Wired on Friday: When you walk into Squid Labs, you can tell you're in the office of a Bay Area tech start-up. There's the obligatory toys, the grinning geeks and the guy asleep by the office library who worked too hard the night before.

But where the toys are usually spread thinly on top of dull corporate life, Squid Labs feels more like a company hastily thrown up underneath its very cool toys.

Squid Labs is Colin Bulthaup, Dan Goldwater, Saul Griffith, Ryan McKinley, and Eric Wilhelm, all recently of the prestigious Massachusetts Institute of Technology (MIT).

They are all engineers - the old-fashioned kind that deal with the physical world - but borrow from the ethos of the software hacker. The five founders weren't happy with the traditional paths that lead out of their MIT education.

Academia has lost of a lot of its appeal in recent years, and the day job with the multinational wasn't going to be interesting for long. So, like many people before them, they piled into a van and headed west to try to make their own scene. Less than a year after founding their company they're doing pretty well.

Between them they had plenty of patents and awards before forming the start-up, but as Squid Labs, they hope to see their first product on retail store shelves in six months. It's a mix of old and familiar and the new and exotic: a "smart rope" woven with sensors that reports strains and fraying via an interface at the end of the tether.

Squid Labs isn't just a development boutique for hardware manufacturers. Their open source-style Instructables website allows people creating objects in the real world to share methodologies in a common and reusable format.

Their Howtoons educational comic series - teaching proto-engineer teens how to make marshmallow cannons, soda bottle rockets, and, thankfully, their own safety glasses, has been picked up for national redistribution.

From eye glass manufacturers to the military and Nasa, everyone who has a hand in hardware manufacture seems interested in Squid Labs' eclectic talents. But with a portfolio that moves from bolo-powered flashlights to augmented reality tablet PCs that overlay data on the Third World to educational cartoons, it's hard to know exactly what Squid Labs is.

The idea for organisations like Squid Labs came from an institution in the aerospace industry called skunk works. Skunk works are small, carefully selected groups in a company's R&D department that are intended to go rogue.

They are generally led by a charismatic insider or star team, unshackled from the corporate rules and budgetary constraints to go out on their own to solve intractable company problems and take existing research in new, novel directions missed by straight-laced company procedures.

Unlike their ancestors, the new hacker-style skunk works companies don't belong to larger companies or research institutes, and don't have a guiding hand telling them what to make.

Squid Labs is in the business of making it up as they go along. What ties all these projects together isn't a field or even a set of disciplines, but a particular aesthetic.

"We generally think about the physical world as a programmable substrate. That lets you think of embedding other systems inside of things like rope - using the fibres themselves, the way things interact, as the data," explains Saul Griffith.

They work with factors of the physical world to find ways to build on the known properties of everyday life. It's a malleable way of looking at what we think of as the static world - hence rope, that oldest of technologies, is a medium full of data to be captured.

But how does someone turn their hacker ethic into a revenue stream? The business of Squid Labs is fundamentally one of creating and licensing technology.

Griffith explains: "We partner with people who do the manufacturing."

This means Squid Labs solves existing problems brought to them by companies and governmental agencies, or takes their ideas to those people and sells their inventions. Instead of going through the cycle of bringing a product to market, Squid Labs is almost more of a service to industries, applying their unique vision to "reinvent other people's industries".

These are the people the patent office was made to serve - they invent and patent, and the financial support from licensing their technology supports the next round of inventing.

The Squid Labs business model takes this to an extreme. They are counting on a continual stream of invention without academic or corporate oversight to provide a living for the researchers and a viable company. Though they have mostly bootstrapped themselves with individual contracting projects, they aren't adverse to a little help.

They're actively seeking angel investment that would allow them a broader range of research options.

The question is: can Squid Labs survive? Its independent and eclectic spirit is a sign of the times, but it's also the trappings of an extraordinary group of people, willing to do bit work to make their lab a financial reality away from the structured methods of Bay Area venture capital.

Griffith notes: "There's some resistance to taking normal venture capitalism because there's resistance to that model."

However, he sees and hopes for the possibility of venture capital-funded specific Squid Labs spin-offs. What no one can know for sure is if that dedication and spirit survives these five people.

A skunk works without a corporate host stands out, but a skunk works as a corporation? That's something a lot of people might pay to see.