The tribal nature of business, education, and technology

I've always thought of EdTech as a tripod with three equally important legs. When you look at the failures of EdTech ventures, it's usually obvious which leg gave out. Universities with impeccable credentials produce great technologies that are not commercially viable.  

When it all works as it should

Venture capitalists pour money into can't-miss start-ups that leverage the latest tech only to find the education market a far tougher nut than they expected. Or a great business full of true educators will limp along with inefficient systems, bungling the user experience with disjointed or out-of-date technology. Why is it so hard to get all three working in sync? I think the answer is more sociological than logical. The three tribes speak different languages, have different cultures, and chase different goals. And they just don't trust one another. 

When business and tech are joined at the hip, they often see the whole educational system as a morass of groupthink without practical, real-world foundations. Where education and technology live together in wedded bliss, the business world can be viewed as a shark tank full of mercenaries who will sell out their own mothers for one more positive earnings report. When education and business are deeply aligned, they can easily view technology as an incomprehensible whirlpool of change for the sake of change, coming with high risk, high cost, and dubious reward. 

When you start looking at the differences among the three in terms of culture, vocabulary, and ideals, it's a wonder any EdTech venture survives. And it's no wonder that those who successfully align all three find great success.

It's easy to get like-minded people to collaborate, but breaking down the barriers between suspicous tribes so that they truly begin to listen, to speak openly, to solve problems together? That's hard work. It requires commitment, patience, and empathy, and creativity. I once solved this in a course development crunch by putting one member (at least) of each tribe onto a course development team, then giving each of them full authority in their own area of expertise, and veto power over every decision. Anyone could call a halt. Resolving disagreements required actually listening, learning about others' concerns, and coming up with solutions that worked for all. 

But that was just a structural solution. It didn't address the underlying distrust. For that, we had to find a common ideal, a flag around which all could rally. And we found one: Every student is also a customer and an end-user. To be truly student-centric, we needed serve all three. A happy end-user needs reliable and suitable technology. A satisfied customer needs to have a great experience and receive value. And a successful student needs to learn and grow and find practical applications without the hassle of user and customer issues. It takes a team of experts working together to deliver on all three.   

We created 10 teams and developed a full master's degree program in just a few months. The product was a huge success. Four of the teams worked smoothly, exactly as hoped. Three of the teams failed at some level and had to be reconstituted. And three of the teams succeeded so thoroughly that they became lifelong friends who wanted all their children to grow up and fall in love and marry one another. Okay, I made that last part up, but the truth isn't vastly different. 

This "EdTech Tripod" principle holds whether you're creating a company, developing product, building software, reengineering processes, downsizing, or scaling up. If you're in EdTech you need the full Tripod in place and operating smoothly. 

Questions? Thoughts? Let me know! I speak all three languages fluently (plus a little French). 

Rethinking the bad rap of rote

You know what gets a bad rap? Rote learning. Memorization. Repetition. In most classrooms, critical thinking smiles winningly and sets an apple on the teacher's desk while rote learning sulks in a corner facing the wall. The higher up the food chain, the less interested teachers seem to be in teaching by memorization. Even corporate trainers, who work in a realm where one would assume only outcomes matter, disdain this tried and true methodology. That's unfortunate, and here's why:

Automaticity. Researchers have found that automaticity, the ability to perform a task or recall a fact without thinking, is a key component of expertise. Benjamin Bloom went so far as to say that overtraining, practicing well beyond the level of competence, is the hallmark of excellence. He called automaticity "the hands and feet of genius." When you don't have to think about the basics anymore, your mind can concentrate on the finer points. The first time you drive on a highway you're just trying to stay alive. The ten-thousandth time you're only thinking about the fastest route home or the best way to get in front of the truck spewing blue smoke. That's expertise based on automaticity. And there's only one way to get automaticity, and that's through practice. Repetition. Doing something over and over until you can do it in your sleep. 

Expertise requires automaticity, and automaticity can be created through rote learning. So in essence, rote shortens the path to excellence. Why don't we take take the short-cut? It's not like this principle is lost to us. It's front and center in the most significant shift in our daily lives since the Internet: artificial intelligence. 

The way you teach a machine to think is by giving it hundreds of thousands of examples and then asking the software to do the same thing over and over while you assess the results and tweak the algorithms. In most cases, neural networks have to be trained on enormous amounts of data, doing millions of calculations over and over in order to learn. That's how we teach robots to think like humans. 

Above: "Teaching a robot to think," generated by AI

I know a lawyer who joined a new firm and was asked to give presentations to potential clients. He was required to memorize these presentations verbatim. As odd as it seemed at first, he became fluent in the main points of this new area of law almost overnight. From there he was able to dig deeper, to ask the right questions, to build on the foundation that was established by rote. That was speed to expertise. 

I led a training development project for a new venture in which front-line employees needed to interact with customers, and get up to speed very quickly. These employees were sourced and paid on the level of fast-food workers, but they weren't flipping burgers; they were dealing with people's nutritional health. We created five "mantras," short, simple, plain-language sentences that could be stated verbatim to customers. There was one for each of the five areas of expertise required: brand promise, nutrition, process, pricing, and product. Then each mantra had two "power phrases," which were slightly longer sentences that explained that mantra. It was one page of memorization that also served as the framework for all the other training to follow. We created short videos showing each mantra and each power phrase in action, to help with both memorization and delivery skills. We drilled and practiced. Voila. Speed to expertise via rote learning. 

It's funny how sometimes a big idea is an old idea with a new application. We have so many elearning tools that support, or could easily support, rote learning. Maybe we should think about why we teach machines the way we do, and draw some application from that. Maybe it's time for the old to become new again.  

