What next for the LMS?

All of a sudden it’s LMS week* in mostly-US higher education. Nudged by the imminent Educause annual conference, there’s a whole pop-up festival of reflection on why we still have enterprise learning management systems—and why we have the ones we have.

Audrey Watters, D’Arcy Norman, Phil Hill, Michael Feldstein, Jared Stein and Jonathan Rees have all contributed to this thoughtful and detailed conversation; anyone who thinks universities just woke up one day trapped inside a giant LMS dome really should read each of these at least. And Mike Caulfield has nailed one of the key problems: LMS features that don’t deliver the function associated with the name—in this case, the wiki tools in an LMS that rhymes with Borg.

As Audrey Watters rightly points out in her look over the wall at what lies beyond the LMS, the natural mode of LMS development is incremental, calibrated to the traditional operations of education institutions. The bottom line is this: content goes in, grades come out, and the whole thing can be flushed and repopulated with new learners the next time it runs. The LMS is particularly efficient at delivering sequential learning, and so it’s learner-centred in the same way that IKEA is customer-centred.

But the LMS story isn’t centrally about user experience. It’s a story about corporations, their investors, and their attention to higher education as a market. This week, George Kroner and his colleagues at the Edutechnica blog revisited their 2013 analysis of four countries in the global LMS marketplace, to see how the market share of key players has shifted over the past 12 months.

This is the state of things as a bar chart:

LMS ‘global’ market share data, edutechnica.com

It’s a flattening visualisation that distorts the dollar value of the Australian market to an extraordinary degree, and it’s triggered a rerun of last year’s polite shoving between George Kroner and Allan Christie, General Manager of Blackboard’s ANZ operations, as to what counts as the Australian higher education market.

Put simply, it is generally accepted that there are 39 universities (38 public, 1 private) in Australia. (Allan Christie)

In short, I do not consider the list of the 39 universities to be a complete representation of higher education in Australia. (George Kroner)

The thing is, the entire Australian market is a hill of beans in comparison to the US. This is why we don’t belong on this misleading chart, but it’s also why our LMS market behaves the way it does, and so strongly favours the existing near-duopoly. In all but three of our generally agreed major institutions, one well known LMS has the advantage of incumbency, and the other well known LMS has the advantage of not being the incumbent, which is unpopular with its users in the same way that politicians are: generically. In a small system where everyone knows everyone, the influence of other institutions’ decisions is direct and intense. It tethers aspiration to conformism, and cautions against risk. Look at the neighbours, we say, they bought a Kia. Or the other one. Either way.

But this year, the disputed inclusion of Australia’s non-university providers is newly significant. The constitution of higher education in Australia is the subject of a substantial reform bill currently under Senate investigation (submissions to the Senate Standing Committee on Education and Employment have just closed, and you can check them out here.) If the Higher Education and Research Reform Amendment Bill passes, it will change the relationship between the generally agreed 39, and the less well understood mix of others who can award degrees but until now have been excluded from Commonwealth funding.

No one’s sure exactly how Australia’s universities will adapt to all this, or how the non-university providers will be able to take advantage of their access to funding previously reserved for university places. But it’s likely that over the next few years LMS selection in the whole higher education sector will be sensitised to the attraction and retention of students who have grown up online, who are facing higher levels of education debt, and who will be vigorously encouraged by price signalling into comparison shopping. They will encounter a university system with more feedback mechanisms, more features, more special offers, and more personalised interventions of all kinds. Even if we’re not yet at the stage of installing lazy rivers, our online environments will become potentially distinctive campus amenities just like our libraries. Their quality, efficiency, and accessibility will become important in new ways, both to students looking to move quickly through degrees and sub-degree programs, and to university leaders looking for ways to expand and secure new markets, while keeping the overheads from teaching as low as possible.

Meanwhile many senior executive decision-makers setting the strategic direction for the use of these systems will still come from the generation whose own undergraduate experience (and perhaps whose academic careers) avoided online learning altogether. This is one reason, I think, why they have a view of LMS use that is far more utopian than most academics or students. It’s also the reason that universities underestimate by a very long way the proportion of academic staff workload that should now be reserved for LMS resource development, not just in exceptional circumstances like LMS change implementation, but all the time.

