Dutch Provinces publish open data, but it always looks like it is mostly geo-data, and hardly anything else. When talking to provinces I also get the feeling they struggle to think of data that isn’t of a geographic nature. That isn’t very surprising, a lot of the public tasks carried out by provinces have to do with spatial planning, nature and environment, and geographic data is a key tool for them. But now that we are aiding several provinces with extending their data provision, I wanted to find out in more detail.
My colleague Niene took the API of the Dutch national open data portal for a spin, and made a list of all datasets listed as stemming from a province.
I took that list and zoomed in on various aspects.
At first glance there are strong differences between the provinces: some publish a lot, others hardly anything. The Province of Utrecht publishes everything twice to the national data portal, once through the national geo-register, once through their own dataplatform. The graph below has been corrected for it.
What explains those differences? And what is the nature of the published datasets?
Geo-data is dominant
First I made a distinction between data that stems from the national geo-register to which all provinces publish, and data that stems from another source (either regional dataplatforms, or for instance direct publication through the national open data portal). The NGR is theoretically the place where all provinces share geo-data with other government entities, part of which is then marked as publicly available. In practice the numbers suggest Provinces roughly publish to the NGR in the same proportions as the graph above (meaning that of what they publish in the NGR they mark about the same percentage as open data)
Of the over 3000 datasets that are published by provinces as open data in the national open data portal, only 48 don’t come from the national geo-register. This is about 1.5%.
Of the 12 provinces, 4 do not publish anything outside the NGR: Noord-Brabant, Zeeland, Flevoland, Overijssel.
Drenthe stands out in terms of numbers of geo-data sets published, over 900. A closer look at their list shows that they publish more historic data, and that they seem to be more complete (more of what they share in the NGR is marked for open data apparantly.) The average is between 200-300, with provinces like Zuid-Holland, Noord-Holland, Gelderland, Utrecht, Groningen, and Fryslan in that range. Overijssel, like Drenthe publishes more, though less than Drenthe at about 500. This seems to be the result of a direct connection to the NGR from their regional geo-portal, and thus publishing by default. Overijssel deliberately does not publish historic data explaining some of the difference with Drenthe. (When something is updated in Overijssel the previous version is automatically removed. This clashes with open data good practice, but is currently hard to fix in their processes.)
If it isn’t geo, it hardly exists
Of the mere 48 data sets outside the NGR, just 22 (46%) are not geo-related. Overall this means that less than 1% of all open data provinces publish is not geo-data.
Of those 22, exactly half are published by Zuid-Holland alone. They for instance publish several photo-archives, a subsidy register, politician’s expenses, and formal decisions.
Fryslan is the only province publishing an inventory of their data holdings, which is 1 of their only 3 non geo-data sets.
Gelderland stands out as the single province that publishes all their geo data through the NGR, hinting at a neatly organised process. Their non-NGR open data is also all non-geo (as it should be). They publish 27% of all open non-geo data by provinces, together with Zuid-Holland account for 77% of it all.
Taking these numbers and comparing them to inventories like the one Fryslan publishes (which we made for them in 2016), and the one for Noord-Holland (which we did in 2013), the dominance of geo-data is not surprising in itself. Roughly 80% of data provinces hold is geo related. Just about a fifth to a quarter of this geo-data (15%-20% of the total) is on average published at the moment, yet it makes up over 99% of all provincial open data published. This lopsidedness means that hardly anything on the inner workings of a province, the effectivity of policy implementation etc. is available as open data.
Where the opportunities are
To improve both on the volume and on the breadth of scope of the data provinces publish, two courses of action stand open.
First, extending the availability of geo-data provinces hold. Most provinces will have a clear process for this, and it should therefore be relatively easy to do. It should therefore be possible for most provinces to get to where Drenthe currently is.
Second, take a much closer look at the in-house data that is not geo-related. About 20% of dataholdings fall in this category, and based on the inventories we did, some 90% of that should be publishable, maybe after some aggregation or other adaptations.
