Calais Wants to be Your Semantic Provider

ScreenshotAny web worker knows that there’s a lot of skill involved in finding information online. Many of us are search engine ninjas, able to dig up what we want with more or less effort – but wouldn’t it be nice if our computers just found things for us? That’s the long-standing promise of the “semantic web.” Now Reuters-owned Calais is trying to make the semantic web a reality, with a growing set of tools for both publishers and consumers.

The core of Calais is a web service that takes in content and returns semantic metadata – lists of people, companies, books, facts, events, and so on in the data. The company says they use a mix of natural language processing, machine learning, and unspecified “other methods” to do this – but however they’re doing it, it seems to work, and quickly (I’m seeing typical response times of under two seconds for the various tools). While the web service is nice for developers (and there’s an API to communicate directly with it), publishers and end users can already benefit from a number of add-ons that the Calais folks have released – entirely for free.For content providers, Calais has three tools available:

  • Tagaroo is a tool for WordPress users. It hooks up your content to the Calais engine, and returns tags that correspond to what you’re writing about. You can use these for simple tagging, but you can also use them to automatically grab relevant Flickr images.
  • The Calais Module for Drupal (built in conjunction with Phase2 Technology) adds Calais integration to any Drupal-based site. It gives publishers fast ways to add additional links and tags to their content.
  • Marmoset is for anyone running a PHP-based site. It automatically modifies pages to insert semantic data in metatags, where it can be picked up by search engines such as Yahoo’s crawler.

ScreenshotAt the same time that it’s catering to publishers, Calais is also working for end users. The current tool in this area is Gnosis, a Firefox extension. Gnosis adds a button to your Firefox toolbar. Click it, and the current page goes through Calais, and comes back with additional links on metadata entities like people or companies. Hovering one of these links gives you context-sensitive choices – for example, you can look for people in LinkedIn or get financial information for companies. There’s also a sidebar to let you drill into just the semantic links, and the ability to specify sites for which you always want semantic markup applied.

Unlike many people who have talked about the semantic web, Calais are actually doing something about it. There are plenty of other startups (like Twine) working in this same area – but of the ones I’ve looked at so far, Calais appears to be doing the best job of delivering value seamlessly.

loading

Comments have been disabled for this post