The result of this failure over many years to recognise the time needed to use an LMS well means that we end up with the situation Audrey Watters describes:

The learning management system has shaped a generation’s view of education technology, and I’d contend, shaped it for the worst. It has shaped what many people think ed-tech looks like, how it works, whose needs it suits, what it can do, and why it would do so. The learning management system reflects the technological desires of administrators — it’s right there in the phrase. “Management.” It does not reflect the needs of teachers and learners.

This is right, but it’s not the consequence of essentially bad design. The LMS is specifically good at what universities need it to do. Universities have learning management systems for the same reason they have student information systems: because their core institutional business isn’t learning itself, but the governance of the processes that assure that learning has happened in agreed ways. Universities exist to award degrees, to the right people at the right time, and to do this responsibly they have to invest in the most robust administrative processes: enrolment management at one end, and lock tight records management at the other. Actual student learning is something they outsource to their academic faculty, who still achieve this with minimal management oversight except through feedback surveys.

But as we move towards a more competitive system, with tighter budgets and higher expectations for quality, we should probably notice that the LMS is also a performance monitoring system for teaching. Minimally this is being introduced through the development of institutional threshold standards for online learning practice, while the attention of analytics tools is technically towards the evidence of student engagement with learning. As more routine teaching shifts online, there is nothing whatsoever to inhibit the development of LMS analytics for staff performance evaluation—including of casual and sessional staff.

This is why even academics who find the LMS a pretty hopeless teaching environment need to keep an eye on its strategic development, and especially to pay close attention when institutions engage in the process of selecting a new LMS. Because behind all the blither about the transformation of the student learning experience, an enterprise level management system is exactly what it says on the tin.

 

* LMS week: it’s like Shark Week, only longer.

 

Learning from failure

The problem with edtech evangelism is that it assumes the most valuable lessons are learned from other people’s success. This is why our lives fill up with stories of exciting tools that have transformed this that or the other thing. Exhausting, really.

Given the importance of failure to innovation, it’s interestingly rare to find blogs, lists, journals, or conferences focused on failure, in any field. The Ten Most Awful Mistakes in Online Course Design.  Ten Tools I’ll Never Try Again.  Seriously, I’d find those useful. But we don’t do it this way.  We don’t fly keynote speakers around the world to tell us what they did wrong, from the cartwheeling Prezi that nauseated the audience to the webinars doomed by lag, and the drop box assignments that were impossible to return—or the MOOC in which experienced and highly engaged learners left class early and set up an alternate public discussion of the misalignment of course purpose to course environment, for all to see.

But perhaps we should promote these experiences, because failure is how most of us learn. The only way to test whether or not a new tool does what it promises to do, for example, is to try it in a populated course. This doesn’t always produce splendid results, as Curt Bonk and his Blackboard partners are currently finding out.  Lisa M. Lane’s original blog on her decision to quit their experiment is worth reading, but it’s the appreciative discussion in the comments between the MOOC’s course leaders and the unconvinced that’s really gripping.

Could this awkward experience have been predicted?  Should either Blackboard or Bonk have given some thought to the possibility that an all night party with an open bar might attract a bigger crowd than would comfortably fit the venue?  Was the suck-it-and-see communication strategy really the way to go, given the sensitivity and confusion around openness (and Blackboard’s investment in open) right at the moment? Were expectations driven too high by the opportunity to interact directly with such a well-known eLearning pioneer?

It seems that for some, the problems were embedded in the environment itself—the unchanging assumptions behind the design of online forum tools (“very 1999”, as Lisa Lane puts it). This is the tiny, familiar failure that’s difficult to avoid. Most LMS discussion tools pass the test of course planning simply by being there.  Blog? Wiki? Forum? Yup, we’re good.  It’s only when you try to use them with 50, 100, 1000 students or more, that you start to see how awful they are.

And this isn’t a trivial limitation.  Just as in the face to face classroom, presence is the hinge on which the online gate swings: it’s the simple, human claim that learners make on one another’s time, that enables us all to feel that we’re exchanging ideas with people, not just testing ourselves against prescribed content. If this is important with a small class, imagine how much more vital it is once enrolment cranks up to the thousands. Without it, really, what differentiates a massive online class from a difficult day at the mall?

