In the fourth of our series on intellectual property issues, Stephen Phillips looks at commercialisation in the US, and Caroline Davis talks to a free information radical.

The United States has become one of the main grounds for the debate over the future of intellectual property rights. In a week in which the private Myriad consortium announced plans to map every protein in the human body and Massachusetts Institute of Technology declared a ten-year project to make most of its course materials freely available online, the differences in opinion could not be more stark. The Myriad move stirs memories of the private-public struggle over the human genome, while MIT's plans aim to "facilitate a new style of national and global collaboration in education" and "challenge the privatisation of knowledge".

Yet things are not so black and white. MIT is also at the forefront of moves to exploit intellectual property rights. Positioned on the doorstep of an entrepreneurial ecosystem, in its case the US's second most advanced enterprise zone, MIT technology has been behind the creation of some 250 start-up companies since 1987. It expects to make more than $70 million (?48 million) from licensing its technology this year and is now exporting its formula to the United Kingdom via the Cambridge-MIT Institute, a joint venture launched last July with ?84 million from the Department of Trade and Industry and business.

"We don't start companies, but we do plug (staff and students) into the robust local entrepreneurial environment," says Lita Nelson, director of MIT's 30-staff technology licensing office. This entails setting up meetings with venture capitalists and helping to negotiate licensing agreements.

ÖйúAƬ

Another university cashing in on its technology know-how is Stanford. Situated in Silicon Valley, it does not need to hard-sell its technology licensing. "Setting up meetings between university start-ups and venture capitalists is unnecessary. We are in the middle of a lot of venture capital dollars and lots of people have done it before," says Katherine Ku, director of technology licensing at Stanford.

Stanford's office of technology licensing has generated gross patent royalties of $450 million over the past 30 years, including the Cohen-Boyer patent on gene-splicing techniques, which generated $250 million (split with co-developer the University of California) before its expiry in 1997.

ÖйúAƬ

Ku describes the main task of her 25-person team as highlighting commercial applications for technologies that the inventors may not have thought of. "Faculty members publish a lot in journals, but it is often not clear what applications (the research) has. It is our mission to bring (these) to the attention of industry."

Stanford and MIT's roles in commercialising knowledge are well known and they have the advantage of being in entrepreneurial hotspots. For other universities, more aggressive tactics may be necessary. The California Institute of Technology, in Pasadena, is one of the new breed of institutions wise to the art of turning research papers into a bottom-line bonanza.

Last year, the tiny, elite institution spawned start-up companies. Mission control for this enterprise is Caltech's office of technology transfer, similar to those maintained by most US universities to generate returns on technologies developed on campus. As well as courting outside companies, however, Caltech has been more active than most in reaching out to on-campus inventors themselves.

A four-strong team pinpoints promising technologies and helps faculty members and students create companies around them. In return for a 5 per cent stake in their start-up, Caltech entrepreneurs can enlist the technology transfer team to take care of patent applications, broker meetings with venture capitalists and even help scout for office space.

Director of technology transfer Lawrence Gilbert characterises the operation as a "virtual incubator" for start-ups. Success stories include the home-grown genomics company Clinical Micro Sensors, bought out for $280 million by electronics giant Motorola last spring, from which Caltech pocketed $5 million.

Gilbert says his team has set out to "develop strong relationships with (academic staff and students) and build rapport and trust". Invention disclosures have risen from 60 in 1995, at the programme's inception, to 160 last year, more than one for every million dollars of Caltech's $150 million annual R&D budget.

Gilbert argues against allowing research material to be made freely available: "You've got a technology you have put a lot of effort into, and believe can make a difference - one way to ensure it gets the best exposure is to develop a product and form a company around it."

Caltech also vigorously pursues licensing agreements with companies. Equity stakes in start-ups and established companies, plus ongoing patent royalty fees, generated $ million last year. This year, a tepid stock market and a straitened venture capital funding climate lead Gilbert to expect a return of $10 million.

ÖйúAƬ

Direct returns, however, are only part of the story. Caltech is banking on its faculty-friendly programme creating a funding circle. "A percentage of the companies will create a successful cadre of alumni who will become philanthropists in the long term," Gilbert says. He is perhaps thinking of Jim Clark, founder of computer workstation vendor Silicon Graphics and internet browser pioneer Netscape, who in 1999 thanked Stanford for letting him pursue commercially oriented research with a $150 million donation.

Programmes such as those at Caltech, Stanford and MIT were made possible by the 1980 Bayh-Dole Act, which authorised US universities to patent the results of federal government-funded research. The aim was to bolster US competitiveness by making university ideas available to industry, while handing universities a revenue source from royalty fees.

