For background, my specific interest is in being able to use the full complement of KSD, KDSK, PE, and MQX to be able to get the most out of the Kinetis collaboration. In particular to be able to take advantage of things like the filesystem, USB, and Ethernet.
So far, I have found two examples of KSD and MQX. The first was in exploring the full combination and learning how to define tasks for MQX. My experience can be found at Defining tasks using Processor Expert where David E Seymour showed me what I was doing wrong. The basic situation to me was that the supporting documentation just isn't up to the current release.
Then later, I found the example at How To: Create a New MQX RTOS for KSDK Project in KDS. However, this does not use PE. It was a cookbook approach and I was able to get it to run and can build on it. However, this doesn't use the PE. To me this is almost the classical approach of having to reference the manuals for all the details while trying each step.
So my question right now is one for general discussion. I am interested in feedback from both aspects of the Community - Freescale employees and users. In looking to the future, which approach would be more productive and why?