Siri’s sibling Trapit tackles web discovery but will it work for you?

Web start-up Trapit, which launched today, uses the same artificial intelligence technology behind Apple’s Siri to help users discover relevant content on the web. Users type in a keyword, phrase or URL and Trapit will find news articles, blog posts and pages that may be related. Give each article a thumbs up or a thumbs down, depending on whether the article met your needs or not, and Trapit can adjust your search results in future to better match your requirements. If you have a topic you’re tracking, you can save your search in a Trap, which then stays updated with new articles. [caption id=“attachment_131899” align=“alignleft” width=“380” caption=“Trapit is more of a competitor to iPad apps such as Zite or Flipboard, which scan the news based on keywords, than it is to Google or Bing.”]  [/caption] Trapit crawls about 100,000 sites at the moment, a tiny fraction of the 120 billion URLs that Google is thought to crawl. But whilst Google undoubtedly rules search, Trapit is more focused on discovery: locating content that is interesting and relevant, rather than answering a particular question. For example, ask Google how many pages it indexes and, after a few false starts, you’ll get an answer. Ask Trapit how many pages Google indexes, and you’ll get a lot of interesting stuff about Google, but you’re unlikely to stumble on the piece of data you need. In truth, Trapit is more of a competitor to iPad apps such as Zite or Flipboard, which scan the news based on keywords, than it is to Google or Bing. Zite bills itself as a ‘personalised iPad magazine’, and works in a very similar way to Trapit: users select a number of topics and Zite finds content that seems to be relevant. Again, users can give content a thumbs up or thumbs down, can block sources they don’t like, and ask for more content related to particular keywords. For users without an iPad, Trapit provides what looks to be a viable alternative to apps like Zite and Flipboard. I used Zite daily, but there’s only one iPad in the house so when it’s on duty elsewhere I can imagine that Trapit could make a suitable web-based alternative. Discovery and personalisation is a thorny problem. Services that relied on harvesting the links shared by a user’s friends on social networks like Twitter or Facebook often fall flat because a friendship doesn’t indicate how much people’s areas of interest overlap. In short, just because I like a link doesn’t mean my Twitter followers will. Says ReadWriteWeb:

While the trendy discovery engines these days are social, trawling your Facebook and Twitter connections and using those to approximate your interests, Trapit goes the other way. It uses only your query, your votes and its machine intelligence. “There’s no concept of crowd-sourcing on here,” [CEO and co-founder Gary] Griffiths says. Trapit shows you featured traps by other users, which you can add as your own, but as soon as you do, they start personalizing for you specifically.

If Trapit can crack the discovery and personalisation nut, they have a bright future ahead of them. Indeed, they have broader ambitions, according to the New York Times:

Trapit is starting on the Web, but the founders said they plan to introduce iPad and iPhone apps next year, as well as share their technology with outside software developers to use to search private corporate Web sites or specific media sites, for example. Trapit plans to eventually make money by licensing this technology and showing ads that are relevant to search queries.

The key to getting the most out of tools like Trapit, though, is being willing to persist in using it and keep rating the articles it gives you. The question remains about whether the average user will get enough value from that effort to bother doing it. Would you use Trapit? Does it solve a problem for you? Or does it look just too much like hard work?