With the overuse of the term cloud computing, many seem to have forgotten that this type of migration is nothing more than outsourcing applications to a hosting service provider. We shouldn't get caught up in a hype cycle that can damage an organization's ability to do business in the near term. It's critically important that there be solid business justification and the risks and rewards are analyzed to determine both the short-term and long-term impact before moving applications to cloud providers.
Earlier this year, we utilized a learning management system (LMS -- sometimes referred to as a course management system) at Westminster College. Our faculty and students supplement and enhance their in-class learning experiences by continuing their instructional engagement beyond the confines of the classroom walls via the LMS. When used correctly and combined with great teaching, an LMS provides students with the best combination of synchronous and asynchronous instructional delivery. It was a critical decision to ensure that the college is running the right LMS. At the same time, the LMS is just like any other business application -- the total cost of ownership has to make sense.
In short, no matter where you are considering outsourcing applications, we have to balance a number of factors:
- Support costs (both implementation and ongoing support).
- Initial licensing costs.
- Ongoing upgrade costs, etc.
For our LMS, Westminster College chose to move forward with the open source Moodle product. Moodle is very popular with widespread deployment and a very active development effort. Moreover, there are a few companies out there specializing in Moodle support and hosting.
Once we made our product selection, I turned my attention to how we would deliver Moodle to the campus community. I considered a number of factors, including:
- How well did we support our legacy LMS product?
- Which hardware would be necessary to support Moodle on-site?
- Which skills would be necessary to support an on-campus Moodle implementation?
- How much would it cost for service providers to support and host the implementation for us?
- Which other potential value-add services might be wrought from a service provider?
In the end, this analysis made it clear that outsourcing the Moodle application hosting to a service provider was an attractive option. As such, we made the decision to migrate from our legacy internally hosted LMS to a service hosted by a provider named Moodle Rooms.
The rationale behind outsourcing applications
Here was my thought process behind each decision point for outsourcing our application:
Support for our legacy LMS product: Frankly, I give my staff members an A+ when it comes to how well they supported the old LMS. In my outsourcing analysis, the staff's previous track record was actually a mark against outsourcing.
Necessary hardware: We've moved to an infrastructure that is about 85% to 90% virtual. From a hardware resource perspective, since we decommissioned our previously virtual LMS, the addition of Moodle would have been negligible.
Moodle implementation and maintenance skill set: This is the first point that was a major benefit of outsourcing. Moodle is a great system -- it's popular and has a huge community constantly creating additions and enhancements. Before implementation and upgrades, we would need to do significant testing in order to ensure that we didn't damage our production environment. By outsourcing Moodle, we eliminated our own internal need to test. Now, my staff members can focus on the user-facing side of Moodle and add more value to the teaching function, since they no longer have to worry about the underlying implementation.
Moodle integration with existing systems: The provider created an integration component that allowed us to fully integrate Moodle with existing campus systems. As students register for classes in our administration system, this information automatically makes its way to Moodle. We've taken the extra step to configure our Moodle instance to authenticate against campus Lightweight Directory Access Protocol servers so that we can provide students with a single username and password for all services.
To me, integration is the key factor in whether or not an outsourcing decision can be implemented. If the decision requires "manual integration," that's a nonstarter.
Costs: The whole package -- hosting and support for Moodle at our provider -- costs less than our former LMS's licensing fee alone. From this perspective, the migration was an obvious win. Further, I didn't think that I would be able to easily or inexpensively add the skill sets needed to properly support and maintain Moodle in-house.
Service provider value-add services: Our provider makes available a number of additional services that we can add on as desired, including e-portfolios and 24/7 support for end users. As Westminster College's needs grow and change, we have a great deal of flexibility with our Moodle provider.
No outsourcing decision should be easy -- the decision certainly means a loss of control over what can sometimes be mission-critical systems. Instead, a careful analysis that includes a cost/benefit analysis, a business process impact analysis and planning must be performed to ensure that the end results of outsourcing applications meets business needs in ways that make sense both today and for the long term.
This was first published in January 2011