July 2016 Update

Welcome to the subscription service and to our first monthly update! In these updates we plan to summarize some of the LMS news over the past month from e-Literate and other sources, take a deeper look at select topics based on our market data, and highlight specific issues that emerge from our discussions.

In the initial phase of this service until Fall 2016, this subscription will be delivered via email service. We have had a few issues with emails going to spam folders – we will see if there are people or organizations not opening these update emails to notify them directly. This fall we plan to start introducing a login-based website to allow subscribers to see archived reports and updates. In the meantime, you can add eLitLMS@mindwires.com to your contact or white label list to avoid spam. 

We would love feedback and suggestions as we move forward, so feel free to send us a note to eLitLMS@mindwires.com!

Select LMS News and Analysis
  • Sometimes we see an event that leads to real-world learnings for both institutions and vendors. Put another way, sometimes we get our teeth in and like to keep chewing. Over the past month and a half we have been running a series of posts at e-Literate looking at a particularly painful LMS outage at UC Davis where the system went down for a full week, the vendor stopped responded with any real updates on the outage, and the school had to limp through finals with two partially-functional systems. You can start with this post that broke the news of the outage and then view this listing of follow-on posts. The series (to date) culminated in our exclusive look at the contract between UC Davis and the Sakai hosting partner Scriba Corp, helping to clarify where the process broke down.
  • Triggered by this UC Davis outage but also by Michael’s recent keynote and attendance at the Apereo Conference, we took a deeper look at open source contributions with an analysis of Github data for Sakai, Moodle, and Canvas, concluding that “Sakai Is Probably Healthier Than You Think”. Despite (or because of?) the loss of several founding institutions, Sakai has even seen a recent increase in the number of contributors to the code base. More on this subject in the next section.
  • We are taking a deeper look at D2L’s recent updates to Brightspace LMS and plan to summarize our thoughts soon. In the meantime, D2L has picked up two significant recent wins with Kaplan University and its 37,000+ students (taking the early lead in gobbling up end-of-life Pearson LearningStudio opportunities) and with the state of New Mexico for all K-12 students. Our focus in this initial subscription is on higher education, but the New Mexico news is the type of “adjacent market” news that we believe is relevant to the higher education community, at least in understanding vendor health and product direction. We don’t plan to re-share all LMS vendor press releases, but these two decisions are important to track.
  • With Canvas gaining the largest percentage of new LMS implementations in the past few years, we thought is was worthwhile to share in public a graphic from our initial report showing where these customers came from. In other words, what were the previous LMSs for schools adopting Canvas, and for that matter, any other LMS over the past five years? This graphic might be our favorite view from our new relationship with the folks at LISTedTECH. There’s just a lot of interesting relationships and analysis possible for the data geek who likes exploring great graphics.
  • Blackboard announced last week that they are partnering with IBM to outsource their data center operations and help accelerate their move to the cloud. This move (along with acquisition from parent company of Higher One / CashNet) are the first big moves with the company’s new CEO and should give some insight into the revised direction for the company. In a blog post, Peter George added that “we have selected Amazon Web Services (AWS) as our strategic cloud services provider and have entered into an agreement with IBM to manage our cloud and datacenter computing infrastructure as we move to the AWS Cloud.”
  • At the MoodleMoot US in Los Angeles, Martin Dougiamas described the four key themes driving the Moodle roadmap: Simplifying UX on Mobile and Web (including a new blockless default theme planned for November), Enhancing Interoperability and Integrations Support, Extending the Mobile Feature Set (planning for 100% of functionality for 80% of users), and Computer-assisted Education (with analytics-based notification to help you teach and learn).
Deeper Look: Assessing the Health of an Open Source LMS

Evaluating the health of any open source project can be tough. With a vendor-owned product, customers generate revenues, and some percentage of those revenues go toward funding developers who are all centrally directed by a company’s leadership. With an open source project, there can be volunteers, there can be institutions putting paid resources on a project, there can be support vendors contributing paid resources, and all of those resources could be working together under any one of a variety of governance models with varying degrees of coordination. We know open source can work overall because we have multiple examples of huge successes like Apache, Linux, and Firefox. But, particularly from a distance, it can be hard to tell why it works, never mind how. In the LMS world, we have three LMSs released under open source licenses in the US/Canadian market with significant market share—Canvas, Moodle, and Sakai. Each has a very different sustainability model than the others. Since we have spent a lot of time this month digging into the details of the UC Davis Sakai disaster, and since there is a popular perception that Sakai is dying, we’re going to take Sakai as a case study.With the departure of some of the marquis named schools that gave Sakai its original credibility—Indiana University, University of Michigan, and UC Berkeley, for example—conventional wisdom is that there is a flight for the exits by the Sakai schools. The adoption of Canvas by Unizin was particularly damaging to perceptions of the project’s health. But in the the blog post mentioned above in this month’s news summary, we showed that the number of code contributions to Sakai has increased dramatically since the departure of those schools.


