The limits and limitations of business requirements

Athabasca University’s Digital Governance Committee recently got into a heated debate about whether and why we should support Zoom. It was a classic IT manageability vs user freedom debate and, as is often the way in such things, the suggested resolution was to strike up a working group/sub-committee of stakeholders to identify business requirements that the IT department could use to find an acceptable solution. This approach is eminently sensible, politically expedient, tried-and-tested, and profoundly inadequate.

horse-carAs Henry Ford (probably never) said, “if I’d asked people what they wanted they would have said ‘a better horse'”.

A design approach that starts by gathering business requirements situates the problem in terms of the current solution, which is comprised of layers of solutions to problems caused by other solutions. For simple ‘hygiene’ tech that serves a hard, well-defined business function – leave reporting, accounting, etc – as long as you do properly capture the requirements and don’t gloss over things that matter, that’s normally fine, because you’re just building cogs to make the existing machine work more smoothly. However, for very soft social technologies like meetings, with potentially infinite ways of using them (by which I mean purposes, techniques, ways of assembling them with other technologies, and so on), no list of requirements could even begin to scratch at the surface. The thing about soft technologies – meetings, writing, pencils, pedagogies, programmable computers, chisels, wheels, technologies of fire, groups, poetry, etc – is that they don’t so much solve problems as they create opportunities. They create adjacent possible empty niches. In other words, they are defined by the gaps they leave, much more than the gaps they fill. What happens as a result of them is fundamentally non-deducible. 

Solving different problems, creating different possibles

Meetings are assemblies of vast ranges of technologies and other phenomena, and they serve a vast number of purposes. Meetings are not just one technology but a container for an indefinitely large number of them. They are, though, by and large, solutions to in-person problems, many of which are constrained by physics, physiology, psychology, and other factors that do not apply or that apply differently online. Most webmeeting systems are attempts to replicate the same solutions or (more often) to replicate other webmeeting systems that have already done so, but they are doomed to be pale shadows of the original because there are countless things they cannot replicate, or can only replicate poorly. Among the phenomena that are the default in in-person meetings are, for example:

  • the immense salience brought about by travelling to a location, especially when it involves significant effort (lost in webmeetings);
  • the fact that it forces attention for a sustained period   (most webmeeting software and ways of using it makes inattention much easier);
  • the social bonding that we have evolved to feel in the presence of others (not well catered for in webmeeting software);
  • the focus and meaning that comes from the ‘eventness’ of the occasion (diluted in webmeetings);
  • the ability to directly work together on an issue or artefact (limited in some ways in webmeetings, though potential exists for collaborative construction of digital artefacts);
  • the inability to invisibly escape (easy in most webmeetings);
  • the microexpressions, postures, movements, smells, etc that support communication (largely lost in webmeetings);
  • the social bonding value of sharing food and drink (lost in webmeetings);
  • the blurred boundaries of entering and leaving, the potential to leave together (usually lost in webmeetings);
  • the bonding that occurs in having a shared physical experience, including adversities such as a room that is too hot, roadworks outside, wasps in the room, etc, as well as good things like the smell of good coffee or luxurious chairs (not remotely possible in webmeetings, apart from when the tech fails – but then the meeting fails too);
  • the support for nuances of verbal interaction – knowing when it’s OK to interrupt, being able to sigh, talk at once, etc, not to mention having immediate awareness of who is speaking (webmeetings mostly suck at this);
  • the ability to cluster with others – to sit next to people you know (or don’t know), for instance (rarely an option in most webmeetings, and nothing like as salient or rich in potential as its in-person counterpart even when allowed);
  • the salience of being in a space, with all the values, history, power relationships, and so on that it embodies, from who sits where to which room is chosen (hardly a shadow of this in most webmeetings);
  • the ability to stand up and walk around together (a motion-sickness-inducing experience in webmeetings);
  • the problems and benefits of both over-crowding and excessive sparsity (very different in webmeetings);
  • the means to seamlessly integrate and employ other technologies, including every digital technology as well as paper, dance, desks, chairs, whiteboards, pins, clothing, coffee, doors, etc, etc, etc. (webmeetings offer a tiny fraction of this);
  • and so on.

A few of these might be replicated in current or future webmeeting software, though usually only in caricature. Most simply cannot be replicated at all, even if we could meet as virtual personas in Star Trek’s holodecks. Of course there are also many things that we should be grateful are not replicated in online meetings: conspicuous body odour, badly designed meeting rooms, schedule conflicts, and so on, as well as the unwanted consequences of most of the phenomena above. These, too, are phenomena that the technologies of meetings are designed around.  In-person meetings are incredibly highly-evolved technologies, making use of technological and non-technological phenomena in immensely subtle ways, as well as having layers of counter-technology a kilometre deep, from social mores and manners to Roberts’ rules, from meeting tables to pens and note-taking strategies. Much of the time we don’t even notice that there are any technologies involved at all (as Danny Hillis quipped, ‘technology’ is anything invented after you were born).

Webmeetings, though, also have distinctive phenomena that can be exploited, such as:

  • the ease of entering and leaving (so breaks are easier to take, they don’t need to last a long time, people can dip in and out, etc);
  • the automation of scheduling and note-taking;
  • the means to record all that occurs;
  • the means to directly share digital tools;
  • the fact that people occupy different spaces (often with tools at their disposal that would be unavailable in a shared meeting space);
  • the captions for the hard of hearing;
  • the integrated backchannels of text chat.

These are different kinds of problem space with different adjacent possibles as well as different constraints. It therefore makes no sense to blindly attempt to replicate in-person meetings when the problems and opportunities are so different. We don’t (or shouldn’t) teach online in the same way we teach in the classroom, so why should we try to use meetings in the same way? For that matter, why have meetings at all?

Dealing with the hard stuff

Some constraints are quite easy to specify. If a matter under discussion needs to be kept private, say, that limits the range of options, albeit that, for such a soft technology as a meeting, privacy needs may vary considerably, and what works for one context may fail abysmally for another. Similarly for security, accessibility, learnability, compatibility, interoperability, cost, reliability, maintainability, longevity, and other basic hygiene concerns. There are normally hard constraints defining a baseline, but it is a fuzzy baseline that can be moved in different contexts for different people and different uses. No one wants unreliable, insecure, expensive, incompatible, unusable, buggy, privacy abusing software but most of us nonetheless use Microsoft products.

It is also not completely unreasonable to look for specific known business requirements that need to be met. However, there are enormous risks of duplicating solutions to non-existent problems. It is essential, therefore, to try to find ways of understanding the problems themselves, as much as possible in isolation from existing solutions. It would be a bad requirement to simply specify that people should be able to see and hear one another in real-time, for example: that is a technological solution based on the phenomena that in-person meetings use, not a requirement. It is certainly a very useful phenomenon that might be exploited in any number of ways (we know that because our ancestors have done it since before humans walked the planet) but it tells us little about why the phenomenon matters, or what it is about it that matters.

It would be better, perhaps, to ask people what is wrong with in-person meetings. It still situates the requirements in the current problem space, but it looks more closely at the source rather than the copy. It makes it easier to ask what purposes being able to see and hear one another during in-person meetings serve, what phenomena it provides, on what phenomena (including those provided by other technologies) it depends, and what depends on it. From that we may uncover the business requirements that seeing and hearing other people actually meet. However, it is incredibly tricky to ask such questions in the abstract: the problem space is vast, complex, diverse, and deeply bound up in what we are familiar with, not what is possible.

It might help to make the familiar unfamiliar, for instance, by holding in-person meetings wearing blindfolds, or silently, or to attempt to conduct a meeting using only sticky notes (approaches I have used in my own teaching about communication technologies, as it happens). This kind of exercise forcibly creates a new problem space so that people can wonder about what is lost, what is gained, reasons for doing things, and so on. If you do enough of that, you might start to uncover what matters, and (perhaps) some of the reasons we have meetings in the first place.

Exploring the adjacent possible

Perhaps most importantly, though, soft technologies are not just solutions to problems. Soft technologies are, first and foremost, creators of opportunities, the vast majority of which we will never begin to imagine. Soft technology design is therefore, and must be, a partnership between the person and the technology: it’s not just about creating a tool for a task but about having a conversation with that tool, asking what it can do for us and wondering where it might lead us. What’s interesting about the ubiquitous backchannel feature of webmeetings, for instance, is that it did not find its way into the software as a result of a needs assessment or analysis of business requirements. It was, instead, an early (and deeply imperfect) attempt at replicating what could be replicated of synchronous meetings before multimedia communication became possible. When designing early web conferencing systems, no one said ‘we need a way of typing so that others can see it’. They looked at what could be done and said ‘hey, we can use that’. The functionality persisted and has become nearly ubiquitous because it’s easy to implement and obviously useful. It’s an exaptation, though, not the product of a pre-planned intentional design process. It’s a side-effect of something else we did – a poor solution to an existing problem – that created new phenomena we could co-opt for other purposes. New adjacent possible empty niches emerged from it.

One way to explore such niches would be to give people the chance to play with a wide range of existing ways of addressing the same problem space. A lot of people have turned their attention to these issues, so it makes sense to mine the creativity of the crowd. There are systems like Discord or MatterMost, that represent a different category of hybrid asynchronous/synchronous tool, for instance, blurring the temporal boundaries. There are spatial metaphor systems with isometric interfaces like Spatial, or Ovice, which can allow more intuitive clustering, perhaps contributing to a greater sense of the presence of others, while enabling novel approaches to (say) voting, and so on. There are immersive systems that more literally replicate spaces, like Mozilla Hubs or OpenSim. I hold out little hope for those, but they do have some non-literal features – especially in ways they allow impossible spaces to be created – that are quite interesting. There are instant messengers like Telegram or Signal, that offer ambient awareness as well as conventional meeting support (MS Teams, reflecting its Skype origins, has that too). There are games and game-like environments like Gather or Minecraft, that create new kinds of world as well as providing real-time conferencing features. And there are much smarter webmeeting systems like Around (that largely solves almost all audio problems, that – crucially – can make the meeting a part of a user’s environment rather than a separate space for gathering, that rethinks text chat as a transient, person-focused act rather than a separate text-stream, that makes working together on a digital artefact a richly engaging process, that automatically sends a record to participants, and more).  And there’s a wealth of research-based systems that we have built over the past few decades, including many of my own, that do things differently, or that use different metaphors. Computer-supported collaborative argumentation tools, for instance, or systems that leverage social navigation (I particularly love Viégas’s and Donath’s ChatCircles from the late 1990s, for instance), and so on. They all make new problems, and all have flaws of one kind or another, but thinking about how and why they are different helps to focus on what we are trying to do in the first place.

Perhaps the best of all ways to explore those adjacent possible empty niches is to make them: not to engineer it according to a specification, but to tinker and play. I’ve written about this before (e.g. here and, paywalled, here, summarized by Stefanie Panke here). Tinkering as a research methodology is a process of exploration not of what exists but of what does not. It’s a journey into the adjacent possible, with each new creation or modification creating new adjacent possibles, a step by step means of reaching into and mapping the unknown. We don’t all have the capacity (in skills, time, or patience) to create software from scratch, but we can assemble what we already have. We can, for instance, try to add plugins to existing systems: it is seldom necessary to write your own WordPress plugin, for example, because tens of thousands of people have already done so. Or we can make use of frameworks to construct new systems: the Elgg system underpinning the Landing, for example, does require some expertise to build new components, but a lot can be achieved by assembling and/or modifying what others have built. Or, if standards are followed, we can assemble services as needed: there are standards like xcon, XMPP, Jabber, IRC, and so on that make this possible. And we don’t need to create software or hardware at all in order to dream. Hand-drawn mockups can create new possibilities to explore. Small steps into the unknown are better than no steps at all.

