The Future of Education
Let’s not fix education; let’s blow it up and start over
If I had told you twenty years ago that almost everyone would trade land lines and laptops for mobile phones, you probably wouldn’t have believed me. And yet, that transition happened without much complaining, because consumers weren’t looking for a better landline phone. They were looking for freedom.
In the same vein, we shouldn’t try to fix education. We should replace it with something better. Just in case it isn’t clear, read this book and then come back to finish this essay:
If that isn’t enough, watch this …
As Tyler Cowen says, “America needs a radical move away from credentialism.” People say it’s easy to criticize. They ask: “What will you do without a college education?”
The answer is: work and learn. By the time you’re 23, you’ll probably be pretty far ahead of your other self just coming out with a degree and its associated debt.
Think of it this way: In the 21st century, you’ll learn 99 percent of what you need to learn after school. So why spend 4 percent of your life at university?
While a four-year college/university education may benefit a few students, I believe it will simply fade away like cordless phones and answering machines. In twenty years, a four-year college education could be the exception, not the norm. I’m sure it will hold on much longer than that, but companies like Google, Tesla, Amazon, and others are already signaling that degrees don’t matter, skills and attitude do. Fewer and fewer companies will outsource their hiring decisions to name-brand educational institutions.
Could high school be the new college?
My big hairy audacious suggestion is to prepare students from grades 9 to 12 to enter the workforce and a lifetime of integrated working and learning. That means throw out all the testing, grades, grade levels, teachers, lectures, problem sets, textbooks, college-prep, SATs, college applications, tuition, and babysitting grown-up kids, and focus on preparing for the real world at age 18. This “no university” approach includes careers in health care, finance, research, aviation, even law. It will take time before the law degree goes the way of the dodo, but I think it will happen in my sons’ lifetimes. I also hope we do away with most professional licensing, but that’s another essay.
I break this new high-school curriculum into two categories: context and skills.
Context
Much of this is learning how to learn. First thing to do is learn to set context. I think it’s helpful to understand these differences:
Complicated vs complex — Also called “ordered vs unordered.” A 747 may be complicated, but if you do enough research on an accident you can pinpoint what went wrong and make sure that problem never comes up in future planes, whereas a horse race or a tornado or starting a new company is impossible to repeat because of the butterfly effect. In the complicated domain we have mastery. In the complex domain, where humans spend a lot of their time, things are never black and white. Cause and effect are very difficult to determine. Plans are often useless. Here, we use tools like models, quantifying uncertainties, Bayesian reasoning, prediction markets, and thinking in bets.
Unadaptive vs adaptive. It’s important to realize when you’re in an adaptive environment vs unadaptive. For example, when you play golf, you’re really just playing against yourself — the course isn’t trying to make you fail. A Rubik’s cube is difficult to solve, but it isn’t trying to prevent us from solving it. As we explore the universe’s mysteries — dark matter, dark energy, the uncertainty principle, a theory of all forces, etc. — the universe isn’t modifying itself as we learn more. But in adaptive situations, like politics, war, ecosystems, evolution, relationships, health, business, markets — models rarely work and predictions rarely come true. What worked last time won’t necessarily work this time. I refer to this quadrant as ecosystems and have a module on that.
Normal vs Pareto outcomes — It helps to understand whether you are looking at a normal or Pareto distribution of possible outcomes. If you’re talking about or making a model of traffic flows, you’re probably looking at normal distributions, whereas if you look at home prices in your city, that’s a Pareto distribution with a long tail. For those interested, I have made a 20-minute video explaining this.
Within those three variables is roughly a 2x2x2 cube that helps us understand which world we are in before we start to even state a problem, let alone begin to explore solutions.
After we understand which world we are in, we need a tool kit for working in that world. In the complicated world, the tools are familiar and usually provide a predictable outcome. This is where skills and mastery come in. But in the complex adaptive world, tools tend to be crude and models always wrong. So we need a different toolkit: quantifying uncertainties, doing controlled experiments, A/B testing, statistical power, bullshit detectors, and more.
Skills
Skills are less important than we may think. The way we do things today is often quite different from the way we did those same things ten years ago. Skills are handy, but the ability to learn a new skill quickly is more important. Therefore, students should be introduced to skills using an agile framework that insists on quick cycle times and immediate feedback. I call it “fast is the new big.” It’s better to acquire skills by seeing real-world results rather than just practicing and memorizing for some future that may never come. A great way to integrate new skills with new understanding is to do projects.
Three activities
In business, there are only three activities:
Process
Projects
Decisionmaking
Each of these has meta, macro, and micro aspects. But that’s it. There’s nothing more. And, in fact, in our personal lives, that’s most of it, though there is also “doing nothing,” which isn’t allowed in business. So to prepare students for a life of working and learning, let’s break this down.
