Wednesday, May 25, 2005

What would we want in a really good CMS? Part 4

Yesterday I had a discussion with a developer at WebCT about their next steps in encouraging sharing of learning objects between different designers, focusing on those objects like quizzes and unlike Word or PowerPoint documents that really can't exist in a functional way outside the CMS. I tried to argue the point that content sharing had to be tied to a community of practice, that the community of practice may very well not map well into the hierarchy that they have inside the Vista software and on which we at Illinois built, which follows our administrative structure (the campus is divided into colleges, the colleges have departments, etc.) and so they really need to enable communities of practice as a separate structure.

In that discussion we focused narrowly on the sharing of the objects themselves. But one can readily imagine moving the discussion to asking the following questions: What tools are essential to enable a community of practice, which might very well engage in a detailed and deep discussion on how to teach and that may be as or more important than sharing objects? When should the tools be inside the CMS (meaning a log-in is required to access them) and when should they exist outside the CMS, perhaps because the same tools are used in other contexts as well? I'd like to advocate for the position that the CMS should not have a version of all types of tools. Asking it to do that will make it excessively bulky over time and harder for it to remain sharp and current on those functions where the market doesn't provide good alternatives. So the CMS must be selective on its toolset and as a consequence it is reasonable to expect that communities will rely on tools outside the CMS to some degree.

Here's a little hypothetical that might illustrate. Suppose we have a community of practice for sharing learning objects inside the CMS and now one of the members wants to set up a Blog that will reside on Blogspot.com so members can have meaningful discussions about teaching. But unlike my blog here where only I can make a post and everyone else can make comments only, suppose in this blog we want a more democratic approach, every member can make a post. So one of us begins by setting up the blog on Blogger.com and then after a fashion reaches the part of the setup where they want to invite members of the community to join the blog. At this juncture they'll come to
a point like this where they are supposed to supply a list of email addresses that are comma separated. (Apologies that the image is so degraded. I hope you can get the idea nonetheless.)

How much of a hassle is it to produce such an email list from within the CMS so the blog site creator can plop that list into the invitation form? If it is a hassle, then only knowledgeable people will do it and hence it won't happen very often. If it is a snap, everyone can do it. Obviously, Blogger.com is only one possibility among many, many external communication environments. Each may have their own way of managing users. A good CMS would enable interaction with any or all of them in the sense that porting the user group over should be a snap.

Let me turn to a different area that has captured the attention of many of my staff, PodCasting. I'm going to ignore the iPod and music sharing issues altogether and note that I've "futzed" with this enough to be able to conclude that the same podcasting clients can be used to download PowerPoint presentations or video presentations or really any file type you can think of and the same RSS feed structure can be used to deliver that type of content. This means that at least in theory we can move from an approach where the student clicks on a link to content on the server and then downloads the content at the time of viewing to the alternative where the students subscribes to content feeds and at regular intervals new content is downloaded automatically. The student can then go to a designated folder on her computer when it is convenient for her and view the new content without any lag due to downloading, because that happened earlier. (Perhaps the CMS is smart enough so that the links there now point to content in that local folder.)

As video content becomes more the norm, this type of content distribution will become increasingly valuable. For TEACH Act or Fair Use reasons, however, we will still need to deliver some content (created by other than the professor and which hasn't been licensed for distribution to the class) in formats either where the content is streamed but is not permanently downloaded to the students computer or if not streaming where it nonetheless resides only in a temporary cache and in a format that is not readily copyable by the student. Indeed, if there were different methods of content distribution depending on who owned the copyright, as I have suggested, this would serve as an excellent mechanism for educating students and instructors alike on copyright issues, something a good CMS should facilitate.

Further, for similar reasons regarding bandwidth and speed, I believe the CMS will have to enable more transactions to occur on the student's local computer rather than back and forth with the server, which has inherent limitations in terms of speed of such things as page reloads. In addition, these local transactions must involve "communication" between CMS content and content from other packages that would be used to do course work. (In my course, that other package would probably be Excel. The CMS must be flexible enough to be able to integrate with whatever software the instructor wants to use.) There will have to be at the end of the series of local transactions some summary transaction with the server so a record is kept. (This would be both for the experiments I talked about yesterday and for individual student assignments.) But otherwise the CMS must get smarter about using the student's CPU.

Now let's cycle back to the question on educating instructors about new teaching approaches. The normal experience with a CMS is the first time through an instructor will use a few features and then the next time, after having gained some comfort in the environment, the instructor will expand use to some other features. Currently many CMS have manuals and built in help on "how" to use these features. What if there were analogous documents written on "why" to use the features and with examples. Many new products on the market provide sample use on the product Web site. Why doesn't the CMS do likewise?

Actually some of the CMS vendors do provide examples of good use. This is WebCT's exemplary course page. The underlying idea for this page is a good one. I think, however, that we are getting little to no use value from this page. (One can find mention of WebCT on our Illinois Compass homepage. (Look inside the Help.) However, there are no links to the WebCT.com site either there or on the EdTech division pages that are devoted to providing user support for Illinois Compass. And it may very well be that if we did this sort of thing we'd want examples from our own campus, as they would have the most relevance.

One last issue to consider is "can you take it with you?" meaning in this context whether an instructor can take the online part of the course to the next institution of employment after switching jobs. The IMS project has helped here, but it is still hard to migrate a course, esepcially one with a fair amount of content. It would be nice to see this problem solved.



2 comments:

Burks Oakley said...

Lanny - Hi! You wrote "I tried to argue the point that content sharing had to be tied to a community of practice..." I agree 100% - and that is exactly the approach being taken by the LON-CAPA project. See:

http://www.lon-capa.org/

They have a "shared content repository" with 60,100 resources -- including:

Content Pages 20,900
Problems 18,600
Images 12,500
Content Assemblies 2,100
Animations 700
Movies 500
Simulations 400

The majority of resources are in the following "communities of practice":

Physics/Astronomy
Chemistry
Biology
Mathematics/Statistics
Geology

It seems that LON-CAPA already has much of the functionality that you would like to see in Vista.

-- Burks

Lanny Arvan said...

Burks - thanks for the LON-CAPA reference. My sense is that environment works well in the sciences and other quantitative fields. It is probably not the right package to use in the humanities. But the sharing and community of practice notions should work in the Humanities. So it would be good if other CMS could imitate LON-CAPA in that respect.