TRANSLATION IN THE 21st CENTURY
It’s become a standing joke in technology crystal-ball gazing: fully automatic machine translation will be available “within five years”, a prediction made regularly since the 1980s. Well, this time it seems to be true. In 2005 TAUS began bringing industry leaders together to promote greater technology awareness in the translation field – an industry historically shy of the glint of machinery in its workflows. In 2008 we published our white paper on Language Business Innovation, identifying translation automation, crowdsourcing, and language-data sharing as key trends for a translation industry ambitious enough to embrace the global challenges of the new century. And today, just two years later, we see the green shoots of such change everywhere.
The speed of technology innovation is clearly accelerating. So in keeping with the venerable tradition of five-year plans, we zoom in on the technologies and visions that will drive the translation industry in the early 21st century.
Estimated reading time: 12 minutes | Download the PDF (180 Kb) and read later
Where are Facebook, Google, IBM and Microsoft taking us?
At the turn of the 21st century it was hard to imagine how much translation would change in a short space of time, or that four companies - Facebook, Google, IBM and Microsoft – would have such an impact on the end-user and business experience of translation.
- Facebook would attract over 500 million users and rely almost entirely on them to localize its products
- Google’s computers would produce ten times more translated words than the entire professional translation workforce worldwide
- IBM would call on their own global workforce to customize their machine translation engines
- Bill Gates would finally identify MT as one of his five strategic technologies for Microsoft.
The activities and visions of these leading IT companies account for much of the technology infrastructure of our digital lifestyles and business concerns. And they are all deeply engaged with translation technology and process.
In just six years, Facebook has developed a social networking community of around 500 million. Facebook would rank as the third biggest country if it were one. In just two years its interface has been localized into 75 different languages. That’s 75 officially sanctioned languages in just two years.
More than half of all Facebook users today are non-English users. Twenty of the 75 languages are supported professionally and the rest by the community. Facebook counts 400,000 volunteer contributors of translations, and one quarter of this community – 100,000 contributors – are active on a weekly basis. We wonder: if 100,000 Facebook users are willing to help Facebook with translations, what will happen if that population is asked to help each other? It is easy to imagine that a community larger than the global professional translation workforce will be mobilized and do volunteer translations.
We include Facebook in our 21st century outlook on translation, not for what they do today, but for the tremendous potential that the largest social networking community brings to a translation solution. We are excited to see whether they, and others, will pick up on this potential.
Google leaves us with no doubt. True to its mission of organizing the world’s information, Google is naturally confronted by the strategic value of written and spoken language, the enabling medium for most of that information. To survive, it also has to maintain explicit linguistic and semantic links between chunks of search content in any language offered, and potential ad words that might address that content. This means that the capacity to handle multilingual versions of user-generated language is critical to its business model. Google Translate has put automatic translation in the hands of literally anyone with internet access, in a range of languages almost unimaginable five years ago. The benefits in terms of the resulting positive perception of the utility of translation is immeasurable. The four-fold increase in search terms including ‘translation’ in the two short years since the launch of Google Translate is a simple indicator of the trend.
Google Translate is a true game-changer for both the translation industry and consumers of information.
IBM is likely to have a bigger impact on enterprise and business users, and is widely recognized as a long-term innovator in machine translation. In 2006 IBM’s "n.Fluent" project kicked off as one of ten corporate innovations sponsored by IBM's chairman. The company decided that the language barrier was a strategic issue, both internally for global businesses and for companies with a worldwide client base. The new partnership with Lionbridge to provide real-time corporate translation is just one manifestation of the ongoing evolution of this project. This cloud-based service will be customizable for specific organizations' content and business processes, promoting “professional” translation for enterprise customers, as distinct from the consumer orientation of Google Translate.
IBM’s commitment to improving cross-language communication with customers, and among employees and partners, worldwide sets a standard for business users.
Microsoft has been dealing with the problem of localization longer than most companies. As the premier global consumer-centric IT company, Microsoft has been obliged to speak the language of its end users, customers, and partners in an economically viable and effective way for decades. In the past, Microsoft’s efforts to develop state-of-the-art machine translation were isolated from its ongoing product localization activities. This changed when fully automatic translation was identified as a strategic technology with great innovative power by the company’s founder Bill Gates. Now Microsoft promotes translation as a solution, with machine translation available on its customer support site.
Recently Microsoft’s MT team was involved in a mission-critical experiment to deliver automatic translation for the Haitian language during the earthquake relief effort. In exactly 4 days, 17 hours and 31 minutes in late January 2010, they built a two-way English-Creole MT system using whatever data was to hand. No surprise: Google did exactly the same thing.
Microsoft confirms the trend: translation is a utility for global consumers and citizens – a solution rather than a problem.
