A report to the European Commission recently called upon Member States to step up their efforts to digitise catalogues of cultural works including books and paintings. This presents some challenges which I think help us to reflect on issues surrounding open data.
The act of digitising works costs money and doing it properly – ensuring that prints are of a high quality and high enough resolution to make them useful – can be expensive. As with any expenditure, public organisations have to demonstrate that their investment is justified by a public need or that they can recoup the money by selling what they produce. This is exactly what Birmingham Museum and Art Gallery has done by selling on-demand prints of their works online. The argument that these works are public property and should be available free of charge may hold in principle, but if all works are made freely available then the funding must come from elsewhere or they simply won’t be digitised in the first place.
I think the same applies to open data: the data may be on paper, or uncollected. The act of collating this data simply for the purpose of making it public costs money and this investment will need to be justified against many other competing demands on the public purse. Where data is already in digital form and it is simply a case of putting a spreadsheet on the internet, the argument is more straightforward; but when it costs money to collect it and make it available it is more complicated, especially if the data is already a source of income. Why would you want to give companies that happily pay handsome sums for, say, mapping data this information free of charge? Why spend lots of money collating data that perhaps nobody will use?
The Localism Bill (see previous article) attempts to solve this by releasing data according to the wishes of local residents. I think we should be exploring other ways of ensuring that data remains a source of income whilst being available for people like community activists and volunteers who want to use data to improve their surroundings.
Public authorities should think about making data available under a Creative Commons licence. This could allow them to preserve their income streams by prohibiting commercial use, whilst allowing people to use the data for personal and voluntary means. It may even encourage more people to buy their datasets by giving them the option to “try before they buy”. Further, there may be some instances where a private company can work with a public authority to collect data for its own purposes and at its own expense. By obliging the company that collects the information to make it available for non-profit public use, this may be a way of covering the costs of data collection whilst retaining the principle of openness.