The comments on Lisa Lane’s blog will ring bell after bell with anyone who has tried to use the communication tools of a standard LMS to engage students in discussion. Somehow, the leading vendors have managed to miss the fact that users want to tailor their online presence, just as they choose how to present their physical appearance. The capacity to craft a personality that works for you is really critical to anyone’s sense of composure in a community of strangers. Users expect to control how their name appears, to use a thumbnail of their choice as their visual avatar, to share images, videos, feeds and links, and to be given some kind of personal site in which they customise, organise and publish what’s meaningful to them, in order to show not only what they think, but to share information about the public online networks through which they move.

This dominos into the second critical issue. Anyone trying to engage with an online class needs to keep up with a snowballing volume of input from others.  Messages need to bundle intuitively, to thread and scroll and quote properly. Users expect to be able to flag, sort and prioritise, to use powerful search tools, and to send quick notes of appreciation with likes, favourites and reposts. Some will save and print, because that’s the way they like to work; and others will need it all to be readable on a mobile screen half the size of a postcard. Everyone will need to skim for new messages, and some will want push notification.  The singular, standout value of learning through online discussion is that it’s a self-transcribing conversation that fosters review and reflection—but none of this works if you can’t find what you’re looking for.

When we undertook an institutional RFP for a new LMS last year, we expected that big companies whose R&D focus must be on the ways in which people use technology and networks to communicate, learn and manage their lives would develop social learning tools that would synchronise with these trends. Any campus LMS has to sit open on a student laptop where the next tab is Facebook (or Tumblr, Twitter and so on).  Students multitask their online learning against a background of rapid, pervasive interaction with friends they genuinely care about; if we’re going to ask them to concentrate instead on staying in the class space, couldn’t LMS designers at least help us out with a social design that’s halfway engaging?

They could—of course they could—so the fact that they don’t is revealing.  My guess is that LMS designers who don’t develop their social tools have made a rational calculation that it’s better for them if we handle this bit. The institutional shift to learning platforms, described so well by Phil Hill in a recent post for e-Literate has really been driven by desperate teachers looking for ways to compensate for awful LMS design. So platforming is a strategy of accommodation that works reasonably well for institutions and exceptionally well for LMS vendors; it’s the admission that L is rapidly uncoupling from M. The inclusion of a few engaging social options like WordPress or Google Apps doesn’t expose the big vendors to much risk providing everything’s bolted to the platform: in fact, it perpetuates dependency on whoever provides the core tools for managing the platform itself.

But as more institutions embrace the platform model, there’s going to be much more churn in social features, tools and options, because this is a wide open space for edupreneurs. Academics will come under increasing pressure to try new things, pushed by articulate, informed student demand. Negative feedback won’t wait for teacher evaluations—it’ll be on Facebook.  This is why it’s so helpful that Lisa Lane, Curtis Bonk and their colleagues are modelling their failure-to-communicate debate in the open, in the truer sense.

Because there’s a whole world of learning through failure up ahead, and we’re going to appreciate these pointers from our peers that show us both how to give feedback constructively, and how to respond to it openly.

Update:

Michael Feldstein’s fresh post that touches on the distinction between edupreneurs and teacherpreneurs (besides a number of other excellent points) really helped me think through some ways forward.  As ever, Instructure  come off well — but will they be able to sustain their personal approach as they grow? Time will tell.

Related articles

• Bonk / Bb MOOC week 1: Motivation & Encouragement (?) (jjulius.org)
• Selecting meaningful #socialmedia tools for a #MOOC or #PLN (downes.ca)
• 4 Reasons Why the Bonk MOOC is So Interesting (downes.ca)

One professor at a time

I’m still worried about the missed potential for edtech entrepreneurs large and small to engage in more substantial dialogue with educators at an earlier stage in their thinking.  At the moment, the pattern of bringing a mostly non-negotiable product to the RFP table involves both parties in an awkward clash of expectations that Joshua Kim has aptly described as a “bake-off”, and that certainly has reminded me of MasterChef more than once.

So the award for persistence in communicating with educators goes to the Instructure team who have been pursuing me from Sweden to Australia over the last month for a chat and a walkthrough of Canvas.