The never-ending challenge of the video lecture

There is nothing worse in online education than a bad video lecture. This is a claim that needs no support, not if you've ever experienced the brain fog and wandering thoughts and heavy eyelids and overwhelming sense of deadening obligation uniquely generated by the monotonous drone of a back-lit talking head swimming in digital artifact while overexplaining an overpacked slide in a tinny, echoing voice. The ongoing challenge, of course, is that a poorly-executed video lecture is dead easy, while producing a quality online video presentation is not. But if you are a teacher, presenter, producer, instructional designer, or product developer in this space and these times, you really have no choice but to put in the work. Downshift into low gear, let out the clutch, put the pedal down and power yourself out of the boring muck.   

Here are my top five key elements for video lectures, from hard-won experience. 

1. Audio first. As counter-intuitive as it may seem, the quality of your audio is actually more important than the quality of your video. Think of it this way, if all else goes south at least a learner can pop in their earbuds and take a long, invigorating walk. But if the audio is hard to take for long stretches, you're sunk before you leave the harbor. Good audio is always the backbone of good video.

2. Bits and bites. Don't ask anyone to watch a 20, 30, or 60-minute video. The only thing people ever watch in their real lives for even thirty minutes straight are things produced with multi-milion dollar budgets. You aren't in a classroom, no matter where your video was recorded. You're on someone's laptop, or phone, or tablet, or television. You're in their space, competing with the next installment of Mission Impossible. Don't compete. Cut your presentation up into small, proccessable chunks that match the content. Make it hard on yourself so it's easy on them.

3. Eye contact. A presenter on video actually has an enormous advantage over a presenter in a classroom in this one, important regard: you can look every person in the eye all the time. Or at least, that's how participants percieve it whenever you are looking at the camera. This is powerful. Take advantage of it, even if it means practicing until you are finally comfortable gazing cheerfully into that black hole of a camera lens.  

4. The right light. Pay attention to what you look like, just as you would if you were getting a portrait taken by a professional photographer. It doesn't need to look formal; it just needs to look good. Three-point lighting is what people expect, even though they may not know it. It's the professional standard we see all the time, the unconscious bar that, when you fall short of it, tells people this is an amateur production: Key light, fill light, back light. In many cases, bad lighting is simply back and fill light without any key light. Add that, and you're golden. Here's what Wikipedia knows.

5. Good graphics. Nothing says "I don't care" quite so eloquently as a long series of text-only slides read aloud. Make sure your slides are visually appealing, and don't overpack them. Don't read them verbatim. Use bullet points that require further explanation. Annotate as you teach, underline, circle, draw stars and exclamation points. Not everything is equally important; accentuate the big ideas. Add graphs, add images. The students pictured above were generated with Microsoft's Copilot. Fun and easy.

The list goes on, but these are at the top. And always remember, the lecture portion will never be the most engaging part of your course no matter what you do. The greatest payoff will come from time, energy, creativity, and resources that you invest in developing activities and assignments that truly spark the imagination.  

Okay, so... AI and eLearning

 A) Artificial Intelligence will improve everything. 

B) Artificial Intelligence will ruin everything.

 C) No one has a clue as to what AI will do to our field. 

Pick one and defend it. 

Or not. Here's the thing about Artificial Intelligence. It is always guided by human intelligence, and unless or until the Singularity occurs, it always will be. That means that wherever AI shows up doing something, somebody somewhere had that thing in mind. And they spent a lot of time and effort to make sure it would succeed. 

I'm not an engineer, much less an AI engineer, but the principles are not all that mysterious. People have invented a way to make machines, which in AI terms means software, that can be programmed to do things that were previously only possible for humans. And once you train the software, it can often do those things better (i.e., more consistently, more precisely, and with fewer coffee breaks) than human beings. But it takes smart people with a particular set of skills and a specific purpose in mind to accomplish this.  

And data. It takes lots of good, clean data to train a machine to produce human-like outputs. The best kind of results come from databases developed in fields with lots of records that have to be scrupulously accurate, like the medical field. But that kind of data is hard for developers to get their hands on. There are laws about such things. Those who have managed to do so have built some very robust AI systems for diagnosis. More often, systems rely on abundant free data, which has to be cleaned up. Like, searching the Internet for written examples of English language usage.  

And that brings us to ChatGPT. This system uses a boatload of... let's just call it unpristine data. ChatGPT had to be carefully trained, and it has to be consistently retrained, in order to generate the desired result. For ChatGPT, the desired result is text that looks, sounds, and feels like it was written by an articulate human being. And that's it. At it's root, ChatGPT is what AI engineers call, in their own highly technical terms, a "people pleaser." Its fundamental purpose is not accuracy. It succeeds when people who read its output are happy with it. "Is this good?" Yes or no. "What would make it better?" People have to answer these questions in order to keep retraining it. 

Google's search algorithm, by way of contrast, is designed for accuracy. Twitter's search has the goal of presenting you with posts that are provocative. X wants your clicks, likes, retweets. It doesn't care if you like what you get--maybe better if you don't. And it is not concerned at all about reflecting reality.

So where is all this leading us in EdTech, in online learning, eLearning, digital education? It actually gives us hope. Look at the academic integrity front. For the most part it will not be students bent on cheating who will be developing deep-learning neural networks, but their professors. In the neverending cat-and-mouse, the cat has the resources. 

More good news on the learning front comes from none other than Sal Khan of Khan Academy. He has the feel-good EdTech AI story of the hour:

Khan Academy announces GPT-4 powered learning guide

Carefully designed AI can do a great job in support of teachers and learners. It just takes smart people with a particular set of skills and a higher purpose in mind to accomplish it.