Stop looking for solutions

Webmeetings that attempt to replicate their in-person inspirations are unlikely to ever afford the flexibility of in-person meetings, because they have fewer phenomena to orchestrate and we are never going to be as adept at using them. The gaps they leave for us to fill are smaller, and our capacity to fill those gaps is less well-developed. However, digital systems can provide a great many new and different phenomena that, with creativity and inspiration, may meet our needs much better. Without the constraints of physical spaces we can invent a new physics of the digital. As long as we treat the problem as one of replicating meetings then it makes little difference what we choose: Zoom, Teams, Webex, Connect, BBB, Jitsi, whatever – the feature set may vary, there may be differences in reliability, security, cost, etc but any of them will do the job. The problem is that it is the wrong job. We already pay for and use at least three major systems for synchronous meetings at AU, as well as a bunch of minor ones, and that is nothing like enough. Those that begin to depart from the replication model – Around being my current favourite – are a step in the right direction, while those that double down on it (notably most immersive environments) are probably a step in the wrong direction. It is not about going forward or backward, though: it is about going sideways.

It is not too tricky to experiment in this particular field. For most digital systems we create our decisions normally haunt us for years or decades, because we become locked in to them with our data. Synchronous technologies can, with provisos, be swapped around and changed at will. Sure, there can be issues with recording and transcripts, there can be a training burden, contracts can be expensive and hard to escape, and tech support may be a little more costly but, for the most part, if we don’t like something then we can drop it and try something else. 

I don’t have a solution to choosing or making the right piece of software for AU’s needs, because there isn’t one. There are countless possible solutions, none of which will suit everyone, many of which will provide parts that might be useful to most people, and all of which will have parts or aspects that won’t. But I do know that the way to approach the problem is not to have meetings to determine business requirements. The solution is to find ways of discovering the adjacent possible, to seek inspiration, to look sideways and forwards instead of backwards. We don’t need simple problem-solving for this kind of situation (or rather, it is quite inadequate on its own): we need to find ways to dream, ways to wonder, ways to engage in the act of creation, ways to play.

 

Can GPT-3 write an academic paper on itself, with minimal human input?

Brilliant. The short answer is, of course, yes, and it doesn’t do a bad job of it. This is conceptual art of the highest order.

This is the preprint of a paper written by GPT-3 (as first author) about itself, submitted to “a well-known peer-reviewed journal in machine intelligence”. The second and third authors provided guidance about themes, datasets, weightings, etc, but that’s as far as it goes. They do provide commentary as the paper progresses, but they tried to keep that as minimal as needed, so that the paper could stand or fall on its own merits. The paper is not too bad. A bit repetitive, a bit shallow, but it’s just a 500 word paper- hardly even an extended abstract – so that’s about par for the course. The arguments and supporting references are no worse than many I have reviewed, and considerably better than some. The use of English is much better than that of the majority of papers I review.

In an article about it in Scientific American the co-authors describe some of the complexities in the submission process. They actually asked GPT-3 about its consent to publication (it said yes), but this just touches the surface of some of the huge ethical, legal, and social issues that emerge. Boy there are a lot of those! The second and third authors deserve a prize for this. But what about the first author? Well, clearly it does not, because its orchestration of phenomena is not for its own use, and it is not even aware that it is doing the orchestration. It has no purpose other than that of the people training it. In fact, despite having written a paper about itself, it doesn’t even know what ‘itself’ is in any meaningful way. But it raises a lot of really interesting questions.

It would be quite interesting to train GPT-3 with (good) student assignments to see what happens. I think it would potentially do rather well. If I were an ethically imperfect, extrinsically-driven student with access to this, I might even get it to write my assignments for me. The assignments might need a bit of tidying here and there, but the quality of prose and the general quality of the work would probably result in a good B and most likely an A, with very little extra tweaking. With a bit more training it could almost certainly mimic a particular student’s style, including all the quirks that would make it seem more human. Plagiarism detectors wouldn’t stand a chance, and I doubt that many (if any) humans would be able to say with any assurance that it was not the student’s own work.

If it’s not already happening, this is coming soon, so I’m wondering what to do about it. I think my own courses are slightly immune thanks to the personal and creative nature of the work and big emphasis on reflection in all of them (though those with essays would be vulnerable), but it would not take too much ingenuity to get GPT-3 to deal with that problem, too: at least, it could greatly reduce the effort needed. I guess we could train our own AIs to recognize the work of other AIs, but that’s an arms war we’d never be able to definitively win. I can see the exam-loving crowd loving this, but they are in another arms war that they stopped winning long ago – there’s a whole industry devoted to making cheating in exams pay, and it’s leaps ahead of the examiners, including those with both online and in-person proctors. Oral exams, perhaps? That would make it significantly more difficult (though far from impossible) to cheat. I rather like the notion that the only summative assessment model that stands a fair chance of working is the one with which academia began.

It seems to me that the only way educators can sensibly deal with the problem is to completely divorce credentialling from learning and teaching, so there is no incentive to cheat during the learning process. This would have the useful side-effect that our teaching would have to be pretty good and pretty relevant, because students would only come to learn, not to get credentials, so we would have to focus solely on supporting them, rather than controlling them with threats and rewards. That would not be such a bad thing, I reckon, and it is long overdue. Perhaps this will be the catalyst that makes it happen.

As for credentials, that’s someone else’s problem. I don’t say that because I want to wash my hands of it (though I do) but because credentialling has never had anything whatsoever to do with education apart from in its appalling inhibition of effective learning. It only happens at the moment because of historical happenstance, not because it ever made any pedagogical sense. I don’t see why educators should have anything to do with it. Assessment (by which I solely mean feedback from self or others that helps learners to learn – not grades!) is an essential part of the learning and teaching process, but credentials are positively antagonistic to it.

Originally posted at: https://landing.athabascau.ca/bookmarks/view/14216255/can-gpt-3-write-an-academic-paper-on-itself-with-minimal-human-input

Weary, old, a little broken, but not letting go of the dream: edtech in the 21st Century

Anne-Marie Scott joins a long line of weary edtech illuminati who have recently expressed sadness and disillusion about life, the universe, and, in particular, the edtech industry (she has plans to do something about that – good plans – but her weariness is palpable). One of the finest antidotes to it all, Audrey Watters, has pretty much given up on trying to do anything about it. Even the usually-optimistic Tony Bates has lost his cool over it (specifically the exploitation of data harvested about students, including children, by cloud-based tools, which I predicted would be a growing issue a while back).

Personally, I burned out long ago, and the remaining embers are barely glowing. My desire to change the world is undiminished, and I still have some ideas that I don’t think anyone else has tried before, but the means, the time, the energy, and (too often) the will left me years ago. I lost most of my passion for most of edtech research long, long ago: so much rehashing of things that we’ve done again and again, so little change apart from for the worse, so many mistakes being made over and over on ever larger scales, so little that’s good getting the exposure it needs, too much that’s awful being over-exposed. The emergency responses to the pandemic just depressed me further, and my own university is devoting pretty much all of its energy and resources into reinventing its infrastructure, leaving little space for my quirky brand of toy making (though the Landing is very slowly, in fits and starts, beginning to get the attention it deserved 10 years ago). But I will not go gentle into that good night. Not yet.

Online learning (e-learning, edtech, technology-enhanced learning, etc), by its nature, has a strong propensity to do ‘human’ badly, which is a pity because education is about very little else than being human with other humans. Edtech (and almost everyone who creates it) wants to control, to measure, to collect, to impose order on disorder. Even its most organic, volatile, social spaces are filled with instrumentality, on the part of both people and machines. Much of the time, human actions are input for the algorithms that seek to control them. Machines try to make automata inside us in their own distorted image. We become what we behold, and what we behold becomes what it has made us, a spiralling loop toward mediocre grey, mirrors reflecting mirrors till all the light has gone.  And the machines, in turn, are cogs in machines, that are cogs in machines, each one turning the next, grinding their gears, oblivious to our humanity, black-boxing what we once did ourselves in uniform, impenetrable digital containers, where efficiency is a measure of what can be measured, and of little or nothing that actually matters.

Back in the 1990s and early 2000s, those of us working in the still-fresh online learning field hoped we’d change the educational establishment but, instead, the educational establishment changed us. It took our monkey-paw rainbows of wishes, chewed them up, and spat them back at us in trademarked beige. It threw away what it didn’t need to reinforce its mediaeval mission, and made what was left into a cyborg prosthesis, an automated monk, each part like the next: efficient, sterile, bland, each human interaction with it a data point, each person a vessel for implementing its measured objectives, ignoring what it couldn’t measure as though it wasn’t there. In the process of putting mediaeval pedagogies online, we lost most of what made them (nearly) work, and amplified the things that make them fail, creating machines (pedagogical and digital) that attempted to control learners more than ever before.

Personalized learning depersonalizes the person. The tools provide a more efficient means of making people who are more the same, as near as possible cookie cutter images replicating the machine’s pre-programmed domain model in learners’ brains.  Increasingly, too, we are learning to be human from machines that learned to be like us from the caricatured curated facades we presented to others in the simplifying mirror of cyberspace. More and more of those facades that are mined by the machines are now, themselves, created by machines. They will be what the next generation learns from, and we in turn will learn from them. Like photocopies of photocopies,  the subtle gradations and details will merge and disappear and, with them, our humanity. It’s already happening. Meanwhile, outside the educational machine, we are herded like sheep into further centralized machines that use the psychology of drug pushers to feed us ever more concentrated, meme-worthy, disposable content, that do the thinking for us so that we don’t have to, that automate values that serve no one but their shareholders, that blend truth, lies, beauty, and degradation into an undifferentiated slurry of cognitive pink slime we swallow like addicts, numbing our minds to what makes them distinct. Edtech is learning from that model, replicating it, amplifying it. ‘Content’ made of bite-size video lectures and pop quizzes, reinforced by adaptive models, vie for pole position in charts of online learning products. These are not the products of a diseased imagination. They are the products of one that has atrophied.

This is not what we intended. This is not what we imagined. This is not what we wanted. Sucked into a bigger machine, scaled up, our inventions turned against us. Willingly, half-wittedly, we became what we are not. We became parts in someone else’s machine.

How can we, again, become who we are? How can we become more than we are? How can the edtech community find its soul again? Perhaps, for example:

  • By revering the idiosyncratic, the messy, the unformed, the newly forming;
  • By being part of the process, not makers of the product;
  • By supporting each personal technique, not replicating impersonal methods;
  • By embracing the complex, weird, fuzzy mystery, not analyzing, not averaging, not simplifying;
  • By appreciating, not measuring;
  • By playing for the joy of the game, not playing to win;
  • By tinkering, not engineering;
  • By opening, not closing;
  • By daydreaming about what could be, not solving problems;
  • By embracing, not rejecting;
  • By making machines for humans, not adapting humans to be parts of machines;
  • By connecting people, not collecting data about them;
  • By owning the machine, not renting someone else’s machine;
  • By sharing, not containing;
  • By enabling, not controlling;
  • By following the learners, not leading them;
  • By looking through the screen, not at it;
  • By doing with, not doing at one another;
  • By drinking from the living stream, unfiltered and unflavoured;
  • By finding softness, not imposing hardness;
  • By asking why, who, and where, not what, how and when;
  • By making learning, not just what is learned, visible;
  • By making learners visible (if they want);
  • By loving the small, the personal, the trivial, the bright seams of gold;
  • By being – and staying – beginners;
  • By grasping the end of the long tail;
  • By living on the boundaries, and tearing down the barriers;
  • By rejecting the central and the centralizing;
  • By engaging with the local, the specific, the situated, the social;
  • By knowing we learn in a place, caring we are in it, and cherishing who we share it with;
  • By searching for the cracks and filling them with light;
  • By doing the dangerous things;
  • By breaking things;
  • By feeling wonder.

We must make playgrounds, not production lines. We must embrace the logic of the poem, not the logic of the program. We must see one another in all our multifaceted strangeness, not just in our self-curated surfaces. We must celebrate and nurture the diversity, the eccentricities, the desires, the fears, the things that make us who we are, that make us more than we were, together and as individuals. The things we do not and, often, cannot measure.

The things that make education worthwhile.

The reasons it matters.

 

Ernst & Young fined $100 million after employees cheated in exams

Not just any exams: ethics exams.

These are the very accountants who are supposed to catch cheats. I guess at least they’ll understand their clientele pretty well.

