1. Can you tell us little about how the just-the-word tool got started?
Phil: Back in grad school, I was working on artificial intelligence and the idea of developing an intelligent thesaurus that could help you choose the right word based on what you had already written on the page. After graduating, I joined Sharp Laboratories of Europe and had the pleasure of working with Pete Whitelock and a great team. The team had developed an intelligent dictionary that could prioritize the definitions of words given the context. Together, Pete and I realized we could use some of that technology and the recent developments in statistical natural language processing to create a tool to help writers.
2. Did you envision it as a tool more for learners or for teachers or both? Why?
Phil: I always saw Just The Word as a tool to help writers, both English native speakers and advanced learners. English has so many different ways to say the same thing, but if you choose the wrong word you might not get your precise meaning across or worse actually imply the wrong message entirely. Even native speakers have trouble finding the right words. I hadn’t really thought about teachers using it, but it’s been very popular. I would like to hear about how teachers use it. Perhaps teachers could share with each other how they use it.
3. Can you give any breakdown of stats for people who visit the site? Numbers? From where? Time of day?
Phil: The site gets around 1000 queries per day from all over the world. The top 10 countries are UK, USA, China, Taiwan, Turkey, Hong Kong, Brazil, Iran, Poland, and Korea.
4. I am particularly interested in having the learner errors features working. What are the chances of that?
Phil: The feature wasn’t used very much according to my logs. Then I found a small bug in it, so I took it down. However, if there is enough interest I’ll put it back; it is useful.
5. What other language/education projects are you working on?
Phil: At Sharp I worked on a number of language and education projects. One of the hardest things in learning a language is to stay motivated. It’s been shown that giving immediate and relevant feedback on a learner’s work and efforts creates focus and engagement. This applies to learning any skill such as driving a car, excelling at a sport, and learning school-based subjects. My team used this idea to develop an app to help beginners learn English by extensive reading of e-books, that gives the feedback by algorithms. The app tracks you while you read, working out your current word level (eg 1000-word or 2000-word). Then it highlights study words in the text at your level that you can focus more attention on. It learns your level by collecting your right/wrong answers to short vocabulary quizzes on any of the words you click in the story. The quizzes are generated automatically using artificial intelligence algorithms, and your level is tracked using a statistical model that weighs the evidence for and against your progress. We worked with a major educational publisher in Oxford and another in Japan to launch the app into the Japanese market. Mobile learning was also important for us since it can be done anytime anywhere, so we called our technology ELMO – English Language MObile. In the Japanese Android Play Store the app is called Tadoku Academy – for Extensive Reading Academy.
6. What are your opinions about corpora based tools for language learning?
Phil: There is great promise in using corpora and real text from the web for language learning. The best way to learn is to study genuine language and use it in real conversations to complete genuine tasks – as you do when you find yourself in a foreign country. Corpora can also be used to automatically create language resources such as dictionaries, Just The Word, and Elmo. But most tools and resources I’ve seen on the web are run as hobbies – Just The Word is no exception. Building a business is difficult and risky.
7. What are your thoughts on the current and future educational technology scene?
Phil: The educational technology scene is exploding. There are so many opportunities now. After many years of challenge to the very premise of using technology in education, it seems we have reached a turning point where now it is just accepted that it is useful and even necessary. Attention in research has turned to identifying and promoting how it can be used effectively and in what context. In the business world established companies and hundreds of startups are finding traction. I believe the most innovative areas will be in educational data from a small scale (like Elmo) up to large scale (like using the student data collected from MOOCs); in statistical modelling to both give feedback to students and to study the efficacy of teaching and learning approaches; in connecting learners locally and globally; in game-based learning; in spreading and improving education in disadvantaged countries; and in new modes of delivery such as mobile learning and connected classrooms.
Many thanks to Phil for taking the time to respond, you can read a bit more about Phil:
As Phil says just-the-word is very much a hobby project, so donations are always welcome to keep the tool online, developed and maintained. I have recommended just-the-word often to my TOEIC students, you can read about one of the ways I have used it. I am very much looking forward to having the learner errors function working and so is at least one other teacher:
Thanks for reading and don’t forget questions/suggestions.