-- AdrianCuster - 26 Jul 2013 Additional comment(s): 1)

Topic: SWG Procedure

Create a more rigorous process/vote during the SWG charter process.

Relations to other topics

Related to Topic..., since ...

Discussion

---

On Behalf of Jeff DLB

Suggested Process:

* Draft announcement/press release about new work, to get agreement on high-level details. Don't issue the announcement yet, just agree on the vision. * CHECKPOINT 1: TC vote on whether to form SWG.

* Draft informational documentation for users and implementers. * Prototype software that implements all, and only, what is covered in the documentation. + Keep log of agreements/decisions that will form basis of standard. * Prototype conformance test tools for all functionality. + Keep log of agreements/decisions that will form basis of standard. * Iterate as needed. * Include 6-12 month period of testing, evaluation, and refinement. * CHECKPOINT 2: TC vote on whether to proceed to writing formal Standard.

* Draft Standard based on agreements/decisions. + All functionality must by covered by documentation, reference implementations, and conformance tests -- no more optional pieces for rare Use Cases that nobody implements. * Iterate as needed. * Circulate full package for public comment: draft announcement, documentation, software, specification. * Revise as needed. * CHECKPOINT 3: TC vote to adopt standard.

* PC vote to approve * Issue press release.

Regards, Jeff DLB, NOAA OGC TC/PC representative

-- TrevorTaylor - 27 Aug 2013

On Behalf of John Herring

Jeff,

I see no problem with any SWG who wants to follow a "different path" could not do so, assuming that path fits the talents of the SWG members. There is no specific policy on how a group does the work in its charter, but it might be valuable to outline the plan of work in that charter. It seems to me that trying this method out for one standard would be valuable first step. Since the internal steps of a SWG are not part of policy (write a charter, go to RFC and go to vote are all external milestones). It may work better for particular types of standards than for others, but that may be a question for the future. Even if there were obstacles in the PnP, the TC can always grant exceptions by a simple vote. I would not foresee any problems in getting support for doing something new and potentially valuable.

Making anything mandatory before we know that this one size fits all seem premature. Best method is either find a group or start one that is willing to do it and then do it.

Regards, John

-- TrevorTaylor - 27 Aug 2013