The lack of an inventory is an obstacle here, but existing inventories should at least be able to point the other provinces in the right direction.
Make the provision of provincial open geodata complete, embrace its dominance and automate it with proper data governance. Focus your energy on publishing ‘the rest’ where all the data on the inner workings of the province is. Provinces perpetually complain nobody is aware of what they are doing and their role in Dutch governance. Make it visible, publish your data. Stop making yourself invisible behind a stack of maps only.
The quality of information households in local governments is often lacking.
Things like security, openness and privacy are safeguarded by putting separate fences for each around the organisation, but those safeguards lack having detailed insight into data structures and effective corresponding processes. As archiving, security, openness and privacy in a digitised environment are basically inseparable, doing ‘everything by design’ is the only option. The only effective way is doing everything at the level of the data itself. Fences are inefficient, ineffective, and the GDPR due to its obligations will show how the privacy fence fails, forcing organisations to act. Only doing data governance for privacy is senseless, doing it also for openness, security and archiving at the same time is logical. Having good detailed inventories of your data holdings is a useful instrument to start asking the hard questions, and have meaningful conversations. It additionally allows local government to deploy open or shared data as policy instrument, and releasing the inventory itself will help articulate civic demand for data. We’ve done a range of these inventories with local government.
1: High time for mature data governance in local and regional government
Digitisation changes how we look at things like openness, privacy, security and archiving, as it creates new affordances now that the content and its medium have become decoupled. It creates new forms of usage, and new needs to manage those. As a result of that e.g. archivists find they now need to be involved at the very start of digital information processes, whereas earlier their work would basically start when the boxes of papers were delivered to them.
The reality is that local and regional governments have barely begun to fully embrace and leverage the affordances that digitisation provides them with. It shows in how most of them deal with information security, openness and privacy: by building three fences.
Security is mostly interpreted as keeping other people out, so a fence is put between the organisation and the outside world. Inside it nothing much is changed. Similarly a second fence is put in place for determining openness. What is open can reach the outside world, and the fence is there to do the filtering. Finally privacy is also dealt with by a fence, either around the entire organisation or a specific system, keeping unwanted eyes out. All fences are a barrier between outside and in, and within the organisation usually no further measures are taken. All three fences exist separately from each other, as stand alone fixes for their singular purpose.
The first fence: security
In the Netherlands for local governments a ‘baseline information security’ standard applies, and it determines what information should be regarded as business critical. Something is business critical if its downtime will stop public service delivery, or of its lack of quality has immediate negative consequences for decision making (e.g. decisions on benefits impacting citizens). Uptime and downtime are mostly about IT infrastructure, dependencies and service level agreements, and those fit the fence tactic quite well. Quality in the context of security is about ensuring data is tamper free, doing audits, input checks, and knowing sources. That requires a data-centric approach, and it doesn’t fit the fence-around-the-organisation tactic.
The second fence: openness
Openness of local government information is mostly at request, or at best as a process separate from regular operational routines. Yet the stated end game is that everything should be actively open by design, meaning everything that can be made public will be published the moment it is publishable. We also see that open data is becoming infrastructure in some domains. The implementation of the digitisation of the law on public spaces, requires all involved stakeholders to have the same (access to) information. Many public sector bodies, both local ones and central ones like the cadastral office, have concluded that doing that through open data is the most viable way. For both the desired end game and using open data as infrastructure the fence tactic is however very inefficient.
At the same time the data sovereignty of local governments is under threat. They increasingly collaborate in networks or outsource part of their processes. In most contracts there is no attention paid to data, other than in generic terms in the general procurement conditions. We’ve come across a variety of examples where this results 1) in governments not being able to provide data to citizens, even though by law they should be able to 2) governments not being able to access their own data, only resulting graphs and reports, or 3) the slowest partner in a network determining the speed of disclosure. In short, the fence tactic is also ineffective. A more data-centric approach is needed.
