Language Trainers Blog

Source: ted.com

Today I watched an absolutely mindblowing video on ted.com, a site of small, non-profit group TED, who are devoted to “ideas worth spreading”. There is a goldmine of interesting stuff in their archives, but being somebody with a keen interest in languages this video seemed truly worthy of sharing.

The talk is given by Patricia Kuhl, co-director of the Institute for Brain and Learning Sciences at the University of Washington and expert on early language and brain development. We all know that babies and young children are better at picking up new languages than adults, but perhaps we forget just how much better. Her studies have shown that during early language learning, they can do things adults simply can’t, and to surprising degrees.

You can view the video below, or go to the page on ted.com.

You may have noticed that some people like to tap the space bar twice after finishing a sentence. Why do they do this?

The double space after a period habit comes from the days of typewriters. The problem with typewriters – besides error correcting, of course – was that due to technical limitations, the text was monospaced, also known as fixed-width or non-proportional. That is to say, unlike when you type on your computer, a character like the letter l or i takes up just as much horizontal space as a wider letter like a D or H. This made things look a little messy sometimes, due to there being a lot of white space between certain letter combinations, and not so much between others. Therefore it became a habit for typewriter users to insert two spaces after a period, to make sentence breaks a little clearer.

With the exception of certain monospaced fonts like Courier, modern day word processing has allowed for proportional fonts, which are easier to read and nicer looking. Since typewriters are never used these days, and since most people write with proportional fonts (with the exception of programmers and web designers, since computer code can be easier to parse when everything can be easily arranged in rows and columns), the double space habit has been phased out. However, there are still people who follow the old faithful – usually older than 30 – who insist that periods should be followed by two spaces, not one.

Don’t listen to them: it’s an archaic habit. Most modern word processors will automatically adjust the size of a space following a period to be slightly larger than the space between words anyway – some will even go so far as to correct a double space to a single space.

With the ubiquity of so-called “smart” phones nowadays, people have struggled with tiny touchscreen keyboards and not-so-agile thumbs. Back in the day (when cellphones had these archaic things called “keys”), people eventually learned to type incredibly fast – the current Guinness world record for sending a 160-character phrase on one of these old-school phones is 37.28 seconds, held by a 24 year old Norwegian, Sonja Kristiansen. At some point came T9, where you only needed to press each number key once and a built-in dictionary would work out the word you wanted, which sped up the process of texting further.

It’s rather harder to do this on modern, touchscreen phones, so naturally technology has arisen to assist us. Apple’s autocorrect is perhaps the simplest – start typing a word, and it predicts what you actually want to type, depending on the nearby keys that your phone thinks you ‘should’ have pressed. It then automatically replaces the word once you start typing the next one: however, this is laden with difficulties, especially when typing in a language with as rich a vocabulary as English. While it recognises most names, I can’t for the life of me get my phone to stop correcting my name (Dave) to “face”. Then there’s Apple’s insistence that they don’t want potty mouths using their devices, to the point where the words “shut”, “duck” and “he’ll” are starting to look like swear words to me. Sometimes it just plain gets it wrong: a fine example of that is the image above. “Disney”, being a proper noun, isn’t in the phone’s vocabulary; and rather than letting it slide, it corrects it to the nearest word, which, unfortunately for the context of the message, is “divorce”. There are entire sites built around these kinds of phone faux-pas (“phone-pas”?), like Damn You Autocorrect.

Google’s Android phones have a few similar tricks up their sleeves, but undoubtedly the most popular is the Swype, which is starting to be installed by default on many handsets. Swype is a simple but very effective way of speeding up your typing speed – instead of tapping each letter like you usually would, you swipe your finger from one letter to the next, lifting your finger at the end of each word. The phone reads the shape you make on the keyboard, and interprets the word you’re trying to write, giving you a choice if there are any doubts. It’s amazingly effective, especially as you don’t need to be particularly accurate as you swipe over the letters.

Phones using Google’s Android operating system also have voice recognition – for any application that uses a keyboard, you can also tap the record button and speak your text. While generally pretty impressive, it’s simply not accurate enough to be able to rely on it fully yet – especially if you’re speaking anywhere with any background noise.

The important thing is finding the happy medium between speed and accuracy – it’s simply a very difficult task to do a good job of mimicing a physical keyboard, especially on smaller screens.

In my last post I wrote about how words like “Google” and “Facebook” used as verbs had made it to Lake Superior State University’s annual “banished words” list. It’s an interesting linguistic shift and with the ubiquity of the internet we see and use these words almost every day. “I friended her on Facebook”, “I spend too much time Facebooking”, “Let me just Google that”… and right now, in fact, I’m blogging.