But how did this happen? There are clues in the article:

“Many of the employees interviewed during the federal investigation said they knew cheating was a violation of the company’s code of conduct but did it anyway because of work commitments or the fact that they couldn’t pass training exams after multiple tries.” (my emphasis).

I think there might have been a clue about their understanding of ethical behaviour in that fact alone, don’t you? But I don’t think it’s really their fault: at least, it’s completely predictable to anyone with even the slightest knowledge of how motivation works.

If passing the exam is, by design, much more important than actually being able to do what is being examined, then of course people will cheat. For those with too much else to do or too little interest to succeed, when the pressure is high and the stakes are higher, it’s a perfectly logical course of action. But, even for all the rest who don’t cheat, the main focus for them will be on passing the exam, not on gaining any genuine competence or interest in the subject. It’s not their fault: that’s how it is designed. In fact, the strong extrinsic motivation it embodies is pretty much guaranteed to (at best) persistently numb their intrinsic interest in ethics, if it doesn’t extinguish it altogether. Most will do enough to pass and no more, taking shortcuts wherever possible, and there’s a good chance they will forget most of it as soon as they have done so.

Just to put the cherry on the pie, and not unexpectedly, EY refer to the process by which their accountants are expected to learn about ethics as ‘training’ and it is mandatory. So you have a bunch of unwilling people who are already working like demons to meet company demands, to whom you are doing something normally reserved for dogs or AI models, and then you are forcing them to take high-stakes exams about it, on which their futures depend. It’s a perfect shit storm. I’d not trust a single one of their graduates, exam cheats or not, and the tragedy is that the people who were trying to force them to behave ethically were the ones directly responsible for their unethical behaviour.

There may be a lesson or two to be learned from this for academics, who tend to be the biggest exam fetishists around, and who seem to love to control what their students do.

Originally posted at: https://landing.athabascau.ca/bookmarks/view/14163409/ernst-young-fined-100-million-after-employees-cheated-in-exams

#AthaU22 – may your journey be rich, gentle, and challenging

In the convocation prayer offered by Elder Maria Campbell each year for Athabasca University graduands, she asks for blessing that their journeys be “rich, gentle, and challenging”. I can’t think of a more perfect wish than this. Each word transforms and deepens the other two. It’s truly beautiful. Every time I hear those words (or, technically, read them – they are actually spoken in Cree) they tumble together in my head for days. I am reminded of these lines (that are about music, but that seem perfectly apt here) from Robert Browning’s Abt Vogler:

And I know not if, save in this, such gift be allowed to man

That out of three sounds he frame, not a fourth sound, but a star.

On this graduation day I wish all our departing students rich, gentle, and challenging lives, and (as Maria Campbell goes on to say, gently acknowledging troubles to come) that the roads they travel are not too bumpy.

And I wish the same to you, too.

Solving The Wrong Problems: Why Online Education Is and Must Be Different from In-Person Education – slides from my invited talk at ICEMI 2022

icemi22

These are the slides from my invited talk at the 11th International Conference on Education and Management Innovation (ICEMI 2022), June 11th. The talk went down well – at least, I was invited to repeat the performance at a workshop (where I gave a very similar presentation today – if you’ve seen one, you probably know the content of the other!) and to give a keynote later in the year.

It’s about how methods of teaching that solve problems for in-person teachers don’t apply online, and it provides a bit of advice on online-native approaches. I’ve talked quite a bit about this over the past decade so there’s not much new in it apart from minor refinements, though I have put a greater emphasis on what goes on outside the classroom in physical institutions because I’m increasingly thinking that this matters way more than we normally acknowledge. Notably, I discuss the ways that physical institutional structures and regulations provide significant teaching functions of their own, meaning that in-person teachers can be absolute rubbish or (in some subject areas or topics) even fail to turn up, and students can still learn pretty well. This helps to explain the bizarre phenomenon that, across much of in-person academia, professors and lecturers are not expected to learn how to teach (and many never do).

Here’s the abstract…

In-person educational institutions teach, at least as much as the individual teachers they employ. Students are taken out of their own environments and into that of the institution, signalling intent to learn. The physical environment is built for pervasive learning, from common rooms, to corridors, to campus cafes; students see one another learning, share learning conversations, learn from one another. Even the act of walking from classroom to classroom makes events within them more salient. Structures such as courses, timetables, semesters, and classes solve problems of teaching efficiently within the constraints of time and space but impose great constraints on how teaching occurs, and create multiple new pedagogical and management problems of their own. The institution’s regulations, expectations, and norms play a strong pedagogical role in determining how, and when learning occurs. Combined with other entrenched systems and tools like credentials, textbooks, libraries, and curricula, a great deal of the teaching process occurs regardless of teachers. What we most readily recognize as ‘good’ teaching overcomes the problems caused by these in-person environments, and exploits their affordances.

Online institutions have radically different problems to solve, and radically different affordances to exploit, so it makes no sense to teach or manage the learning process in the same ways. Online, students do not inhabit the environment of the institution: the institution inhabits the environment of the student. It is just one small part of the student’s physical and virtual space, shared with billions of other potential teachers (formal or not) who are a click, a touch, or a glance away. The institution is just a service, not the environment in which learning occurs. The student picks the time, the space, the pace, and virtually all the surrounding supports of the learning process. Teachers cannot actively control any of this, except through the use of rewards, punishments, and the promise of credentials, that force compliance but that are antagonistic to effective or meaningful learning. In this talk, I will discuss the implications of this inverted dynamic for pedagogy, motivation, digital system design, and organizational structures & systems for online learning.

 

 

 

Interesting product: Bionic Reading

https://bionic-reading.com/about/

Having spent a while researching the literature on ways that visual landmarks and other text enhancements (and deliberate obfuscations) affect comprehension and recall, I am a little sceptical about the underlying theory for this patented product that is based on the assumption that we read better if the first chunks of each word are bolded, like this. The primary foundation appears to be a 1980 paper that uses gaze duration/eye fixation to predict readability of text. The Bionic Reading product creates artificial fixation points at the start of each word, so the theory seems to be that we can read faster, and recall more as our eyes are guided through the text. I don’t see any mention of any other research on the Bionic Reading site that supports its claims apart from the 1980 paper, but (ironically) maybe it’s because I missed it.

The assumptions may be a bit over-simplistic: we don’t read everything the same way, there are differences in ways that different people read, subject matter matters, intent matters, and so do many other factors. I found that I could grasp the meaning of the sample plain text that they provide on the home page far quicker than I could the bionic text equivalent: it was a small enough chunk that I could absorb the gist of it in a second or so, whereas I had to read the bionic text word by word in order to understand it, which took several seconds. Familiarity matters, though: there are recognition mechanisms at work here, both in making unadorned text easier to grasp (for me), and in learning to read the bionic text. I suspect that, after a while, the (possible) benefits would diminish as we learn to recognize whole words more easily in their modified form. It makes me wonder whether the benefit is similar to that of making a font more difficult to read, for which there is some (contested) evidence that it can improve recall. When we have to try harder to read the text, for some but not all kinds and lengths of text, we tend to recall more. In fact, anything that makes it more likely for us to read something word by word – as long as the flow is not lost – can aid comprehension and recall, under some circumstances.

The product is interesting, though. It provides an API that can be called to convert any text to bionic text, for use (in principle) in any app. It might make an interesting variation on the ways that we are using to modify text in our Landmarks application (for which I claim prior art, having written about this in 2012). Landmarks is intended to make chunks of e-text more recognizable, especially when text reflows, so it isn’t trying to compete in the same territory. However, the ways that the Bionic Reading app make passages of text more distinctive from one another might play a useful part in overcoming the big problem with most e-texts: that everything looks pretty much the same, and there are very few navigational cues, so it’s harder to remember what you read and where you read it.

Originally posted at: https://landing.athabascau.ca/bookmarks/view/13620551/interesting-product-bionic-reading

The problematic metaphor of the environment in online learning (update: found a publisher!)

This is a preprint draft of a paper that has been translated by the exceptionally talented Junhong Xiao (he always gives the best and fastest feedback I’ve ever received on any of my work, and he does the translations) for publication in a forthcoming (likely August) edition of the open Journal of Distance Education in China. I’ll be touting it for publication in English so, if you’ve got an open journal that might want it or something like it, drop me a line: it’s a 10,000 word paper but I could shrink it to fit journal needs if that’s too long (thanks to editors of the OTESSA Journal for taking this on!). The paper is in fact mostly a mashup of a couple of two of my earlier blog posts – Nobody has ever learned anything at a distance, and no one ever goes to a distance institution  and A few thoughts on learning management systems, and on integrated learning environments and their implementation though it comes to some slightly different conclusions and emphasizes a few different things (and it has more references!).

I was reminded to share this because I attended an excellent and thought-provoking opening keynote yesterday by Martin Weller at the OTESSA 2022 conference, in which he discussed themes and ideas from his forthcoming AU Press book, Metaphors of Ed Tech. Martin takes a much broader (and really interesting) perspective on uses of metaphor than I use in this paper: I’m really looking forward to reading the book. This paper is largely focused on some of the more obvious spatial metaphors, notably that of the ‘environment’. I’m releasing it as CC-BY-NC so do as you wish with it but, if you do, please give credit both to me and to the Journal of Distance Education in China, where it will be published in Chinese (trans. Junhong Xiao).  Sorry for any weirdness caused by copy-and-paste from the original.

The problematic metaphor of the environment in online learning

Jon Dron, Athabasca University, jond@athabascau.ca

Abstract

In online educational systems, teachers often replicate pedagogical methods, and educational institutions replicate systems and structures used by their in-person counterparts, the only purpose of which was to solve problems created by having to teach in a physical environment. At the same time, a great deal of the development and use of learning technologies has focused on creating virtual learning environments that attempt to replicate features of their physical counterparts, thereby weakly replicating in software the problems that in-person teachers had to solve. This has led to a vicious circle of problem creation and problem solving that benefits no one. In this paper I argue that the term ‘environment’ is a dangerously misleading metaphor for the online systems we build to support learning, that leads to poor pedagogical choices and weak digital solutions. I propose an alternative metaphor of infrastructure and services that can enable more flexible, more learner-driven, and more digitally native ways of designing systems (including the tools, pedagogies, and structures) to support learning.

Keywords: online learning, learning environment, learning management system (LMS), Next Generation Digital Learning Environment (NGDLE), personal learning environment (PLE), learning infrastructure.

Introduction

Outside the walls of educational institutions, for those with adequate Internet access, intentional learning using online systems is almost certainly more popular than its in-person counterpart, as at least the first port of call for learning almost anything and, often, as the primary means through which it occurs. From Google Search to Wikipedia, from MOOCs to Twitter exchanges, from YouTube videos to Khan Academy tutorials, people with online access are swamped with  learning opportunities. However, many academics and students still see online education as a poor second-best to in-person learning (e.g. Protopsaltis & Baum, 2019; Bouygues, 2019; Tichavsky, Hunt, Driscoll, & Jicha, 2015). In this paper I will argue that the distinction between online and in-person learning is far less significant than it appears, because all learning is in-person and never online, and most learning that is labelled as ‘in-person’ actually occurs at a distance from the teacher. Problems emerge, however, when  institutional online teaching inadequately attempts to replicate features and forms of in-person teaching, many of which:

  1. Exist to solve problems caused by the distinctive physical, temporal, psychological, and economic limitations of material spaces, and
  2. Are successful mainly as a result of the features and forms of the physical setting, not as a result of intentional teaching.

As a result, the systems we develop may not take full advantage of the medium, may not take advantage of the physical context of the students, and may attempt to solve problems that should not exist for those using them to learn because they are the result of in-person constraints. In many cases, online teaching may therefore actively militate against effective learning. Some of the problems may be solved using pedagogical adjustments in teaching and organizational changes at an institutional level, discussed in the first part of this paper. However, many emerge from the electronic systems that we use to teach, that poorly mimic the functions of their in-person counterparts in software. In this paper, I suggest that this is, to a significant extent, due to the misappropriation of spatial metaphors that cannot and should not be applied to online systems. I propose a different approach to the construction and conceptualization of tools for online learning, that better reflects the innate benefits of the medium, and that more fully supports the needs and circumstances of both online students and online teachers. I conclude by putting the pieces together and suggesting ways that, in combination, pedagogical, organizational, and digital changes may co-evolve to achieve the potential transformation of education that is afforded by digital networked devices.