The third fence: personal data protection
Mostly privacy is being dealt with by identifying privacy sensitive material (but not what, where and when), and locking it down by putting up the third fence. The new EU privacy regulations GDPR, which will be enforced from May this year, is seen as a source of uncertainty by local governments. It is also responded to in the accustomed way: reinforcing the fence, by making a ‘better’ list of what personal data is used within the organisation but still not paying much attention to processes, nor the shape and form of the personal data.
However in the case of the GDPR, if it indeed will be really enforced, this will not be enough.
GDPR an opportunity for ‘everything by design’
The GDPR confers rights to the people described by data, like the right to review, to portability, and to be forgotten. It also demands compliance is done ‘by design’, and ‘state of the art’. This can only be done by design if you are able to turn the rights of the GDPR into queries on your data, and have (automated) processes in place to deal with requests. It cannot be done with a ‘better’ fence. In the case of the GDPR, the first data related law that takes the affordances of digitisation as a given, the fence tactic is set to fail spectacularly. This makes the GDPR a great opportunity to move to a data focus not just for privacy by design, but to do openness, archiving and information security (in terms of quality) by design at the same time, as they are converging aspects of the same thing and can no longer be meaningfully separated. Detailed knowledge about your data structures then is needed.
Local governments inadvertently admit fence-tactic is failing
Governments already clearly yet indirectly admit that the fences don’t really work as tactic.
Local governments have been loudly complaining for years about the feared costs of compliance, concerning both openness and privacy. Drilling down into those complaints reveals that the feared costs concern the time and effort involved in e.g. dealing with requests. Because there’s only a fence, and usually no processes or detailed knowledge of the data they hold, every request becomes an expedition for answers. If local governments had detailed insight in the data structures, data content, and systems in use, the cost of compliance would be zero or at least indistinguishable from the rest of operations. Dealing with a request would be nothing more than running a query against their systems.
Complaints about compliance costs are essentially an admission that governments do not have their house in order when it comes to data.
The interviews I did with various stakeholders as part of the evaluation of the PSI Directive confirm this: the biggest obstacle stakeholders perceive to being more open and to realising impact with open data is the low quality of information systems and processes. It blocks fully leveraging the affordances digitisation brings.
Towards mature data governance, by making inventory
Changing tactics, doing away with the three fences, and focusing on having detailed knowledge of their data is needed. Combining what now are separate and disconnected activities (information security, openness, archiving and personal data protection), into ‘everything by design’. Basically it means turning all you know about your data into metadata that becomes part of your data. So that it will be easy to see which parts of a specific data set contain what type of person related data, which data fields are public, which subset is business critical, the records that have third party rights attached, or which records need to be deleted after a specific amount of time. Don’t man the fences where every check is always extra work, but let the data be able to tell exactly what is or is(n’t) possible, allowed, meant or needed. Getting there starts with making an inventory of what data a local or regional government currently holds, and describing the data in detailed operational, legal and technological terms.
Mature digital data governance: all aspects about the data are part of the data, allowing all processes and decisions access to all relevant material in determining what’s possible.
2: Ways local government data inventories are useful
Inventories are a key first step in doing away with the ineffective fences and towards mature data governance. Inventories are also useful as an instrument for several other purposes.
Local is where you are, but not the data pro’s
There’s a clear reason why local governments don’t have their house in order when it comes to data.
Most of our lives are local. The streets we live on, the shopping center we frequent, the schools we attend, the spaces we park in, the quality of life in our neighbourhood, the parks we walk our dogs in, the public transport we use for our commutes. All those acts are local.
Local governments have a wide variety of tasks, reflecting the variety of our acts. They hold a corresponding variety of data, connected to all those different tasks. Yet local governments are not data professionals. Unlike singular-task, data heavy national government bodies, like the Cadastre, the Meteo institute or the department for motor vehicles, local governments usually don’t have the capacity or capability. As a result local governments mostly don’t know their own data, and don’t have established effective processes that build on that data knowledge. Inventories are a first step. Inventories point to where contracts, procurement and collaboration leads to loss of needed data sovereignty. Inventories also allow determining what, from a technology perspective, is a smooth transition path to the actively open by design end-game local governments envision.
