How to Surf the Multilingual Web

Fact: According to an official press release issued yesterday by a Chinese-government-related organization, “1,500 translators and scholars from 76 countries and regions attended the opening ceremony of the eighteenth International Federation of Translators (FIT for “Fédération Internationale des Traducteurs”) World Congress today at the Shanghai International Convention Center. The theme of the congress is translation and cultural diversity.”

Analysis:  Sure, you’re tempted to scoff at these carbon-based units and their old-fashioned “human translation skills”  – because you  like technology so much – but in the realm of language translation, computers are still lagging behind. There’s a long-running debate over the promise of “machine translation” (MT), as a Wikipedia entry argues:

Relying exclusively on machine translation ignores that communication in human language is context-embedded, and that it takes a human to adequately comprehend the context of the original text. Even purely human-generated translations are prone to error. Therefore, to ensure that a machine-generated translation will be of publishable quality and useful to a human, it must be reviewed and edited by a human.”

With the profusion of non-English content on the web, multilingual approaches are absolutely necessary, whether machine, human, or hybrid.  GigaOm yesterday featured a piece by Brian McConnell, founder of Der Mundo, a “multilingual blogging service and translation community that combines human and machine translation.” McConnell writes, “I remain convinced that a multilingual web will be a reality in a short time, and that a menagerie of tools and services will emerge over the next few years… As these emerge, the web will begin translating itself, and within a short time, we’ll be able to read content from sources worldwide just as we currently explore the web in our own language today.”

Click for full-size screenshot

Well, guess what?  You can pretty much do that right now, with MT at the state of its art, using a nifty Internet Explorer browser plug-in. The Windows Live Translator web-service gives you the ability to surf and read across multiple languages, with simultaneous side-by-side translations.

Install this one-click Windows Live Translator button instead of just going to a Google or Babelfish page and having to copy and paste in a chunk of text to be translated. It’s much cooler to install the translator button right in your browser, and the toolbar is also good because of the other advanced maps/mail/search options it gives you.

I use it to turn on “Bilingual Browsing” – and now as I surf around foreign-language sites, the browser provides a split-pane, side-by-side with the English translation (or whatever target language) displayed alongside the original page. The speed is pretty impressive; depending on the amount of text on the screen, it usually takes 5 seconds or less to do a full page.

There are a dozen or so languages handled currently, including Arabic, Russian, Chinese (2 kinds), etc.  I believe the quality is pretty darn good for pure MT (no human in the loop).

Multilingual websurfing is really very empowering – if you’re on a site you want translated, just click the Translator button on the toolbar. From then on, you can surf from one page to another by clicking any of the links on the page, as normal, and the translator will translate each page you’re visiting.

I find this to be an extremely helpful research tool, as I expand my web reading beyond just English-language sites and the college French, Russian, and Arabic I remember.  You can keep track of the emerging MT capabilities at the Microsoft Research MT Team blog here.

(For bonus points: here’s a study that explores the startling question I began wondering about when I was a teenager first learning foreign languages: “Are you a different person when you speak a different language?” Fascinating.)


Email this post to a friend

AddThis Social Bookmark Button

4 Responses

  1. […] How to Surf the Multilingual Web […]

  2. Hi there,

    Good subject for debate.

    I work for a large UK translations company.

    There are a few points to make regarding automated translation systems such as those found online.

    Software can be used for translation projects, but I would advise against it if what you’re translating is intended ultimately for publication.

    There are often so many cultural nuances to take into consideration that cutting corners by using translation software can render your message practically impossible to understand for your target audience.

    And beyond popular European languages like Spanish (that being 2nd only to English and 4th overall in the world), you run the risk of disasterous misinterpretations.

    Often the only way to go is to utilise the services of a professional translator.

    Great blog, keep it up!

  3. Hi Maxwell, thanks fo the excellent comment, and I agree with you entirely. While Microsoft is committed t improving the quality of machine-translation output in terms of semantic quality and comprehensibility, that will likely always be sub-optimal. So we’re experimenting with social-software approaches to making the marriage between the two, human and machine, in providing expert-quality translations. More on that in a future post soon!

  4. Lewis,

    As you so ably stated, we are still a ways off from the semantically correct and publishable versions of MT output. Its value, in some worlds that see extreme volumes of foreign language content, is the ability to make an assessment of what needs to go to a human and leverage the tools to assist in reducing some of the translation costs. We tend to lose site of this ROI when educating our workforce or those that make decisions on enterprise implementation. we will find that happy balance someday. I applaud your company on doing some of the backbreaking (and budget breaking) R&D work in this area.
    Nick

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: