IM+M (Jim Farmer’s company) has published some survey data on VLE adoption by higher ed institutions in the UK from 2001 to 2005. Interestingly, the data looks bad for almost everyone. There’s some limited good news for Moodle and Boddington, but as institutions are voting with their feet, the general sentiment seems to be “A pox on all your houses.”The report, which appears to be extrapolated from several different surveys as part of a larger JISC-funded project, starts with an extremely broad definition of “Virtual Learning Environment”. It notes that there are 102 VLE products on the consolidated list compiled from Edutools, EdTechPost, and UCISA (the UK-based University Colleges and Information Systems Association). Many of the products on the list are not systems that you might traditionally think of as learning environments, e.g., Lotus Domino, Microsoft Sharepoint, DesignersEdge, DazzlerMax, etc. Farmer et al acknowledge that their working list is extremely broad, and I don’t think that the approach hurts their conclusions at all. To the contrary, their results are actually strengthened by it.
The first headline to note is that the total market share for proprietary VLE’s dropped from 93% to 57% from 2001 to 2005. That’s a decline of nearly 10% a year, with the combination of Open Source and homegrown VLE’s now commanding over 40% of the total market. Now, this change might not be quite so scary for vendors if the total market were expanding rapidly. But it’s not (at least in the UK). The study found that the total number of institutions with at least one VLE grew by only 14% during the same four-year period and that the number of VLE’s per institution did not change in a statistically significant way. Worse, 95% of the institutions surveyed now have VLE’s. The market is largely saturated, which means that loss in market share translates pretty directly into loss of clients (as opposed to simply not growing as fast as your competition, which can happen when the total market is expanding). The proprietaries are apparently in trouble.
As you might expect, platforms with shaky futures have declined in market share the quickest. Lotus Domino and WebCT have taken the biggest nose dives. This is consistent with a classic industry consolidation pattern. But erosion of the total proprietary market share is fast enough that consolidation isn’t really helping the winners get ahead as much as you might think. For example, the combined market share of Blackboard and WebCT declined from 56.8% in 2001 to 43.7% in 2005. So contrary to the fears of many (including me), the merger does not appear to be a slam-dunk move allowing Blackboard to dominate the market (again, at least in the UK).
With the proprietaries on the decline, Open Source should be picking up…right? Well, yes, but not as much as you might think. Moodle and Boddington had the biggest growth stories, each going from 0% to 4.4% market share in the four-year period. There were a couple of others that surged a percent or so and a bunch (like Sakai and ATutor, for example) that didn’t even garner enough market share to make the breakout list. The total market share for Open Source solutions went from 0% to 11%, with 8% of the gains happening in the last 2 years.
But the real shocker is that the number of homegrown VLE’s jumped from 7% in 2001 to an eye-popping 30% in 2005. There are close to three times as many homegrown systems as there are Open Source systems and the growth in market share is more than twice as fast. These numbers are particularly startling given that they don’t include systems based on generic groupware such as Domino or SharePoint as “homegrown”. This about “roll your own” in the purest sense.
So what can we make of this trend? We can’t really draw hard and fast conclusions with out more detail on why institutions are making their selection decisions. But I do think there are some preliminary inferences we can make. First, the barriers to developing a basic virtual learning environment have apparently dropped low enough that even universities, with their limited IT resources, can consider taking on the challenge. This may be due to the increasing availability of course-grained development components in general and Open Source components in particular. Give me a discussion board (e.g., JForum) and a system for sharing files (e.g., Alfresco), and you’ve given me almost everything that 70%+ of university instructors currently use in their VLE.
Second, the trend is a hopeful sign that the battle of the marketing bullet points may finally be ending. It is fairly safe to assume that the majority of the homegrown systems do not have as many features to list on Edutools as the leading proprietary and Open Source VLE’s listed on Edutools. So adopting institutions must be making their decisions based on other criteria. What we really need to know is what those criteria were.
Which brings me to my final inference. I strongly suspect that cost was not the deciding factor for institutions choosing to build their own VLE. After all, how would that be cheaper than adopting, say, Moodle or Boddington? So if it’s not number of features and it’s not cost, what could the decisive criterion be? My guess is that, in at least some cases, institutions felt they had very particular needs that weren’t being met by any of the pre-existing options. In other words, the VLE is becoming situated software. That points to a useful future for a standards-based framework that can act as a VLE erector set–i.e., a Learning Management Operating System.
Very interesting report. Thanks for the IM&M crew for this information.
The reports states, 14% of colleges and universities that had no VLE in 2001 had one or more in 2005. I wonder how many of those first time implementations where fee-based, OS or homegrown.
Also if fee-based VLE’s dropped by 33.7% and locally developed grew 23.2%, I wonder which platforms (Angel, Blackboard, WebCT) those who invested in local development migrated from? WebCT seems to have seen the biggest drop.
So I wonder if there is a “maturity level” that campuses reach. That is, those who are new to online learning assume fee-based applications provide stability (code quality, services, support), then after some time and experience running a VLE, realize the costs associated with commercial support are greater than the assumed risks of either OS or just doing it themselves. As Michael points out, if campuses see faculty primarily using a discussion forum and content repository (while supplementing grading techniques just as they do for offline classes), then why pay for all of the other tools, e.g. portfolios, portals, e-commerce? And since your not running an “enterprise level” LMS anymore, why by all of the service/ support (if you where even using or happy with it anyway)?
Here is another thought. Many of the commercial applications (and now some of the OS) are extending the scope of their products/projects into non-LMS functionality, as noted above with portals and ecommerce. Campus administrations must be struggling with replication of services. If my LMS (Bb) has a portal, and my SIS (SCT Banner/Luminis) has a portal, and my DB has a portal (Oracle/CampusEAI) while my LMS (Angel) has a calendar, my email (Outlook) has a calendar, and my portal (Academus) has a calendar, how do I integrate these or push all my users to just one? And if I am only going to have one interface, why pay for and support all the ones I am not using? Perhaps LMS functionality is now modular enough (how many discussion forums are there now anyway?) that it’s easier to find the educational tools you need and deploy those as a bundle, and if that bundle just happens to look like an VLE, great.
So the next question is, looking at the locally developed VLEs in Great Britain, how much of their functionality is original (written from scratch by the institution) and how much is components based (modules of existing applications bundled together)?