At the HP Labs conference today, Lab director Prith Banerjee presented in his keynote this new "blueprint" for HP's coporate research arm. HP Labs is now going to focus on "big bets" all linked to 5 main areas: Information explosion, dynamic cloud services, content transformation, intelligent infrastructure and sustainability. More on these in HP's press release here.

What's interesting is the deep consequences this new plan has on HP Labs structure: going from hundreds of projects to 20-30 big bets. The Labs' more than 600 researchers will now be "regrouped" into 23 separate labs around the world, of 20 to 30 researchers each: smaller teams which should make them more dynamic and responsive to change. In terms of investment, HP Labs will now spend 1/3 of its $150 million annual in exploratory research (versus 10% before), 1/3 in applied research and the last third in advanced product development. More on that later.

HP to foster "open innovation", collaborative research and input from VCs, start-ups, academia, government, etc... To inject start-up DNA into HP Labs

"We realise that not all the smart people work for HP Labs", said Banerjee.

HP's open innovation programme has 3 aspects:

1) an innovation research programme with universities that will start with a call to proposals for projects in HP's 5 focused areas. HP will fund 100 students per year who will work on these projects during Summer and that will also form a pipeline of recruits, future talents, for HP Labs;
2) the Entrepreneur in Residence Programme will form partnership with VCs, like Foundation Capital, in hope that they will tell the HP researchers what's hot. In return, HP will share its research expertise with VCs and their portfolio companies;
3) Idealab, a website that offers a peek into certain early-stage software projects inside HP Labs, designed to encourage feedback from consumers and the developer community who can download the software. I've taken a "peek" at it and although the projects look innovative (but I'm not an expert), I don't think there's anything that was not already out there.

HP Labs will incubate research projects and license core and non-core technologies

"This has been a challenge in many corporate research labs. We do all this cool work and we have difficulty in coming out with products and services based on the fantastic research we do", admits Banerjee.

So on the foot steps of Stanford University and other leading academic institutions, HP is creating a technology transfer office that will help quickly transfer the scientific discoveries of the lab into commercial products. It will do that in 2 ways: 1) By incubating the research project, combining 1/2 a dozen of researchers with 1/2 dozen of product engineers, as well as sales and marketing staff to build a prototype and try to sell it in the marketplace; 2) And for the technology that HP chooses not to productise, it will make it available through its licensing office. The first 4 technologies coming from the labs that is available for licensing including the cool Memory Spot project and Nanowires.