Many will view this new year with a sense of foreboding: money worries prevail, and fretfulness pervades the media - will economic recession becoming a depression?

We at Mloovi, however, feel it pays to be positive. Take a moment to think about what you have in your life. Then take a moment to think about what you lack. For most of us, the former will outweigh the latter. For the rest? Well, we hope that 2009 brings a chance to change that around.

Good luck, and we wish you a happy and healthy new year!

Controlled authorship offers improvements to the writing process that make for better efficiency, and it reduces repeat work. Cost of human translation is often cited as the biggest bug-bear. However, by construction textual pieces with a proper process in mind, over time we can alleviate the cost of repeated translation of the same pieces.

There are four areas for attention…

Modularisation
The aim is to reduce the number of times that text is translated, saving time and potentially greatly reducing cost. With modularisation we break text into areas that could feasibly be used elsewhere - much like a computer programmer would write a general routine and call it multiple times. This is particularly easy to do when writing international instruction manuals for different models of similar products. It’s easy to understand that many instructions will apply to all models.

Standardisation
This is simply the process of ensuring that product or object-specific names are removed from text. For example, “Superputer THX1138 has capacity for extra RAM expansion” should be written as “The computer has capacity for extra RAM expansion”. The product name could then be confined to a section heading, and the translated text could be used for different computer models.

Simplification
This is very straighforward: why use long-winded text when short, punchy (but grammatically correct) statements will translate better, and be better understood?

“When using the computer, make sure all cooling vents are uncovered and unobstructed, to allow efficient case cooling.”

…could be better written as…

“Keep the case cool by not obstructing vents”.

Reduction
Reduction is very similar to simplification in effect, the difference being is that this last pass should focus on the removal of words that add nothing to the sense of the statement.

“To make the sentence structure more efficient without changing the meaning, we can remove words and sections that add nothing to the piece.”

…could be expressed by writing…

“Make the sentence more efficient by removing words that add nothing to the meaning.”

…or even better…

“Improve sentence efficiency by removing redundant words.”

By applying these simple steps, and remembering the guidelines noted here, we can drastically improve the quality of both machine and human translations.

Currently we are employing a so-called ‘machine’ translation. That is, the translation spurted out of the end of Mloovi has been performed automatically by a computer. The upside is that the translation is free, and quick to achieve. The downside (sometimes) is that the translated text may be difficult to read.

At a future point we aim to offer higher-quality human translations to premium registered users but, until that day, there are some things you can do to achieve a better translation.

The concept is simple: by using something called ‘controlled language’ we can drastically reduce mistranslations and awkward sentence results.

What do we mean by ‘controlled language’? Simply that the author should treat text as if it were a precision instrument, and not a stream of thought. Good technical authors already employ these methods: they standardise, simplify, correct and reduce their text.

Clearly, every language will require a unique set of rules, if we were to consider a complete solution. However, there is a core set of guidelines that apply to most world languages…

1. Ensure that your spelling is correct.
2. Use plain language. For example…
   Good: drive the car
   Bad: facilitate the locomotion of the motor vehicle
3. Use short sentences, of around 25 words or less.
4. Keep to one idea per sentence. For example…
   Good: Writing clearly and plainly will make machine translation more reliable. A reliable translation will be more easily understood.
   Bad: Clear and plain writing makes machine translation more reliable, and therefore it will be more easily understood.
5. Use proper, complete sentences when you write. For example…
   Good: Would you like cream in the coffee?
   Bad: Cream?
6. Use consistent language when expressing the same ideas. For example…
   Good: I travel to work by car when it rains.
   I travel to work by bicycle when it is sunny.
   Bad: I travel to work by car when it rains.
   When the sun shines, I cycle to work.
7. Repeat a noun rather than use a pronoun. For example…
   Good: Put a stamp on the letter before posting the letter.
   Bad: Put a stamp on the letter before posting it.
8. Use active rather than passive sentence forms. For example…
   Good: My father will post a Christmas card to you.
   Bad: A Christmas card will be posted to you.
9. Use articles to identify nouns. For example…
   Good: Drive the car.
   Bad: Drive car.
10. Do not use Yoda-speak! Construct your sentences in a logical, progressive manner. For example…
    Good: Use a vacuum cleaner to collect dust from your carpet to make it clean.
    Bad: By the use of a vacuum clearner, to make a carpet clean, collect dust from it. (Hmmmm. Carpet clean, you will…)

Please remember that these are only guidelines: this kind of rigour does not suit some styles of writing. In fact, if I were to apply these rules to this blog I would have to tear down much that has been written. In doing so, I would lose the conversational tone of the blog - and that’s a price too expensive to pay!

