
As mentioned in a previous blog post, one of the major goals of Kutano is to help bring the right information from Twitter to you at the moment you want to see it. More specifically, we like the idea of Twitter as a powerful “thought” database, an idea brought to light and discussed by both Neicole Crepeau and Erick Schonfeld. Of course, understanding the best ways to mine and filter the data also depends on where you think the true value of it derives from. As mentioned in a previous blog entry, we believe the first step is putting tweets in the right context, an idea also brought up in Erik Schonfeld’s post through a quote by John Borthwick:
“I think context is the next hurdle. Social context and page based context...”
In the case of Kutano, context is given to tweets by placing the tweets on the web page they are associated with. That said, a further necessary step is finding the best way to filter through and rank these tweets.
The current version of Kutano filters tweets in three ways:
By “alike tweets”: Kutano currently groups tweets that are similar to each other. This way, when browsing to a new page, users see unique tweets instead of the same tweet repeated over and over from multiple or the same account. That said, for those who are looking to connect with those who tweeted these more commonplace tweets, it is still possible to view all the different people that tweeted this similar tweet with just a click.
By RTs: Similar to “similar” tweets :P, RTs are another layer of Twitter communication that are grouped together by Kutano but can also be examined in more depth by interested users. RTs are interesting because they are one (but by no means a sole) indicator of the influence of the twitterer or the value of the tweet.
By website or page: tweets can be seen through their association with the web page or with the general website. This grouping is especially useful on websites with multiple sub domains such as news sites, shopping sites or company websites. Using “tweets on a website” as opposed to “tweets on a page” is a good way to see what pages or stories on the website people are talking about.
Ultimately the way that the tweets and information is filtered determines what parts of that information come most easily to you automatically. What parts of that information would you like to see?


