Joichi Ito does not have the kind of background that would normally catch the eye of an appointment committee searching for someone to head a prestigious university research lab. To start with, he is not an academic - he is an internet entrepreneur, a venture capitalist and a former disc jockey. And, if that were not enough against him, he dropped out of university. Twice.

But not every lab is like the Media Lab at the Massachusetts Institute of Technology. The lab in Cambridge, Massachusetts, which turned 25 last year, is world famous for its "renegade" research environment and creative and wacky projects that combine design with cutting-edge technology. It is responsible for, among other things, the electronic ink technology that e-readers use to simulate printed paper; for Guitar Hero, the hit video game in which players simulate playing the guitar in rock songs; for Lego Mindstorms robotics building kits and for the XO-1 laptop, a budget computer designed to be distributed to children in developing countries around the world as part of the One Laptop per Child project.

Nicholas Negroponte set up the Media Lab to explore human-machine interaction and the life-enhancing possibilities of new technology. He led it from its start until he stepped down as director in 2000. Ito, who took over in September, is the third person to head the lab since the departure of its founder; he took the reins from Frank Moss, professor of the practice of media arts and sciences at MIT.

"I was surprised as well," Ito says of his appointment. The 45-year-old, who recently served as chief executive of the open licensing technology non-profit organisation Creative Commons, was an early investor in more than 40 technology start-ups, including Twitter and Flickr. But he admits that, until now, his experience of universities "hadn't been great".

ÁñÁ«ÊÓƵ

Ito says he dropped out of university - once from studying computer science at Tufts University in Massachusetts and once from studying physics at the University of Chicago - because the learning environment did not suit him. This was not because he lacked respect for formal education but, as he sees it, because his brain was wired to learn by pursuing passion and interest rather than by attempting to absorb the contents of lectures and books. For this reason the Media Lab is a good fit, because it is a "whole class of people who are kind of like me". The appointment committee wanted someone whose sole project was going to be the lab, Ito thinks.

Thus far, there has been high praise for the choice. But the new director has plenty to prove and much to do, for the feeling among some is that - after a series of high-profile successes - the Media Lab has lost something of its cutting edge.

ÁñÁ«ÊÓƵ

Larry Smarr is the founder and director of the California Institute for Telecommunications and Information Technology (Calit2), a partnership between the University of California, San Diego and the University of California, Irvine that has been described as the West Coast's version of the Media Lab. MIT's Media Lab "is still a global brand that is widely recognised and admired", Smarr says, "but I would say it is not what it was in its heyday under Negroponte...They may very well need a high-risk appointment if they are going to get the Media Lab to shine at the level that it has in the past."

Negroponte, who was involved in Ito's appointment, says that Ito's highest priority must be to "re-energise the place" and create a new buzz both inside and outside the lab. "My vision [for the lab] is more about revolution than evolution," Negroponte explains. Hiring a director without a university degree was deliberate. "I believe we have become too 'academic'. I hope that changes back."

Bringing new creative spark is important not just to those at MIT, says Smarr. The Media Lab, he thinks, acts as a beacon and a symbol of the creativity of US universities. If it cannot be revitalised, it is not just the Media Lab that is in trouble. Such a failure would call into question the "Wild West" aspect of American academic culture, which allows the rules to be broken in the name of innovation.

The lab moved into a stylish new home last year and its current crop of projects - there are about 350 at any one time - span everything from prosthetic feet to robotic opera (in which robots and people perform music together) and wallets linked to bank accounts that become hard to open if you are low on cash. The 25 research groups, which each consist of a faculty member and about six students, have a combined total annual budget of roughly $30 million (?19 million), of which slightly more than 60 per cent comes from industry - a remarkably high proportion for a university lab.

The Media Lab, which came to wide public notice through books such as Stewart Brand's The Media Lab: Inventing the Future at MIT (1988), has been compared to the legendary industrial research and development operation Bell Laboratories. For Negroponte, there is a similar unconstrained nature, but this is where the likeness ends. "Think of it more like the Bauhaus than Bell Labs," he says. "The Media Lab is a style of thought, one characterised by...iconoclasm, irreverence and plurality."

Uniquely within MIT, it is both a department and a research lab, a status Negroponte thinks is crucial. The Media Lab has the power to admit students, award degrees at postgraduate level, hire faculty and offer tenure. About 25 per cent of its 150 students are new each year, and this keeps it a "fountain of youth".

There are other features that set it apart. For one, as at a design or architecture school, the culture is based on demonstration, critique and iteration (indeed, the lab falls under the auspices of MIT's School of Architecture and Planning). This is encapsulated by the motto "demo or die" - in contrast to the "publish or perish" philosophy of most university labs.

The whole thing functions "somewhat like a kindergarten", explains Mitchel Resnick, the academic head of the lab and Lego Papert professor of learning research. "We have more advanced technologies than wooden blocks and fingerpaint, but the approach - playfully designing and creating things in collaboration with one another - is the same. It is a good strategy for innovation and learning."

ÁñÁ«ÊÓƵ

The staff and students are also drawn from across disciplines. Those working there call it "anti-disciplinary" because designers, engineers, artists and scientists work side by side. The Media Lab was a pioneer of the university "interdisciplinary innovation centre", a concept that has been widely emulated over the past decade, says Smarr.

