It´s a fad! 7 Reasons against Open Data

Claudia Schwegmann - Gründerin der OpenAid-Initiative, die sich für Open Data und Transparenz in der Entwicklungszusammenarbeit einsetzt - hat einen Artikel für Digital Development Debates geschrieben, in dem sie 7 oft genannte Argumente gegen Open Data untersucht. Voilá:

The interesting thing about arguments against open data is, that is is really similar to arguments against innovation in other areas, the first personal computer, the invention of telephones and early research on aeroplanes. Reflecting about the many discussions on open data and particularly the International Aid Transparency Initiative (IATI) since 2009 I have identified 7 very popular arguments against innovation. The following article describing these arguments was first publised in the Digital Development Debates Magazine:

It’s Just a Fad!

“I think there is a world market for maybe five computers.” In 1943 this rather dire prediction for the IT branch was made by no one less than the Chairman of IBM, Thomas Watson. Looking back on such erroneous predictions of the short-lived nature of products so familiar to us today can be very amusing. Are we currently experiencing the same phenomenon with the purported swan song of open data?

In his day, William Preece was a true techie. He studied civil and electrical engineering in the 1860s and took a post as consulting engineer for the British Post Office a few years later. William Preece was not a bureaucrat. In fact, towards the end of the century he made important contributions to the development of wireless telephony and telegraphy. Still, the reason that many people today have heard of William Preece is not because of his technological accomplishments, but rather for his drastic misjudgement of the potential of new technological ideas. In 1876, Sir William Preece famously claimed that “the Americans may have need of the telephone, but we do not. We have plenty of messenger boys.”

“No balloon and no aeroplane will ever be practically successful.”

How can this be? How could the later Engineer-in-Chief of the British Post Office and inventor of wireless telephony and telegraphy have failed to see the potential of his own area of expertise? And William Preece is not alone. In 1902 William Thomson, the first British scientist to be knighted by the Queen and a specialist in electricity and thermodynamics, predicted that “no balloon and no aeroplane will ever be practically successful.” Seventy years later Ken Olson, the Chairman and founder of Digital Equipment Corp, made history by totally misjudging the added value of personal computers: “There is no reason anyone would want a computer in their home.”

The least that we can conclude from these examples is that we should be careful not to prematurely reject the value of new technologies. The big challenge of innovation, it seems, is not to come up with a new idea, but to gather sufficient evidence of its utility and garner broad support from relevant stakeholders (bureaucrats, funders, users, customers). Only this kind of support will allow an idea to be adopted widely enough to be a true innovation. What we can also learn from these and similar quotes on technological innovation is that many new ideas face very similar types of objections. Seven very popular objections seem to be:

1. We don’t need it - it has no economic value!
2. We don’t have time for it!
3. It is not safe!
4. We have it already!
5. We don’t work like that here
6. It does not solve the problem
7. It’s a fad - I just don’t like it!

One big innovation in recent years, both in business and in public administration, is open data. And like many other innovations, the seven arguments listed above are all part of the debate on open data as well. The remainder of this article will describe and address these seven objections regarding open data one by one and hopefully help us become less sceptical about this new frontier in information management.

The big challenge of innovation, it seems, is not to come up with a new idea, but to gather sufficient evidence for its utility and garner broad support from relevant stakeholders.

1. We don’t need it: In the field of open data there are at least three versions of the “We don’t need it” argument. The first is “we don’t need data”. Admittedly, there are not many people who seriously hold this position, though many organisations still seem to work on this premise and don’t establish a clear strategy for collecting and using data. Many people may subscribe to the value of data but don’t act on it. For the real data sceptics, the video The Joy of Stats by Hans Rosling is a very entertaining way to appreciate its value.

The second version of this argument is “We don’t need open data”. Given the popularity of buzz words like “knowledge economy” and “information society”, this argument may come as a surprise. But again, many administrations and organisations still work on the premise that open data is not a priority for them. One of the organisations that does see the value of open data is the European Commission. Since its 2003 Directive on the Reuse of Public Sector Information, the EC has continued to build an evidence base and policy framework to promote the use of open data for economic growth and public accountability. In August 2011 the EC published a comprehensive report on the economic value of open data. According to this study, the economic gains from open public data amount to 40 billion euros per year in the European Union. Following this assessment of the economic importance of open data, the European Commission published its own open data strategy.

