From Foreign Language, to Programming Language, and Back Again

It’s true, I’m not studying to be an educator in the traditional sense of the word. That may make it seem like I’m different from the other arcade players since my goal isn’t to be the influence in a classroom. Despite that though, we’re not that different. I do want to be an educator, by making smart software that teachers. While I may not be an  influence in the classroom, I hope my software will one day be.

I’ve had a chance to look back at how I came to the conclusion that software can improve the way we learn, and it all started with a fellow who goes by the name of Khaztumoto. Some years ago, after a fateful post-new years hangover, I had a strange introduction to Japanese culture by way of a DVD that I accidentally obtained. That’s another story, but it ends with me deciding that I absolutely had to learn the Japanese language.

I didn’t know where to begin, but the google-fu was strong in me and I turned up Khaztumoto’s blog, All Japanese All the Time. Before even reading about the study techniques he was advocating, I was sold on the opening of its about page:

I am your host, Khatzumoto. My zits have been photoshopped out of that picture. I learned Japanese in 18 months by having fun. In June 2004, at the ripe old age of 21, all post-pubescent and supposedly past my mental/linguistic prime, I started learning Japanese. By September 2005, I had learned enough to read technical material, conduct business correspondence and job interviews in Japanese. By the next month, I landed a job as a software engineer at a large Japanese company in Tokyo (yay!).

Khaztumoto improved upon a method (perhaps invented) by a couple Poles who essentially went from zero English ability to fluent simply by learning 10,000 sentences (it turns out there might be something to that number). He did so in an obsessive way in which he brought Japan to the US, immersing himself completely in Japan and Japanese in everyway imaginable. But what I was most keen on was how he went about actually studying 10,000 sentences. With software, of course.

The software was basically a simple flash card tool, I don’t remember the name, what was important was that it was using an interesting algorithm by P.A. Wozniak (who coincidentally hails from Poland), which was producing incredibly promising results in his own flash card-like program, SuperMemo.

Sure it would be possible to have thousands of index cards. Sure it would even be possible to implement the algorithm without a computer. But realistically the algorithm is too complex to be calculated in one’s head, so traditionally index card based flash cards used much simpler, less efficient algorithms like the Leitner system.

Animation of the Leitner system sorting

Animation of the Leitner system in action. Box 1 contains cards that need to be reviewed sooner rather the later. Box 3, contains cards that don’t need to be reviewed for a while. In practice there can be any number of boxes and review intervals. It might help to see Box 1 as items that need to be reviewed today; Box 2, three days; Box 3, one week, and so on.

But now that we have computers practically all the time, be it on our desks or phones, why not use the more complicated and efficient algorithm that offers up to 96% retention rates? Not only does learning with software make save time studying over using less efficient methods that are brains are capable of computing, but it saves space, adds convenience and never makes a mistake.

I was never able to follow through on Khaztumoto’s methodologies, but I did begin wondering about how software in general can be used to improve the learning process for any given subject. More than that really. I created a web-app to help students learn kanji in context, and eventually ported it to iOS. It’s not what I want it to be though (only 3½ stars in the App Store), and don’t have the time to improve it or add some of the new e-Learning ideas I’ve come up with right now — however, since this is my final class, depending upon how the job market is, I soon may.

It turns out Khaztumoto and I have somethings in common: we both studied computer science, have Kenyan heritage somewhere down the line. Because we have similar interests, it’s not surprising to me that we both turned to software to make learning more effective. So, I still read his blog occasionally, and follow and unfollow him on twitter (depending upon how psyched up or bummed out his realist motivational tweets make me feel).

If I am to ever make an application that makes language learning easier, it will have been in-part to the ideas that Khaztumoto continues to share with the global community.

Whether it’s for my software iterations, or iterations of myself, I’lll end wtih one of the most valuable lessons he’s shared: