Open data

By Tessa Marano on 3 March, 2015

Since 2013, Australian governments at federal, state and territory levels have talked about the benefits of open data. Each government is taking steps to open up more and more government data to the public. The common rhetoric is that making this data readily available will lead to innovation and build the digital economy in Australia. I agree.

PSMA Australia is 100% supportive of maximising access to data for the benefit of the economy.

In fact, this was the catalyst that led the governments of Australia to conceive our organisation in 1992, and ultimately led to the formation of PSMA as a company in 2001.

From this perspective, PSMA would like to see the conversation move beyond the ‘free or fee’ argument to a deeper discussion focused on what needs to be done for an open data framework to deliver the benefits espoused in this rhetoric – benefits we all seek.

Not all data is the same

The concept of open data is a great conversation starter. It gets people from all industries talking about data.

But data isn’t homogeneous. And since it’s not, why would we expect the same approach to its release, to work equally well for all data?

Some data (and it’s common across most types of location data) continually change; sometimes slowly, sometimes much more quickly. Think of your in-car navigation device. How many times have you gone searching for a new address that’s not in your car’s system? The real world is constantly changing, so the data that represents it must change equally quickly so that you don’t end up driving the wrong way down a one-way street in the Sydney CBD.

But at the other end of the scale, if you look at something like financial reports, that data relates to a specific point in time. This means that once it is published, it stands alone and does not change. Think of this as static data. The value of static data is the snapshot it provides on that point in time.
Even when we narrow down to location data, we may not even be talking about the same kind of thing. After all, location data encompasses infrastructure, properties, addresses, facilities, demography, natural resources, social services and public amenities such as public transport infrastructure and the road network.

A common phrase used about location data is ‘to unlock value’. However, dynamic data differs from static data in many ways, and those differences mean that not all data can be treated the same when it comes to maximising the benefits of open data initiatives.

The aim of open data

At Locate14 Conference, Minister Malcolm Turnbull said :

“To extract the most value from data held by government, we need to make it readily available to the private sector and citizens to make it truly open.

“We are committed to regularly publishing the data we hold and that we publish much more of it in machine-readable form.”

PSMA supports the minister’s enthusiasm for open data including location data, but before we try to catch up to the 200,000 datasets released by the UK government and mentioned in this speech as the government’s target, let’s get some things clear.

The idea of open data broadly recognises the attainment of three benefits through its adoption: transparency, social and commercial value, and participatory governance (open government). I enthusiastically support these three pillars of a strong democracy. However, I think that it is fair to say that the data that assists with achieving transparency of government, participatory governance, and even social value, is rarely the same data that achieves commercial value. For this reason, we shouldn’t make the assumption that all data published under open data policies will generate commercial value. Consequently, a high number of datasets accessible in a machine-readable form need not be the best measure of the achievement of commercial value, although it may well be a reasonable measure for the other objectives of open data.

Targeted approach required to achieve commercial value

To realise the desired outcomes of industry innovation, reduced costs and improved service delivery – the commercial value objective – we certainly do need to make more data more accessible. However, it is not realistic to assume that simply releasing data freely to the public will automatically deliver commercial value. It is a very important step, but it is not a complete solution to maximising the value, especially for the class of data described above as dynamic data.

Maximising access to data needs to be about providing the type, quality and format of data at a price point that meets the needs of the broadest section of the market. These are the true barriers to uptake.

The open data policy framework needs to expand and deepen the open data policies to appropriately deal with all the barriers to achieving commercial value, rather than simply maximising the number of government datasets released to the public.

Quality and value comes at a cost

Minister Turnbull also acknowledged that the “value of data, of location information, is latent. It requires innovative analysis to realise the benefits”.This is an insightful comment but again, we need to go deeper into the issue.

The main source of raw location data in Australia is state and territory governments. These governments have a major responsibility for land administration and the legislation and policy associated with this role drive much of the data capture and management requirements. Consequently, they each have their own standards and formats for raw data, which almost invariably require further work to achieve uniformity and consistency at the national level. The national datasets developed in this way and available to the market today are collected, standardised, enriched and delivered at a considerable cost.

So here is an interesting policy question: if data requires further processing before it can be used to gain benefits in the way described by the minister, who should do that work?

