Just a question to the author: Perhaps it is just me, but is the use of the term "exemplary" correct in this context? When I read "exemplary", I think of something worthy to imitate or model? When I read several of the sections of this document, I get the idea that the "exemplary" application is something scaled down and not quite what one would want in a commercial application. In fact, I even helped write one of the sections that clarified how we can use scope to limit the commercial qualities.
This is not just a word nit to pick with the original author. It is incumbent upon all of the contributors to come to grips with this positioning. It is one thing to purposefully scale down a set of features due to schedules, resources, and similar constraints. It is another to set the project on a separate course where the stewards purposefully hobble or limit function so as to stake out their own commercial claims.
I understand that some companies may not want to pursue a project where they commoditize some of the features of their own products. And as stated by some, it is true that there are already open source candidates for many of the areas of systems management products and creating yet another may not make a whole lot of sense. The alternative, somewhat hobbled open source tool that dances around commercial interests appears to conflict with the original COSMOS project proposal.
So it would seem that the project is at a crossroads. Focus on developers and take a supporting role to the build to manage projects within the Eclipse foundation (focus on SML and SML tooling) or build a platform and management tools (including the SML tooling) upon the Eclipse frameworks.