jump to navigation

Powerset – no pretender to the throne, but the real deal June 29, 2007

Posted by John in ask.com, google, john mckinley, powerset, rdf, semantic web, triple store, yahoo.
trackback

Last night was the first step by Powerset, the new search company that there had been so much conjecture about, to shift from its stealth mode to a much more open stance about its technology and its position in the overall search ecosphere.

The event was the kickoff of PowerLabs, their creative program to leverage outreach to a broad community of developers, domain experts, and bloggers to both gain early access to the service and to serve to shape shape and refine Powerset as it marches to its major milestone of release of a more functionally complete service and its expanded initial index in the early fall timeframe. This dramatic flip in transparency was made possible by the team’s progress to-date to demonstrate to naysayers that, in a space with multiple roadkill over the last 2 decades, we finally see a player who has ridden the wave of cheaper infrastructure, advances in high performance indexing, and natural language processing to create a real step-functional advance on the state of the art.

Did they achieve their initial goals last night to outline the science behind their solution and change people’s minds about the perceived impossibility of the task? From the initial feedback (e.g., here and here and here) I see in the blogosphere, I think the answer is a strong “yes”. Powerset had all the key leadership across each function there (with the exception of Barney), and there was a really open dialog, ranging from questions about the underlying technology to more esoteric issues raised by some of the computational linguistics folk attending. I think Powerset did a great job in answering the questions without being coy or too over-reaching about the progress to-date.

What did we see?

  • A great team, with folk who have solid credentials building search experiences at scale at Yahoo, Ask.com, and other places.
  • A combination of a state of the art NLP engine from Xerox Parc and a really advanced indexing system
  • A bet on platform investment well in advance of user experience efforts, which sometimes scares folk off who worry that without early stage use cases, platform efforts might not have sufficient focus (in this case, I think the bet is paying off in a major way – this isn’t just an NLP veneer on the somewhat static state of the art in traditional indexing techniques used by each of the majors today).
  • An incredible outpouring of desire by the tech community to help the cause. There are already 10,000+ members of the Powerset Labs community that the company is just beginning to engage and leverage
  • Very nice work in the RDF/triple store space – maybe the best I have seen (and that includes some of the triple store infrastructure pure plays out there) – that’s secret sauce I won’t go into at any point moving forward, other than to say that all players in the semantic web space will or are already facing huge challenges to create something in the area that can credibly scale, and this is where only the strong will survive.
  • Real opportunity to influence the broader consumer community to stretch themselves beyond the Google muscle memory of 1-3 word keyword searching – the results payoff potential really is that much better
  • A real platform play (as a business model), that allows a company like this to have “Google-esque aspirations” in the application of their fundamental IP
  • A neat approach (via PowerLabs) to also stretch the state-of-the-art in broad community engagement at a pre-beta stage (in roles well beyond just stress-testing the bits)

I have had the great fortune over the part three months to see some of the best-in-class early stage players in a number of exciting categories in mobile, infrastructure, LMS, search, and other categories, an experience that is serving to shape my overall world-view of the opportunities over the next 3 years (the pragmatic outside limits to my planning horizon), and this is about as good as it gets. Best of luck to Barney and Steve and the whole team – you are off to a great start!

About these ads

Comments»

1. Powerset Meets the Press » Backwhack - July 1, 2007

[...] John McKinley of Great Falls Ventures (Powerset – no pretender to the throne, but the real deal) [...]

2. The Singularity Institute Blog : Blog Archive : Powerset’s first public demonstration - July 2, 2007

[...] The natural language search mashup platform Powerset: No pretender to the throne but the real deal Powerset: Open source approach to beat Google Powerset is not a Google killer! This entry was [...]

3. První "veřejné" demo vyhledávače Powerset - July 8, 2007

[...] CNET), Kevina Burtona na TailRanku, Dona Dodgeho z Microsoftu, Arringtona (TechCrunch), CTO v AOL Johna McKinleyho nebo Matta a Kyle z Uncov. 9. 7. [...]

4. Andy - July 29, 2007

You should take a look at http://www.linguisticagents.com. It’s a start-up company that has developed a natural language understanding technology that will be used in many applications in addition to search. This technology uses a deep parsing algorithm that is based on nano-syntax technology.

5. BuzzMag » První “veřejné” demo vyhledávače Powerset - August 3, 2007

[...] CNET), Kevina Burtona na TailRanku, Dona Dodgeho z Microsoftu, Arringtona (TechCrunch), CTO v AOL Johna McKinleyho nebo Matta a Kyle z [...]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 208 other followers

%d bloggers like this: