September 27, 2016

Is Estonia afraid of Open Data?

Only days after Estonian Open Knowledge activists organized series of discussions about status and challenges of Open Data reforms in Estonia, where at the house of Chancellor of Justice of Estonia all interested parties were invited to participate in an open discussion, [1] an opportunity that none of the current government officials responsible for the digital issues decided to use, we find them preaching the success of Open Data reforms in Washington.

Although Estonia developed its digital infrastructure rapidly in late 90s and early 00s, the development has mainly served reducing administrative overload for the government and not so much empowering citizens and democratic participation. [2] So for an insider this is no surprise, that Open Data law in Estonia was ratified in 2012 only after European Commission launched breach proceedings for Estonia not following re-use of public sector information directive from 2003.

In making the data collected by the state accessible as Open Data, Estonia has been “modest” and data published is “mostly not shared or re-used” as National Audit Office diagnosed the situation in 2014. [3] Audit office also comments that from “accessibility, use and socioeconomic impact” point of view Estonia was 14th in Open Data Barometer index in 2013 with score of 49.45.

The indicators of Open Data Barometer change year by year although in reality the process is quite still

Presumably due to some charitable expert reviews Estonia suddenly rose to 60.18 in 2014 in Open Data Barometer only to fall back to previous level of 50.63 in 2015 which is 24th in overall ranking. [4] An insider might argue that even this position might be too optimistic, for example one can hardly see any basis for improved rating for “Citizen and civil rights” indicator which has moved up to 95/100 for 2015.

That things have not improved for the end of 2016 was also the general spirit of the debates we had at the house of Chancellor of Justice with couple of dozens of activists, entrepreneurs and legal experts. The recurring motive in presentations and discussion was striking incoherence of laws, policy and real practise.

This might have some short term advantages, for example not too eagerly enforcing laws that are relics of information policies from the times before Internet might be actually good for society, but in the long run civil society also needs that laws that are meant to commit to the free and open information society are properly enforced. Sadly this is not the case for Open Data paragraphs of Public Information Act. [5]

Open Data additions were adopted in December 2012 and were supposed to take full effect from January 2015, however the government action plan from beginning of 2014 already failed to encapsulate this deadline and the deadline was moved further to 1st of February 2016 only in late December 2015, when the the law had been already in effect for one full year.

“Modest” progress in Open Data is arguably only a symptom of narrowly focused development of information society in Estonia, which has for many years failed to accommodate international criticism on e-voting by OSCE/ODIHR election observers [6] or independent research groups [7] or take seriously the recommendations of privacy advocates [8]. Escapist tendencies have earned Estonia a title of e-Narnia: place where miraculous things happen, which are neither explainable for outsiders nor scalable for use in other countries. [9]

The problem with Open Data in Estonia is not the laws, which are pretty decent and most of the government institutions and municipalities could actually be sued for at least not providing Open Data for all the information systems built after January 2015. The problem is more about the public awareness and that even the people being responsible for setting course for our digital infrastructure don’t minimally understand the concept of Open Data.*

One of the common misconceptions is to think that X-Road, our government information exchange system [10] could somehow be a substitute for Open Data. To present such claims is to ignore the core meaning of Open Data, which by definition means data published and accessible to everybody for whatever use without asking permission. X-Road is an opposite of that and is neither accessible for laymen nor for goverment officials and even academia has to struggle to get access to data they need for research.

Open Definition has become a default framework for the concept of Open Data [11], which consists of two parts: technical openness and legal openness. Technical openness is about not using proprietary file formats or CAPTCHAs to make it complicated or costly to use the data. Data should be linked and machine-readable with semantic metadata in order to be Open Data of best quality standard, but just avoiding common technical restrictions is good enough for a start. Legal openness is about right to get the data, use it, share it and build on it without making any restrictions. And not only for non-commercial use, Open Data should be also available for business and innovation without any legal restrictions.

The same definition is clearly echoes in Estonian Public Information Act as well as Open Data implementation guides by Ministry of Economic Affairs and Communications, however Estonian law adds some extra requirements, for example that Open Data has to be accessible through “Estonian information gateway” created specially for that purpose. The gateway currently publishes humble number of 53 datasets. [12]

In the background of this, Siim Sikkut, government strategy unit leader for information policy issues without any hesitation endorses X-Road and proposes that we should “reconsider how we do open data” and “not be stuck in current definitions” [13], but fails to show up to discussion about future of Open Data at the house of Chancellor of Justice with the founder and current president of Open Knowledge International, when opportunity is offered.

Better doesn’t do also deputy secretary general for state information systems Taavi Kotka, who is leading the same domain on the part of Ministry of Economic Affairs and Communications. Mr Kotka is supposed to present a keynote tomorrow September 28th at a conference organized by White House titled – “Estonia Runs on Open Data”. [14]

It’s not hard to imagine, what Mr Kotka will be stressing on, but the reality of Estonia running on Open Data yet remains to be seen. However, looking at the history of past few years one could rather describe the process as Estonia running from Open Data.

X-Road is an information exchange system for government and has in itself hardly anything to do with concept of Open Data.

Replace X-Road with another information storage and exchange technology, for example MySQL, and think if that makes sense to say: since we use MySQL, this is our “different way” [15] of providing Open Data. Actually rhetorically MySQL would function even a bit better, since this is at least Open Source. Although this doesn’t help to get nearer to Open Data.

X-Road as governmental information exchange layer is by default closed even for government institutions themselves and is used mainly as technical backbone for Estonian e-services. Although X-Road might provide us with zero bureaucracy for e-services [16], it will be also full dystopian technocracy without openness, digital rights and control over the use of my own personal data built into heart of it.

This is only one example of the many reasons why we need open and international debate about information society in Estonia which I hope organizations like Open Knowledge International would very much help to conduct.

* Although this cannot be said about officials working at the ministries, who have written very reasonable and thorough guides how to implement Open Data. However, despite the efforts currently no applications [17] have been filed for 5 million Euros Ministry of Economic Affairs and Communications is giving for providing Open Data starting from May 2016. [18]

[3] pg 54 ff
[5] §§ 3¹, 8, 29, 58²
[16] cf

This is an article I didn’t publish, because my colleagues from Open Knowledge Estonia considered it too critical at that moment.