But science policy expert Sheldon Krimsky, professor of urban and environmental policy and planning at Tufts University, argues that such technology licensing programmes allow corporations to hitch a cheap ride on publicly funded research by turning public risk into private gain.

MIT's Nelson counters that industry does bear risk in commercially developing university intellectual property. "Getting products and jobs out there only happens if industry is willing to take the risk of going from the gleam in the professor's eyes to making it real," she says.

The Association of University Technology Managers adds that the commercialisation of academic research makes a major contribution to the public purse. It estimates that sales of products derived from university intellectual property generated tax receipts of $5 billion in 1999.

Nevertheless, while technology licensing may be viewed as a clear-eyed acknowledgement of the commercial imperatives facing universities, many US academics view with misgivings the growing cosiness between universities and industry that it fosters.

The increasing incidence of university staff serving as consultants for companies and taking equity stakes in them, which arises from technology licensing agreements, poses a threat to the role of academic institutions as seats of disinterested learning and research, Krimsky says.

He believes commercial considerations are already imbuing universities with a proprietary approach to knowledge. "The tradition of science is as a communitarian enterpriseI If you want to succeed in industry, you have to be controlling of information."

After the death in 1999 of a patient who took part in an experimental gene therapy operation at the University of Pennsylvania, it emerged that researchers had failed to report adverse reactions from previous trials and that a company founded by the lead researcher had partly funded the research.

Such incidents crystallise the importance of robust conflict of interest policies. Although Caltech, Stanford and MIT all claim to have rigorous regimes, governing the amount of time staff can devote to and the financial interests they can take from extra-curricular enterprises, this is far from standard across the US.

ÖйúAƬ

A recent poll of 89 leading medical research universities, published in the Journal of the American Medical Association , found, for example, that only 19 per cent specified limits on faculty financial interests in corporate research sponsors.

Crusader for freedom

When the co-operative hacker community at the artificial intelligence lab at MIT fell apart in 1981, programmer Richard Stallman faced a choice: join a software company and sign non-disclosure agreements or start again by writing his own operating system that he could make available to his fellow hackers.

He chose the latter - and the Free Software Movement was started in 1984. He also worked on the system that became GNU/Linux, now a household name in the information technology industry.

Stallman believes in freedom - not giving software away for nothing, but allowing everyone access to the source code so they can change and redistribute it. He sees helping other people as the basis of society and free software as a moral issue.

Stallman uses recipes to explain his idea of intellectual freedom. With a cake recipe, the user is free to substitute butter for margarine if they feel it tastes better or to pass on the recipe to friends.

He objects to the term intellectual property. "It carries a hidden assumption - that the most natural way to think about the issue of copying is based on an analogy with physical objects and our ideas of them as property."

Stallman believes this overlooks the difference between material objects and information. "Information can be copied and shared almost effortlessly, while material objects can't be," he says.

The term "intellectual property" also lumps together copyright, patents and trademarks, which Stallman sees as very different entities.

For Stallman, copyrighting and patenting should be allowed only if they permit progress. Making money does not come into it.

"Maybe you'll need to have a day job. How much money programmers make is not as important as our freedom. The current method is wrong and obstructs the progress of science."

Stallman has come up with a way of protecting the freedom of software that he calls Copyleft. Its slogan, dreamt up by a friend, is "Copyleft - all rights reversed".

"Instead of a means of privatising software, it becomes a means of keeping software free. We give everyone permission to run the programme, copy the program, modify the program, and distribute modified versions - but not permission to add restrictions of their own. They become inalienable rights."

Stallman also distinguishes three copyright situations. Functional works such as recipes, software and textbooks must be free so that people can modify and copy them as they choose, improving or individualising them as required.

People should be able to copy works that state opinions such as scientific papers, memoirs and speeches but only if they do so verbatim. Aesthetic or entertaining works where the content is intrinsically of interest, such as novels, plays and poetry, are the only ones where he sees the need for copyright, but only to enable publishers to continue to publish them.

Stallman treats patents differently. They are unacceptable for software, but for pharmaceuticals and engineering, which rely on physical materials, he believes they are useful.

Because software is built from a great many abstract mathematical ideas, it is relatively cheap and easy to put together large complex systems. If any of these elements is patented, this would prevent development in the field.

But developments that use matter are costly and face logistical hurdles. Patenting enables them to progress.

Stallman drew his inspiration from the scient-ific model of cooperation. Therefore, he is angry that universities on both sides of the Atlantic are bowing to government pressure to commercialise their research.

"It's important for universities not to be dependent on corporate money to do research. University administrators should realise that this is not a viable approach. He who pays the piper calls the tune."

  The ownership of knowledge, back to contents

ÖйúAƬ