How Wowd Leverages the Power of Human Attention to Build a Better Search Engine

For the past 15 years, search engines have relied on the same basic model: crawl the web to find as many pages as you can, pull the page content into a bank of central servers, create an index from that content, match the user’s query terms against this index, and rank the results using as many proxy signals for human interest as possible.  This approach has numerous flaws: Crawlers eagerly import useless and spam-filled pages, proxies of human interest, such as inbound links, lag the increasingly real-time nature of the web, and the search index is only useful if you know what you are looking for.

A radically different architecture is needed to break through these limitations. Wowd, a new real-time search engine, employs a distributed peer-to-peer system powered by the users themselves to create a search user experience with significant advantages:

1) Pages are indexed as soon as they are found by any user and made available in seconds.

2) People are much more discriminating than crawlers in the pages they will visit, so a people-generated index has significantly less spam.

3) Statistical analysis of the browsing habits of the users provides real-time data on what people are actually interested in, significantly improving results ranking and therefore quality.

4) The real-time “Attention Frontier” of the user community helps those users find what is interesting and popular on the web right now.  Users help each other to find interesting trends, breaking news and popular web pages.

5) User privacy is protected. Because searches are run on the local computer and data is stored throughout the distributed cloud, searches and browsing data are kept on the user’s computer, not in centralized servers.

This talk will explain how Wowd is using this unqiue approach to give our users a better search experience.

Presenter(s)

Bill York is the VP of Engineering and Operations for Wowd. Prior to joining Wowd Bill was a Founder, CTO, and SVP of Product Development at Comergent, an enterprise e-commerce software company. Bill led Comergent’s product development efforts from its founding through to its sale to Sterling Software. Prior to Comergent, Bill was at Xerox PARC where he worked with the UIR group on the Web Forager. He was also part of the founding team of InXight, a company spun out of PARC. Prior to this, Bill worked at several companies in the Artificial Intelligence space. Bill has a Bachelor's degree in Computer Science from MIT and holds various patents. Bill’s favorite search topics include: cars, movies and gadgets.