Viewing and reuse
By the early 2000s New Zealanders could already see a surprising amount of geographic information on the web. had a free Census WebMap by March 2002. Councils were putting property and planning information into browser viewers. Government agencies were learning that a useful map no longer had to sit on a specialist workstation. Yet being able to look at a map was not the same as being able to take the data away and use it in another GIS. Open-data policy later addressed permission to obtain and reuse the underlying datasets.
March 2002 product material records the available products and their access conditions. A public user could view census geography through a free WebMap, but the reusable GIS products underneath it still carried prices. The complete Meshblock Database was listed at NZ$1,200 plus GST, with multi-organisation use at NZ$5,000 plus GST. The browser had opened a window onto the information, but the underlying data were still being treated as a product supplied under particular commercial and licensing terms.
For a GIS practitioner, that difference was practical rather than philosophical. A web map could help answer a question on screen. It could show where a meshblock was, what a council parcel looked like or how a statistical pattern was distributed. But if the job required combining the same geography with local data, running analysis, producing a new map series or building an application, the organisation needed the actual data and permission to use them. That usually meant finding the supplier, understanding the product, paying the required charge and working within the licence.
Access to New Zealand spatial data changed through successive releases and licensing decisions. Some prices collapsed before licences became standardised. Some data could be viewed freely before they could be downloaded. Some datasets were distributed cheaply while reuse remained controlled. By the early 2010s, important national datasets were increasingly shared across organisations.
The price of geography
The shift had started before the phrase open data became common. On 21 September 1999 the Government announced a major change to the pricing of LINZ topographic information. Copyright reproduction fees for LINZ-held digital data, topographic maps and aerial photographs were to be abolished, and charges for digital data were to reflect dissemination rather than the much heavier cost-recovery regime that had gone before.
The numbers were striking. The Minister said a large map producer could previously have paid about NZ$100,000 a year in copyright fees. More dramatically, the cost to data distributors of buying the entire 1:50,000 digital topographic database was expected to fall from almost NZ$2 million to around NZ$1,500. Printed maps would still be sold, but the economics of reusing the digital national topographic database were being radically altered. That is the sort of price change capable of turning a dataset from something discussed in a budget meeting into something a developer might simply decide to use.
The 1999 decision was one stage in changing access to government data. The terminology, licensing framework and online services of the next decade were not yet in place. Pricing and licensing could restrict access to publicly funded information and its reuse. Even when an organisation had GIS software and capable staff, the price of foundational data could determine whether a project was practical.
The earlier model also shaped everyday project planning. A GIS team could know exactly which national dataset it needed and still have to treat acquisition as a budget item in its own right. A project manager might have to decide whether a full national database was essential, whether a smaller geographic extract would do, or whether an older or less suitable source could be substituted. Data purchasing therefore influenced project scope in ways that are easy to forget once foundational layers become routine downloads.
The dramatic reduction also separated two costs that had often been bundled together. Producing and maintaining a national topographic database remained expensive public work. Copying an existing digital dataset for another user was becoming comparatively cheap. Once distribution costs fell, charging each downstream user as though the whole national database had to be created again became harder to justify. That economic distinction sat underneath much of the policy change that followed.
This had consequences for who could participate. Large central agencies and established commercial firms might be able to absorb data costs that were prohibitive for a small consultancy, a university researcher, a community organisation or a local project. High prices could also encourage organisations to maintain their own partial copies, work from paper products or avoid using a dataset altogether. Lower prices widened the set of projects that could realistically be attempted.
Census access changes
confronted the same issue from a different direction in 2002. On 9 May the Government announced that detailed results from the 2001 Census would be made available free of charge through the website from late May. The announcement was unusually explicit about the old cost barrier. Access through Supermap had cost a minimum of NZ$3,300 and could rise to about NZ$25,000 depending on the information required.
Those figures explain why the effect reached well beyond a nicer website. Census data were basic inputs to planning, health research, local-government analysis, education, business and social policy. A public-health analyst calculating disease rates, a council planning services or a researcher studying population change needed demographic information tied to geographic areas. GIS could perform sophisticated spatial analysis only when the population data required to interpret the results were available as well.