In-person teaching

Although they may have physical or virtual windows to the world outside, the walls of the classroom provide clear boundaries that define the space and, because participants must be co-present,  the time in which activities intended to bring about learning occur. This is also true of most of the other buildings, rooms, and spaces that are provided by in-person institutions for students, including corridors, student accommodation, meeting rooms, common rooms, cafes, halls, quadrangles, staff offices, libraries, gymnasia, and examination halls. These spaces are not just support structures for the classroom, but active participants in the learning process (Dron. 2021). Even the act of physically walking to the classroom, especially with other people, creates a salience and value to the activity that is very different from that of clicking a link to an online resource. Most significantly, they are social spaces where learning happens as a result of direct and indirect interactions between learners (who are one another’s teachers) and, often, with their designated teachers. Simply seeing others learning makes a difference, as do the fliers and leaflets on the walls, and the spaces intended to support clubs and societies, where much academic discussion often occurs. Cafes, bars, and canteens are rich in learning dialogue. Student rooms, dormitories, and (especially) their kitchens, are powerful seed beds for learning, where much sense-making discussion occurs. Many universities provide purpose-built study areas. Even and perhaps especially, areas cordoned off for smokers provide an extremely fertile space where students from different subject areas and disciplines can and do share and construct their knowledge. Similarities in their design the world over speak to the fact that these are highly evolved spaces, supporting a learning process that extends far beyond the classroom.

These countless diverse learning opportunities in physical spaces, perhaps counter-intuitively, speak to the fact that there is a distance component to virtually all in-person education. Indeed, almost all learning is distance learning, in the sense of occurring somewhere and somewhen other than where and/or when deliberate, instrumental teaching occurred. People who learn with teachers in a physical space are almost always also interacting with other participants in the teaching role at a distance, usually in time and space, such as textbook authors, classroom designers, editors, illustrators, creators of timetables, and curriculum designers. And, for ‘in-person’ institutional learners, much of the learning itself also occurs at a distance, outside the classroom. This is most obvious in the form of assignments and homework but just as much learning can occur in conversation and interaction with others. Even when alone, if teaching works, sense-making connections always occur after the lesson is over, and continue to do so long after (sometimes decades after) the teaching event, almost never in the same place that the lesson originally occurred. In-person students do not have one teacher: they may have thousands. Weaknesses in in-person teachers can often be compensated for by these many other teachers, including the learners themselves and the institutions that provide the framework and resources for learning. This is amplified by that fact that, although credentials and grades are highly antagonistic to persistent intrinsic motivation (Kohn, 2011; Blum & Kohn, 2020; Ryan & Deci, 2017), they do encourage compliance. In search of good grades, students will therefore make use of whatever means they can – including those many other teachers as well as cheating or satisficing – to achieve the marks they seek. The physical environment of an in-person institution provides many supports to make this possible.

Online teaching

The in-person teacher, by default, controls learning in the classroom because it is a self-contained environment of which they are in charge for the duration of the lesson. Relinquishing control must be an active choice, or the result of an error. In contrast, the online teacher cannot, without a great deal of concerted effort, control the online student, any more than a writer of a book can control a reader. Online students can always choose when, where, with whom, how, for how long, and with what tools, media, and resources they learn (Dron & Anderson, 2014). It would therefore be surprising were online pedagogies to closely resemble their in-person counterparts, because they have different problems to solve. Most notably, without the requirement to share a single environment, with all the many rules, norms, structures, and constraints that entails, and without the need for the teacher to fill every moment of classroom time with learning activities, there should be no need for teachers to exercise the same level of control over their students.  However, online teaching evolved from in-person teaching, and online institutions must continue to interoperate with the in-person educational systems of which they are a part. As a result, many online teachers assume that they should dictate the learning process as much as their in-person counterparts and, usually, it becomes a partly self-fulfilling assumption through coercive methods like frequent grading, draconian scheduling, and tests. They consequently often make use of very similar pedagogies to those of their in-person colleagues, struggling to find simulacra or workarounds for the affordances of physical spaces that are no longer available, vainly believing that the learner is going to follow the path that they have determined for them and, too often, imagining that this is the sum total of the learning experience. To make matters worse, educational institutions impose other structures that are purely the result of constraints of teaching in physical classrooms, such as fixed-length (or multiples of fixed lengths) courses, deadlines, and perhaps most perniciously, the concept of failure. As any game-player or musician knows, failure is part of learning: it cannot ever be its end but, because of the constraints of having to run a course with co-present students and a beginning and end, failure becomes a potential outcome, not just part of the process. When all these factors are put together, the online student may have little more independence than their in-person counterpart but, at the same time, may lack the countless structures and forms of physical institutions that support in-person students. Rather than being immersed in learning opportunities, they must actively seek them within their own physical and virtual environments.

One obvious solution to this problem would be to create an online learning system that provides much of what is lost in translation from the physical environment, for example through a custom-made social media platform or an informal discussion area within a learning management system. However, this is not as easy or effective as it may seem, especially if it remains tightly coupled with other institutional policies, norms, and teaching methods. Partly, this is because of the too common focus on explicit outcomes and grading found in most institutional teaching together with failure by students and teachers to recognize the critical role of in-between spaces in learning. Thanks to the extrinsic coercion of marks and credentials,  if it makes no direct contribution to a grade, then it is seen as less valuable. Mainly, though, it is because it is not just there: students will not pass it on their way to somewhere else or be there for other reasons (like a need for rest or refreshment). They have to intentionally visit, typically with a purpose in mind. However, as the main value of it is its purposelessness (or, at least, that it supports a very broad set of purposes), that is rarely going to happen. Online systems are not environments in which students dwell: they are parts of their own environments.

This speaks to the central phenomenon around which this paper revolves: that nobody actually learns anything at a distance. We are always learning it where we are now. All learning is in-person learning, and it all takes place within a physical environment, part of which (but only a part) may include whatever technologies we might be using to talk with people, read, watch, listen, and learn from: books, computers, pens, emails, learning management systems (LMSs) and so on. Some of these may extend into other physical spaces occupied by other people, perhaps at other times, connected by online means. The broader learning environment is highly distributed in time and space, but learning itself only occurs locally. What we describe as ‘distance learning’ or ‘online learning’ is thus, in fact, nothing of the kind. It may involve distance or online teaching but the learning is always in-person. Online students exist, because the word ‘student’ only has meaning in relation to an online teacher, but online learners do not.

The promise of online learning environments

It is understandable that, when we teach in person, we have to occupy and make different uses of the same or similar environments like classrooms, labs, workshops, lecture theatres, and offices. There are huge financial, physical, and organizational constraints on making the environment fit the task, so it would normally be madness to build or even to substantially reconfigure a whole new classroom every time we wished to run a different class. Rooms may be built for flexibility, with moveable partitions and furniture, and that is much to be wished for, but there are physical limits such as walls and property boundaries that prevent this from going too far. Instead, our pedagogies and processes are normally made to fit the affordances and constraints of the classroom: they are another problem that our pedagogies have to solve, and/or an opportunity that our pedagogies can take advantage of. We may, sometimes, have some choice between classrooms that offer different facilities but, for the most part, our options are limited by what has already been built.

Online, there are countless tools available and, if none are suitable, it is not too hard to build them or to modify them to suit our needs, at least when compared with the costs of creating new physical spaces. There are few significant physical limits on how many can be used or how many people may use them: there are none of the limitations of physical space that constrain the use of buildings. Once they are built, moving between virtual tools just takes a tap of a screen or the click of a mouse or keyboard. It is even possible to use several of them at once, especially with a large high resolution monitor or more than one device. The learner’s environment may contain countless tools and systems, any of which may support learning, including physical books, instruments, and other people around them. And yet, for the most part, online teachers tend to make use of only a handful of possible tools: most consist of no more than a learning management system , email, and perhaps a webinar system.

There are many mutually reinforcing reasons that online teachers rarely provide the perfect application or combination of applications for the context of study:

  • Teachers’ lack of knowledge of the options (it takes time and effort to discover what’s available).
  • Teachers’ lack of skill in using them (most interesting tools have a learning curve, and that gets steeper in inverse proportion to the softness and diversity of the toolset, so most teachers don’t even know how to make the most of what they already have).
  • Lack of time and/or money for development (an application is just a shell for the content it contains and the interactions it supports, and it is not always as easy to add existing materials to a new tool as it might be in a physical space: for example, an in-person lecturer only needs to talk, whereas an online teacher must master the complexities of the hardware and software needed to record, edit, and share the same thing ).
  • Costs and difficulties in management (each tool adds costs in managing faults, configuration, accounting for use, performance, and security).
  • Cognitive load involved for learners in adapting to the metaphors, signposts, and methods needed to use the tool itself.

All of these are a direct consequence of the very diversity that would make us want to use different applications in the first place. This is a classic Faustian Bargain (Postman, 1998) in which the technology does what we want, and in the process creates new problems to solve.  Every digital system must establish rules of engagement that its users must learn, such as the ways that navigation occurs, the ways to make it perform its functions, the terminology it uses, and so on. In effect, every application invents its own metaphorical physics. That makes virtual systems harder to find out about, harder to learn, harder to develop, costlier to manage, and more difficult to navigate than the static, fixed facilities found in particular physical locations. They are all different, there are few if any universals, and any universal today may become a conditional tomorrow. In the case of cloud-hosted systems, the owners of which may unilaterally make changes to the software or configuration, this may be literally so.

Learning management systems

The learning management system (LMS) addresses all of these problems, to some extent. Almost every LMS essentially automates the functions, though not exactly the form, of traditional classrooms. Indeed, they are typically seen as environments, or are referred to as ‘platforms’, underlining the physical metaphors that inform them. In some parts of the world people prefer to use the term ‘managed learning environment’ (MLE), and the LMS/MLE is, in most vocabularies, the most dominant representative of a larger category of systems usually described as virtual learning environments (VLEs) that also includes things like MOOs (multi-user dungeons, object oriented), immersive learning environments, and simpler web-based teaching systems that replicate aspects of physical teaching such as Google Classroom or Microsoft Classroom. The use of spatial metaphors for the names of such systems reflects a deep-held belief or tacit assumption that the virtual systems can provide the boundaries within which actions occur, in ways that tend to be seen as analogous to those of physical spaces. In a few limited contexts, notably through immersive systems, this belief may be partly justified,. However, it matters that even the most immersive system occurs in a physical space. For instance:

  • when participants leave the immersive system they exit into different rooms, losing the natural opportunities for incidental or continuing chat that are innate to physical spaces;
  • participants are at the mercy of dropped network connections, glitches, and issues with the machines that run the immersive environment, leading to potentially quite different experiences for different participants;
  • different participants experience different temperatures, background sounds, smells, and opportunities for interruption in their own physical environments.

LMSs differ from physical environments to a much greater extent than immersive systems, in ways I will describe over the rest of this section. This is not a trivial issue of nomenclature. I will be arguing that the misconception that they are meaningfully analogous to physical classrooms lies at the heart of many weaknesses and failings in both the design of the tools and their use, reinforcing the belief that online teaching closely resembles in-person teaching, and blinding us to essential differences between the two.

The building metaphor

Creators of early LMSs and VLEs back in the 1990s (including the author) based their designs on the functions and entities found in a traditional university because that was the context from which they sprang, and that was the context in which they had to fit. In the eyes of its designer, an LMS could be thought of as a big university building with rather uniform classrooms. It may have extensions built onto it using plugins or standards such as LTI (the learning tools interoperability standard), and it may have a few doors and gateways (mainly in the form of hyperlinks) linking it circuitously or in jury-rigged fashion to other similarly weakly connected ‘buildings’  such as ‘places’ to register, to seek support, to talk to an advisor, to complain, to find books, and so on. For the most part, though, its fundamental organizational metaphor is that of a university, college, or school.