Instructure are the other company emphasising disruption to the standard LMS at the moment.  Anyone who has dealt with them seems to agree that this hard yards approach to consultation is their signature: they’ve been wearing out their shoes walking the hallways of higher education trying to figure out what educators actually want. To this, I can add that they’re really open to criticism of their product from someone with no technical background, and when they don’t have an answer to a question, they say: “We don’t know.”  This is actually a very good way to talk to people who teach, and it’s quite a bit less frustrating than the approach that’s been taken by several big edtech market announcements recently, where a curiously warped vision of education driven by content, analytics and automated workflow has been sculpted in haste by marketing communications.

The problem facing edtech marketing is that this vision might sell in the university boardroom or the engine room of strategic planning and quality assurance, but it’s one that most individual educators end up pinning to their mental dartboard.  So it’s just as hard for major corporates as it is for bootstrappers to figure out, I think, whether their intended audience sits at the institutional level, with the CIO, the CFO, and CEO and the hand that signs the paper, or whether their appeal should be pitched at users–teachers and their students.  What seems to happen is an awkward amalgam of the two, and this is never more offputting than in the corporate video spruiking the vision of the imaginary college experience.  We’ve all seen them.

So this brings me to Adrian Sannier’s response to Joshua Kim’s invitation to Pearson to do better at communicating with both their potential customers and those who are more sceptical about whether or not OpenClass is really aiming to change anything.

I’ve been genuinely impressed by Sannier’s willingness to front up to the criticism that OpenClass has been premature in its marketing announcements.  The OpenClass webpage has come in for sustained criticism from the education community, which is hard on Sannier as he is the only substantial piece of content on it.  And the unfortunate “see you at EDUCAUSE” answer to most questions hasn’t really worked for those of us not within a hemisphere of the northern American trade fairs, although I agree with Phil Hill that EDUCAUSE itself probably prompted the jumpy timing of the premiere.

Still, when I’m not seeing Sannier herding cats in a cowboy hat, I’m uncomfortably reminded of Quade Cooper; it’s the flipside of both corporate and sporting celebrity that there are times when the camera holds its tight close-up for what must feel like a very long time.

So, to be clear, I think his approach to taking these questions wherever they pop up is helpful, and I just want to follow up on one comment that has caught my eye.  In response to a question about the positioning of OpenClass in relation to other Pearson products, Sannier writes:

Pearson LearningStudio and OpenClass serve different markets. Pearson LearningStudio is the de-factor standard for fully online programs at scale, allowing programs a great deal of control over the academic experience. By contrast, OpenClass is designed for the campus market, where curriculum decisions are made one professor at a time. We understand the needs of these markets are quite distinct and have made OpenClass with that in mind.

We recognize that there is more than one set of institutional requirements around the world for a LMS. OpenClass complements Pearson’s other platform offerings very effectively.

This fleeting attention to the “one professor at a time” model of what actually happens in the classroom is rather lost in the subsequent strong focus on institutional buy-in that infuses the rest of Sannier’s answers, but the flicker of recognition is nevertheless there.

What this brings into the conversation is the most important shift in educational technology, that pre-dates the current potential for modes of managed promiscuity at the institutional level.  Individual educators have been quietly sneaking over the back fence like teenagers for some time, searching for the tools and applications in social media that will enable them to compensate for the limitations of big LMS solutions focused on analytics, workflow and content management (and frankly, we could sum all this up as educational CRM).

Most of us have found our tools through word of mouth—I know I came to Ning and Voicethread via Jane Hart’s excellent annual list of Top 100 Tools recommended by other educators; I started using Flickr memory maps as a learning tool after a bit of searching for something else entirely.  Diigo for Education is a list that provides me with a daily hint to think of something new that might work.  But sometimes the ideas come over coffee with colleagues: what really works for me is when someone who has tried something can tell me what I should avoid.

Educators are naturally good at this kind of crowdsourcing for social learning, and this seems to be the wave that OpenClass is trying to catch.  This is precisely why they’ve garnered so much attention without showing so much as a screenshot: the desire for change is really there.  So even if it’s not clear that OpenClass have the answer, or that anyone actually wants the future of education managed by big publishing, the passionate level of frustration with OpenClass should be giving other edtech entrepreneurs claiming to be disruptive something to think about in terms of their audience: the keenest advocates for disruption are also those who really don’t respond well to a conventional sales pitch.

