I wrote this piece on Open Data for the Irish Times, which was published on the 24th June last. In it I observe some of the challenges entrepreneurs may face in building their businesses on Open Data, and how these may be addressed…
“A naturally occurring DNA segment is a product of nature and not patent eligible merely because it has been isolated” – so wrote Justice Clarence Thomas of the US Supreme Court, a couple of weeks ago. The Utah company Myriad had claimed 20 year patents on the BRCA1 and BRCA2 genes, which are risk indicators for ovarian and breast cancer. Based on testing by the same company, the actress Angelina Jolie recently disclosed that she had an 87% chance of developing breast cancer, prior to her elective double mastectomy. Various medical groups, researchers and patient lobbyists argued that patenting naturally occurring DNA hampers research and innovation, including for diagnostics which currently – based on its patents – only Myriad offers.
Information arising from naturally occurring phenomena is generally considered available for all. Professor Robert Merton, the father of the sociology of science, asserted in 1942 that the result of scientific research should be freely accessibly by the community and, by surrendering any intellectual property rights, each successful researcher enables his scientific field to advance. The 2009 Nobel Prize Laureate for Economics, Elinor Ostrom, observed that “information of the commons” is similar to public goods, because use by one person does not impede use by others. She also noted that not only does use of information of the commons preserve the common stock (as with public goods), but in fact further enriches it.
According to OpenDefinition.org, “Open data is data that can be freely used, reused and redistributed by anyone – subject only, at most, to the requirement to attribute and/or share-alike”. Protagonists assert that public data, especially data whose discovery was paid for by tax-payers, are a common asset in the same way as publicly funded scientific discoveries. In December 2007, an activist group for Open Data met in Sebastopol (north of San Francisco) to lobby the then US Presidential candidates. A year later, the newly elected President Obama signed a number of presidential memoranda to promote open government, including open data.
Today, there is a myriad of open data, and a large number of open data smart apps across a disparate range of topics: just as examples, car parking optimization (availability and price); live train movements; public information about companies; mapping carbon dioxide emissions; and the evolution of legislation. The PublicData.EU web site lists many of the European open databases. OpenData.ie lists various Irish related datasets on a number of topics, including Oireachtais and Central Statistics Office information. Fingal County Council took a national lead in making open data, and now has over 90 datasets – for example, conservation areas, budgetary information, community childcare providers, and disabled parking areas. Dublinked.ie is a collaboration between Dublin’s local authorities and NUI Maynooth to exploit open data and encourage its use.
With the wealth and variation of sources rapidly becoming available, is open data an opportunity for entrepreneurs? Should we anticipate some new high growth companies as a result?
In practice, there are a number of challenges with open data. At the technical level, the quality of the information is not always as well presented as one might wish: there are sometimes duplicated and mal-formatted data, which requires filtering and repair to clean. Many public agencies note that although in theory they support open data, in practice they do not have spare resources to do so to a professional level. At the commercial level, there are a number of concerns. If data, already paid for by taxpayers, is freely published, then is it legitimate for a commercial organisation to charge its customers for exploiting this data for financial gain? For example, if the data is anonymised public health information, is a pharmaceutical company allowed to exploit the patterns and trends therein? Should a national postal address database be open data (a debate in the UK, which its Government has recently concluded by deciding against)? Is it legitimate for open data app developers to gain royalties by wrapping their work with adverts? Can a commercial organisation build a market advantage if one of its main inputs is open data equally available to its competitors?
In the software industry, there are some business models which may be helpful for the open data community. Sometimes software is freely published in its original source format, and is free at least for experimentation and prototyping. However it is common that professional and commercial use of such software requires a license, and payment: open does not always imply free.
In the world of open data, free access to publicly available information at a minimum provides a forum for amateur software developers to cut their teeth into, and gain experience developing (sometimes very) useful apps for the public at large. By contrast, any commercial organization will require reassurance that any open data on which their business is built has a stable and consistent supply: that the open datasets concerned are always available, are always current and accurate. It is therefore quite reasonable that a commercial organisation will require and proactively seek a commercial license under which the supplier (e.g. a local authority or national agency) guarantees, under financial or legal redress, that the relevant data continues to be available in an appropriate manner. In turn, there is a potential new income stream for public sector.