 

The LMS is, however, an impoverished school. It has no metaphorical corridors, halls, common rooms, canteens, yards, libraries or any of the other parts of a typical university environment where students gather to (amongst other things) learn. Students rarely get to even be aware of other classrooms beyond those they are in. Some teachers may give classrooms informal-sounding names like ‘the learning cafe’ but it is still just another classroom that works in the same way as the rest. Students teleport from one classroom to the next because what happens in between is not perceived by the designers as a useful classroom function to be automated or perhaps, more charitably, they could not figure out how to automate that. But there are other differences that are, perhaps, even more pernicious, to which we turn next.

Centralized code bases

In a physical environment, every object is discrete, occupying its own space. Physical classrooms can and do change – new furnishings, equipment, and so on – but the effects are local to that particular space, and they seldom prevent teaching from occurring across an institution.  Learning management systems, on the other hand, re-use the exact same code to generate all of the virtual classrooms of which they consist. Instead of a number of courses occupying the same physical spaces, and there being many such spaces to choose from, every course gets its own instantiation of a single centrally hosted toolset. There may be options to switch features on and off within any given course instantiation, options to configure each component differently, and a choice between components may be offered, but everyone gets exactly the same set of features, determined by the developer and the system administrator. This means that one set of features has to suit everyone. If, say, a teacher wants a discussion component that does things the default discussion component does not support, then it has to be installed or integrated in the centralized code base. While the LMS may technically support this – through plugins, LTI integrations, OKI components, and so on – system administrators are usually rightly reluctant or unable to allow it. Every component is another potential source of failure or (often) security holes, incurs management costs, uses system resources, creates a significant maintenance burden, and increases the complexity of the system for everyone. To allow unfettered installation of alternative components would be completely unmanageable. As a result, most available features must be a compromise, that can be bent to suit the needs of (typically) thousands of courses and teachers, but that are unlikely to be an ideal fit with any of them. Unlike the physical classroom, changes to the underlying application affect everyone, at once. When the LMS goes down, it takes the whole institution with it, and when changes are made, they are made for everyone, often affecting hundreds or thousands of courses and tens of thousands of students.

This is particularly problematic in cloud-based systems where administrators are not even part of the same organization, and where the system must support hundreds or thousands of institutions. Few of us who teach using cloud-based systems have not experienced difficulties when the systems on which our courses run change without warning or consultation, disabling or altering things that disrupt the design, sometimes rendering it inoperable. Even when they work, the fact that they use a single code base limits the potential for customization. Because most LMSs based their designs on what was presumed to occur in an average university, they rarely fit well with any actual university, because virtually no universities are average. Sometimes, the problems may be relatively minor. For example, Blackboard calls its organization elements  ‘courses’, whereas many other names for such things are common, including modules, units, and papers, and ‘course’ may refer to what others around the world might call a ’program’. Even this may disrupt and cause confusion (Dron, 2006). Other problems can run deeper, to which we turn next.

Reified roles

The typical LMS is a very controlled environment where everyone has a programmatically enforced role (typically at least partially reflecting traditional educational roles), that may vary according to the ‘room’ in question, but that are far less fluid than those in physical spaces. There are strong hierarchies, and limited opportunities for moving between them. Some of those hierarchies are native to the online learning system: the system administrator, for instance, has far more power than anyone in a physical university to determine how learning happens, like an architect with the power to move walls, change the decor, add extensions, and so on, at will. The programmers of the system are almost god-like in their command of its metaphorical physics. But the ways that they give teachers (or learning designers, or administrators) control, as designers, directors, and regulators of the classroom, are perhaps the most pernicious. In a classroom a teacher may lead, and that is the default, but they may and usually should choose to at least share leadership with their students, often fluidly and in response to how students are learning. In an LMS, a teacher (or someone playing that role) must lead,

Tools such as discussion forums may seem to be more egalitarian, but teachers’ power to control events in them is usually far greater than that of their in-person colleagues, often including the means to delete unwanted messages, prevent replies, stop conversation threads stone dead, and many other things that would be superhuman capabilities in a physical space. In a physical classroom, a determined enough student can always make themselves heard. In an LMS, the teacher can silence them. There is thus less of the soft flexibility found within in-person classrooms that allows for conversational pedagogies that adapt to the interests and needs of learners. At the same time, though, it should be (though too rarely is) remembered that the teacher’s power is confined to a small part of the learner’s own environment, not to a whole classroom. In practice, teachers still tend to treat the forum as an analogue of the classroom and, recognizing the value of dialogue in such contexts, often resort to coercion to make it happen online: marks for discussion contributions are far more common than in in-person settings, even among experienced online teachers. This combination of hard, role-based digital authority and hard, reward-based pedagogical authority is fundamentally different from its physical analogue. It creates both a social and a power distance that compounds what is already a less immediate relationship between student and teacher.

Within the LMS the teacher sees things that students cannot, and controls things that the students may not. A teacher configures the space, and determines with some precision how it will be used. With a lot of effort and (usually) high risk to the security and stability of the system, it can be made to behave differently, but it almost never is, because doing so usually involves promoting students to roles with similar capabilities to that of the teacher. In many cases, especially when it involves the use of plugins or other tools that extend across the system, this cannot be localized, so the risks to every user of the system must be considered. This is beyond the capabilities or rights of most teachers, and so it usually falls to system administrators, reinforcing their already substantial power to affect the teaching process.

Functional design

An LMS is typically built along functional lines. Rather than attempting to be a precise mirror of the in-person context, its functions are mostly based on loose, superficial observations of the things that teachers and students seem to do in physical classrooms, analysed to their component parts. Mostly, they are structured by teaching functions: presenting, discussing, assessing, guiding, and so on. For instance, in most LMSs, if you want to talk with someone, you normally need to go to a separate discussion area inside the classroom or, metaphorically,  to leave a note on the teacher’s desk in the form of a direct message. Unlike a physical classroom, dialogue is seldom possible everywhere. The same is true if you want to take a test, or to share your work with others: it rarely occurs within the context of learning, but in a separate screen, often separated from its context by a hierarchical set of links. Indeed, in many architectures, it will be handled by a different component than the rest, with its own tables in the database and its own distinctive interface.

Similarly, lectures are either literally that (video recordings of lectures) or (more usefully, from a learning perspective), text and images to be read on screen. This results from the erroneous assumption that the only function of lectures is information transmission, which is perhaps their least useful role, given that we have known for almost a century that it is far more effective to read a book (Greene, 1928). Lectures can and do have value as physical and temporal signposts, as motivators to pay attention, as events that demand attendance and thus have greater salience than simple reading, as well as providing opportunities to engage with others, sometimes within but always outside the lecture hall. Online, there is seldom a chance for students to even put up a metaphorical hand to question the teacher, and ‘joining’ a lecture is no more salient than clicking a link to a Facebook post. There are limited opportunities to be aware of what other students are doing, including for the teacher (although teachers do usually have access to system logs that offer an impoverished caricature of what students are doing, albeit one that is blind to anything they do beyond clicking and tapping keys on a machine). Much of the ‘space’ may as well be unpopulated, given the little students see of one another. Learning resources are normally static and designed in advance, and so the teacher cannot nimbly adjust to student reactions to them. Notices can usually only be pinned on the ‘wall’ by teachers, often with names such as ‘announcements’, further emphasizing the controlling nature of the teacher-student relationship. Classroom timetables are embodied in software despite the fact that a rigid and unforgiving timetable makes little sense in a medium that supports learning anywhere, any time. Some LMSs may allow you to break up the content differently, but it is still another timetable; just a timetable without dates. It is always the teacher (or one to whom the role is delegated) who sets the order, pacing and content.

Robot overlords

The LMS provides a high-tech classroom, populated by metaphorical robots.

Some of the robots may be programmed to attempt to force students to behave in ways determined by those higher in the hierarchy (sometimes teachers, sometimes administrators, sometimes the programmers of the software). For instance, adaptive systems might act as gatekeepers that prevent students from moving on to the next section of work before completing the current one, or they might prevent students from submitting work before or after a specified date (Martin, Chen, Moore, & Westine, 2020), or they might limit their access to a specified time period.

Some of the robots might even mark your work (Keuning, Jeuring, & Heeren, 2018). Human beings have grown up with other humans and therefore understand the context of the work, the motivations of the students, and the many different ways that things can go wrong, as well as creative and unexpected ways they can go right. Robots – even those that are employ deep learning and similar AI approaches – do not. While hard, mechanistic systems may be useful for providing feedback when students must play their role correctly in hard, mechanistic systems (in hard, ‘right answer’ subjects), those mechanistic skills are seldom the most important part of what they learn. Human teachers do not (or should not) just judge success or failure: they should model practice,  remedy misconceptions, provide encouragement, and so on.

There are metaphorical surveillance cameras everywhere, recording students’ every move (in very low resolution), often only accessible to those with more powerful roles, though sometimes a robot or two might give them a filtered view of it, such as through learning analytics traffic-light interfaces (Verbert, Duval, Klerkx, Govaerts, & Santos, 2013). Though the perpetrators of these tools may claim to have student interests in mind, and will often talk of ‘personalization‘ by way of justification, it is not personalization at all: it is system-enforced customization done to, not by the students (Kohn, 2015). These are all tools that are designed to enforce compliance: an attempt to embody in software the control that is demanded of an in-person teacher due to an accident of physics, not for any pedagogical purpose.

Beginnings and ends

The fundamental social form of the classroom that provides the primary metaphor of most LMSs is the formal group (Dron & Anderson, 2014). Formal groups are technological entities – inventions that are designed to address problems – at least as much as they are social. Among their many technological features are names, roles, procedures, rites of joining and leaving, rules of behaviour, schedules, beginnings and ends, almost all of which arise from the constraints of in-person learning, such as the need for people to be co-present, problems when people talk at once, limits to the capacity of classrooms, directionality of hearing and sight, and so on (Dron, 2016). Unsurprisingly, many of these features are embodied in code, not only in the reified roles already discussed but in processes of joining and processes of leaving.

A student cannot usually go back and visit when their course is over because most online courses have opening and closing enrolment dates. Perhaps their designers assumed that, when teaching was done, the learning was done which, of course, it never is. Learning keeps on evolving long after explicit teaching and testing occurred. Again, this is because physical classes are scheduled and terms come to an end because they must, not because it makes pedagogical sense. And, like almost everything, it is possible to override this default, but hardly anyone ever does, partly because it brings back those Faustian bargains, especially in manageability, but mainly because most people accept defaults (Kelly, 2009, Dron, 2006). LMSs embody enrolment technologies as much as they do teaching technologies and, in the process, they unnecessarily limit potential for learning.

Because the primary metaphor of almost all LMSs is the classroom, they can be a particularly poor fit with ways of teaching that have no classes, such as self-paced courses and MOOCs, individual projects, or flexible networked ways of learning such as those underpinned by Connectivist, or Rhizomatic models of learning. This is not to say that such uses are impossible. For example, assumptions about class schedules that are embedded in software (such as that all students must submit work by a certain deadline) can be disabled, or bypassed by setting a deadline in the far distant future, then manually informing students of when to submit their work. However, the fit with self-paced models of learning is typically poor. Among the many peculiarities that result are students who engage in discussions with ‘classmates’ who no longer have access to the provided forum, and the impossibility of collaboration when every student is at a different point in the course. More challengingly, and unlike teacher-paced courses in which the teacher can modify almost any aspect of the content or curriculum at will, knowing that the whole class will be affected in the same way, much confusion and even dismay can arise when changes are made to materials that may be in use by existing students.

Imperfect caricatures of physical spaces

In summary, most LMSs provide an automated set of metaphorical classrooms that harden many of the undesirable side-effects of educational systems in software, in ways that have little to do with how best to teach, and that inappropriately apply spatial metaphors in ways that conceal rather than illuminate their functions. Each bit of automation and each navigational decision hardens pedagogical choices, at least as much as the walls, doors, and physical limitations of physical spaces and, often, more. Programmers do not replicate physical classrooms but instead create or enlist new laws, new kinds of structure, and new kinds of hardened process that can be embodied in code. Classrooms solved problems of physics for in-person teaching and form part of a much larger structure that has evolved to teach reasonably well. LMSs just focus on a limited subset of teaching roles, and empower the teacher in ways that caricature their already excessive dominance in the classroom, that only occurred because of the nature of the physical space and the constraints it imposed.