Clearly, not everyone who uses the data should do this work. It would make much greater economic sense to do this work once. The effort saved in not needing to perform this activity then creates opportunity to invest that saving in innovation.

There is likely to be one proviso: the investor wants confidence, or at least as much as possible. If the supply of the content is not consistent in terms of availability, update and quality, they may well choose to perform that work themselves.

So, to some extent, the delivery of commercial value and economic benefit will rely on the confidence instilled in the consumer by the provider. Or put another way, data with a higher quality and certainty is more likely to instil greater confidence in a consumer. The consumer is then comfortable to rely on the supply from the provider and utilise the savings from not doing the work themselves in other innovations.

Invest not reinvent

It shouldn’t come as a surprise then that the governments have already considered this powerful economic argument and established a mechanism to refine location data (and, in particular, dynamic location data) to the point where the extraction of benefits occurs with considerably less effort for the consumers.

This mechanism is PSMA Australia. An organisation that has been quietly operating for more than 20 years, PSMA has established and nurtured relationships with data custodians across all jurisdictions to be able to collect, collate and provide transformed data, negating the need for the market to undertake a large amount of pre-processing that would be otherwise required.

While the broader economy may not have heard of PSMA, it has certainly benefited from the dramatic reduction in effort required to extract value from the government location data. PSMA’s work reduces the complexity of the data supplied to us, thereby increasing the number of organisations that can make use of it. This benefit is key to achieving the accessibility being sought by governments.

PSMA’s role in transforming raw custodian data into national assets requires significant investment in people, infrastructure, technology and innovation. However, the costs of doing this are shared broadly across all beneficiaries and the resulting benefits far outweigh the costs, both at the individual business level and the whole-of-economy level. Data quality is higher, confidence in continuity is strong, and considerable commercial value results.

‘Open data’ is a means, not an end

When we talk about ‘open data’, we are not talking about data that is free per se. In fact, we too easily get bogged down in debates around fees, when we should really be focused on the outcomes and the best way to sustainably deliver them. This is especially the case with economic activity through data-driven innovation.

I call on our industry peers to drive the open data debate to consider what things should be included in the policy, how they should work together to achieve significant economic benefit, and how to ensure that the conditions that deliver this economic benefit are sustainable.

Before we can talk sensibly about ‘who pays’, we need to work out what needs to be done to achieve these highly desirable economic benefits for Australia.

The elephant in the room: the perfect open data partner

But the conversation inevitably leads to the question ‘who pays?’ Or, more commonly, should PSMA be charging a fee for our asset of data?

Here’s the thing: geospatial data integration is an expensive business. The role of PSMA Australia in delivering national datasets, costs. And as PSMA Australia was established (quite purposefully) as a non-government funded, self-sustaining, full-cost recovery entity, those costs must be recovered from the market via commercial activity.

Under PSMA’s existing commercial model, it is the user – or the beneficiary of the data – that pays but that cost is far less than the alternative of performing all the work yourself.

What’s more, in many cases PSMA is the first step in the value adding chain that begins with open government data. Not only are the activities of PSMA Australia entirely compatible with open data initiatives, they are essential in providing the market with consistency, quality, and the confidence required to derive the maximum commercial and economic benefits from that open data.

Let the magic be in action not just words

As people who live and breathe location data, PSMA fully understands the government’s zeal to deliver on its open data policy.

But we should not allow instant gratification to come at the price of long-term benefits.

Simply releasing location data into the open without the investment and mechanics to ensure that data is useable and sustainable will not deliver the highest savings, efficiencies and innovation the government seeks.

As the mechanism established by the Australian governments to deliver a nationally consistent, geospatial information infrastructure for use within the economy, and as the means to surmount some intrinsic structural and demographic challenges, PSMA sees it as part of our mission to drive this open data debate beyond rhetoric and towards outcomes.

Let’s talk about what data is needed to build the digital infrastructure that will underpin our digital economy. Let’s work together to ensure a sustainable supply of that data. Let’s invest in the mechanics, the education and the support the market needs to access and use this data.

I say ‘let’s’ because this is not a job for a single stakeholder. Government, industry, end users and everyone else in between all have a role to play.

So let’s start working towards those outcomes.

Dan Paull is the CEO of PSMA Australia.