Searchers would never say no to something that would make their work a lot easier. With the coming of Sphere ? the latest blog search engine, it promises to help users discover high-quality, relevant and timely blog posts with the use of its advanced search algorithm. It aims to deliver results far better than other existing blog search services.



The People behind Sphere

Sphere was founded by Tony Conrad, Martin Remy, Steve Nieker and Toni Schneider to assist readers in their discovery of relevant blog content as well as encourage more people to read blogs or be bloggers themselves. Investors who helped Sphere get started include Doug Mackenzie, David Mahoney, Kevin Compton, Mike Winston, Phil Black, Vince Vanelli and Will Hearst. Its partners and capitalists, on the other hand include Darcy Bentley of Hearst Publishing and Venetia Kontogouris of Trident Capital.

This blog search engine had a very successful first private beta release in late 2005. Its success can be attributed to Martin Remy and Steve Nieker who have worked on blog and content matching technology for a long time. It also has for its advisors the likes of Matt Mullenweg, founder of Automatic/WordPress and a blog content tool leader; Mary Hodder, founder/CEO of Dabble and a blog user experience expert; and Scott Kurnit, founder and former CEO of http://About.com who was instrumental in creating a work community there.

The feedback gained from the well received private beta was used to add some advanced and intuitive features. Sphere is based in San Francisco but members of the team work virtually from other places like Seattle, Denver, Vancouver, New York, Phoenix and the Bay Area. It highly recommends its own Sphere Blog for information regarding ongoing updates.

Tony Conrad was a general partner at VSP prior to founding Sphere. This was where he led consumer-tech and marketing software and services investments. He has also served as a director for Oddpost, Iconoculture, MusicNow and Centive. He has likewise been very active in managing investments in Post Communications and Stonyfield Farms. At present, he serves on the Board of Directors for Automatic/WordPress.

Martin Remy and Steve Nieker have worked together for a long time and are both considered technologists extraordinaire. They were together when they founded Think Tank 23 during the so called dot-com boom. It was able to survive largely because they were able to build a lot of great technology and sold a few software projects to big publishing companies in spite of being funded by venture capitalists. They have extensive roots in information retrieval that goes back to the end of the last century. Their NAV 4 which is a document-analysis genome that drove precise information retrieval across broad networks without onerous manual overhead is baked into Sphere. This has provided many benefits including the base technology driving the related news stories links.

Toni Schneider worked at Yahoo! and at Oddpost which was subsequently acquired by Yahoo! He created the Yahoo! Developer Network. He comes from Switzerland and only came to California to study Computer Science for a year. At present, he is happily based in San Francisco while serving as the CEO of Automatic and a venture partner at True Ventures. He remains an adviser to Sphere.

The Relevant Blog Search

Sphere is being touted as a search engine that has made blog search much better. Its interface is designed by Adaptive Path and comes out quite clean and understandable. One of its more notable features is the profile link that produces a small DHTML overlay with some vital stats for each blog such as average posting frequency, posts per week and blogs recently linked to among others. Its technology is believed to be more resistant to spam blogs than other search engines. It doesn't actually filter out such spam blogs but is able to rank them so low that readers rarely encounter them.

The right combination of factors that produces a more robust ranking technology of blogs and posts is the reason why Sphere seems to work better than other blog search. It pays attention to the ecology of relationships between blogs and credits a higher weighted value to links that have more authority. This will ensure that the work of an author who goes off-topic and engages in rant will rank considerably lower compared to that written by a reputable blogger who regularly writes about the topic.

The metadata about a blog is also considered by Sphere. It would look into the frequency of postings, the length of the posts and the average links a post gets. Sphere will not rank blogs well if they are not part of a larger ecosystem of linking.

The system has made it possible to have spams fall at the bottom of the ranking. It also gives ways to filter results either by date, relevance or by language. A new version of the services has some neat interface hacks along the lines of a time axis.

Blogs are definitely big, with millions of existing blogs multiplying by the second, relevant information can be very hard to find. Sphere helps readers to find relevant and intelligent blog posts on a specific topic based on authority. It should be noted that the authority being mentioned here is not an arbitrary decision of a human community but rather a collective effort.

Searching the Sphere is very easy. The user needs to enter some keywords and the system shows related posts, blogs, profiles, related news, photos, books and podcasts. It sorts the results by relevance or time and refines the date range with dropdowns right on the results page. It also supports quotes for phrase searching, Boolean operators and minus for exclusions.

The entry of multiple words or phrases in a query finds posts that match all these words and phrases. A query comprised of two words will find posts that contain both words. Posts that contain the words together usually turn up first while posts that contain the words separately and in different places will show up much later. If the search is to be limited to an exact phrase, only results that mention the words together are obtained. For finding posts that match either of the words, the OR operator can be used.

Excluding posts can be done by putting a minus before a term or phrase. Parenthesis is used to group terms and expressions together. A search can also be confined to the blog's domain.

Sphere has taken a new approach to blog search as it uses variables to understand both individual blog posts and the nature of the blog they appear on. It attempts to understand link structures. In the process of analyzing these structures, it attempts to understand who is starting or linking discussions instead of bloggers who simply comment on conversations.