LMSs leave much to be desired, but the metaphors on which they are based bear enough resemblance to physical reality to be readily understood by teachers and students. They usually provide just enough configurability and flexibility to more or less adequately work as teaching tools, for everyone, almost no matter what their level of digital proficiency might be. They more or less address the Faustian bargains listed earlier, albeit they normally do so by stifling what we wanted and should have been able to do in the first place with online tools, In the process they create new and quite extensive problems, as well as failing to replicate most of what makes physical universities work in the first place. Virtual learning environments are not like physical learning environments: they are only ever parts of them. There are other electronic ‘places’ to escape from them, such dedicated social media, or even just plain old email, but then all those Faustian bargains come back to haunt us again. They occupy space within the learner’s own physical environment, but it is rare for pedagogical designs to even acknowledge that, let alone to consider it in the design.

Improving the LMS

It is tempting, faced with these problems, to assume that they could be solved if only we made the LMS more closely resemble the physical environments on which it is modeled. However, this is a poor solution because, as we have already seen:

1)    physical environments create constraints and problems to solve that are unnecessary and avoidable in virtual systems; and

2)    it is not practical nor is it within our technical reach to replicate all the many incidental benefits of physical environments.

That said, there are lessons to be learned from physical spaces. Among many improvements that could be made would be:

1)    To make every part of the system at least potentially social: to allow synchronous and/or asynchronous dialogue to occur on every page or screen of the system. This is the default in all physical spaces: talking has to be prohibited if it is not wanted.

2)    To allow at least some parts of the system to be free of roles, or with more flexible roles, allowing all members of the system to create and share posts and resources using discretionary access control (so it is the poster’s responsibility to choose who can see it, and who can change it). Even in highly controlled physical environments, we choose what we reveal and to whom.

3)    To support social networking and the blurring of boundaries between areas, tools, and features of the site, so that courses are just one of many kinds of organizational unit, with selectively permeable boundaries through which others can pass, or with which they can overlap. Again, this is a default in physical spaces, that leak information through walls, floors, windows, and doors, that exhibit continuity of engagement when people enter or leave classrooms, that allow teachers to open doors to others, that admit a multiplicity of primary uses.

Though these improvements appear simple to achieve, adapting an existing mainstream LMS such as Blackboard, Canvas, Moodle, or BrightSpace to support them is fraught with difficulty.

By far the easiest of these improvements to make within an existing LMS is to make it more social. Achieving this within an existing course structure is a simple programming problem that can readily be solved in countless ways. In most LMSs, it could be built as a plugin. Existing architectures, in which courses and roles play a primary structural role, make it somewhat more difficult to extend such dialogue beyond the boundaries of the course. The metaphorical walls of a course are, for the most part, more of a barrier to engagement beyond it than those of a physical classroom because their metaphorical physics can be (and is) enforced in code. It is not, however, an insoluble problem. For example, a context-aware embeddable discussion system such as Disqus or Isso, hosted locally or remotely, could fairly easily be added.

Making the system free of roles is much more difficult because, in most LMSs, they underpin almost every function and structure of the system, and they cannot be made to work with an open, discretionary access- based model of permissions: the two approaches are, architecturally, mutually exclusive. One way of dealing with this would be to follow the lead of the Drupal content management system to support ‘organic’ groups: limited areas of the LMS where everyone has the same rights to create shared content or social areas, and where anyone can control who can see what they post. These areas could be as large or as small as desired but it would be difficult to make them extend beyond a course, or to encompass one or more courses. It would not be impossible, but to do it safely and reliably (without giving everyone a single, very powerful role) would require a major rewrite of the underlying LMS.

For all of the LMSs of which I am aware, the most difficult of all these improvements would be to blur the boundaries of the tools, features, and courses. The course is such a fundamental architectural unit of most, if not all, LMSs that changes to its operation would demand a significant redesign. It could be done, but it would not be the same kind of system any more.

It is for these reasons that, wishing to support all of these features and realizing the extreme difficulty of modifying the LMS without compromising some or all of its existing functionality, a group of us at Athabasca University created The Landing (Dron & Anderson, 2014), as a separate system to the LMS, linked only by a single sign-on and tenuous hyperlinks and, to a limited extent (only supporting public posts in either direction) RSS feeds. Further efforts to design deeper integration proved too difficult, for both technical and organizational reasons. Unfortunately, The Landing suffers from the same Faustian Bargain that besets all attempts to expand the range of systems available. The maintenance burden of a system with many thousands of users is too much to sustain for a system with very limited central support and even more limited funding. Pedagogically, the system fulfills an important need and so it has survived for more than 12 years but, technically and from a management perspective, its future is in jeopardy. Similar issues are playing out the world over. The more control and diversity that we enable, the more difficult and expensive it is to manage it.

Alternative approaches

Incremental improvements

Athabasca University is currently building an Integrated Learning Environment (ILE) that centres around very conventional elements of a institutional teaching system: an LMS, some relationship management tools, a student records system, an enrolment system, an examination management system, and so on. These are tightly integrated, but it is intended that the ILE will also embrace many other tools and systems that are far less institutionally bound, from the aforementioned Landing, to other social media (such as WordPress), to portfolio tools, to shared software repositories. This is an approach that starts with replicating existing structures and services by building a tightly managed administrative core, but that is intended to grow to support more open, diverse, and rich approaches to learning and teaching, co-evolving with methods and pedagogies that are more in keeping with the different problems and needs of distance learners. However, though it provides a managed approach to supporting change, this approach carries many risks.

A design approach that treats online systems as environments invariably makes the assumption that it is where everything associated with what goes on inside it happens, and (for online systems) this creates quite unnecessary restrictions on what can happen. Athabasca University’s design approach for its ILE was highly participative, engaging most of its teaching, technical, and administrative staff and asking for what they needed. However, inevitably, their requests were based on assumptions formed by their existing practices and, especially, by the existing environmental metaphors of the LMS and associated systems with which they were already familiar. In essence, they were asked what kinds of spaces they needed, and what kinds of stuff needed to be in those spaces for them to do what they currently do.  ‘Space’ and ‘stuff’ are what Stewart Brand (1997) describes as inevitably being the fastest-changing, most volatile parts of any physical building, after site (its physical limits), structure (what holds it up), skin (mainly the external walls), and services (electricity, gas, network wiring, etc). More abstractly, this is a solid structural principle that applies as much to ecosystems and educational systems as it does to buildings. As Brand himself observes, drawing from O’Neill, DeAngelis, Waide, & Allen (1986), the larger, slower-changing elements of any system affect the smaller, faster-changing more than vice versa. In physical spaces, these naturally tend to be bigger and/or more difficult to change, but the same is true in virtual spaces, where size seldom matters that much, but hardness (inflexibility, brittleness) has the same effect. The more difficult it is to make changes, the more an element of the system determines the behaviour of other elements in the system that interact with it. The ILE’s structure, skin and services have been designed based on needs determined by perceptions of the space and stuff within it, that were in turn very strongly determined by the LMS and other systems that went before, with all the inherited baggage that they inherited from in-person environments. Hence, the ILE’s fundamental design model is really no more than an extended LMS, and it inherits most of its weaknesses. The main way in which it differs is that it is designed to be extendable, but those extensions will still – in terms of how they are treated and used – be part of that same environment, with all the aforementioned problems that this entails.

Integrated learning infrastructures

I have argued that a better name for the system being developed at Athabasca University is not an ‘integrated learning environment’ but an ‘integrated learning infrastructure’ (ILI). In metaphorical terms, it should be like the utilities, services, and mechanisms that make an environment possible, but it should never be thought of as the environment itself.

Stripped to their essentials, digital systems intended to support the educational process provide services, consisting of tools that may be used to support learning, teaching, accreditation, and other roles and functions of an educational system.  Such services are many and various: discussions, presentations, file sharing, assignment submission & grading, quizzes, blogs, scheduling, wikis, bookmarking, real-time communications tools, enrolment systems, identity management systems, support systems, and much more. There is no good reason that these should be confined to loose approximations of their physical counterparts, nor is there any good reason that teachers or system administrators should be the only ones to control them, though it is important that each of them is owned by someone, otherwise the resulting free-for all would be difficult to manage. Microservice architectures that support such systems are quite mature, and widely implemented in different fields, if not so much within the educational sector. From the point of view of end users, these can be thought of as assemblable components, and the assemblies can be performed by anyone, including students. Ideally, it should be possible to integrate them with other applications and services offered beyond an institution, including on the desktop of individual students.

Ideally, it should be possible to assemble them into units with value in the system, that can themselves be assembled into other components. This provides a path for evolution from existing approaches because those units might include courses. There may be a need for additional services to support non-teaching functions associated with educational systems, such as administration or credentialling.

Such services are not so much environments as they are infrastructure that exists within and between the different environments that learners, teachers, administrators, and technicians occupy both virtually and in person. Non-exclusively, such infrastructure may minimally support needs such as:

  • Dialogue and interactions between participants
  • The presentation and curation of content
  • Assessment, formal and informal
  • Sharing of words, images, video, audio, and other document types
  • The formation of groups,  networks, and sets (social gatherings around shared interests or other commonalities)
  • Sharing of tools and resources
  • Etc.

What matters most is that all of these services can be combined in indefinitely many ways, by anyone.

This is not a new idea. In the early 2000s, the ELF (e-learning framework) and OKI (Open Knowledge Initiative) both attempted to provide ways to assemble services (ELF) or components (OKI) in many different ways. However, for the most part, both of these initiatives were firmly focused on building centralized systems that replicated the functions of an LMS, so they carried forward the assumption that what would be built from the components would be teaching environments; a better LMS, but still an LMS.

Around the same time as ELF and OKI were being developed, and driven by similar intents,  the notion of the personal learning environment (PLE) became popular, though with very many quite radically different interpretations (Martindale & Dowdy, 2010), ranging from institutionally controlled systems that were often described as ‘platforms’ (Yen et al, 2019) to collections of applications and services assembled by a learner on their own desktop in an ad hoc fashion (Wilson, 2008). Though some of the promoters of the concept saw the environment as extending beyond virtual systems, the vast majority of these interpretations considered only the digital tools, not the physical and social environment of the learner, nor the pedagogical and technical skills used by learners to create and manage those tools. Again, the ‘environment’ metaphor was inadequate and misleading. The PLE was also, for the most part, a concept, not a technology, though efforts were made in some circles to create standards for mashing up those tools, most notably through work on ELF which was, by some, seen as the VLE of the future (Wilson, 2005, cited in Martindale & Dowdy, 2010), and a number of systems were built that were described as PLEs, but that were essentially another kind of institutionally managed server, much like Athabasca Landing, referred to previously.  A more promising set of standards that did focus on the development of standards-based widgets that could be assembled by individuals as well as within an LMS or other system (Wilson, Sharples & Griffiths, 2008), failed to gain enough momentum, despite endorsement of the widget specification by the W3 Consortium, and implementations within all major operating systems. Meanwhile, the term ‘PLE’ itself became such an amorphous concept that even conversations about it were difficult to sustain, let alone useful implementations.

In more recent years, the Educause organization has vigorously promoted the Next Generation Digital Learning Environment (NGDLE), which is essentially very similar in purpose and approach to the earlier ELF initiative, but that:

1)    Takes into account the possibility of learners assembling their own digital toolsets;

2)    Incorporates developments in analytics and artificial intelligence, and

3)    that is largely agnostic to standards used for its implementation, although it does recommend standards and protocols such as xAPI, LTI, learning record stores, and Caliper to help bind them together (Brown, Dehoney, & Millichap, 2015).

Combining the best ideas from service-based systems and work on PLEs, the initiative shows promise. While, once again, the ‘environment’ metaphor fails to extend into the actual spaces that it is intended to be deployed, the initiative is a genuine move beyond the teacher-centric, classroom-inspired models of the LMS and towards a student-oriented service-provision model. There are now some implementations of the concept. For example, the OERu aggregates a wide assortment of open source tools systems providing services such as discussion, microblogging, blogging, wikis, social bookmarking, and so on, that can be used independently by students or as part of the university’s own system (Lane & Good, 2019). While these are still largely perceived as an environment composed of environments, the potential for such a design approach is to free us from the traditional classroom metaphors of the LMS.