Today many companies across Europe already offer products based on open data. And countries like the UK and Denmark have published assessments of the value and use of open data. In Germany alone, the market for geo-information was estimated at 1.4 billion euros in 2007 and in the Netherlands the same sector accounted for 15,000 full-time jobs in 2008 [1]. It is true that more research and experience is necessary to identify areas and sectors where open data is most valuable and how its value can be increased. But to claim that there is no need for open data is simply wrong.

The third version of this argument is: “They don’t need open data”. “They” in this case refers to regular citizens who are not familiar with raw data or, in the context of international development for example, people in poor countries who surely have other priorities and lack the skills to analyse data, such as from the national budget. It is, of course, true that most people are not very comfortable with raw data. But sites like data.gov.uk are already making an effort to become more accessible to non-technical users. What is more important, however, is that most citizens will benefit from raw data through intermediaries. Already now there are countless examples in which organisations and individuals have used raw data and rendered it useful for the public: civic hackers use open transport data to create apps that allow people to optimise their use of public transport, organisations like the Open Knowledge Foundation build visualisations of national budgets and help citizens understand how their taxes are used. Other initiatives take hospital data, geographical data and national health data to increase transparency and improve service delivery in the health sector or create online tools for election monitoring [2]. These projects are happening in London, New York, Stockholm and Berlin, but also in Mumbai, Nairobi, Dakar and Lagos. Obviously apps and online tools are used more in big cities, but numerous organisations specialising in mobile technology are already demonstrating how the urban bias can be overcome and how rural and poor people can be included in information and feedback loops [3].

“The wireless music box has no imaginable commercial value. Who would pay for a message sent to nobody in particular?” – David Sarnoff’s associates in response to his urgings for investment in the radio in the 1920s.

If you stop running your business, you die today. If you don’t innovate you will be dead tomorrow.

2. We don’t have time for it: Given the evidence of the economic value of open data, the argument regarding lack of time is like saying we don’t have time for innovation, increased efficiency and growth. Nevertheless, lack of time seems to be one of the major arguments against open data strategies within organisations. Lack of time is probably one of the main arguments against any organisational change initiative. There is never a perfect moment to lay aside the work that needs doing on a daily basis to improve internal processes. Instead, those responsible for innovation need to find a balance between keeping the system functional on the one hand and improving the system on the other. Both elements, the smooth running of daily activities and improvement to the system, are vital. If you stop running your business, you die today. If you don’t innovate you will be dead tomorrow.

3. It is not safe: Professor Gunter Dück has been Chief Technology Officer at IBM in Germany for many years and one of his core responsibilities was convincing colleagues and partners of the potential of new technology. In his experience for every innovation about 10% are pioneers and early supporters and about 40% have open minds and would support the innovation if all practical problems were solved. But there are also at least 40% who have closed minds and just don’t want the innovation, who don’t want to use telephones, or computers, smartphones, cloud computing or Facebook. And the favourite argument from this group against an innovation is, according to Dück, “it is not safe”. Unfortunately this pretext against innovation seems very credible from the outside, even if it is not backed-up by sound evidence and real security risks are often highly overestimated. The two most common versions of security concerns related to open data are, in my experience, the fear of data being misused and the fear of data errors and negative publicity.

Open data is not illegal hacking or data theft.

Open data is not Wikileaks: The fear of data misuse is very well illustrated by the fact that many people I have talked to in the last two years somehow associate the topic of open data with Wikileaks and the fear of the uncontrolled release of confidential data. In this respect the only way to counter this fear is by raising awareness about open data: open data is not illegal hacking or data theft. Open data is the planned and conscious release of structured, machine-readable and openly licensed data by the data holders themselves. Obviously, it is possible that the publication of open data, e.g. by public administrations, might expose the misallocation of funds or bad policy decisions. But who would seriously advocate for bad decisions to remain secret and for misallocations to continue? Open data may make some people more uncomfortable, but it will probably make the majority of citizens and taxpayers more comfortable.

Open data reduces the risk of misuse: Yes, people can take data that is openly available and misrepresent it. But this is not a new risk. If the German Ministry of Finance publishes a pdf document with national budget data, anybody can use and misquote this data. In fact, anybody can produce a data visualisation with totally fake data and cite the Ministry of Finance as a source. Misuse cannot be curbed by unhelpful formats or restrictive licenses. In the world of 2.0 the misrepresentation of data is most likely to be corrected and curbed by the crowd - by the owner of the data and other people pointing out that the data presented is wrong. In his book “Here comes everybody” Clay Shirky often cites the corrective power of the crowd in Wikipedia: if you write an article in the online encyclopaedia with faulty information, it usually takes only days, if not hours, before the errors are corrected by the crowd. As long as data is not available in open and machine-readable formats, it is much harder to identify such misuse because it takes so much effort to re-calculate and verify the data.

