Whereas my first post fixated on the role of MOOC’s and the transformative power of the internet, in this post I hope to step away from MOOC’s in order to focus on a disruptive technology that I am excited to see finding its way into digital learning.
It has to be fun. It has to be relevant.
To set the stage for the next wave of educational technology that I see coming, it’s useful to start with a baseline of where we are now. Let’s start with two essential pieces of why educational technology is attractive to use in our classrooms: technology is relevant in students’ lives and it offers a capacity to offer immediate feedback. In fact, when Margaret and I started LinguaZone.com almost seven years ago, these were two of the driving forces that fueled our interest in creating customizable language games. (The site has expanded in its scope a bit since then, but the central focus of the educational games certainly starts here.)
Relevance: Teaching tools like LinguaZone connect with students where they are. They offer a fun, relevant way to approach class material. And that is a powerful thing: I believe effective learning needs to happen in a way that is joyful and meaningful to our students.
Immediate feedback: When a student engages with these kinds of tools, they can get feedback on their progress immediately. No waiting for a teacher to collect their work and get it back to them later.
There is much more to say on the value of these two things (and others), but let’s leave it at that for now in order to take a peek further ahead.
Intelligence and learning, human and non-human
The next disruptive technology that I see having a profound impact on our learning is, perhaps ironically, machine learning. Machine learning is at the core of artificial intelligence, and in the past few years has started to pick up on the promise that futurists at first expected when it made its theoretical debut a few decades ago. For a quick crash-course into what machine learning is, check out Sam and Aqeel’s screencast about pattern recognition and machine learning using Connect Four (below). These two high school seniors made this 3 minute video for a self-directed research project in my Computer Science 2 class at Friends’ Central School in December. (For a deeper dive, check out their blog post on the same subject.)
The big idea here is that computers are great at harnessing huge amounts of data. Now we are developing more sophisticated methods to recognize patterns in this data. And once we develop systems to recognize patterns, we can then develop systems to make accurate predictions. And that’s a powerful idea. Think about what you see on Netflix: You enjoyed Sports Night and The Office. Maybe you would like to watch Arrested Development? We see the same thing on Amazon: You bought X and Y. Maybe you would like to consider buying Z? It took a big investment of time, energy, and skill to get to this point (the $1,000,000 Netflix Prize is an interesting example to note), and we can now see this technology spreading into lots of different places. How is it that Apple’s Siri is becoming better at interpreting your commands over time? It’s because Apple is collecting huge sums of data, getting better at recognizing patterns, and then adjusting Siri’s ability to predict what it is that you are actually saying. (And if you find Siri to be frustrating and unhelpful, check out Google Voice Search. Google in fact is much more advanced when it comes to the machine learning that powers these systems.)
Coming to a school near you
Machine learning is still in its infancy, but as a disruptive technology it will soon make its way into many different fields (not just marketing or entertainment) and transform them. How will this affect education? Imagine a student practicing exercises to develop proficiency with a certain skill. If that skill can be expressed on a computer, each response from the student can get stored as a data point in the system. Now imagine that this is happening over the course of millions of exercises with millions of students. Over time the system will be able to suggest which exercise the student should do next. With all sorts of incredible content making its way online, this idea — that a system can accurately and immediately offer suggestions on what exercises will be an appropriate next challenge — is game-changing. The work that Khan Academy is doing in this space is thrilling (and wildly sophisticated, if you would like to read a bit about some of the specifics).
2 + 2 = 5
So here we are: the internet has democratized access to information, and machine learning will someday offer us digital personal trainers to help us build all sorts of skills. Does information + skills = a well-educated person? Most would say that it does.
Where does that leave forward-thinking brick-and-mortar institutions? We must do more than equip our students with information and skills. The challenge upon us is either to identify our x factor or to make 2 + 2 = 5.
The good news is that we do this already. The whole is greater than the sum of its parts. So what’s next? I suppose we need to find our own way to understand and explain that.
[Continue reading the next post in this thread: Mistakes, empathy, and serendipity]