The use of nouns as verbs has been notable in recent years, and is especially so in the fact that, well, it is no longer considered particularly notable. People just do it, even if they are not particularly conscious of the fact that they’re doing a process grammarians refer to as denominalisation. In fact, being able to convert nouns to verbs easily is something that defines English as the language it is – this cannot be done easily in many other languages, where verbs take different forms or have a variety of different stems or endings depending on their purpose in a sentence. These changes, or inflections, are not really employed in English, besides things like adding an s to denote a plural noun, or adding -ing to a verb stem to form the present participle.

For example, in French, regular verbs have 3 different types, those with infinitives that end in -er, -re or -ir. If a French speaker wanted to convert a noun to a verb, they’d have to alter the word – in English, we can just use the same word in a different context, and most people will pick up on it.

Without any official kind of regulation, the English language is free to go where it likes, often dependent on popular usage. It’s interesting to watch new words and phrases come and go, but some words are permanently added to our collective vocabularies – whether they’re in the dictionary or not.

Source: questvisual.com

Every day we seem to move a little closer to developing the equivalent of Douglas Adams’ legendary Babel Fish: a device that allows instant translation, transcending the language barrier and enabling universal communication. The idea of the Babel Fish has fascinated me from the first time I read The Hitchhiker’s Guide To The Galaxy, and I find it incredible that just 15 or so years later, what I originally thought to be a quirky yet impossible invention is fast approaching realization.

Of course, machine translation has been around us for a while – from Altavista’s Babelfish service and Google Translate, Google Goggles and speech-to-speech translation – but never before has it been realized so clearly and so simply.

The software in question is called Word Lens, marketing itself as “a dictionary, evolved”; and the idea is simple. Using augmented reality (a rather complicated sounding term meaning taking visuals from a camera and overlaying other data on top of it) together with translation algorithms, the concept is that you point your phone at some foreign text, and you the translation replaces the text on your screen. Instantly.

Here’s a video of Word Lens in action – believe it or not, this is actually how it works:

It may not recognize handwritten or particularly stylized printed text, but it’s good enough for signs, menus and the like.

Unbelievably the software is available for free from Apple’s App Store, though you have to purchase certain language pairs in order to enable translation (currently only English <-> Spanish is supported, but more languages are on the way). It works with all Apple devices with a camera with autofocus (iPhone 4, 3GS or the new iPod Touch with video camera), with firmware version 4.0 and above.

The demo version reverses the letters in words instead of translating them, but it still demonstrates that the software recognizes the words (and it does it well, too). Adding the translation layer is pretty simple from there, but you have to pay for it.

This kind of thing would be invaluable while on vacation, on a business trip, or in any situation where you need to be able to understand your surroundings.

Admittedly, it’s not quite a Babel Fish, but we’re getting closer every time… it won’t be long now!

Source: pleco.com

As someone who is living in China and has only recently started learning Chinese, I am constantly flummoxed by Chinese characters. They might call it “Simplified Chinese”, but it still looks pretty complicated to me!

I have been using a dictionary on my phone called Pleco, which has been a fantastic tool. It’s packed with features, including flashcards, audio pronunciation (very useful for a tonal language like Mandarin), handwriting recognition (which is so accurate it’s almost eerie).

However, their latest update has brought along with it something that makes reading Chinese a complete breeze – an OCR recognition module, using the phone’s camera.

OCR stands for Optical Character Recognition, and has been around for a while – it basically takes an image and turns it into text. This has a wide variety of uses, but for Chinese it is invaluable: now when I want to know what something means in Chinese, I can simply point my camera at it and let Pleco do the rest.

Here’s a video of this amazing technology in use:

Naturally I will still be learning Chinese as I can’t be entirely dependent upon my phone, but with so many Chinese characters, this has really helped me decode menus and signs when I needed it.

Isn’t technology fantastic?

Named after creator Stephen Wolfram, Wolfram Alpha is a different kind of search engine from Google, Yahoo and all the others – it brands itself as a “computational knowledge engine” with hundreds of practical uses. You can input math calculations, get in-depth weather reports, study socio-economic data, or even find out how many calories there are in a cubic light year of fried chicken (hint: a lot).

However, it also has linguistic uses, and one that I’ve found myself using a couple of times is a kind of “crossword puzzle solving” feature. You know when you have a word on the tip of your tongue, but just can’t get it out? For example, you know the word you want starts with a “T” and ends with “-ation”, and it means to represent a word in another alphabet… but you can’t for the life of you remember the exact word.

Wolfram Alpha can help out: simply type in T and the ation parts, and fill in the gap with underscores (_). It will then calculate that you’re looking for a word that fits the template, and give you a list of possible results.

