A world-first open data dump from Australia’s patent office last week promises to grease the wheels of innovation and herald a new era in policy research around intellectual property rights.
IP Australia’s dataset includes highly detailed information on inventions, brands, designs and plant breeders’ rights from 1906 to the end of 2013, from application to the granting of rights, and will be updated annually. Importantly, the majority of the data has been matched with details of the entrepreneurs or companies that applied for those rights, including their size, location and the type of firm.
As well as speeding up innovation by making it quicker and easier for anyone to get an overview of the key players in a particular area of research and development, the dataset should be a major boon to policy researchers.
IP Australia chief economist Dr Benjamin Mitra-Kahn told The Mandarin he hoped it would allow better informed and therefore more productive debate around the future of the intellectual property framework.
“The sheer amount of data and information means that [academics] should be able to build good evidence bases for us, to help us get better insight into what IP does in the economy,” Mitra-Kahn said.
In time, the dataset should lead to a better understanding of whether society is getting what it pays for in the fundamental transaction at the heart of intellectual property rights.
“The question that we wrestle with is one where we provide a monopoly right on an invention, or a name, a design or a plant, in return for the inventor telling everyone how they did it,” Mitra-Kahn explained. “The calculus is that should benefit society on a net basis. Now the evidence for that is not as strong as one might want it, and it differs from sector to sector.”
The kind of linked-up data IP Australia has put online is usually unavailable to researchers, meaning that very little work is published about the impact of IP rights, and on the benefits of the framework to society.[pullquote] “There’s only a handful of papers in the world on that, and the main thing that stops that sort of work being done is the availability of data …” [/pullquote]
“There’s only a handful of papers in the world on that, and the main thing that stops that sort of work being done is the availability of data, and the cost of creating the data,” said Mitra-Kahn, explaining that linking the rights to details of the applicants and rights-holders was the most valuable aspect of the dataset.
“That exercise is relatively costly, and so very few universities undertake it in the world. And when they do they tend to keep the information to themselves … By making it available freely and easily, the hope is that people will start thinking: ‘Ah! What can we do? What questions can we answer? What can we use this for?'”
The process took IP Australia 15 months; according to Mitra-Kahn, about 10 of those were spent extracting the information. “It’s all about wrestling with IT,” he said.
“IP Australia’s been around 110 years and in that time we have computerised various functions at various times, and we’ve updated them separately at various times, so there’s a rather large IT legacy system that you need to sort of get through to get the data out, because none of these systems were designed for that.”
He hopes the release will push other people in government to see the value of putting “big, value-adding” tranches of information on the government’s open data portal, in useful formats.
Mitra-Kahn says the world-first project has attracted attention from other patent offices around the world. “I actually had some very nice emails over the past few days from the United States, the World Patent Office, the French office, the United Kingdom office and the Swiss office, congratulating us, saying this is what [they] want to do as well,” he said.