I just stumbled across this thread and found it very interesting. I am a big fan of DNN, and admit that I can see both sides.
For my personeral website I want a clear upgrade path, so me modifying the source can be a cause of many headaches. Since I've enjoyed the rich object module, I have been trying to find the right "fit" for DNN within my corporate environment. For this, I would of course need to access the code to make updates/changes ( 25+ developers to continue, back to the community ). Having said that I fully understand the branching issue, but ultimately I think you'd be hard pressed to find any member of SourceForge that doesn't understand that process.
I hope as DNN becomes more of a platform, and less of a development community that it's roots are not lost. If it wasn't for Microsoft releasing the code after all DNN wouldn't exist. I hope that the reduced chatter in the forums around module development is not a sign to come or I fear that the grassroots development community that has been built over the years will fade back to ASP.Net
As a corporate member of a few larger open source projects I've always struggled with how the code code has been delivered to the community. In my experience the Mozilla projects seemed to have been the closest to getting it right. But, at the end of the day I think that to engage corporate support you need tor release all code with a public release, and offer "builds" of future releases, so that the community can see progress, determine impact, and help with the development process. After all if only 1% of the community members are "developers" that provides you with a build community of over 4,000 programmers. In my opinion access to such a mass group is what has allowed Linux to grow at it's current rate.
Ultimately I believe that DNNCorp will steer the product in the right direction.