In this case, it found my word (transliteration).

This is a very basic feature of WA (it can perform some dizzyingly complex mathematical operations in a matter of milliseconds), but still worth mentioning as it can be incredibly useful if you want a quick solution to your vocabulary woes. You can find a gallery of other uses of Wolfram Alpha on the site. Pretty impressive if you ask me!

Even for (or rather, especially for) people like me who write often, a decent dictionary can be critical – not simply as a tool to check that you’ve spelled a word correctly, but also to continue to expand your vocabulary.

Of course, thanks to the internet you no longer have to leaf through a hefty tome of definitions to find the word you’re looking for – simply typing your word into your favorite dictionary web site will take you to the definition within a second, and even if you’ve spelled it wrong it will still intelligently take a guess at what word you meant.

However, as someone who uses a dictionary often, I’ve found my method of looking up words evolving further. I used to go to dictionary.com and search for the word. I later found out about Firefox’s address bar shortcuts, and discovered that I could simply type dict in the address bar, followed by the word I wanted to define.

However, upon discovering the nifty online dictionary NinjaWords, it has now become my main dictionary.

Why? Well, firstly because it’s fast. Really fast. It doesn’t reload the entire page to find your definition, it gives you the results inline. That saves about a second in itself, and makes the whole experience more instantaneous.

Secondly, it leaves a trail. Sometimes I’ll look up a word, then look up another, and look up another, and then decide that I preferred an earlier word. Rather than traipsing back through my history with the browser’s ‘back’ button, I can just scroll down and see my history, still complete with definitions. It even keeps a history of words you’ve looked up in previous sessions.

Thirdly, like so many things nowadays… there’s an app for it. Much as I am loathe to jump on Apple’s bandwagon, the iPhone app for NinjaWords is what it claims to be (and what ninjas are): smart, quick, and deadly accurate – and the biggest bonus of all is that it doesn’t require an internet connection to work.

Highly recommended!

In a previous post I extolled the virtues of Google’s visual search feature built into Android smartphones – Google Goggles. The idea is that instead of going through the effort of typing to search for something, you can simply snap a photo of it with your built-in phone camera, and upload it straight to Google, who do the rest of the work for you. Want to know what a painting is? Just take a photo of it and Google Goggles does the rest.

Recently Google also built in a translation feature, allowing you to take a photo of a foreign text – be it a book, a menu, anything – and in combination with Google Translate, Goggles can provide an instant (machine) translation. I was excited over this development, because it seemed like the first steps of something similar to Douglas Adams’ Babelfish from his genius Hitchhiker’s Guide to the Galaxy series of books.

Apple users can now rejoice, as Google Goggles has recently hit the iPhone, via the Google app. You can grab it from the iTunes store here.

YouTube user evan1965 is currently teaching an entirely self-contained audio/video Latin course using the popular video sharing site – but not, like most Latin courses (and like every one I ever studied), a course that simply enables you to read and understand existing Latin texts. This Latin course, or rather “cursum latinum”, teaches Latin language, as you would learn any other language, training you to read, write and speak in the ancient tongue. Besides a little English in the first couple of introductory lessons, the entire course is taught in Latin, meaning that it’s open to everybody, no matter what your native language!

Of course, since we have no surviving ancient Romans around and there were no audio recording capabilities in those days, we are not entirely sure just how Latin was spoken (though we are sure that it was not in a thick American accent as in later episodes of the TV show LOST). However, from piecing together various ancient texts on the subject of Latin rhetoric and grammar, modern day experts have some idea – though there are, of course, some models of pronunciation that not everybody agrees with. Then, of course, there is the common issue of how exactly to render modern inventions such as “computer”, “internet”, etc. in a language with no notion of the concepts.

Evan says in the course description:

The course is taught in Latin, based primarily on the work of J.A. Comenius (1657) and J.G. Adler (1856). This is not a translation course, the goal is to get you reading and writing and thinking in Latin. A version of the restored classical pronunciation of Latin is used.

Suitable for beginners, Latin is taught slowly from the ground up, using spoken Latin, while keeping a firm grip on grammar. The first 2 lessons have some English. The rest are only in Latin. You will need to study the course in sequence, as each lesson follows the preceding one in a progressive order.

As the course is taught totally in Latin, you need no other language.

You can access the videos on his YouTube profile. There are For anybody who has ever wanted to study Latin, this could be a very interesting (and more importantly, free!) way of fulfilling that aim. To date, Evan has made 323 videos, so you may have a little catching up to do!

There is also an audio-only version of the course available in podcast format.