The first question that arises in relation to this piece of legislation is why it has been drafted - what perceived need does it fill? And why the need to draft so widely - and even inventions that might conceivably become patents some day?

As far as I can establish, there are two separate areas that the government feels needs addressing. One is the perception that the universities are not performing well enough in delivering value for the money that is being invested in public research in the country. The other is that South African knowledge resources and intellectual property - as is common across the developing world - risk being pillaged by patent-seekers from the global North, particularly from the USA. In the later view, unless we protect ourselves with a strong IP regime, we will risk losing the exploitation of our intellectual capital to more powerful Northern pirates and raiders.

As South Africa's National Research and Development Strategy (2002) said: 'These are valid concerns. More South African research needs to be more effectively disseminated and exploited for the national benefit. And the risk of predatory raids by US bounty hunters is real enough - the Rooibos case is the most high-profile recent case in this regard and there are genuine concerns about how best to protect traditional knowledge from appropriation. The problem is in the solution being proposed, which, I would suggest, is in fact contrary to some of the DST's most enlightened - and most central - policy-making and might well be the wrong cure for the disease.

I was concerned to see in an ITWeb article that Matlu Mabokano, manager of hydrogen and energy at the Department of Science and Technology (DST), is quoted as saying that the Bill is heading for Parliament this week even as comment is being sought. He is quoted as being dismissive of the fact that there have not been many comments submitted yet, accusing South Africans of being chronic last-minuter responders. This seems an opinion based on a blithe assumption that the issues in the Bill are not problematic and are simple and straightforward to respond to. This is not the case -the issues at stake are very complex and it has taken the Australian government, for example 800 pages to summarise the outcome of its consultation on the same issues in the Productivity Commission Report published two months ago. Moreover, as the DST itself wrote in the National Strategy for Science and Technology: 'International thinking on legislation is as fluid and fast-moving as the new technologies themselves'. Yet Mabokano's apparent assumption of simplicity and obviousness is not an uncommon view among those who propound proprietary models of IP protection. The Copy/South Dossier, which reviews the global IP regime from the perspective of developing countries, argues that the 'dominant discourse around intellectual property - whether legal or sociological - starts from some largely unexamined assumptions'.

The assumption that a strong IP regime on its own fosters development and economic growth is one that is being increasingly challenged worldwide. Policy-making needs to be forward-thinking. As NEPAD argues in its discussion document on science and technology indicators, policy-makers need to be able 'to discern, based on their expert knowledge, the future trajectories of the subject and the interventions which might improve its development'. The future does not look as if it will be one of proprietary IP systems only.

The Department of Science and Technology has, as I noted in my last blog, published a Draft Bill on IP in Publicly Funded Research for comment - and comments have to be submitted by 18 July which is next Wednesday. This is causing a scramble among those of us who have an interest in IP issues and particularly those of us who support commons and non-proprietary approaches to knowledge dissemination, especially for developing countries.

First, what the Bill says. What I offer here is a quick and crude summary but Andrew Rens will let us have a human-readable version very soon. I will set out the provisions of the Bill - as I understand them - in this blog and will then go one to reflect on various aspects of this legislation in further blogs. The issues are complex, the time for discussion and submissions very limited, so any responses, arguments and reflections would be very welcome.

Briefly, the Draft Bill requires all publicly funded institutions (which includes all universities) to have an IP Management Office and provides for the creation of a National IP Management Office. (Because my interest is in the universities, this is how I will frame my discussion from here on.)

The Bill provides for the IP in all patentable inventions to reside in the university. If the university does not want to patent a particular invention, then the right passes to the National IP Office. Only if the national office does not want to patent an invention do the rights pass back to the researcher concerned. The Bill also provides for the sharing of rewards in the patented invention.

In other words, it appears to be modelled on the Bayh-Dole Act in the US. But there are also important differences. One is that the SA Bill is more stringent and less flexible. There seem to be fewer rights for the inventor and the Bill extends beyond the protection of patents to 'copyrights in any work related to patentable inventions' which also become the IP of the university. This seems to me, as a publisher, to be very wide, as it looks as if the university would have IP and would therefore control for publication purposes a very large number of potential publications. Would a researcher be able to publish, without university permission, a conference paper or journal article on her research even after the patent is registered?

