Inspiration
The Kiffets project (2008-2010) was about organizing news on the web. It grew out of earlier projects at PARC on information foraging, especially the “ScentIndex” project which organized search results for a book according to curated categories from the back-of-the-book index.
I was inspired by that project because it brought a conversational intelligence to search. In conventional search, a user gives a short query and just gets back a list of matches. Since users do not know exactly what information is available, they do not know detailed they need to be in their query. In the scent index, a user could query a book about user interfaces for information about (say) “Ben Bederson”. Instead of simply returning a list of results mentioning “Bederson”, the scent index would essentially ask — “Do you mean information about “Pad++” or the “Human-Computer Interaction Lab” at the University of Maryland, or perhaps his other projects like “Fish-eye menus” and so on. In this way, the index provided a framework that informed the reader and organized results.
Could the idea of an index be made evergreen to cover the expanding material of the web? Could it draw on the expertise of multiple curators, rather than just the author of one book? Our best idea for where to apply the idea was for personalized news — making it possible for people to get the news that mattered to them, curated by experts that they trusted.
The Kiffets Project
Lance Good and I built an a personalized news system where readers could subscribe to organized channels of information that were curated by experts. Our design goals were to radically reduce the work load of curators, and to efficiently present information to readers. The system went through three implementation cycles and processed over 16 million news stories from about 12,000 RSS feeds on over 8000 topics organized by 160 curators for over 600 registered readers. (We were joined by Barbara Stefik, Ryan Viglizzo, Sanjay Mittal, and Priti Mittal — who created channels and carried out usability studies).
See the 2012 AI Magazine article cited below for a summary of the technology, early user experiences, and business issues. Although we attracted the interests of two of the top five U.S. newspapers, we and the PARC business team were not able to commercialize Kiffets satisfactorily.
Videos
You can see a demo in the following video overview.
Getting the News that Matters to You
When we launched Kiffets for beta users in June 2010, we created four 1-2 minute videos to help readers understand how to get personalized news, and to help curators understand how to create news channels. These videos are available on YouTube.
- Getting the News that Matters to You (video)
- Making Your News Time Count (video)
- Building a Simple Channel (video)
- Building Multi-Topic Channels (video)
Papers
Below are some papers about social indexing. I include a chapter Focusing the Light from my 1999 book The Internet Edge because it frames and describes the problem solved by Kiffets (but without solving it).
Stefik, M. & Good, L. (2012) Design and Deployment of a Personalized News Service. AI Magazine, Vol 33, No. 2., pages 28-42.
Stefik, M. (2011) We Digital Sensemakers in Bartscherer, T. & Coover, R. (eds.) Switching Codes: Thinking Through Digital Technology in the Humanities and the Arts. Chicago: The University of Chicago Press. pp 38-60.
Stefik, M. (2008) Social Indexing. CHI 2008 Sensemaking Workshop, Florence, Italy. April 5-10, 2008.
Stefik, M. (1999) Focusing the Light Internet Edge. Chapter 5 in The Internet Edge: Social, Technical, and Legal Challenges for a Networked World. Cambridge, Massachusetts. The MIT Press.
Patents
Several patent applications have been filed about the social indexing technology of Kiffets. The U.S. patents for personalized news and social indexing started issuing in 2011.