Tuesday, November 15, 2011

Introducing TrapIt, Your Other New Assistant

My friend John Blossom recently turned me on to a new internet facility making its debut today. It seems Siri, the iPhone assistant, has a relative called TrapIt. Underneath this web site is the same intelligence originating from the DARPA funded project called CALO which gave rise to Siri and other AI tools.

Having played a bit with Siri, I wanted to discover more about TrapIt, so as soon as it was available this morning I went to the site. Registering for the service was really easy. It was facilitated by the acceptance of Facebook or Twitter credentials. More and more sites are doing this bringing the nirvana of a near-single sing-on environment to the web.

After a few introductory screens, the service begins to display topics. Initially, it looks around and reports on what is "trending" and top stories by the usual set of categories. However, you can "tune" searches with key words and phrases and TrapIt will continue to find more.

Asked to find information about itself, TrapIt cooperated by displaying a dozen recent postings. They are organized into a neat, grid configuration, with three boxes across all the way down the page. You can sort results by most recent or most relevant. Floating over any of the results will display an excerpt of the text, while clicking on the box will bring up the whole  article.There are convenient next and last buttons on the right and left of the screen making it easy to move through all the results.

It is no longer humanly possible to monitor and consume all of the information generated each day. The fire hose of content is directed at you through so many passive channels you clearly need someone or something to filter and curate, delivering only the most relevant information. TrapIt promises to learn over time what items of news it should bring to me. It will be interesting to see if my TrapIt feed could become a replacement for my Twitter and Google Plus feeds which take literally hours of time to scan.

It is supposed to be like RSS but with a rating system that drives more of what you like into your view. You define the topics with keywords and TrapIt constantly trolls the internet looking for new content to add to the results called traps. Moving through the results, a simple thumbs up or down indicates your degree of satisfaction and this is how TrapIt is going to learn about you.

No doubt it will take weeks of experimenting to determine how well TrapIt works and if indeed it can figure out what I am truly interested in reading about. Given that I am not always sure what to read about , this could be a real challenge for both of us. But, if it delivers on the promise, it could be a real time saver.

I'll lay a few traps and see what we catch in the coming weeks. If you try it, let me know how you make out with the service.

Captain Joe

Follow me on Twitter @JPuglisiLLC