Open data as a policy instrument
Where local governments want to use the data they have as a way to enable others to act differently or in support of policy goals, they need to know in detail which data they hold and what can be done with it. Using open data as policy instrument means creating new connections between stakeholders around a policy issue, by putting the data into play. To be able to see which data could be published to engage certain stakeholders it takes knowing what you have, what it contains, and in which shape you have it first.
Better articulated citizen demands for data
Making public a list of what you have is also important here, as it invites new demand for your data. It allows people to be aware of what data exists, and contemplate if they have a use case for it. If a data set hasn’t been published yet, its existence is discoverable, so they can request it. It also enables local government to extend the data they publish based on actual demand, not assumed demand or blindly. This increases the likelihood data will be used, and increases the socio-economic impact.
More and more new data is emerging, from sensor networks in public and private spaces. This way new stakeholders and citizens are becoming agents in the public space, where they meet up with local governments. New relationships, and new choices result. For instance the sensor in my garden measuring temperature and humidity is part of the citizen-initiated Measure your city network, but also an element in the local governments climate change adaptation policies. For local governments as regulators, as guardian of public space, as data collector, and as source of transparency, this is a rebalancing of their position. It again takes knowing what data you own and how it relates to and complements what others collect and own. Only then is a local government able to weave a network with those stakeholders that connects data into valuable agency for all involved. (We’ve built a guidance tool, in Dutch, for the role of local government with regard to sensors in public spaces)
Having detailed data inventories are a way to start having the right conversations for local governments on all these points.
3: Getting to inventories
To create useful and detailed inventories, as I and my colleagues did for half a dozen local governments, some elements are key in my view. We looked at structured data collections only, so disregarded the thousands of individual once-off spreadsheets. They are not irrelevant, but obscure the wood for the trees. Then we scored all those data sets on up to 80(!) different facets, concerning policy domain, internal usage, current availability, technical details, legal aspects, and concerns etc. A key element in doing that is not making any assumptions:
don’t assume your list of applications will tell you what data you have. Not all your listed apps will be used, others won’t be on the list, and none of it tells you in detail what data actually is processed in them, just a generic pointer
don’t assume information management knows it all, as shadow information processes will exist outside of their view
don’t assume people know when you ask them how they do their work, as their description and rationalisation of their acts will not match up with reality,
let them also show you
don’t assume people know the details of the data they work with, sit down with them and look at it together
don’t assume what it says on the tin is correct, as you’ll find things that don’t belong there (we’ve e.g. found domestic abuse data in a data set on litter in public spaces)
Doing an inventory well means
diving deeply into which applications are actually used,
talking to every unit in the organisation about their actual work and seeing it being done,
looking closely at data structures and real data content,
looking closely at current metadata and its quality
separately looking at large projects and programs as they tend to have their own information systems,
going through external communications as it may refer to internally held data not listed elsewhere,
looking at (procurement and collaboration) contracts to determine what claims other might have on data,
and then cross-referencing it all, and bringing it together in one giant list, scored on up to 80 facets.
Another essential part, especially to ensure the resulting inventory will be used as an instrument, is from the start ensuring the involvement and buy-in of the various parts of local government that usually are islands (IT, IM, legal, policy departments, archivists, domain experts, data experts). So that the inventory is something used to ask a variety of detailed questions of.
We’ve followed various paths to do inventories, sometimes on our own as external team, sometimes in close cooperation with a client team, sometimes a guide for a client team while their operational colleagues do the actual work. All three yield very useful results but there’s a balance to strike between consistency and accuracy, the amount of feasible buy-in, and the way the hand-over is planned, so that the inventory becomes an instrument in future data-discussions.
