I agree with Frances - it's both/and. But there may be an issue with ones definition of "educational flexibility." Most educational programs are course-centric. They have been for decades, and will likely remain so for a long time to come. Within that overall context, both COTS and OSS systems provide a great deal of pedagogical flexibility. I've been using Blackboard for close to seven years, and there's very little that it structurally prevents me from doing.
However, many "forward-thinking" educators view the course model itself as outdated and restrictive. They argue that it encourages silo thinking rather than holistic understanding, and so forth. So to that group of users, a course-centric system may be automatically viewed as lacking flexibility because it assumes that students are enrolled in courses.
Alice MacGillivray raised a related point in a recent discussion on SMEs. She's asked me to reference it here, since she seems to be unable to post in this thread. http://scope.lidc.sfu.ca/mod/forum/discuss.php?d=61 The gist of her question is:
It seems to me that the intent of any course shapes the -- dare I say ontology and epistemologies -- behind the course. However, other courses I have designed/taught are about facts and best practices and are structured and assessed very differently. How do we manage the tension behind a drive for consistency in everything from structure to look-and-feel with the inherent differences in the nature of knowledge and ways of knowing in different fields?
IMO, it's fundamentally a moot point. It doesn't matter. Regardless of the content domain, regardless of the method by which learners/participants will learn/construct ever the content/knowledge, before you begin the process, you will have (or should have) defined the condition for a successful outcome. That's going to be true whether you're teaching a set of facts to be memorized, or facilitating a vision/values/mission session.
There are simple-to-use tools that can help clarify the priorities of different stakeholders, for example, the Preference Matrix. Draw a grid with all the potential features along the top and down the side. Each square in the grid represents a forced choice between the option at the top and the one at the side. The options that garner the most votes "win."
It is crucial that ALL stakeholders have input, though, if for no other reason than to be able to say that everone got to have a say, even if they didn't get their way.