To Host or not to host your VLE? That is the question.
The following is a draft chapter excerpt from Distance Learning Management.
When looking at a new platform for your distance learning program, inevitably the question of hosting will come up, that is where is the platform stored, and who is responsible for maintaining it? While the choices in VLE (Virtual Learning Environment, a.k.a. LMS) now exist in commercial and open source versions, where your platform is hosted and how it is delivered to your learners can bring up all sorts of issues surrounding who owns the data, how the platform is managed and maintained, and who ultimately ‘owns’ it. This ‘owning’ of data has been mentioned previously, and speaks to the separation between technology for institutional operation and technology for teaching and learning. Each of these technologies requires a different approach, a different set of priorities and being aware of this categorization of technology can help your institution deliver different platforms in the best way possible.
If your institution makes this distinction, if there is separate staff for managing Learning Technologies than technologies for institutional operation, then the choice will be different than a school where the same department in charge of email and networking is in charge of learning. The difference comes down to a focus on priorities. In the traditional IT Administration mindset, issues such as stability, security and control come to mind, but in the Teaching and Learning mindset, flexibility, experimentation, motivation, improvisation and the human connection are the priorities. These two different ways of seeing the way we approach our work, can either ensure that a learning platform is a way to transform learning experiences for the better through technology, or reduce the learning platform to a file sharing website where teachers dump PDFs of their PowerPoints.
As we move into an era of more cloud based solutions (storing data and applications on computers that are off-campus), the trend is toward outsourcing the hosting and management of servers, as is the case when universities turn to Google or Microsoft to manage their email or file sharing systems. A learning platform is no different.
Earlier in this this chapter we asked the question who the learning platform was for. The correct answer is ‘teachers and learners’, obviously, and while we all know this to be true, do are we making sure that the choices we make are devoting the most resources to supporting what happens on a VLE, namely teaching and learning. The choice to self-host or not, and even the type of third party hosting that is available will determine the resources you have available to let your VLE do what it’s supposed to do.
Having any online platform for offering classes and programs requires human resources. They need to manage the system, enhance the platform, they need to support its use for teachers, learners and anyone else who may want to use these platforms. There are also the folks who provide new user training, conduct research on the use of the system by collecting data. Finally, we have the instructional support staff, including instructional designers and materials developers that work with the platform and instructors and students in tandem to really get to the core of making the platform create amazing learning experiences. Much of how staffing to support learning online at your institution will be guided by your school’s philosophy of teaching and learning, which could be very concrete, or non-existent.
The point here is that no matter what, your institution will be devoting people to make this thing work. Work technically, as well as work pedagogically, and the question of whether to self-host or not should be guided by which of these is the priority. If you you decide that the integrity of systems at your institution from a technical perspective is more important than teaching (which is not really a choice, but more of a workplace history/culture thing) then you should invest more in technical management of the VLE. In some situations, local or federal laws require information to be hosted locally because of privacy concerns surrounding the data. In this case, you’re going to have to self-host, the the technical management will be the priority. If teaching and learning is the priority (pssst, which it should be), then the investment you make in managing the system should be less, pointing you towards a hosted solution, and more investment in supporting the pedagogy behind online learning. In either case, it’s safe to say that if your institution is investing more in just making it work from a technical standpoint, ensuring that the ‘lights stay on’ so to speak, over staffing to support sound pedagogy, you’re simply not going to see the innovation in teaching and learning happening because you don’t have enough boots on the ground in devoted to the right mission. The question at your institution is to look at how the VLE is perceived. Is it a system to be administered, like email servers, file servers, the cables and routers and switches that make the institution operate, or is it a learning tool, like a blackboard, teacher training, a classroom?
