
is just a way to stage the parts to maximize the utilization of available 
resources given the constraints of the review process.

Please let me know if anyone has concerns with this.

Regards,

-Rob