Powerset isn't yet gunning for Google however. It's aiming only at this point to prove that its natural language-based search makes it easier to find things on Wikipedia and Freebase, an open community database. And even that ambition may be overreaching, given the uneven technical response of the site.
Such opening day slowness might be less of an issue were it not that Powerset relies on Amazon Web Services infrastructure, which is supposed to be able to cope with large surges in demand.
In an article on Amazon's Web site, Barney Pell, founder and CEO of Powerset, is quoted as saying, "By using Amazon EC2, Powerset is able to match the infrastructure of large-scale search companies on a startup budget."
On Monday at least, Google's site was far more responsive than Powerset's. There may be a good reason for this however: On his blog, Pell says that Powerset was hit with a denial of service attack two hours after it launched.
In contending for the search crown, or at least a trophy, PowerSet is primarily concerned with search relevance.
"Instead of being limited to keywords, Powerset allows you to enter keywords, phrases, or questions," the Powerset blog explains. "Instead of just showing you a list of blue links, Powerset gives you more accurate search results, often answering questions directly, and aggregates information from across multiple articles."
Powerset does appear to deliver more accurate search results than a search using Wikipedia's search box. But a Google search is arguably more accurate than a Wikipedia search too. Powerset can answer questions directly. But so can Google. Given the question, "Who is the president of China?," Powerset did not return a visible name on its search results page, whereas Google returned "Hu Jintao" as its top result.
That's not to say Powerset's technology isn't useful, because it does offer some real user interface advantages over Google and other search engines. And the way it segments search results by concept can be very useful. But it does suggest there are probably a fairly narrow set of applications where natural language search provides an edge.
A lot of sophisticated computing goes into understanding search queries. The question is whether a user might get equally useful results from a less linguistically savvy search engine like Google by putting a little thought into his or her query, instead of expecting the machines to supply the absent brainpower.
Or perhaps the question is whether Google or Microsoft will acquire Powerset, as rumors suggest.
Stay connected and informed by visiting our Enterprise IT Community!

Become a member today for instant access to free InformationWeek research, expert advice, peer perspectives, and more on the following topics:
- Application Performance Management (APM)
- Security Management
- Mainframe 2.0
- IT Automation
- Service Assurance
Also, visit our Government, Retail and Financial Services groups to see how these technologies apply specifically to those industries.
NOTE: Offer valid for U.S., U.S. possessions, & Canada only.