What comes out as raw numbers is itself often counter-intuitive to local government. Some 98% of data typically held by Dutch Provinces can be public, although usually some 20% is made public (15% open data, usually geo-data). At local level the numbers are a bit different, as local governments hold much more person related data (concerning social benefits for instance, chronic care, and the persons register). About 67% of local data could be public, but only some 5% usually is. This means there’s still a huge gap between what can be open, and what is actually open. That gap is basically invisible if a local government deploys the three fences, and as a consequence they run on assumptions and overestimate the amount that needs the heaviest protection. The gap becomes visible from looking in-depth at data on all pertinent aspects by doing an inventory.
(Interested in doing an inventory of the data your organisations holds? Do get in touch.)
As we are living in a networked world, increasingly government bodies execute their tasks while collaborating in networks of various other stakeholders. This also happens when it comes to collecting, providing or working with data as part of public tasks. One of the potential detrimental side effects is that it quickly becomes unclear who can decide to open such data up. Or whether a government entity, who wants to publish data as part of a policy intervention, still feels able to do so. This ability to decide over your own data, I call data sovereignty. I think without proper attention, the data sovereignty of public institutions is under pressure in collaborative situations and a threat to the freedom of public entities to decide and act on their own open data efforts. This is especially problematic where the lack of data sovereignty hinders public entities in deploying open data as a policy instrument.
I have just completed an inventory of the data sets that a Dutch province holds and the visible erosion of data sovereignty was the main unexpected outcome for me.
This erosion takes different shapes. Here are a few examples of it, encountered in the Province I mentioned:
Data collection on businesses locations and the number of people they employ (to track employment per municipality per sector) is being pooled by all provinces (as a national level data set is more useful). The pooling takes place in a separate legal entity. It is unclear if this entity still falls under FOIA and re-use regulations. This entity also exploits the data by selling it. Logical at the organisational level perhaps, but illogical in comparison with the provincial public task (and maybe not even legal under the Re-Use law). Opening up the data needs to be done through that new entity, meaning not just convincing yourself, but all other provinces as well as the entity who has commercial interest in not being convinced. The slowest will thus set the speed.
Data collection on traffic flows, collected by the Province, is stored directly in a national data warehouse (NDW). Again pooling data makes it more useful, but the Province cannot store cleaned data there (anomalies filtered out, pattern changes explained etc.), so always needs to redo that cleaning and filtering whenever they want to work or access their own data. Although the publicly owned NDW now publishes open data, until recently they saw themselves as a commercial outfit, adverse to the notion of open data.
Data collection on bicycle traffic, done by the Province, is stored in the online database of a French service provider active in the entire EU. Ownership of the data is unclear. The Province only accesses the data through the French website. If a FOIA request came, it would be unclear if providing the data runs counter to any rights the service provider is claiming.
Data collection on the prevalence of bird species is being collected in collaboration with nature preservation groups and large numbers of volunteers. The Province pays for the data collection, but the nature preservation groups claim their volunteers (by virtue of their voluntary efforts) are the rightful owners of the data. Without seeking internal legal advice, the discussion remains unsolved and stalls.
None of these situations are unsolvable, all of them can get a definitive answer. The issue however is that nobody is clearly in a position, or has the explicit role to make sure such an definitive answer gets formulated. Because of that, uncertainties remain, which easily leads to inaction. If and when the Province wants to act to open data up, it therefore easily runs into all kinds of questions that will slow action down, or ensure action does not get taken.
It is entirely logical that public entities are collaborating in networks with other public entities and domain-specific stakeholders for the collection, dissemination and use of data. It is also certain, given our networked society and the drive for efficiency, the number of situations where such collaboration takes place will only rise. However, for the drive towards more openness it is detrimental when ownership of public data becomes unclear, gets transferred to an entity that potentially falls outside the scope of FOIA, or falls under the rights of a private entity, just because nobody sought to clarify such matters at the outset.
Public entities should learn to strongly guard their data sovereignty if they want to maintain their own agency in using opening up data as a policy instrument. Moving to open by design as a default for the public sector, requires stopping the erosion of data sovereignty.