Not too long ago, I introduced the notion of a Learning Management Operating System, or LMOS. I'm going to be blogging about this concept quite a bit in the coming days and weeks, so I have created a theme heading for it on my blog. My first few posts on the topic will focus on the motivation behind the concept, starting with the need to address the long tail of teaching and learning applications.There are a small handful of applications that the vast majority of online courses will use--discussion board, file sharing, some kind of testing facility, etc. These applications are all fairly well understood and are beginning to commoditize--e.g., one discussion board is beginning to look much like the next. (This process is taking much longer than it should, but that's another topic entirely.) One of the next frontiers for LMS development will therefore be to look out to applications that are focused on narrower teaching niches. Take, for example, the Horizon Wimba voice tools, which are great for teaching foreign languages, or GenChemLab, the Open Source chemistry lab simulator. Tools like these can have a strong positive impact on teaching and learning for smaller slices of the teaching and learning world. Over time, the number of these niche applications will dwarf the small handful of tools that everybody uses. This is what is meant by the "long tail" (as in the long tail of a logarithmic curve), because if you graph out the learning applications with number of users served on the vertical axis of your graph, there are many, many tools out on the "tail" of the graph that serve a relatively few number of users. In an ideal world, teachers will be able to pull the specialized tools that they need into the learning envrionment for their courses.
In addition, there are some applications that, even if everybody uses them, are so personal and ideosyncratic in how they tend to be used that it is impossible to get just one tool that serves everyone's needs. For example, consider a gradebook. I have never seen two teachers who set up their gradebooks in quite the same way. Likewise, I have never seen a single gradebook software application that meets the needs of all teachers. In an ideal world, teachers will be able to choose from a range of gradebook applications designed to meet different record-keeping styles, all of which would integrate appropriately with other applications in the learning envrionment.
Today's monolithic LMS designs only provide this capability to a limited degree. You can get a clue on this by looking at the terminology used by WebCT. "Powerlinks" is the term they use for extending their system. You "link" (with "power", I guess) to new teaching and learning applications. They are separate from the system. Of course, terminology isn't everything. Despite the fact that Blackboard calls its similar capability "Building Blocks", the fact remains that these systems are designed first as monolithic learning management systems, with integration of external applications considered to be a supplement or an afterthought. FOSS LMS's like Moodle, dotLRN, and Sakai, aren't much different in this regard.
We need a system that, from technical, business, and developer culture perspectives, privileges integration of external tools over development of internal functionality that may be redundant to tools that have been created elsewhere. We need a system that is optimized toward slotting in new pieces as they become available, not as an after-thought or an add-on, but as a fundamental characteristic of the system. We need a system that lowers the barrier to innovation of new learning tools.