Process
Even within a project you need reliable processes designed to give consistent results. Learnings from process improvement have led us to important advances in manufacturing, coding, and other production environments. This is the domain of lean, Kaizen, Kanban, and other tools students should be familiar with. When you do something over and over, you want to study it rigorously and optimize it. I wouldn’t put too much emphasis on this in school, but I would dedicate maybe ten percent of the last two years to building these “muscles.”
Decisions
Decisionmaking is a skill. There is a lot to learn about decision science, but mostly there’s a lot to practice and keep getting more and more sophisticated. I would would want to design about twenty percent of students’ time around it. This includes modeling, statistics, framing, defining problems, exploring solution spaces, learning how to actually make decisions, and learning how to design experiments, analyze results, measure effectiveness, quantify uncertainty, and generally develop critical-thinking skills. An important part is looking back on past decisions and learning whether they could have been improved at the time, knowing what they knew then.
Projects
This is the core of my 4-year curriculum. I believe that, starting in ninth grade, there should be a market for projects. Let’s say it looks like this:
This is just my initial suggestion, but I’m sure it will be different for various student communities. The idea is to work up to be able to propose, win, plan, and execute a 6-week project as a senior.
Experiments are projects. You can do a project in as little as an hour. Most projects are done in groups, though they could be done by pairs or solo, depending on the student. Maybe a student wants to compose music or write a screenplay— that would be a solo project with a deadline and a deliverable. There wouldn’t be a strict hierarchy of grades, so students can cross those lines easily if they want to. I see ninth grade as a “transition year,” and then the next three years blurred together.
Proposals
Projects start as proposals. If you’ve ever seen an unconference, that’s a good model — people put up index cards with their ideas, and people can vote their interest with dots. The projects with the most dots move to the planning stage, though individuals and pairs can still move forward under certain circumstances. A team has from one to four days to work on their proposal. Naturally, some students will be leaders, others fill in various skills and duties, and some will be more focused on marketing and selling the concept to others. If your project isn’t chosen, work on it for the next proposal week.
You can learn about a similar process in a fantastic book called Joy, Inc., by Richard Sheridan.
There should be hundreds of ideas, maybe 20 or so proposals, and then some small number are chosen, and all students must either chose a project or, perhaps, choose two projects to be in half-time each. This internal market encourages students to sell their ideas to others. Sometimes they lead. Another time they follow. Another time they add their special skills to one or more projects. This microcosm is much more like the real world and in fact can be used to run student government and other activities. If projects need money, then fundraising is a key part of the project. If they don’t get much money, they still have to improvise. This is life. In one of my books, I proposed that the outgoing senior class should take on the selection of the incoming ninth grade class as a project, doing most of the heavy lifting for the admissions team. Students should make their own tools and pass them to the next class to build upon. This is the kind of ecosystem I think is valuable for students.
Learning
I wouldn’t envision any classes, but I would envision lots of workshops. Each person would choose workshops and study materials to best support his/her objectives within a project. If a student learns whatever she needs on her own, without a workshop, that’s fine. The project deliverables are required, nothing else is. The goal is to weave together learning and work, a combination we will use for the rest of our lives. We will continually need to learn new concepts, use new tools, write white papers or research papers, teach others what we have learned, etc. Initially, these projects would not be overly ambitious. They would be guided by staff to help students go step by step from simpler to more complex projects, keeping the chance of success around 70 percent each time.
The pattern here is: Set up project, start executing project, different people learn what they need to learn during the project, and then the whole thing has to come together at the end, there should be deliverables, documentation, a presentation, and a retrospective. So let’s say the second-to last school day of every month is presentation day, where students present their projects and their results, and the last day is housekeeping and learning from the process.
Projects should gradually become more sophisticated. There may, for example, be money available for projects that win, and the founders may want to interview prospective team members. In fact, you’d want a form of “blind hiring” for at least a few projects, so team members don’t just choose their friends. There are a lot of dynamics I’m not mentioning here. The goal of any schools doing this initially is to learn what works and pass it on to other schools. In fact, doing projects with students in other schools and having a larger inter-school project market should become the norm, not the exception. We will need to develop the right tools for that. Google can probably help — they have a similar internal project market for their employees who get 20 percent of their time to do anything they like.
Imagine a group of students who want to study particle physics or European history. That could take several projects. But it wouldn’t be book study or tests. It would have to be in the form of deliverables, whether they are experiments, papers, simulations, presentations, etc. At the scale of many schools across the country, this could become pretty exciting. The collaborations could be epic. They could even start new businesses.
