Google Launches ‘Cross-Language Information Retrieval (CLIR)’

As promised at its “Searchology” event last week, Google has launched cross-language information retrieval (CLIR). It’s now been added as a new feature to Google Translate and went live yesterday.

As Chris Sherman previously wrote, Google is playing a bit of catch-up with this initiative; Yahoo has been doing this on a more limited basis since 2005. Regardless, it appears to be a useful service, chiefly for non-English speakers seeking information from English-language sites.

As Google Engineering VP Udi Manber said, when he introduced the service last week, the goal is to open up the Web “universally to the whole world.” And although it works from 12 non-English languages into English, there’s also reciprocal functionality allowing English speakers to translate non-English pages.

In other words, I can now search for hotels in Paris on French-language pages or bars in Moscow from Russian sources.

The languages currently supported are: English, Arabic, French, Italian, German, Spanish, Portuguese, Russian, Japanese, Korean, Chinese (Traditional), Chinese (Simplified). The service is based on automated (machine) language translation and native speakers or those conversant enough in the non-English tongue will undoubtedly find awkward translations. But it’s a helpful tool (especial the side-by-side presentation of results) to compare pages.

In addition to the new Search Results translation feature, Google Translate continues to offer its long-standing Text and Web translation (again like Yahoo’s Babelfish), so it can equally be used as a foreign-language dictionary.

Here’s the official Google Blog Post explaining the service, Search without Boundaries.