New search hurdles are not technical

One of my Informatics classmates is working at Kosmix, a next generation search engine company. Unlike the Google and Yahoo! models, the results are intentionally presented in a way that highlights the user motivation for search rather than the technical output.

Another search option, being developed at Indiana University, leverages knowledge embedded in local practice. 6S (“Success”) is a network approach which relies on individual search indexes and intelligent local networking through member nodes to amass collective knowledge about what links to present.

In both cases, adoption is going to depend on more than just technical and usability know-how. Perception is the biggest obstacle.

I grew up on Alta Vista and, briefly, OpenText before using Google exclusively for search. The main reasons I like Google are:

• A sense of reliability. Most things I am looking for are on the first page. That is a perception that only comes from lots of use.
• A sense of modernism. At a basic level, I understand Google to be cutting-edge, a better way of finding information than what came before.
• A sense of stability. Google isn’t going anywhere anytime soon, so the search techniques I have learned will likely be applicable in the future.
• A sense of service. While Google doesn’t present the search results in the manner you/other newer tools do, they provide many different channels that help filter based on my needs (i.e. Google Scholar and Blog searches).

Kosmix or 6S could be better tools, but the above predispositions toward Google are going to have to be overcome. Both are works in progress. Kosmix will present a new face to the world soon that should be very different from the current site. 6S is a technological brainstorm which needs to emphasize usability. Under the hood, though, the results still have to reach Google levels for reliability, stability and service.

To get a better gauge on some of those technical differences, I did a search for “Will Dennis Kucinich be the next president” using both the new and the old of search tools.

My peer search using 6S turned out to be identical to a Yahoo! search:

1. Kucinich Vlog
2. Kucinich for President 2008
3. Dennis Kucinich 2004
4. Dennis Kucinich Blog and News
5. Dennis Kucinich parody site
6. Dennis Kucinich: ‘We have to bring our troops home’ (October 2003 article in CNN)
7. Op-ed from Ben Cohen on Dennis Kucinich (Jan. 2004)
8. Op-Ed from ‘Spirit in the Smokies’ Magazine
9. CQ Politics Interview with Dennis Kucinich (Dec. 2006)
10. The Dirty Rag commentary about Kucinich’ 2004 presidential bid announcement

By comparison, Google gave pointers to the current Kucinich site, his blog and the parody site. But the remaining links included several more recent items as well as Dennis’ 2003 posts on the Lawrence Lessig blog and a meetup in Georgia for the current campaign. Google’s search results also had an extra page listed from the main site (which showed up first in the results, rather than the vlog).

Kosmix — whose search results at the moment appear a bit cluttered, looking more like a crowded blog than a search tool — was very different. There were some overlaps (43 Things, Kucinich parody, and the Spirit in the Smokies commentary), but the top results also included: a Blame Bush blog post (Feb. 2004), the Official House site for Dennis Kucinich (querying 2002), a campaign sticker shop from Irregular Times, a 2003 message from a listserv, a Draft Dennis petition site following the last Presidential election, an anti-Kucinich blog that stopped at the end of 2003, and a broken link from 2003.

What does this prove? Nothing. 6S is currently recruiting participants in their initial testing of the system, which improves thorough frequency of use and number of users. The strength of Kosmix will likely be personal relevance and ability to browse meaningfully to other searches. Neither are currently living where their vision resides.