Open Source FTW or, a Small Anecdote of a WPMS LTI Integration Plugin

Back at Domains 2019 Andy Millington came all the way from the University of Edinburgh to Durham, North Carolina to share the work of his team to create an LTI  that integrates WordPress Multisite with Moodle. This is a project Anne-Marie Scott wrote about extensively, and I can think of few more eloquent and ardent supporters of open source in higher ed, so in many ways this post is for her–big fan!

I’ll be honest, LTI is not necessarily the sexiest edtech acronym I’ve used on this blog. In fact, for many it’s a more restrictive API that is designated for the worst of teaching tools: the LMS, or VLE, or what have you.* That said, Jon Udell made a pretty compelling argument in defense of the LTI (although I will not forgive him his LMS love) which is very much inline with his thinking through light-weight system integrations for decades now. What’s more, companies like Hypothesis and Lumen Learning have listened to the Dead Moocmen, and they know the LMS is here to stay, and it will never die. So LTI integrations into learning management systems of all kinds is a key part of their success, and while I find the continued dependence on the LMS sad and pathetic, I do understand the need for them. Such are the compromises of an aging edtech.

But if I can pull myself out of the depression this line of thought plunges me into, one silver lining is open source code that makes these LTI integrations more broadly applicable and freely re-usable. And here begins my quick anecdote that I hope Andy and Anne-Marie can appreciate. In early December I was on a call with a university that has a legacy WordPress Multisite that has been around since 2009 and has 17,000+ sites.† What’s more, it’s integrated with their LMS, which in this case is not Moodle but Sakai, and in order for them to offload the hosting they need to re-work that integration. They asked us if we do development work, which is a hard no. We have folks we can recommend, but we realized early on that development is not our game; it’s a totally different skillset and long-term maintenance is always more work than one could ever imagine. That said, during the meeting I believe Tim recommended they take a look at the code on Github for the LTI plugin developed at Edinburgh before going the often expensive and time-consuming custom development route.

When we met again right before the holidays one of the agenda items was regarding custom development for LTI integration for their WPMS into Sakai, which Lauren and I were sure was going to be a deal breaker. So as we got to that bullet point the developer said this was no longer a concern, they looked at the LTI plugin from Edinburgh on Github and with some slight customizations for Sakai reported it worked brilliantly. In fact, it was even better than what they had been using previously. YEAH!

I’ll be sure to follow-up and see if that  modification can be shared somewhere for other folks using Sakai and wanting WordPress LTI integration. But in the interim it just seemed important to tell the story because those universities like University of Edinburgh that are leading by giving back, and putting the talent they have locally to work for a much broader global community is a facet of the power of open that made me fall in love with that whole concept way back in 2004 or 2005. Avanti!

________________________________

*I still hate the LMS as much as I ever did, and dealing with it tangentially last semester as Antonella started teaching again re-surfaced all the old wounds and that deep-seated loathing of a true tool of teaching oppression.

†As it turns out, this WordPress Multisite instance was a result of a visit and consultation I made in that same year. It has been amazing to me how many sites I helped folks get up and running a decade ago are now are hosted with Reclaim, it’s truly a long game/con I have been running all these years 🙂

Questioning the Conformity Curve

Image credit: Alan Levine’s “Converging and Crossing Lines”

The folks involved in Cal State University, Channel Islands’s domains project known as CI Keys, wrote a series of really thoughtful posts about the projects from a variety of vantage points. Michelle Pacansky-Brock wrote about the project as a catalyst for teaching on the open web. Jill Leafstedt wrote about the possibilities for faculty at CSU to create connected learning spaces for their courses. Jaimie Hoffman breaks down lessons learned after a year of building numerous classes out using CI Keys. And Michael Berman offers a broader rationale as to why an investment in CI Keys made sense. I love that each of them explained the impact of this pilot through their own lens, and it captures a nice cross-section of the possibilities and limits of such an initiative.

One of the things I think Berman totally nails is the willingness to question the adoption curve of a project like CI Keys. He notes the following:

I am coming to question the usefulness of the innovation diffusion curve in Ed Tech. First of all there’s an implicit value judgment that early adopters are better than late adopters – not to mention the infamous laggards. Not all technology adoption is useful, to say the least, and some is downright harmful. Second, why is success measured as universal adoption? If 20% of the faculty at my campus find CI Keys to be a useful and even transformational tool for encouraging student learning, does that necessarily mean that the other 80% are missing something by not using it? Perhaps, but I’m not so sure. It’s nice to think that we can provide a single tool for everyone to use but we can see where that’s gotten us. Instead, some will use institutional tools, some will use open source, some will use commercial tools, and faculty and students will use different tools (really, media) to accomplish different things. Is that hard from an ed tech support position? No doubt! But I think that’s the world we live in, not one where we always think in terms of scale-up and universal adoption – that ship has sailed.

I can’t possibly have said it better, and it really frames beautifully the predicament at many campuses. Someone throws out the stat that 85% of faculty are using the LMS and the conversation stops there. The various constituents who need resources beyond the LMS are poorly served, if at all. The adoption curve is actually a conformity curve used to justify supporting fewer and fewer tools on campus. So what should be seen as a pretty basic resource like web hosting/publishing are all but absent on many campuses. As Brian Lamb noted in the “Reclaiming Innovation” piece for EDUCAUSE Review:

…institutional leaders may refuse to support alternative systems….lest they draw attention and users away from the “serious” enterprise learning tool, diverting resources and endangering investments. If a technology is sufficiently large and complex, it can dictate policy, resource allocation, and organizational behavior far beyond its immediate application.

And the investment-based logic that can breed an aversion to alternatives often fails to comprehend that they’re not only significantly cheaper than any given system, but often complementary to that system. So, rather than endangering investments, it provides alternatives that make the system that much less monolithic. What’s more, it serves a portion of a campus community that has been forced to fend for themselves for almost a decade.