Things to Consider
When it comes to self-hosting or not, the question of how to store learning resources becomes an issue as well. Files take up space, physical space on a hard drive or SSD somewhere. When your platform is self-hosted, you’ll need to ensure you have enough ‘space’ to hold all this stuff. This means servers, which are expensive, and also leads to limitations on storage. If you have a class that uses lots of video, how does this affect performance and delivering data for other classes? Does it affect performance of the platform as whole? If you go with a hosted solution, nowadays these are scalable, meaning the amount of space you use will scale up and down based on your use. If your hosting provider is still charging for storage space (ex. 20TB per year), tell them that the 1990s called and they want their hosting philosophy back. If you have a good hosting company, with specific experience in education, then most likely they’ll be charging per user, which might sound expensive in the beginning, but it removes the limits of storage, which will enhance teaching and learning
Persistence of Data
Related the storage of data, is the persistence of data. If you are self-hosting or hosting to a size limitation, then there will also be limits in the persistence of your data, meaning at some point, stuff has to be deleted. Remember when you were in school and you could go back through your old binders to look at the work you did? Sometimes for a project you’d need to pull out old assignments and refer to them and sometimes it was just nice to look back for review purposes if you forgot something from the previous year. By allowing access to older classes online, your students can go back in this same way, reviewing materials, interactions and assessments, and bringing them into their current learning experiences. By having to erase old classes and cut off this access for students, they are being limited in the sources they can use for review, and for collecting evidence of their learning.
Of course, one could argue that they should be saving all this material offline, or that they can use an ePortoflio platform to save the data as well, which is true. The issue here again is how long do they have access to this, how easy is to manage, this downloading of everything they do? If it presents barriers, or if it takes more than than a reasonable student is willing to invest, then they won’t do it. If your hosting provider has no storage limitations and your institution is paying per user, then persistence of data can only enhance the learning experience, by integrating them across classes and allowing for review. Being able to look at classes from years ago, can also help with a myriad of other issues, from data collection for course improvement, student success, as well as a record of interactions between individuals and with the platform itself.
Another issue to consider is uptime, that is, how long is the platform up and available for use. In a self-hosted situation, upgrades and platform optimizations need to occur, and most times this means the system needs to be taken down, unless your institution has invested in redundant servers that allow for upgrading on alternate servers and switching them over when complete. With a hosted server, this should not be an issue. Of course you’re at the mercy of your hosting provider, but this should be part of your decision making process. What is your provider’s record for uptime? How do they handle updates and outages when they do happen? This could mean the difference between students being able to hand in assignments and failing a class if something goes wrong. Also, remembering that the self-host question brings in human this issue of uptime might require technical staff at your institution to be on-call in case of an outage, coming into work at 3am on a Saturday morning just because a server crashed.
If you find a bug on your learning platform, meaning something doesn’t quite work as it should, who is responsible for fixing it? If your institution self-hosts, your staff is, whether they dig through thousands of lines of code to fine the solution themselves, or they connect with a community or outside company to assist them. If you go with a hosted solution, you’re at the mercy of the hosting company to fix it for you, which again comes down to reputation and research before you commit. Assessing your hosting provider for how fast they can fix problems, respond to user input or complaints and squash a bug should be taken into account when making this choice.
Enhancement and Customization
One of the added ‘bonuses’ of self-hosting a learning platform is the ability to customize it, which has its pros and cons. On one hand you’re able to dive into the code of the platform, hack it to your hearts content and make it look all pretty. Customization also leads to enhancements, with the addition of plugins (depending on the platform) and adding features that can enhance teaching and learning. This ability to enhance and customize should be looked at critically though, as far as how this can be achieved. With self-hosting, chances are you have to change the core code by installing a plugin, or adding to existing code to deliver new features and integrations to your platform. This is quite frankly a very old way of enhancing a website or any platform. If you’re working with a modern platform, the core code should never have to be edited or ‘enhanced’ at all. The best online platforms, whether it be for learning, content management or blogging should include robust systems for enhancement and customization that don’t require ‘hacking’ of any kind.
