Of a Different Twine? Hands-On with What is Supposed to be the Semantic Web.

I have to thank Marshall Kirkpatrick of ReadWriteWeb for breaking the silence on Twine. We have been awaiting the “Semantic Web” service ever since we did a Q&A with its founder Nova Spivack in the February issue of LAPTOP.

So when we received our press invite to the Web service last week, we were psyched to try it out. LAPTOP editor in chief Mark Spoonauer stopped by my desk a few hours after playing around with it and said , “So, pretty unimpressive, huh?”

I agreed with Mark’s first impressions. And apparently so did Kirkpatrick. When I interviewed Spivack a few months back he took me through an on-screen demo of the “first mainstream semantic Web application.” I was really interested in how the service could disseminate information from around the Web and make search smarter.

Spivack explained in the interview back in January that Twine “will manage personal information and bring it all together into an easily searchable database. Currently there’s a large quantity of information you use every day, but there’s no easy way to show everything in one place. You have to look in many different places for some information, including your personal e-mail, calendar, and Google. Twine will take this data and interpret it in one place to create your own Web.” Basically, Twine takes info on the Web and turns it into understandable search terms and ideas.

Twine collects this information by providing a plug-in for your browser. When visiting a Web page, you can click the “Twine This” button on your toolbar and a pop-up window appears with information about that page or article. Since it runs on semantic Web algorithms, it’s supposed to be able to figure out that information on its own, including what genre the content falls under and what the text is about. However, that seems to be the biggest problem with the service right now.

When I went to a blog post on Gizmodo and GigaOm, it didn’t collect any tags or information. I had to put in my own info. I had better luck when I went to NYTimes.com to an article about Spitzer’s impending resignation. It picked up tons of information including tags and a short summary. It understood this was an article about politics and a governor.

But like Kirkpatrick says:

It doesn’t consistently grab summary text or tags for pages you save in Twine, it doesn’t recognize article authors as relevant people and it often captures summary information about the domain you’re on instead of a particular page’s content.

You can check out Spivack’s reaction to this on his blog.

I had no problem saving the page to my Twine. But I did have a problem retrieving the information from my Twine portal. I found the interface a bit confusing, and I am not sure I know the difference between “MyTwines” and “MyItems.”

I did, however, like searching for information across Twine. When you input Web pages and the information is scanned, it becomes searchable to anyone on Twine. So when I simply searched for “Gmail,” articles about the topic showed up. Of course very little is saved in Twine now, but you can imagine as more and more people hit that “Twine This” button how intuitive search in this tool could become.

Sure Twine isn’t a fully baked Web 3.0 tool, but it sure is a step in the right direction in making the Web smarter.