We found that other contributors stepped in with the departure of those early adopter institutions:

It is not unusual for this sort of pattern to emerge in open source projects that have significant adoption and use when founding developers leave. Unlike a company, where there is a single point of control, some open source governance models are more resilient in the face of a crisis of leadership because there is no single point of control and therefore no single point of failure. (Note that the governance models of Moodle and Canvas are very different from Sakai as well as from each other; they almost certainly have different resilience characteristics.)

One of the reasons that the development of new leadership is not obvious to many in the US/Canadian market is that, while Sakai’s US adoption trend has been stagnant to downward in recent years, it has managed to grow overseas adoption substantially. The non-US institutions that have adopted Sakai as their primary LMS is approaching 50% of total adoption:

Some of the new leadership in Sakai development has come from Spain. We now live in a world in which the Spanish LMS market can influence the viability of the choices available to US and Canadian institutions.

Furthermore, our data do not support the implications in a recent Edutechnica blog post that a very large percentage of Sakai schools are looking to move. Edutechnica lumped schools currently running RFPs or pilots, which are strong indicators of possible change, with those schools running another LMS as a secondary system, which is a weak indicator. When we exclude secondary systems from our early warning analysis, we don’t see evidence that there will be a large drop in Sakai adoption this year. The caveats are that RFP and pilot data are both hard to get and seasonal. We may be missing some schools that are looking and we may see a surge when the 2016-2017 North American school year starts.

The last piece of information we can add to the picture is qualitative. The highlights of the new release of Sakai suggest that the developer community is aware that ease of use has become a critical competitive factor and has prioritized improving usability. Further, the community has produced an external-facing marketing video for the first time in recent memory. Open source projects that are entering a death spiral typically become very inwardly focused. Sakai does not appear to be in maintenance mode.

Overall, the current data suggest that Sakai’s health is fragile but not in crisis. It is too early to tell whether global adoption will stabilize at the current level, take another leg down, or begin to move back toward the peak it achieved in 2014. But the larger point is that the surface indicators that the market uses to judge the health of an LMS, such the departure of prominent adopting schools, are not always strong indicators of health and stability for open source LMSs like Sakai.

Spotlight: Disaster Recovery

As called out in the series of posts on UC Davis and Scriba, the school has been paying explicitly for Disaster Recovery services. Even though Scriba failed to deliver on the contract, the terms raises an important issue of what kind of services should a college or university request from a managed-hosting or cloud-hosting LMS vendor.

What we find as the default these days are claims and contractual terms of 99.9% uptime, measured annually or monthly. Vendors tend to include this type of language in a contract and often add a term that if they do not meet this metric, the client is entitled to a prorated refund. The catch is that clients are often asked to sign that this is their only remedy to a vendor missing uptime requirements.

But 99.9% uptime leads to just under 9 hours of unplanned outages each year. There are three problems to consider:

  • Losing 9 hours at once could mean losing an entire day during finals. Is this acceptable without remedy?
  • As seen as UC Davis, an extended outage can threaten the entire academic operation of a university – an issue that warrants more than just relatively small refunds on a service.
  • How much data is or will be lost with an outage? Getting a system back up after 4 hours while losing a day or more of course content may be less of a problem than getting back up in 9 hours but without losing any course data.

Other items that schools should consider are definitions of maximum data loss in outage, maximum response time per incident, and an explicit disaster recovery plan. Does your vendor have a plan to minimize your losses when the unexpected happens? For additional fees, there may be methods for a school to ensure there is a hot spare system in place or at least a synchronized backup database and a fully-built application environment ready if needed.For vendors, this is a real opportunity to put your money where your mouth is, and you can even create competitive advantage by offering these terms proactively. Times are changing – LMS is mainstream now and a core part of a college’s operations.

Coming Up . . . 

Again, thank you for your subscription. We look forward to building out this service not just for you but also with you. Please send us your feedback and questions to eLitLMS@mindwires.com.

What to expect: In a month, Michael and Phil will have attended the major user conferences for Sakai, Moodle, Schoology, Blackboard, and Instructure. In addition, we have been given private, in-depth demos from D2L on their product direction. The August monthly update will feature our coverage and analysis of these events. And we will ask why four major commercial LMSs hold their conferences in a 11-day span based in Miami, Las Vegas, Washington DC, and Keystone, Colorado. Really? I already know the biggest request we have of our corporate subscribers 🙂

From the Shameless Commerce Department 

If you are finding the report and this monthly update to be valuable – share with a colleague! Do you know an institution that could use this level of insight and analysis? Go ahead, remind them gently to subscribe or even browbeat them. Either way works for us. But the value of the subscription will increase with more subscribers – especially those on campus. Thanks for getting the word out.

Until next month . . .