Projects do not have to have immediate commercial value. Projects work for anything: poetry, plays, bringing a novel to life, dance, painting, history, philosophy, calligraphy, etc. It’s up to the students to get other students behind their ideas. Imagine 1,000 schools hooked to a market like this — kids would be making films, music, videos on all kinds of obscure topics, and much more. My son is really into stoicism. He’d love to create a story or a video on stoicism that helps others learn what he has learned.
One core skill should be modeling. There’s no need to learn math by doing exercises and using text books. Learning how to model, how to “think in spreadsheet” is something students should learn early and keep improving all the time. They should do projects with simple models and then increase the complexity, understanding what models are, what they aren’t, what they can do, what they can’t, and how to use models to help ask the right questions rather than give answers.
Learning by Failing
The majority of projects really shouldn’t succeed more than 75 percent of the time. In the real world, that figure is about 40 percent, and that includes mega-projects that cost over $1 billion. You learn more from failure than from success. If most projects succeed, then they are not challenging enough.
Think of it this way: over four projects, two should succeed, one should fail, and one should go sideways and need improvisation. If that isn’t happening consistently, then you need to dial up the difficulty factor. We don’t want to continue the “everyone gets a medal” mentality into our new schools.
Not only should a reasonable number of projects fail, but students should learn how to internalize those failings and find ways to prevent similar failures in the future. It has to be okay to fail — students should learn over and over that quick failure is a great way to learn.
Simulations
I believe many schools will create simulations to add depth to their project portfolios. Imagine a group of 200 students in 20 different schools committing to put a small team on the moon. They can create a simulation of a rocket inside a small room and plan to put the small team into that room for the duration of the mission. Everyone has a job to do, from engineers to mission planners to scientists to back-office. This simulation would probably take about 6 weeks, with the first 5 weeks all preparation and for the last week the team would be locked into the room (there’s a bathroom, of course) with all the food and manuals and supplies they need to get to the moon and back. The mission would take place in real time. There wouldn’t be any weightlessness or zero-g experiments, but they would go orbit the moon, send down a lander, come out and explore the new territory, take samples, take photos, then get back in and blast off to return to earth. In reality, they are in cots and air mattresses with crates of food in plastic bags.
As this develops, there would be a growing number of simulation “manuals” available, and students would continue to extend and refine those manuals. There could be simulations for crossing the ocean, descending into the depths, climbing Mt Everest, constructing the Statue of Liberty, discovering the double helix, using telescope time to look for new galaxies or comets, etc.
One of the most valuable skills students can develop is not to put together a presentation that gets a good grade but rather a presentation that gets other kids to put their time, energy, and resources into their project. You can imagine a pitch event where students try to convince their parents to support their project. Raising $500 from supporters to build something is a huge deal to young people. They can even use Kickstarter. This kind of activity builds important skills for the real world.
What about humanities? Anything a student will find useful in the real world will find its place in this scheme. There will be more need for journalists than novelists. There will be a huge need for short explainer videos and pitch decks. There should be essays, white papers, and investigative journalism. All students will need to become effective communicators. Some will do it more with writing, while others do it more with video or voice. The more students practice, the more they will automatically pick up skills they need to succeed. There would still be student plays, but they would be produced by students, and they could charge admission so they need a back-office to support the project.
Skills vs Learning
In this environment, some kids will become specialists. Some will become marketers, video editors, accountants, some will focus on food and supplies, some will be project managers. Students should specialize if they enjoy what they are doing, and they should be encouraged to try new things rather than get stuck doing one thing. This kind of emergent doing vs textbook learning is far better than classes that study these careers or “case studies” that show others doing the work.
Also in this environment, some kids will naturally rise and others will naturally fall back. This is what school is for — to give them a chance to find a good balance and improve their skills. A student who is a follower most of the time can still be a leader in other situations. Students have four years to get up to speed, find their weaknesses, find how to get help, figure out how to succeed.
Retrospectives
Once students have finished their projects, there should be time for other students to learn from those projects. That could be, for example, a day when students are presenting their deliverables to other students in different sized groups. These project deliverables could have a life of their own that exceeds the project by years. Others should be able to build on what they deliver.
Then, I think 2–3 times a year there would be something the whole school does, like a winter-sports week or school trips, fair, fundraiser, etc.
After 18, we’ll see programs that continue to support adults in learning as they work.
Tools
As this project-centered ecosystem grows, new tools will sprout or be discovered to help. Students will need project management, collaboration, production, presentation, decision frameworks, and other tools to become more competitive. Part of the process is borrowing other tools and modifying them, sharing the mods, and creating repositories of projects, so everyone can see what others have been doing.
A few rules should govern the choosing and allocation of resources to projects. I think some schools will try pitch days and widen those to include parents and community. But in practice, a huge board, sticky notes, index cards, and dot voting is very effective.