The modern landscape of web applications is all about fostering interoperability, and through sharing of data, systems talk to each other very easily and integrate parts from other systems when needed. Any new web platform that launches typically has an API (Application Program Interface) that allows code hosted outside the core application to pull and push data into that platform. Another core technology that is now ubiquitous is OAuth, which you’ve probably already used. This allows users to login to one website, using a different site’s credentials, such as Google, Microsoft or Facebook. This sharing of data is obviously the future and should be a part of any VLE down the road. In the learning space, the core technology that your VLE should support is LTI (Learning Technologies Integration). This standard allows to web applications to, with little effort, integrate, share data with and build upon existing features of the core VLE. Many platforms now allow integrations through LTI in a purely graphical interface, meaning no code needs to be touched in order to enhance. With regards to customizing the interface this should also be easy to accomplish. If you want to change the theme or look of your site, you shouldn’t have to do this by delving into folders upon folders on a server and ‘hacking the core’. You should be able to do this through a web interface quite easily, or if you need to provide your own code, have it exist outside of the core platform’s code.
The choice of learning platform that is made is definitely interrelated with how your institution chooses to host and support that system. If the VLE you choose only affords the opportunity to enhance it though ‘hacking’ of core code and features, then you may opt for self-hosting, but if you opt for a VLE that has APIs for sharing data, a robust system for theme customization and support for LTI applications, then a per-user hosting solution is probably the best.
When your teachers and learners have issues with the learning platform you choose, they need to get support (something that will be covered in more detail later in this book). How you choose to host the platform will also affect how users get support, because it can allow your institution to delegate responsibility based on what type of support request it is. If your school has a self hosted solution, and a technical support request comes in, then staff hours need to be allocated to help the student and fix the problem. If, you go with a hosted solution, then the host provider is responsible for providing this type of support, leaving the institution solely in charge of helping out with pedagogical and ‘how to’ issues. This simple difference between how support mechanisms are delineated and delegated can affect how the platform is supported and in what ways. If the goal of your institution is to enhance teaching and learning, devoting more human resources for instructional design and how-to questions can be maximized by de-investing in ‘tech support’.
Ultimately the choice of whether to self-host or go with a hosting provider will be, in part, a financial decision. If you look at the line items for cost pay for storage-based hosting vs, user-count hosting, or for self-hosting with or without support from the authoring company or external third party support, the cost is not simply inclusive of the platform itself. The decision has to factor in all costs associated with the platform, and your institutions entire learning and teaching agenda.
The Effect on Pedagogical Innovation and Improvement
Progress is limited to that of the slowest in our population. This is definitely true of innovation and the technologies that facilitate it. While the choice in what learning platform your institution adopts is a big decision, married to that decision is how that platform is delivered to your teachers and learners. Human and Technology resources are always limited, because no one school has an infinite budget. Taking into account what you want to achieve when it comes to your distance learning goals, the speed of progress can change based on how these platforms are delivered. If your platform is currently supported from a technical standpoint by your institution, then human and technology resources will be invested in making sure this button works, or why this font is bold and that one is italic. It will be invested in programmers writing custom integrations for plugins that were never meant to work on your version of your VLE. It will be maintained by systems administrators that care more about system integrity, security and the infrastructure side of the technology, more than the teaching and learning. It will therefore will mean that when a teacher wants to try something new, using a tool they haven’t used before, they may have to fill out forms, justify their choice, put it to a committee and wait a year for testing, integration work and security assessments before they know if they’re ‘allowed’ to use this tool.
Using a VLE that includes APIs, LTI, is hosted by a good company that handles all the technical and security aspects of having a learning platform reduces these barriers. It could potentially allow an instructor to find a new tool at 9am on their favorite blog, identify it as useful, click a few buttons to integrate it into their class and use it later that day. With or without pedagogical support, an instructor is more free to do what good teachers do, and try new things, instead of having to ask permission, and wait to make sure it’s safe.
In a nutshell, when deciding on whether to self-host, the more you invest in the technical aspects of having a learning platform, the less money, time and energy there will be for investing in the pedagogical aspects.
*Originally published in edtechtattler