Institutional teaching beyond virtual environments

A distance learner’s environment is never digital, though digital tools and services can comprise important parts of it. A learning environment is not just comprised of physical or virtual structures but also the social, pedagogical, organizational, personal, and other dynamic elements that determine how the parts of the structure evolve and interact. It is not just physical matter, or virtual systems, but also the people and what they do together. It is not just how teachers teach, but how learners teach themselves, and teach one another, and are taught by the countless teachers who create the websites, interactions, tools, and structures of the broader internet, and the many teachers who inhabit their own physical spaces, from family members to people in the street. How, therefore, should teachers in institutions teach, when they are just parts of someone else’s environment, co-players in the process, and what kinds of digital tools and systems will be needed to support that?

Perhaps one of the reasons that it is too easy to fall into the trap of thinking of the digital tools and systems as an environment is that it they are  an obvious class of things around which to put a boundary. However, an infrastructure is not just the digital tools but also the human-enacted methods, rules, protocols, and standards that accompany it. It is not just what we use, but the ways that we use it.  It is natural to focus mainly on the software and hardware when designing an online system to support learning, and thus to come to think of it as providing the learning environment itself. If, instead, we remember that we are only building tools to use in the learner’s own environment, and that we are just providers or curators, not controllers or managers of that environment, then a critical and oft overlooked design principle becomes clear: that online students are the primary orchestrators of their learning rather than, as in the physical classroom, their teachers.

An integrated learning infrastructure should therefore not attempt to replicate the form and structure of a traditional classroom, nor should it solely support teachers in assembling the tools needed for their teaching. Instead, the focus – both digitally and pedagogically – should be on making it possible for learners to assemble the services into their environments themselves, in order to avail themselves of the support they need, when they need it, for the purposes they intend. The processes, methods, techniques, tools, and structures that students bring with them are at least as important as those created by their teachers. An integrated learning infrastructure needs to support these aspects at least as much as the interconnections between software tools. Again, it is necessary to think of the environment as considerably more than just a set of digital components that it uses but one that includes the people, the spaces they inhabit, and the things that they do.  Pedagogically as well as technically, there may be a need to support students in making the best use of all of that, for instance to search well, to find people that can help them to learn, to organize their own learning process, but such support Is, again, a service on which students may draw, not a teacher-determined requirement. And those pedagogies themselves need to adapt: for example, those that rely on rewards and punishments to enforce compliance must be excised, while those that provide learners with autonomy should be amplified. New pedagogies will be needed that acknowledge the many teachers in a learner’s environment, that help them to traverse the complexity of it, to leverage the advantages and to avoid the pitfalls. Teachers will need to let go, but stay close.

Tools that involve engagement with others – the means to share, the means to discuss, the means to work together, schedule meetings, and so on – are connection points in learners’ environments that cannot usually be completely controlled by any one of them, because of the need to at least agree protocols through which to engage and, in many cases, the systems which they will use to interact . One way to deal with this problem is to make a decision to use a  small range of tools, ideally in consultation with students. A better approach is to use tools that give students a choice of toolset, using protocols or standards such as SMTP, Jabber,  iCal, WebMention, ActivityStreams, or NNTP. However, few new standards have gained traction in recent years thanks to the dominance of closed social media monoliths intent on locking users in to their systems, so this may unnecessarily limit the range of systems that may be used. Another approach, commonly used in Connectivist approaches to learning, is to aggregate what learners provide themselves, using standards like RSS or Atom, or proprietary APIs offered by tool providers, or mailing systems to collect what students have shared elsewhere. If that is impossible, even simple copy-and-paste by human beings (students, teachers, or others) may be sufficient to connect multiple systems: not everything in an ILI needs to be implemented in software. For example, student blogs may be shared through flexible  technologies such as email and messaging apps, then copied by themselves or by their teachers into shared wikis. One  interesting benefit of such approaches is that they can support both diversity and manageability, inasmuch as the management burden may be shared by the participants rather than taken on by a single teacher or institution. Students may choose which tools they use, rather than having them chosen by the teacher. This is the principle used by Connectivist MOOCs (Downes, 2008), in which one site aggregates the shared artefacts created in many different learner-managed systems.

A learner’s environment consists of much more than the digital tools and systems offered by an institution. While, to a large extent, much of this environment may be unknowable to their designated teachers, there is much value to those who seek to support student learning in discovering how they are learning, and what constitutes their learning environment. Learning – the process, the tools, and the ways of learning, and not just the products – must be made visible if teachers, including other students, are to help learners to learn (Hattie, 2013). Much use can be made of pedagogical approaches such as shared learning diaries or blogging, and some careful use may even be made of automated systems that indicate presence, or that record traces of visits, as long as their role is to provide support for understanding student learning, and not to provide the teacher with means to control of the student. Beyond individual courses, there may be much pedagogical value in encouraging learners to share their learning experience through media such as blogs, microblogs, and other online tools, which may (as long as means are available for the student to control their privacy as needed) be aggregated and shared across their whole distributed, diffuse environment. Rather than replicating the necessarily closed and time-limited nature of the classroom, the artefacts of learning and the relationships that are developed in the process may persist indefinitely. Connectivist MOOCs provide a useful model for this. For example, Cormier (2014) talks of ‘Zombie MOOCs’ in which learning and interaction persist long after the course itself is over.

Bringing about such changes at an institutional level requires both bottom-up and top-down support. Teacher’s pedagogies are normally  more malleable than digital tools, because they can adapt rapidly to any tools: they are, in Brand’s terms (Brand, 1997), the ‘stuff’. However, they are therefore also the most constrained by the structures into which they must slot, and the least able to significantly impact things at structural level.  A single teacher, or even a small group of teachers pressing for change is therefore unlikely to sway either institutional policy or the design of the LMS because, as we have seen, one LMS must address the needs of all, so anything that changes it must suit everyone.

From the top down, replacing the LMS with an integrated learning infrastructure is a necessary step towards breaking out of the vicious loops that prevent the pedagogies and structures from evolving. At first, an ILI will naturally resemble the LMS it replaces, because its boundaries will continue to be largely determined by the less flexible layers above it: the institutional forms and structures such as courses, credentials, legislation, and teachers’ employment contracts. It is important to remember that the LMS was originally designed not just to replicate classroom behaviours but to fit into the larger, slower-changing structures and systems of institutions, and that significant changes in how we teach will not occur unless those structures and systems also evolve. They create the boundaries within which the ILI operates and, to a large extent, are not just containers of it, but part of it. An infrastructure is not just the digital tools but also the human-enacted methods, rules, protocols, and standards that accompany it. It is not just what we use, but the ways that we use it.  However, unlike the LMS, in an ILI those boundaries will be malleable. This opens up opportunities for the structure, skin, and services to in turn change.

The opportunities for change may not be taken, at first, at least in part because the signals (such as qualified students, their credentials, and so on) that pass in and out of the boundaries of the university will go to and from governments, employers, and other institutions that may not be prepared for radical change, even if the institution itself is committed to it. If, say, other institutions insist on grade point averages for standardized courses, then it will be difficult to completely avoid providing them, or something that is recognizably equivalent.  However, the adjacent possible empty niches (Kauffman, 2019) that an ILI supports will inevitably be filled by those who see the opportunities it entails, from courses whose lengths are pedagogically determined, to integration of lifelong and workplace learning, to new forms of credentials and learning. Perhaps, if enough institutions start to adopt such practices, we may break free of the insular single-institution model of education altogether. Out of this may grow a truly learner- and learning-driven future, in which learners draw on services from multiple educational providers, leading to a vast participative system in which institutions meld or blend to offer support for learning not just any time and any place, but every time and every place.

References

Blum, S. D., & Kohn, A. (2020). Ungrading: Why rating students undermines learning (and what to do instead). West Virginia University Press.

Bouygues, H. L. (2019). Does Educational Technology Help Students Learn? https://reboot-foundation.org/does-educational-technology-help-students-learn/

Brand, S. (1997). How buildings learn. Phoenix Illustrated.

Brown, M., Dehoney, J., & Millichap, N. (2015). The next generation digital learning environment. A Report on Research. ELI Paper. Louisville, CO: Educause April, 5(1), 1-13.

Cormier, D. (2014). Community learning – the zombie resurrection. http://davecormier.com/edblog/2014/05/25/community-learning-the-zombie-resurrection/

Downes, S. (2008). Places to Go: Connectivism & Connective Knowledge. Innovate, 5(1). http://www.innovateonline.info/pdf/vol5_issue1/Places_to_Go-__Connectivism_&_Connective_Knowledge.pdf

Dron, J.  (2006). Any color you like, as long as it’s Blackboard®.  In. Hawaii: AACE.

Dron, J., & Anderson, T. (2014). Teaching crowds: Learning & Social Media. AU Press. http://teachingcrowds.ca

Dron, J. (2016). P-learning’s unwelcome legacy. TD Tecnologie Didattiche, 24(1), 72-81. http://www.tdjournal.itd.cnr.it/article/view/891

Dron, J. (2021). Educational technology: what it is and how it works. AI & SOCIETY, 1-12. https://doi.org/10.1007/s00146-021-01195-z

Greene, E. B. (1928). The relative effectiveness of lecture and individual reading as methods of college teaching. Genetic Psychology Monographs.

Hattie, J. (2013). Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement. Taylor & Francis.

Kauffman, S. A. (2019). A World Beyond Physics: The Emergence and Evolution of Life. Oxford University Press.

Kelly, K. (2009). Triumph of the Default. http://www.kk.org/thetechnium/archives/2009/06/triumph_of_the.php

Keuning, H., Jeuring, J., & Heeren, B. (2018). A systematic literature review of automated feedback generation for programming exercises. ACM Transactions on Computing Education (TOCE), 19(1), 1-43.

Kohn, A. (2011). The case against grades. Educational Leadership, 69(3), 28-33.

Kohn, A. (2015). Four Reasons to Worry About “Personalized Learning”. http://www.alfiekohn.org/blogs/personalized/

Lane, D. C., & Good, C.  (2019). OERu’s delivery model for changing times: An Open Source NGDLE.  In World Conference on Online Learning, Dublin, Ireland. https://docs. oeru. org/s/fXQk2rJbzWCk8ia.

Martin, F., Chen, Y., Moore, R. L., & Westine, C. D. (2020). Systematic review of adaptive learning research designs, context, strategies, and technologies from 2009 to 2018. Educational Technology Research and Development, 68(4), 1903-1929.

Martindale, T., & Dowdy, M. (2010). Personal learning environments. Emerging technologies in distance education, 177, 193.

O’Neill, R.V., DeAngelis, D.L, Waide, J. B., & Allen, T. F. H. (1986). A Hierarchical Concept of Ecosystems. Princeton University Press.

Postman, N. (1998). Five things we need to know about technological change. Denver, Colorado, 28.  https://student.cs.uwaterloo.ca/~cs492/papers/neil-postman–five-things.html

Protopsaltis, S., & Baum, S. (2019). Does Online Education Live Up to its Promise? A Look at the Evidence and Implications for Federal Policy.

Ryan, R. M., & Deci, E. L. (2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. Guilford Publications.

Tichavsky, L. P., Hunt, A. N., Driscoll, A., & Jicha, K. (2015). “It’s Just Nice Having a Real Teacher”: Student Perceptions of Online versus Face-to-Face Instruction. International Journal for the Scholarship of Teaching and Learning, 9:2. http://digitalcommons.georgiasouthern.edu/ij-sotl/vol9/iss2/2/

Verbert, K., Duval, E., Klerkx, J., Govaerts, S., & Santos, J. L. (2013). Learning analytics dashboard applications. American Behavioral Scientist, 57(10), 1500-1509.

Wilson, S. (2008). Patterns of personal learning environments. Interactive learning environments, 16(1), 17-34.

