Wednesday, 10 July 2013

Meet the EdTech entrepreneurs - An interview with Luis von Ahn

Launched in June 2012, Duolingo's core idea is simple: you learn a language for free while helping to translate the web.
The Good MOOC interviews Duolingo’s founder and CEO Luis von Ahn.

  • Can you introduce yourself?

I'm a computer science professor and the CEO and co-founder of reCAPTCHA (sold to Google) and Duolingo. In general, I like making systems that don't suck and that are used by millions of people. :)

  • How did you first think about the concept behind Duolingo and why is this important to you?

The whole thing started as an academic project at Carnegie Mellon between me and my PhD student +Severin Hacker. I had just sold my second company to Google, and we both wanted to work on something related to education. Education is very general, so we decided to concentrate on one kind that is in huge demand everywhere (except the US): language education.

It turns out there are over 1.2 billion people learning a new language around the world. Now, the majority of these people, like 800 million of them, are learning English, and don’t have very much money. In fact, the reason they are learning a language is to be able to get a better job, or a job at all. The majority are not learning French because they want to get ready for a trip to Paris over the summer. They’re learning a language to get a job at a call center.

Here’s the crazy part about this. There’s a huge number of people that want to learn English, and most of them can’t pay, but the ways to learn a new language typically require them paying, because somebody has to make money. It’s either that you learn a language in middle or high school, which means you're pretty wealthy because you go to a school that offers foreign languages; or you learn it in college or adult classes, in which case you’re also pretty wealthy because you can pay for those; or you buy some sort of software like Rosetta Stone or Open English, in which case you have $1,000 to spare.
So, the largest part of the market was not being addressed because there was no great way to make money from them. Most people who wanted to learn a language couldn't really afford the best ways of doing it.
We wanted to have a way to teach people languages for free. But not just free. We wanted to have the best quality of language education, and offer it for free.

But making the best language learning system (or the best of anything) requires a lot of money. You have to hire the best people to develop it, improve it, and keep it alive. This is why students typically have to pay: somebody has to finance the operation. So the question became: how do we fund the best language learning system while keeping it 100% free to the students.

And this where an idea similar to reCAPTCHA came to us: Is there a way in which we can generate something valuable out of the mental energy spent by students learning a language? And it turns out there is, and that’s what Duolingo does. Instead of paying tuition or a subscription fee, our students actively translate a range of documents for organizations as they practice what they have learned. Things like news articles, websites, Wikipedia, and blog posts. They hone their language skills with real world applications, which gives their effort a purpose and offers an opportunity they previously couldn’t afford.

It’s a mutually beneficial arrangement for both parties. Our students receive a high-quality, completely free, language education, and organizations are given translation services powered by the students.

We now have 5 million people across the globe learning a language for free and growing by the day.

  • Can we expect more advanced courses coming?


  • You've stated before that you will allow users to create their own lessons for new languages. Do you have any updates on that?

We're only starting to work on this, so it will be a few months until it's deployed.

  • What's your analysis of the current state of the language learning market?
I don't know this for a fact since I don't have exact numbers on our competitors, but in terms of number of *active* users, I believe we're already the #1 service. I think people pay for inferior products because there is a lot of money spent on marketing them, whereas we don't spend any money on advertising (we spend it all on making the system better!).

  • How are you doing with your plan of monetizing your users' translations?

Very well. We're signing a number of deals to translate content of major content producers. We'll make an announcement in the next few weeks.

  • Duolingo is complex idea yet the website feels intuitive. How did you go about designing your user experience?
We've spent *a lot* of effort on that. We have 4 very talented designers, and we test every change we make (both by watching video recordings of strangers using our system and by measuring engagement on the website).

  • What is your current split between mobile and PC usage?

Mobile is 75%.

  • You’ve hinted at the possibility of matching users after they've reached a given mastery in the learning trees (for instance an American learning French with a Algerian learning English)... Any news on that since your Reddit AMA?
Not yet :(

  • Have you had any surprises in the way people use Duolingo compared to what you envisaged when designing it?

We didn't expect that people would become so attached to the way things work. We constantly improve Duolingo, but almost every time we make significant changes we get a number of students asking for it to return to how it was. The funny thing is that every time we make a big change we test it very thoroughly, so by the time we make it we're certain the new version is significantly better.

  • Can you share an example of something you learned about how people learn languages that you didn't expect?

We're writing a paper about this, and many of the results will be there. For now, we're not sharing them yet.

  • Where do see higher education going in the medium to long-term future with the ongoing trend of free and open resources?
I think we will see more and more people learning things by themselves. I don't believe that universities are going away any time soon, but I think people will be able to learn more effectively with these online resources. Also millions of people who wouldn't otherwise learn things (for example because they can't afford going to a university) will now be able to learn them by simply having access to the Internet.

  • Have you taken a MOOC?

I have not. :-(

Write a comment below or share this post!
Post a Comment