The following is a list of general comments and observations:

• Is the team size relevant? My impression was that this protocol would work well for a team size between 5 and 25 people, maybe. Is that correct? If so, this should be mentioned in the document.
• I would not talk about Second Life meetings, but call them virtual world meetings. This seems more generic and there is nothing specific to SL in the protocol as far as I can see.
• I'm not really sure what constitutes a "problem dimension". It is defined as an "issue in the Case", which does not really make it clear to me. But this terminology might be familiar to SMEs?
• The protocol assumes that all team members can meet in SL, which may not be true if SMEs are distributed in far apart time zones. Could this be addressed by an addtional level in the hierarchy, if required?
• The templates are all of the type that you would print out and complete by hand. Is there a better way of providing the templates? I'm wondering whether some CMS (Drupal) pages might be used here? This would bring up the question of document ownership and editing though. 
• The step preceeding the second meeting involves the collection of information that does not depend on the outcome of the first, e.g. personal background. Wouldn't it be usefult ocollect this information asap?
• The Gatekeeper role seems to be something you would only want for larger teams.
• In the Integrated Team Experience Matrix, I would initialize the Team Members column with the names of the people who brought up the dimension. Then people who feel their name should not be there could be asked to suggest an expert, either from within or outside the team.
• Is there a "Plan" template. I would be most interested in that since (AI) planning is my area of expertise :-)
• Should there be a closer tie-in with technology provided via openvce.net for this project?
• Shouldn't plan development be an iterative process? And how much guidance do LOE participants need for the planning stage? Will they have enough planning knowledge to just go away and come up with a plan? - I guess this is more a question for us than a comment on the protocol.

I have a few overall and structural comments which could usefully be addressed in an early revision of Draft 1 so we have a better basis to work from in our commenting.

The sections should all be numbered into main and sub sections, to make review and commenting easier.

Throughout, I do not think we should mention Second Life or Second Life meetings.  We should introduce the term Virtual Collaboration Environment (VCE) Meetings and then use VCE throughout.  Second Life should not appear as it is ONE possible platform, which could immediately e substituted for Opensim for secured behind-the-firewall meetings without any technical changes for example.  And in any case, we are working hard to ensure that alternative and backup VCE meeting modalities are available such as Harmonie web Adobe Connect pro, and even Skye on technical issues with the 3D Spaces.  We always refer to 3D spaces when we want to specifically say the Second Life style virtual worlds., and VCE when we mean these any all types of backup.

In the early stages, you launch into the role of the process coordinator.  I would like to see a section just before that covering Mission and Team Formation. 

Mission and Team Formation

1. Who owns the problem, mission or task and who is the stakeholder in that sense.
2. Who appoints the process coordinator and assigns them their task, and states the relationship/responsibility to problem owner.

3. How are team members and specialists identified, invited or persuaded to join, and what is the relationship/responsibility (which might be none) of them to the problem owner. Brokering mechanism? Reward (not necessarily monetary).

Process Coordinator Role

The problem maps are mentioned in the "Expectations of the Role" bullets 2 and 3 before they are really explained.  Suggest some info before this detail is given.

At 10:32 21/08/2009, Robert Cross wrote:
> Looks like you can do some neat things with Status and Prioritization
> in a quick pass through the paper.  Both of those would be of great help
> to people and I look forward to seeing how this can support the experiment.

We have a methodology that will allow us to help you with the way the protocol is stated. I have suggested to the team here that we create am OpenVCE.net book section for the VCP that has two things:

• VCP process, checklists, procedures and forms
• A template for a specific case/mission

These would be cut and paste of your protocol initially, but then we could quickly refine that to be rooted in a solid ontology of processes, plans and process products.  This uses the <I-N-C-A> methodology, but also related to international process modeling standards such as NIST PSL.

Some features that we would try to provided to refine the protocol include:

1. clear separation of the C3I collaboration, planning and analysis activities and the workflow of the agents involved in the collaboration itself (i.e. those typically involved with the Virtual Collaboration protocol and using OpenVCE.net to support that), from the target domain of activities and performing agents related to the mission objective.
2. clear understanding of the "purpose" (i.e. objective or mission tasking) and the stakeholder/purpose holder for those.
3. clear understanding of the "performer" of activities.
4. use of verb <noun phrase> ... format for all activities and checklist items, at all levels of abstraction.
5. clear description of (meta-level of) "process product(s)" refined by the activities - such as documents, plans or other artefacts.  Use of Object-Attribute-Value  (OAV) for this.
6. clear relationship of activity <noun phrases> to change of value in OAV description of the process products.
7. separation of agent performers, resources, time and other constraints from the activity description itself to allow flexibility of performance and recovery.
   
   

I would like to observe that the protool is in my view far too linear and centred on the three meetings in sequence. I envisage a MUCH more parallel and dynamic async and sync process than that for real uses of the VCE.