The Verge about Esperanto on Duolingo!
The Verge just published quite a long article about Esperanto with prominent mentions of our Duolingo course! You can read it here: http://www.theverge.com/2015/5/29/8672371/learn-esperanto-language-duolingo-app-origin-history
Great article Chuck! I will be part of the Esperanto baby boom :)
I've been interested in the language for about 15 years now and the Duolingo course makes it too convenient to pass up this opportunity. Thank you so much!!! Here's to 200,000 others like me by the end of the year!!!
"This would be great for computers to learn, was the first thought that Chuck Smith had about Esperanto. It was 2001, and he was in college, taking a class called Models of Mind. With its logical structure, Smith thought Esperanto could work as a bridge language for translation, especially between two languages like Finnish and Turkish, that are unlikely to have a large overlapping dataset that a machine translation program could use. Just as Zamenhof intended, Esperanto would be the metamedium of communication across tongues."
This is actually what my senior thesis is going to be about. I was really happy to see it in this article. :D
About using Esperanto as a pivot language in statistical machine translation. We're on break right now, so I'm spending the summer gathering some resources and learning Esperanto. In the fall I'll be setting up my tests (so working with aligning the parallel corpora) and hopefully running some tests. I'll be writing (the rest of) the report in the spring.
I was not---thank you for that bit of history. (:
It seems that what I'm doing is a good follow-up from where that paradigm shift from more AI-based machine translation to the newer data-driven approach (the "statistical" part of statistical machine translation) left off. If you're interested in this stuff, look up a 2010 paper called "Climbing the Tower of Babel: Unsupervised Multilingual Learning" by Benjamin Snyder (and check out his PhD thesis---this is more about how computers learn languages) and then read a few papers by Phillip Koehn (who writes about statistical machine translation).