Waiting for disruption

This week’s excitement has been the announcement by Pearson of their shakeup of the LMS experience.  On the OpenClass website, where we’re told in very big letters that this is all Open, Free, Easy and Amazing, the promotional video starts with Adrian Sannier, Senior VP, making the big claim that the LMS “as you know it” is dead. Sannier brings serious university research and administrative experience to Pearson’s push into the edtech market, and I’m confident that he knows what he’s talking about when he says that the standard LMS has “only ever been an ineffective administrative tool … it’s closed, it’s clunky to use, it’s costly.”

But his claim is a bit of a heartstopper for all the institutions who’ve woken up contractually handcuffed to the corpse of one or other dead LMS, for several years to come. It’s such a bold prediction that I’ve been distracted by visions of Sannier delicately blowing the powder residue from the barrel of his Colt 45 as he enters the darkened saloon where the frightened townsfolk have been cowering.  Yup, the LMS as we knew it won’t be bothering us no more, no sir.

OK. But before we start lining up the shots at the bar, there’s one thing that seems strangely familiar about our new situation: the things that are promised in the brochure are going to take a while to arrive.  So although the promotional video tells us that we can join OpenClass today, in fact most of us can’t.

If the product really is free and open, surely the hustle isn’t necessary, given the way things are across the LMS market. A genuinely free, open, amazing and scalable social LMS with strong integration with Google Apps has a good chance of selling itself. But over eager marketing is creating a problem that wasn’t there before: if you announce that something is now thrillingly and game-changingly available, and this turns out not quite to be the case, this is actually going to remind people very much of their experience in the game you’re proposing to change. As Joshua Kim puts it in his thoughtful summary of the things that could limit the disruptive potential of OpenClass:

There is no need to “sell” the LMS, only a need to get as many people as possible in the EDU community full access to the platform, and to share every detail about the technical specifications, cloud infrastructure, and product roadmap. … Where is the screencast walking me through OpenClass? If it is cloud based, why can’t I make an account now and play like I can on Blackboard’s CourseSites? Where is all the technical documentation and all the other information I’d need to start my research?

In other words, if we can’t take a look at it, there’s no need to tell us that we can.  Far better just to say that beta testing continues with the pilot institutions that were under the original non-disclosure conditions, and then the first round of others that will be allowed into the lab now that the secret’s out. The original launch of Gmail and more recently Google+ have shown that users are prepared to wait in line until invited in, and understand that there are benefits to robust testing as things scale up. So why not be clearer that whatever OpenClass might be in the future, you cannot join it today? Why not say straightforwardly when the door might open?  Even Disneyland tells you how long you’ll be standing in line from this point.

And the second familiar aspect is that it’s not exactly clear what the real cost of the gunslinger’s favour will prove to be. Audrey Watters has written this up very well, and says that we should be looking much more closely at the small print.  Jonathan Rees is more blunt, when he says that we should beware giant publishers bearing gifts.  Certainly, we’re all wary of the free lunch.

This is why I found it so helpful to read George Siemens’ explanation of the business context for “free, open, easy, amazing”, with his analysis of the rise of platforming as a way of simplifying and streamlining the need for individuals (or institutions) to pull together different tools for different little tasks. The way that platforming seems to make life easier for the consumer helps explain the pattern of competitive acquisition that has been the hallmark of edtech for a while. As Siemens says, “educators don’t want to think about the platform. They want something easy to use – simple, effective, and extensible – so they can get on with teaching and research.”  Justin Bathon points out that this isn’t exclusive to OpenClass, given the way in which other LMS designers like Instructure are integrating with existing public cloud social media rather than wasting design time on their own poor copies, particularly in terms of communication and collaboration tools.

The attraction of platforming over original product development makes sense, especially in a market crowded with solutions in which institutional decision-making is still slow. This slowness means that anyone’s enterprise level opportunities to change the game are limited by the number of times an education enterprise is actually interested in changing anything at all, a point I’ve been debating recently with Phil Hill.  Platforming is a good strategy in response to this sluggishness.

But the platforming model does mean that conservative educational institutions are going to need to get used to contractual LMS arrangements that might best be described as organised promiscuity.  This is the really big disruption on our side, given the fact that all our planning is based on the model of serial monogamy.

Meanwhile, speaking of open, those of us who are genuinely curious to know whether the “open” of OpenClass signals anything more than an opportunistic nod in the direction of open education, will just have to wait in line.

Related articles

• Freeing the LMS (downes.ca)
• architectures of control: Learning Platform Edition (bigthink.com)

Patched

Tenured Radical is asking the question that should be required reading for every CIO, CTO, LMS suitor, and higher ed tech commentator: if computers are such a blessing to higher ed, how come their use is actively extending the working lives of their users to such a damaging degree? The series of events she describes will have elicited low moans of sympathy from all of us working in writing-intensive courses online: an innovative learning task set up to use a system well, driven by good teaching principles and student-centredness and all the rest of it, all going swimmingly until, with a clunk:

Students suddenly became unable to upload the papers that they were expected to share with each other for the following day.  Since I became aware of the issue around 6 P.M., when the earliest finishers started trying to deposit their work,  I ended up spending an evening on it that I would otherwise have spent preparing the class itself (of course, given that it was long after working hours, I would have preferred catching up with Boardwalk Empire, reading a novel or staring glassily into space.) No technophobe, I opened up the settings to see what was wrong, and there went the rest of the night, with a quick break to warm up something from Trader Joe’s.  At one point, I found myself displaying the platform on three separate browsers to compare the settings that had worked to the settings that hadn’t, creating exact parallels between the settings on each browser, and testing each (as if I were a student) to see if what we were dealing with was a browser rather than a software issue.

You need to read the whole post to find out what went wrong and how it was eventually fixed, but the point that can’t be hammered home stridently enough is that this experience is … so … routine.

Something I think many university leaders find hard to grasp as they’re searching for the magical online cloud that will save them from the infrastructure costs of more buildings and services, is that clouds seem easier and quicker to establish, but they really don’t stay still for long.  That’s what makes the metaphor so effective: these clouds are moving.  Software is constantly being upgraded, and systems are constantly being patched.  Each tiny fix and patch risks causing one part of the overall structure to fall out of love with another, a situation made far more acute by the fact that higher education institutions can’t for one second guarantee browser compatibility among their thousands of users.  Meanwhile, security, storage and disaster recovery are exceptionally volatile areas of institutional risk: people who are bothering to spend their time engaged in malicious hacking aren’t going to wait for everyone to catch up calmly. It’s a jungle out there.

All this means that what looks like a stable system at the interface, both for your CEO and your average user, is in fact just how things are patched up today.  Most of the situations like the one Tenured Radical describes result from one of these tiny background fixes that has a butterfly effect somewhere else in the ecosystem. And this is so often discovered only late at night, on weekends, when students work intensively and academics really should be advised not to try to use this time to catch up on a little online reading or class preparation because this is exactly how we get caught over and over with the emails of panic about assignments that won’t upload, systems that lock out or freeze, or content that won’t download or open.

The realisation that this level of failure is so routine has added to my thinking yesterday about the lack of real competitive energy in the higher education technology market.  The leaden nature of contract timing means, frankly, that vendors have become used to being able to disappoint us.  We’re locked in and they know it.  So again, the significance of the recent wins that Phil Hill has tracked in the emerging LMS vendor market isn’t the size of the institution or the number of users, even though these are truly impressive: it’s the length of the contract, and the time it will take for that institution and its users to come back on the market. Until then, the vendor is more or less safe from the consequences of poor design, and incomplete or stalled development.

Sadly, the real cost of this is swiftly transferred onto the higher education institution trying to manage its own customer relations. Our IT departments are flat out with the running repairs, and not all of this is done in normal office hours, as the systems teams try to find the time of the week when they can take the whole thing down safely for a few hours. In too many cases, this is at 6am on a Sunday morning, but even with this effort and cost, they still can’t save the tech-friendly student-centred academic who’s up late on Sunday night trying to figure out and communicate Plan B to her frantic students.

Momentarily I’ve forgotten why it is that we have all become so used to this, and so accepting of it.  If the bricks and mortar showed a similar pattern of instability, I’m not sure we’d go back into the building on Monday, would we?