Wilson, S., Sharples, P., & Griffiths, D.  (2008). Distributing education services to personal and institutional systems using Widgets.  In Proc. Mash-Up Personal Learning Environments-1st Workshop MUPPLE 8 (pp. 25-33).

Yen, C.-J., Tu, C.-H., Sujo-Montes, L. E., Harati, H., & Rodas, C. R. (2019). Using personal learning environment (PLE) management to support digital lifelong learning. International Journal of Online Pedagogy and Course Design (IJOPCD), 9(3), 13-31.

Informal Learning in Digital Contexts | Handbook of Open, Distance, and Digital Education

This is the second of two chapters by Terry Anderson and me (the other being on the topic of pedagogical paradigms, that I shared a week or two ago) from Springer’s Handbook of Open, Distance, and Digital Education.

The ‘paradigms’ chapter more or less wrote itself – we’ve churned those ideas around for long enough now that we both know the topic rather well – but this one caused us a lot more trouble. Our difficulties were largely due to the fact that we started out with roughly as much idea about what the term ‘informal learning’ means as anyone else. In other words, we kind of recognized it when we saw it, but could come up with no plausible definition that was not either simply wrong, incomplete, or vaguely defined as ‘not formal’ (sometimes adding the utterly circular cop-out notion of ‘non-formal’). As we later figured, ‘formal’ is no better defined than ‘informal’, so that didn’t help. Faced with the need to cover a fairly representative sample of work in the area, we therefore made a mess of it. Our initial draft consisted mainly of a set of examples culled mainly from Terry’s encyclopaedic knowledge of the literature in the field, bound together in loosely connected themes. Because the literature we were citing was based on a large, vague, and often mutually contradictory variety of understandings of ‘informal learning’ the chapter reflected this too: the parts were fine, but the whole was quite incoherent. We needed a better framework.

So we started to brainstorm a few different ways of thinking about the problem, looking at as many ways the term was used as we could find, identifying common patterns and frequently associated concepts, trying to distinguish necessary from sufficient conditions, and consequently finding a much bigger mess than the one we had started with. The amount of fuzzy thinking and loose, almost arbitrary terminology found in the field of informal learning turns out to be quite staggering. It’s not a field: it’s a jungle.

Not for the first time, though, I found Michael Erault’s work in the area to be an inspiration and source of clarity. Erault doesn’t try to come up with a single defining characteristic, instead recognizing that there is a richly variegated continuum of informal-to-formal ways that people learn from and with one another (at least in the workplace settings he has studied). Although (as far as I know) he didn’t  explicitly use the term, the sets of characteristics that Erault uses to identify relative degrees of informality seemed to me to imply that he was thinking in terms of what Wittgenstein described as Familienähnlichkeit (family resemblances). No single cluster of characteristics define learning as informal (or formal, for that matter) but, if enough are present, we can usually recognize it as one or the other, or somewhere in between.

This gave us a useful starting point, but it still left a lot of vagueness, and  Erault’s focus on informal workplace learning did not fully address all of the meanings and instantiations of informal learning that are particularly significant when examining digital contexts – all the stuff that happens in exchanges through social media, for instance, from Quora to YouTube tutorials and back through email, Reddit, and Twitter. Also, it seemed to gloss over the formal stuff which (as we noted) is as poorly defined as ‘informal’, and that almost never occurs in anything resembling a ‘pure’ form: there is hardly ever any formal learning without informal learning lurking close by. It would be a lot easier if we just talked about formal teaching, because that does refer to a much clearer set of better-defined activities, but teaching is not at all the same thing as learning. Indeed, sometimes the relationship is very oblique indeed, notwithstanding Frere’s claims that you cannot call it teaching unless learning occurs. And then there’s the complex role of credentials of various kinds in both assessing and influencing learning. We wanted to find a way to capture the richness of that, but could find no existing work that worked well enough for us.

We went through a lot of different concepts and representations (yes, there were Venn diagrams!) before finally hitting on the notion that it is not so much a two-dimensional continuum between formal and informal, but a multi-dimensional spectrum defined in terms of relative degrees of dependence/independence and intentionality/non-intentionality.

 

Informal learning as a 3D continuum, with dimensions of dependence/self-direction and incidental/intentional

We (tentatively) reckon that we can situate at least most existing work in the field within this framework, and that it provides a helpful way of thinking about whatever is happening in a particular moment of a learning trajectory (another concept from Erault that I’ve found very useful in the past, especially when talking about transactional control in my first book). An individual’s learning trajectory will constantly wind around this space and, when other individuals are involved (not just formal teachers), their paths will affect one another in interesting ways. After we’d worked this out, the rest of the chapter fell more or less into place. You can read the result here.

Here’s the chapter abstract:

Governments, business leaders, educators, students, and parents realize the need to inculcate a culture of lifelong learning – learning that spans geography, time, and lifespan. This learning has both formal and informal components. In this chapter, we examine the conceptual basis upon which informal learning is defined and some of the tools and techniques used to support informal learning. We overview the rapid development in information and communications technologies that not only creates opportunities for learners, teachers, and researchers but also challenges us to create equitable and culturally appropriate tools and contexts in which high-quality, continuous learning is available to all.

Reference

Dron J., Anderson T. (2022) Informal Learning in Digital Contexts. In: Zawacki-Richter O., Jung I. (eds) Handbook of Open, Distance and Digital Education. Springer, Singapore. https://doi.org/10.1007/978-981-19-0351-9_84-1

Pedagogical Paradigms in Open and Distance Education | Handbook of Open, Distance, and Digital Education

This is a chapter by me and Terry Anderson for Springer’s new Handbook of Open, Distance, and Digital Education that updates and refines our popular (1658 citations, and still rising, for the original paper alone) but now long-in-the-tooth ‘three generations’ model of distance learning pedagogy. We have changed the labels for the pedagogical families this time round to ones that I think are more coherent, divided according to their epistemological underpinnings: the objectivist, the subjectivist, and the complexivist. and we have added some speculations about whether further paradigms might have started to emerge in the 11 years since our original paper was published. Our main conclusion, though, is that no single pedagogical paradigm will dominate in the foreseeable future: that we are in an era of great pedagogical diversity, and that this diversity will only increase as time goes by.

The three major paradigms

Objectivist: previously known as ‘behaviourist/cognitivist’, what characterizes objectivist pedagogies is that they are both defined by assumptions of an objective external reality, and driven by (usually teacher-defined) objectives. It’s a paradigm of teaching, where teachers are typically sages on the stage using methods intended to achieve effective learning of defined facts and skills. Examples include behaviourism, learning styles theories, brain-based approaches, multiple intelligence models, media theories, and similar approaches where the focus is on efficient transmission and replication of received knowledge.

Subjectivist: formerly known as ‘social constructivist’, subjectivist pedagogies are concerned with – well – subjects: they are concerned with the personal and social co-construction of knowledge, recognizing its situated and always unique nature, saying little about methods but a lot about meaning-making. It’s a paradigm of learning, where teachers are typically guides on the side, supporting individuals and groups to learn in complex, situated contexts. Examples include constructivist, social constructivist, constructionist, and similar families of theory where the emphasis is as much on the learners’ growth and development in a human society as it is on what is being learned.

Complexivist: originally described as ‘connectivist’ (which was confusing and inaccurate), complexivist pedagogies acknowledge and exploit the complex nature of our massively distributed cognition, including its richly recursive self-organizing and emergent properties, its reification through shared tools and artefacts, and its many social layers. It’s a paradigm of knowledge, where teachers are fellow learners, co-travellers and role models, and knowledge exists not just in individual minds but in our minds’ extensions, in both other people and what we collectively create. Examples include connectivism, rhizomatic learning, distributed cognition, cognitive apprenticeship, networks of practice, and similar theories (including my own co-participation model, as it happens). We borrow the term ‘complexivist’ from Davis and Sumara, whose 2006 book on the subject is well worth reading, albeit grounded mainly in in-person learning.

No one paradigm dominates: all typically play a role at some point of a learning journey, all build upon and assemble ideas that are contained in the others (theories are technologies too), and all have been around as ways of learning for as long as humans have existed.

Emerging paradigms

Beyond these broad families, we speculate on whether any new pedagogical paradigms are emerging or have emerged within the 12 years since we first developed these ideas. We come up with the following possible candidates:

Theory-free: this is a digitally native paradigm that typically employs variations of AI technologies to extract patterns from large amounts of data on how people learn, and that provides support accordingly. This is the realm of adaptive hypermedia, learning analytics, and data mining. While the vast majority of such methods are very firmly in the objectivist tradition (the models are trained or designed by identifying what leads to ‘successful’ achievement of outcomes) a few look beyond defined learning products into social engagement or other measures of the learning process, or seek open-ended patterns in emergent collective behaviours. We see the former as a dystopic trend, but find promise in the latter, notwithstanding the risks of filter bubbles and systemic bias.

Hologogic: this is a nascent paradigm that treats learning as a process of enculturation. It’s about how we come to find our places in our many overlapping cultures, where belonging to and adopting the values and norms of the sets to which we belong (be it our colleagues, our ancestors, our subject-matter peers, or whatever) is the primary focus. There are few theories that apply to this paradigm, as yet, but it is visible in many online and in-person communities, and is/has been of particular significance in collectivist cultures where the learning of one is meaningless unless it is also the learning of all (sometimes including the ancestors). We see this as a potentially healthy trend that takes us beyond the individualist assumptions underpinning much of the field, though there are risks of divisions and echo chambers that pit one culture against others. We borrow the term from Cumbie and Wolverton.

Bricolagogic: this is a free-for-all paradigm, a kind of meta-pedagogy in which any pedagogical method, model, or theory may be used, chosen for pragmatic or personal reasons, but in which the primary focus of learning is in choosing how (in any given context) we should learn. Concepts of charting and wayfinding play a strong role here. This resembles what we originally identified as an emerging ‘holistic’ model, but we now see it not as a simple mish-mash of pedagogical paradigms but rather as a pedagogic paradigm in its own right.

Another emerging paradigm?

I have recently been involved in a lengthy Twitter thread, started by Tim Fawns on the topic of his recent paper on entangled pedagogy, which presents a view very similar indeed to my own (e.g. here and here), albeit expressed rather differently (and more eloquently). There are others in the same thread who express similar views. I suggested in this thread that we might be witnessing the birth of a new ‘entanglist’ paradigm that draws very heavily on complexivism (and that could certainly be seen as part of the same family) but that views the problem from a rather different perspective. It is still very much about complexity, emergence, extended minds, recursion, and networks, and it negates none of that, but it draws its boundaries around the networked nodes at a higher level than theories like Connectivism, yet with more precision than those focused on human learning interactions such as networks of practice or rhizomatic learning. Notably, it leaves room for design (and designed objects), for meaning, and for passion as part of the deeply entangled complex system of learning in which we all participate, willingly or not. It’s not specifically a pedagogical model – it’s broader than that – though it does imply many things about how we should and should not teach, and about how we should understand pedagogies as part of a massively distributed system in which designated teachers account for only a fraction of the learning and teaching process. The title of my book on the subject (that has been under review for 16 months – grrr) sums this up quite well, I think: “How Education Works”. The book has now (as of a few days ago) received a very positive response from reviewers and is due to be discussed by the editorial committee at the end of this month, so I’m hoping that it may be published in the not-too-distant future. Watch this space!

Here’s the chapter abstract:

Building on earlier work that identified historical paradigm shifts in open and distance learning, this chapter is concerned with analyzing the three broad pedagogical paradigms – objectivist, subjectivist, and complexivist – that have characterized learning and teaching in the field over the past half century. It goes on to discuss new paradigms that are starting to emerge, most notably in “theory-free” models enabled by developments in artificial intelligence and analytics, hologogic methods that recognize the many cultures to which we belong, and a “bricolagogic,” theory-agnostic paradigm that reflects the field’s growing maturity and depth.

Reference

Dron J., Anderson T. (2022) Pedagogical Paradigms in Open and Distance Education. In: Zawacki-Richter O., Jung I. (eds) Handbook of Open, Distance and Digital Education. Springer, Singapore. https://doi.org/10.1007/978-981-19-0351-9_9-1