What will a diploma look like? Not “What have you learned?,” but “What have you done?.” It will probably involve skill certifications of some kind, but I believe we should all start making our own accomplishment portfolios (new tool needed here) and keeping them online. If accomplishments were verified, it would probably be more useful than a resume or LinkedIn profile.
Teachers
What’s the role of teachers in this new reality? It will evolve over time. As Michael Strong, founder of The Socratic Experience says, it’s not his job to know the answers, it’s his job to help students ask good questions and go find the answers themselves. The answers may be wrong, but a) experts don’t do much better, and b) the questions are more important.
Can teachers help? Some teachers can. Teachers need the proper training in statistics, critical thinking, and the Socratic method to be good facilitators. But even that isn’t a very necessary role in my project-based scheme. Clearly, teachers will have to help students formulate useful projects, so that not every project is learning how to make candy. But I think teachers will be surprised at how motivated and relevant this project market will be with just a little mechanism design and a few nudges.
I think students should seek and engage mentors. Maybe mentors are the new teachers. If you can convince a busy person to mentor you, you’re probably worth mentoring. Your mentor isn’t on the list of approved teachers. Your mentor doesn’t get paid a teacher’s salary. You choose your mentors, you sell them on the idea of mentoring you, and then you SHOW them why they should keep spending time with you. If they are patient, they’ll help you achieve your next level. If you do that, they will probably open their network to you, so you instantly have more options and can keep scaling up your journey to independence. That’s pretty much the opposite of teachers.
I don’t know the future of teaching in this scheme, but I think it results in far fewer teachers. I think students could help take over much of the job the school staff does today, and in that way build skills they will need, like bookkeeping, scheduling, admissions, donor relations, hiring, firing, budgets, and more. Every school will see how far they can take this and on what schedule. I can only hope that “best practices” for teachers is radically different ten years from now and that students come out more ready for the real world, without heading off to university.
How to measure success
Tests and grades are useless. They play no role in the real world, and they should not be a part of education. I hope they are largely gone in twenty years. To me, there are two metrics:
Do students have marketable skills? I’m fine with a student graduating high school and starting as a video editor. That person is every bit as likely to become a media tycoon as the person who gets a media degree and an MBA. And I don’t measure success in units of tycoons. Working and learning is the only way to adapt to the changes this century will throw at us. The sooner we figure it out, the better.
Are students critical thinkers? I’ve been studying and teaching critical thinking for a decade. It takes years to become aware of cognitive biases, master most of the skills required, and build a lifelong dedication to the scientific method. I teach my kids about Richard Feynman. We study his lectures, his life, and his philosophy. There are many paths toward rationality and away from magical thinking. It isn’t easy. In my view, the true test is when a person has a solid foundation for when to defend the status quo and when to challenge it — when to go with the trend and when to be a contrarian. This is hard. It goes far beyond labels and belief systems. You have to take each topic or question and look at the fundamentals, look at the data, and do your own rigorous analysis. I don’t expect many students to master this by age 18, but I do want them to be on the path and understand where it leads and whom they should follow.
What about credentials? What about academic careers? What about publishing and basic research?
Are you sure you want to ask those questions? Do you think the system we have now is optimal? I don’t think it can be fixed. I think we’ll need to re-invent it. I’m not sure exactly how, but I think markets can be part of the solution.
Summary
The greatest challenge facing mankind is the challenge of distinguishing reality from fantasy, truth from propaganda. Perceiving the truth has always been a challenge to mankind, but in the information age (or as I think of it, the disinformation age) it takes on a special urgency and importance.
-Michael Crichton, 2003
These ideas are in their infancy. I hope they stimulate conversations about the direction of education. There should be many experiments, online, in-person, and hybrid. Successful models will evolve alongside a number of new tools to help people learn as they work. It’s hard to see, because that ecosystem doesn’t exist. But the more we transition to a world where most people start working at age 18, the more resources will spring up. By 2040, I hope some of this in some form is working its way into mainstream education.
I believe by not fixing education but rather creating something more effective from scratch, costs of education will drop dramatically. No more student debt. No more ultra-expensive schools based on brands, ratings, and real-estate. Far fewer universities. Anyone who wants to help this transition could help us get the message out to employers — it’s time to stop hiring by looking at resumes and credentials. It’s time to start hiring young people and training them as they grow.
Eventually, this lays the foundation for an agile career, where we keep reinventing ourselves as automation comes for the routine jobs and we continually have to reskill to find the new creative ones. You can learn about that in my essay, The Machine Economy.
Finally, if anyone is interested in talking with me and Michael about this, or if you know any investors or venture capitalists looking into this space, I would be happy to explore any ideas in this direction. Contact me via the form below.