The lab also has a funding model that is very different from most. The core of its industry funding is not for directed research. Rather, the cash - from more than 70 companies whose businesses range from electronics to toys to finance - goes into a pool for the lab to work on things no one has thought of yet. Sponsors, which include Google, Hasbro and Bank of America, and other companies pay between $200,000 and $750,000 a year to draw on the results as they choose, taking prototypes, inspiration and ideas back to their own labs. Sponsors can make use of the lab's intellectual property without having to pay a licence fee. BT is one company that has had a long and productive relationship with the lab stretching back more than 20 years.

A lot of companies "don't have as much long-term R&D", says Resnick. "Part of the idea is they can look at the Media Lab to get a vision of the future and a sense of what some of the new possibilities are."

As to whether its products are intended to benefit humankind or be commercially useful, the answer appears to be both. "Making the world a better place is key and drives everybody," says Negroponte, "but that need not be limited to eliminating poverty, creating peace and saving the environment."

Walter Bender, a former director of the lab, committed the "Seven secrets of the Media Lab" to paper in 2004. They included the idea that there should be no compartmentalisation of ideas, that the investment should be in the passion of people rather than their projects, and that design and engineering should be valued equally.

Despite the lab's success, attempts to franchise the model overseas have failed. Media Lab Europe, a collaboration with the Irish government, shut its doors in 2005 after operating for less than five years. Media Lab Asia, a partnership with the Indian government, also broke down. Similar efforts were made in France, Japan, Germany, Korea and Sweden but did not work out. "You would think we would get it after a while," says Negroponte. "The reason is that the US, [the] Boston/Cambridge [area in Massachusetts] and MIT all create a setting that has far more to do with the culture of the Media Lab than any of us would have thought or believed."

American culture - for good or ill - worships youth, loves risk-taking and does not see failure as a problem, says Smarr, and this means that there are "big differences" between the US and Europe or Asia. "I think it is very difficult to translate something like the Media Lab into other countries, because they just don't have the cultural framework."

This does not mean that there is nothing UK universities can learn from the model. Negroponte highlights the "anti-disciplinary" approach. "The UK is a culture that respects idiosyncratic people, which is a wonderful start...[but] tradition has created compartments that are not meaningful, and the very notion of departments is silly."

ÁñÁ«ÊÓƵ

Locating the key to re-energising the lab is not easy. Total income is "probably as high as almost ever", says Negroponte, although the rate of growth is "currently flat and maybe a bit downward in the past five years". But times have been tough for everyone. As for intellectual productivity, the lab averages 80 publications a year, according to data from Thomson Reuters, and it is granted between three and six patents annually, although it files many more applications.

One persistent criticism is that the lab's products lack substance. Mark Sandler, director of qMedia, the Queen Mary, University of London's equivalent of the Media Lab, visited it earlier this year as part of a "fact-finding" trip organised by a UK research council.

"I wouldn't say that there was no substance, but the work seems to be a bit 'gee-whiz' and somewhat superficial," he says. "I certainly felt the work in the Computer Science and Artificial Intelligence Laboratory [another of MIT's most revered labs] was more substantial."

Negroponte also has reservations about some of the lab's output. "But that is OK...We can take a lot of risks, we can shoot high. I prefer somebody who shoots for the moon and misses than small, incremental [gains]."

The problem, as Smarr sees it, is that the Media Lab's focus "has become a little bit fuzzy". The new director needs to impose "some kind of intellectual order" on the chaos that fosters all the innovation and identify what exactly the lab is pioneering.

And while the MIT Media Lab is undoubtedly the one-of-a-kind "grandaddy" of the field, other university-based interdisciplinary innovation centres focused on the human dimension of technology have sprung up around the world.

Atau Tanaka directs Newcastle University's version, the Culture Lab, set up in 2006. He has studied the rise of "media labs", recently penning a chapter in the book A Blueprint for a Lab of the Future (2011). "The Media Lab is not the only game in town any more...there is a network of these labs worldwide," he says. These include not only university labs but also media art labs and industry labs. The democratisation and lower cost of technology mean that "you can make quite a powerful media lab in your bedroom".

"When the Media Lab [came into being] they were inventing the future," adds Tanaka. "But the future has arrived - society has become digital - and it is the things that we do with that technology that now becomes more important than the technology itself, which is a fundamentally different perspective."

As for Ito and his prospects for success, Smarr worries that the public sector can be a tough place for an entrepreneur, while Tanaka wonders if the appointment could spell a more instrumentalist direction for the lab.

It is early days. Ito has bravely initiated a review of the lab's intellectual property policy, despite its success. "It's not clear that anything has to be changed," he says. "[But] I want to make sure that we have the best policy possible and that we're not encumbering ourselves...IP policy often gets in the way of innovation, learning and collaboration."

He also plans to get the lab and companies working more closely and to inject some Silicon Valley mentality - what he calls its "agile entrepreneurship". "The Media Lab has always been irreverent. But there is a certain style of irreverence that Silicon Valley start-ups have, and I am trying to bring some of that DNA here."

For Resnick, the biggest challenge is ensuring that the Media Lab remains "an incredibly creative environment", and this will come from the right combination of freedom and structure. The fact that labs with similar aims have sprung up elsewhere can only be "good for the world", he thinks - but the Media Lab must "continue to do it better".