Language Mentor Update

Our shiny new user forums have received their first post – chrisd has asked "where are you in the development process?". This is a big enough question to justify a blog post – on this shiny new blog.

This update includes:

• Language Mentor Development
• Lesson Development
• Authoring Tool Development
• Evangelizing the Language Mentor Platform
• How You Can Help

Language Mentor Development

Our Language Mentor language study tool has just been released in a public beta version for Android phones. This means that it's finished enough to be usable and useful, but still has some wrinkles that need to be ironed out. I'm seeing a few small bugs but they're "non-fatal", i.e. they are either small inconveniences or can be resolved by restarting the app.

Currently Language Mentor can only be downloaded from our website. Once we've tested and polished a bit more – probably by the end of this year – we'll release it on the Android App Store.

As this is Language Mentor's first release, it is missing some non-essential features which will be implemented at some point in the future. For example:

• In a future version users will be able to select which lessons they want to download. In this version, as we only have a limited number of lessons available, I've decided that its both easier and better to simply give users the option to "download all new lessons". This will work fine until enough lessons exist to make it this approach impractical.
• In a future version users will be able to enter the URLs of learning lesson libraries located anywhere on the web. This is a key feature – Language Mentor is designed to be an open platform that anyone can create and publish lessons for – but given the fact that at present we are the only people creating lessons, it isn't needed in this version. Most of the "infrastructure" code for this is already in place so it will be easy to implement.
• In a future version, users will be able to download lessons in the background while they do other things. At present our download screen asks users to wait until the downloads are finished.
• Currently, Language Mentor is "hard wired" for learning Mandarin Chinese. Once lessons become available for other languages users will need to be able to select a preferred target language. Or we may take a different approach and offer a separate version for each target language, i.e. "Mandarin Chinese Language Mentor", "English Language Mentor", etc.
• And there will probably be a number of other features that we'll implement in future versions. The ability to delete lessons comes to mind. But one of our primary goals is to keep Language Mentor as simple as possible – so that it's really easy to use – so we don't envision adding a lot more features or major new functionality.

Currently Language Mentor works on Android phones. Because we created this app with Adobe Flex and Adobe AIR technologies, it should be fairly easy for us to also create an iPhone version. Look for a public release in the first quarter of 2012. And Adobe's goal is to make it possible to release apps that are created with these technologies on many platforms. We're hoping to be able to release versions for Blackberry, Windows phones, etc. within the next two or three years.

Lesson Development

Our primary goal is to create a platform which others can use to publish lessons – for many different languages – not to create the lessons themselves. But we obviously need some lessons in order to show how the platform works so we've been working with a few collaborators to help them create lessons. At present three Mandarin Chinese lessons have been released by Zhang Qiwei, and she, Xi Aina and Ren Huibo have a number of other lessons in the pipeline.

In general, we've found lesson development to be a more time-consuming process than expected. This is partly because we've been doing "team translation". Team translation consists of two people, one fluent in one language, one fluent in the other, working together to create high quality translations. While our collaborators' English level is quite high it isn't quite at the "native speaker" level, and my Chinese skills don't even approach that level. We spend a lot of time discussing the nuances of Chinese and English words and expressions. A person who is truly fluent in two languages could probably translate lessons in a fraction of the time that this process takes us. We plan to explore the possibility of hiring such a person or persons some time soon.

Besides translation, we've also found that the process of writing scripts has been quite time-consuming. Surprisingly, we've found that once we have a finished script the process of obtaining voice recordings, editing the audio, and doing the other tasks necessary to publish a lesson go fairly quickly. Why has writing the scripts taken so much time?

• We want them to be fairly perfect. After all, these are demonstration lessons, intended to show how lessons can be created, so we want them to be fairly polished.
• We use a very specific format for our scripts, so that we can use an automated process to convert them into voice scripts for voice talent.

Authoring Tool Development

As we've developed lessons it's become clear that the process can be made more efficient through the development of authoring tools. We currently have a fairly simple in-house tool (which we call ScriptHelper) which a) checks script formatting, and b) converts lesson scripts into scripts for voice talent.

But it's clear that we can do a lot more along these lines. I'm not going to go into detail on all the ideas that we have in this post, but we plan on making this a primary focus over the next few years. Stay tuned.

Evangelizing the Language Mentor Platform

Our primary goal is to make this platform available for others to use – as a platform. While we like creating lessons, and plan to continue to do so, this is a secondary goal for us. Why is this?

Let's step back for a moment and look at the big picture. In our world today there are immense unmet needs for language learning materials:

• The world needs language learning materials that are free.
• The world needs materials that can be studied while the student is busy doing other things.
• The world needs language learning materials for many different languages. Wikipedia estimates that there are between 3000-6000 active languages in the world today, and lists 80+ languages that have over 10 million speakers. The SIL Ethnologue lists 473 languages that are nearly extinct.
• The world needs language learning materials whose "vocabulary scope" extends far beyond what is currently available. For example, Pimsleur's "comprehensive" courses typically teach about 500 words. While this is a great start, students need to learn 2000 or more words in order to attain even basic proficiency.

We believe that Language Mentor can provide one important piece of the solution to this unmet demand – simple, free, software that evolves and improves upon current CD-based and MP3-based approaches.

It is important that such software be an open platform because the task – as outlined above – is a "world sized" task requiring the contributions of many.

And it's important that such software be open source so that contributors can have confidence that the software will continue to be available. In other words, the fact that our platform is open source provides a kind of insurance – it gives contributors the option to download its source code and tweak and maintain it if that becomes necessary.

Okay, so that's the big picture. Let's return to what we – The Language Collaborative – hope to accomplish over the next few years. Clearly, we can't hope to make more than a tiny dent in the above outlined "world needs" – in fact, it would be silly for us to try. Instead, we'll continue to produce a limited number of lessons for one or two languages (Mandarin Chinese now, English soon). These lessons will help some learners to learn language, but more importantly they will serve as models that demonstrate the simplicity and usefulness of the Language Mentor approach.

We'll also be doing our best to get the word out. How will we do this? We haven't fully figured this out yet. Do you, dear reader, have any suggestions to offer? The problem isn't a lack of ideas – the problem is that there are so many possible avenues, and we need to identify which ones will be most effective. We need to find communities of interest where word of mouth will come into play. We need to find organizations that are interested in using our free tool rather than creating something similar from scratch. And we need to learn how to communicate effectively with these communities and organizations.

How You Can Help

If you think that Language Mentor is a worthwhile approach, here are some ways that you can help:

• At present, we need beta testers. If you are a student of Mandarin Chinese and own an Android phone please give Language Mentor a try and send us your feedback.
• Going forward, we'll be focusing more and more on evangelizing our platform. If you have suggestions re how we can do this we'd love to hear them. And, of course, if you might be interested in creating lessons for Language Mentor please be in touch – we're eager to help.

Thanks!