The next move is that the legislation also covers 'the protection of basic scientific research results that are capable of forming the basis of a patentable invention but are not yet capable of protection under the Patents Act..' The Minster can extend these categories further by notice in the Government Gazette. In other words, this is a very wide-ranging definition of the IP that will become the property of the university or the State. Because these definitions are vague, it looks as if just about everything could fall under these provisions.

All employees will be deemed to have assigned their IP as defined by the Bill. Employees are defined as including students undertaking research in the institution. The universities and their employees (and students) seem to be obliged to exploit commercially any research that is capable of commercialisation. If they do not, they can be subject to disciplinary action. Any potential IP from research carried out in the university has to be reported to the IP office within 30 days of identification by the researcher.

Then - and this startles me as a publisher - the university IP office has to screen 'all publications from the institution for potential IP that through publication might lose protection in terms of the Patent Act.' All publications? I ask. That would mean journal articles, conference papers, chapters in books, research reports, etc, before they are published. Because a 'publication' is not defined, it probably means blogs, website, online discussion forums as well. In other words, researchers are constrained from communication until their work has been vetted by the IP office to ensure that they are not revealing something about a potential patent. I cannot imagine this being carried out unless the university hires teams of specialist people to scan all publications at pre-publication stage. This would surely have s seriously chilling effect on publication and this is turn would impact negatively on the revenues that the universities get from their publication subsidies, which are an important revenue stream for them.

It is the institution, in consultation with the national IP office, that decides the licence conditions for an invention, not the researcher concerned. Exclusive licensing is preferred.

Where research is co-financed, there are very specific provisions as to how the IP for these funded research projects shall be managed. It appears that funders would not be able to stipulate that IP should be exploited any other way other than commercially, through patenting. The funder can only become a co-owner or IP holder if they are in a position to commercialise the the IP. This section does not appear to provide for non-profit bodies and I question whether this is not going to be a serious disincentive for donor funders who currently invest in South African research.

It also looks as if, given the copyright provisions in this Bill, that stipulations and mandates from funders for Open Access dissemination would be seriously constrained.

These are serious issues, but before we go ballistic and start taking all this apart, we need to step back and consider why the very good department enacting this legislation is considering such draconian provisions. That is another story - - - next thrilling instalment in my next blog.

We have known for a while that there was a bill in the offing on the management of IP in publicly funded research and this Draft Bill is now available for perusal on the website of the Parliamentary Monitoring Group. The Deputy Minister of Science and Technology, when he visited CET, said that there would be a period for comment on the Bill and, as this draft Bill does affect universities and researchers in universities, I am providing a heads-up for those of you who have a particular interest in the management and ownership of the IP in the research that you carry out.

I was half expecting a Bill on the rights of public access to publicly funded research, along the lines of discussions in the UK, the USA and the EU, among others, for access to research publication. South Africa is a signatory of the OECD Declaration on Access to Knowledge from Publicly Funded Research, so probably needs to enact provisions of this kind at some stage.

This Draft Bill is not along those lines at all. It appears to be about institutional and government control of the commericalisation of research and provisions for any research that is potentially patentable. I have not had time to peruse it properly nor think through its implications - these in any event probably need to be teased out by an IP lawyer. However, it would be interesting to get reactions from researchers at UCT and other universities as to how they perceive this Draft Bill and how it might affect them.

The Australian government's Productivity Commission has recently undertaken a major exercise on returns from public investment in research and there is much discussion in the 800-odd pages of its report, issued in March 2007, about the kind of issues faced in this Draft Bill. From my perspective, as someone who deals in copyrights - the dissemination of research - rather than in patents, the following statement struck a chord:

Ultimately, in terms of community wellbeing, it is the transfer, diffusion and utilisation of knowledge and technology that matters. The social return from public
investment in R&D depends on: whether knowledge and technology are transferred out of universities (that is, whether they see the light of day); how fast and widely the knowledge diffuses among potential users; whether the knowledge andtechnology is developed into some form of practical application (that is, whether it is taken up in some form or other that is welfare enhancing); and how widely the resulting innovation is utilised. There are multiple pathways for achieving these benefits (p.280).