So, controlled language works particularly well for technical documents, news items and pithy, factual pieces. It won’t work well if applied to stories or poetry, or ranting commentary! However, by bearing these recommendations in mind, you can be selective and still achieve a better translation.

For more information on the notions of controlled language, and CLOUT in particular, visit here.

Next time I’ll discuss the difference between ‘controlled language’ and ‘controlled authorship’.

MlooviTweet is very much a prototype at the moment, not least of which is the visual look. Expect some tightening up of this very soon.

In the meantime, why not trot over to MlooviTweet and give it a try? If you’re not signed up to Twitter, then consider doing so now!

The site seems to have started in April 2007, and typically features several daily posts - so there’s no lack of activity!

The site features regular guest spots by internet business experts, and is a thriving area for business development news. Additionally, business events are featured, and often entry fee discounts are offered.

Importantly, there’s a startup jobs section which is evidently regulary updated.

The tone of the site is very much as you would expect: business-like, terse and to the point. It’s all the better for it, as the blog entries are kept short and sweet, and stick to the pertinent facts.

Translated site here. Original site here.

Since October 2006 OpenLearn has provided free access to Open University study materials to over 2.5 million learners and educators, supporting the University’s mission to open up access to education for all. Working with key partners, such as unionlearn, the Workers Educational Association (WEA), the Third Age Trust and Lifelong Learning Networks, the website has been used to reach learners from under represented groups in higher education.

Billy Khokhar, Assistant Director of Regional Collaboration and Widening Participation for the OU in Yorkshire has been using OpenLearn to introduce people to learning. He said, “We’ve been recently working with a group of Asian women in community centres in the region. Many of these women would be lost to the system because family, cultural and traditional ties often make their participation at traditional campuses difficult. OpenLearn introduces and provides a flexible method of study and we work to ensure that they have the IT skills necessary to be able to continue e-learning on their own.”

OpenLearn has also been used to train staff working with disadvantaged children. The Children’s Workforce Network in Plymouth City Council is using OpenLearn to provide a common learning experience for a team of practitioners from different disciplines, based at over 15 locations around the city.

Well, depending on which source you read, there are around 6,900 spoken languages in the world today! However, before we task Google with the job of coding the translation protocols for all of them, we must take into account that many are spoken by fewer than 1,000 people.

Most popular languages? That subject is, as they say, quite a can of worms. Do we count all speakers of a language (for instance, when English is retained as a second language we add that speaker to the English count), or just primary, or native, speakers?

Taking broad estimates results in large table to be found here. You’ll note that although Mandarin ranks as number one, the estimate ranges wildly between 873 million and 1200 million people! The Top Ten are…

1. Mandarin
2. Hindustani
3. Spanish
4. English
5. Arabic
6. Portuguese
7. Bengali
8. Russian
9. Japanese
10. German

This makes for some interesting reading, and certainly highlights the fact that Google’s choice for language roll-out is not purely based on popularity, given that, for example, Czech is spoken by ‘only’ 12 million people and yet is already in the Google list.

Clearly, then, the choices for the good folk at Google are based on internet usage, and economic status. Not entirely surprising criteria for a commercial operation.

So, I’m pretty sure that the 60,000-strong Scottish Gaelic community wields a fair amount of internet clout. Perhaps we’ll see this language added to the Google list before, say, Haitian Creole?

Why so many pages?
Well, every time a blog’s RSS feed is submitted for translation we create a result in all 35 languages, regardless of which language you have chosen.

Why is this good?
Since these are permanent HTML pages they are then indexable by every global search engine that sends out a spider! Over time, and without any effort on your part, you will find that your blog will come up in searches made in lots of different languages. Is that cool, or is that cool?

Let us say thanks to you Mr, Mrs or Miss BlogOwner for using Mloovi, and for having the foresight to want to speak to your world!

The YouTube page is here.

Kris uses Mloovi to translate RSS feeds, and has written a great two part tutorial that firstly talks us through some RSS basics, then secondly focuses on the application of Mloovi.

If you’re still a little unsure about RSS feeds and Mloovi, then nip over to Kris’s blog for a second opinion. Thanks Kris!