Open data increases data quality: The experience of data publishers so far suggests that open data increases data quality. For example the UNDP’s Multi-Donor Trust Fund began releasing its transaction data in real time more than two years ago. Prior to the launch of the new website, many staff members were very concerned about data errors and negative publicity. The experience so far has been that more eyes on the data have helped to improve the data quality and the public response has been overwhelmingly positive. The British Department for International Development (DFID) has had a very similar experience with the publication of their data in the open data standard of the International Aid Transparency Initiative (IATI). Additionally, in the case of the DFID, the main beneficiary of this improved data has been DFID staff itself, who have a better overview of the activities of different departments.

Promoting a beta culture: While open data has the potential to increase data quality, increased transparency might still lead to negative publicity in the media. Partly it is in the nature of transparency and open data efforts to highlight shortcomings in public policy or unjustified expenditures and thus improve the situation. However there is also a need to promote a “beta culture” in administrations and policy, where continuous learning and improvement is the standard and mistakes and wrong decisions are part of a learning process. Promoting a beta culture probably also requires a changed attitude from the media. It takes media representatives who scrutinize public administrations critically without exploiting shortcomings for sensational and unwarranted coverage.

Open data is not only about the transparency of one organisation. It is about the transparency of the whole system.

4. We have it already: One objection to open data is that open data is not necessary because a given administration or organisation is already very transparent. This argument is very common in development organisations. Transparency has been on the agenda of these organisations for some time and many aid agencies have started to publish more comprehensive annual reports or project information on their websites. The problem is that there are thousands of development organisations in both donor and recipient countries. Who has the time and the resources to click through all these websites to find out who is doing what and who is related to whom? Open data is not only about the transparency of one organisation. It is about the transparency of the whole system. Without data that is machine-readable, standardised and legally open, data that can be disentangled and combined with new data sources, it is simply impossible to achieve the transparency of the aid system as a whole.

5. We don’t work like that here! One argument against open data is that it will change how things are done within an organisation. In a way, this argument is valid, because the whole purpose of open data is to change processes. And the changes will not be the same for everybody. Some will welcome them. Others will object. Within an administration for example, the IT department may welcome an open data strategy, because it changes its internal role from a rather unimportant service provider to a key player with policy relevance. Additionally many IT people favour the open data movement and are interested in data applications. Other departments may worry about their role losing its importance, such as internal units for statistical analysis. Yet other departments may worry about some of their decisions being subjected to public scrutiny. So, yes, open data can change how things are done within organisations - that’s the point!

6. This does not solve the problem: Open data is not enough to establish accountability, ensure efficiency and curb corruption. Many critics of the open data movement have repeated this criticism time and time again. Of course open data is not enough - nobody ever claimed it was! Fair elections, rule of law and a good education system are not enough either, but that’s no reason to abolish fair elections, rule of law and a good education system. Open data is but one prerequisite for a more transparent, efficient and accountable society. How can citizens hold governments to account for spending decisions if they don’t know the national budget? How can we curb the influence of business lobbying in politics if we don’t know who finances whom? How can we increase coverage of social services if we don’t have an overview of e.g. health service providers in a given area? Openness is not enough, but that’s not an excuse to reject openness.

7. It’s a fad: This is the ultimate argument against any innovation: “It’s a fad and I don’t believe in it.” But open data is not a religion. Nobody has to believe in it. Open data is already happening in different countries, at different administrative levels, in different sectors and in different types of organisations. And with the growing pool of experience with open data, there is also a growing body of evidence of how open data can shape our societies and our economies.
   Instead of fighting innovation with short-sighted arguments, we now need to further explore the evidence on what works and what doesn’t. The UK is not only pioneering the release of open data, but also research into open data. For this purpose the UK government has decided to create the UK Open Data Institute to promote research on open data. It is ok to be sceptical about open data, but it is not ok to disregard the evidence and remain ignorant.

Footnotes:
[1] http://ec.europa.eu/information_society/policy/psi/docs/pdfs/directive_proposal/2012/open_data.pdf
[2] Some examples: http://omvard.se/; http://samabaat.com/temoignage/
[3] For example http://ureport.ug/ uses mobile technology to allow feedback from rural and poor people to government services in Uganda and http://www.aidmonitor.org.np/ is working to involve poor communities in rural Nepal in the aid effectiveness discourse.