The 2002 announcement identified cost as a major barrier for communities and other users with limited budgets. Census Table Builder was intended to let users construct tables from hundreds of census databases and compare 2001, 1996 and 1991 census information where available. Its scope was census tables; boundary data and datasets retained their separate access and licensing arrangements. The policy still marked a clear commitment to wider public use of statistical information.
The intended users were revealing. The Minister referred to businesses, universities, local authorities, community boards, health boards, schools and community organisations. These were precisely the kinds of users for whom a several-thousand-dollar subscription could distort whether spatial evidence was used at all. The policy therefore widened the practical audience for demographic analysis without requiring every user to become a specialist customer of .
This also changed expectations inside government. If one agency could make a nationally important information resource available without transaction charges, users could reasonably ask why other foundational datasets were still sold or supplied only through bespoke arrangements. Agencies had different statutory functions, business models and privacy constraints. The ease of obtaining public information became a policy question.
Free access did not always grant reuse rights. A user could obtain information without paying a subscription and still encounter conditions around copyright, redistribution, confidentiality or specialised products. Statistical information also carried privacy constraints that could never be solved merely by making files downloadable. Census geography had always involved aggregation partly because individual responses carried confidentiality constraints. Open access widened legitimate use while leaving the obligation to protect people intact.
For GIS practitioners the practical effect was cumulative. If population data became cheaper or free, spatial analysis became easier to justify. If statistical boundaries were also available, demographic information could be joined to maps without constructing the geography independently. Census information did not suddenly become simple. One of the recurring transaction costs around using national population data nevertheless began to recede.
Permission to derive
The next stage concerned what a user was legally allowed to do with government geography once they had it. OpenStreetMap provides a useful New Zealand example because its model allowed contributors to incorporate, modify and redistribute geography inside another database.
On 18 March 2008, LINZ gave OpenStreetMap permission to use its information if the required Crown attribution and disclaimer accompanied the data. The permission allowed LINZ information to be included, modified and derived in OSM with suitable attribution.
That was a different kind of access from a free web viewer. OpenStreetMap contributors were incorporating and maintaining geographic data inside another database. They wanted to bring authoritative geographic information into a separately maintained public database, edit it, combine it with community contributions and redistribute the result. Doing that safely required more than technical access. It required permission that fitted the downstream use.
Users still needed to understand open-licence terms. When OpenStreetMap moved to the Open Database Licence in 2011, LINZ gave further consent on 23 August for relevant CC BY data from the LINZ Data Service to be relicensed under OSM’s terms. Reuse therefore required additional permission as well as a download. A dataset could be easy to obtain while downstream rights still needed attention.
The OpenStreetMap case also illustrates the practical terms of data reuse. LINZ had to authorise reuse separately from making the data visible or downloadable.
National topography becomes downloadable
On 23 September 2009 LINZ launched the Topo50 map series. The new series replaced the older NZMS260 system, moved national 1:50,000 mapping onto NZGD2000 and NZTM2000 and aligned official topographic mapping more closely with modern GPS use. The series also changed how users obtained and reused national topographic information.
At the launch, the Minister for Land Information stated that technology now allowed LINZ to deliver topographic data free of charge to third parties and members of the public in electronic form. The same national mapping organisation that had once charged very large sums for digital topographic databases was now presenting free electronic distribution as part of the normal launch of a new national map series.
Digital topographic data already existed: New Zealand had been building and maintaining national mapping databases for decades. The significant shift lay in the relationship between custodian and user. Electronic data could now circulate widely as reusable infrastructure rather than remain a scarce specialist product whose price recovered a large share of production cost.
The timing coincided with changing public expectations. By 2009 ordinary users were already familiar with Google Maps and Google Earth. Government and council browser mapping had been established for years. The expectation that maps should be easy to find online was no longer unusual. Making the underlying national topographic data freely available extended that expectation from viewing into production and analysis.
Free digital Topo50 also created a cleaner connection between official mapping and downstream innovation. Tramping applications, emergency systems, planning tools, research projects and commercial products could use the same national topographic foundation without beginning with the older transaction costs. The paper map remained useful and continued to be sold. Open digital access added another mode rather than erasing the old one.
The release also illustrates a broader change in what a map series was becoming. Topo50 was still a recognisable national sheet series, but the same underlying information could circulate as digital images and data outside the printed-sheet workflow. A user could treat the official map as a product, the underlying topographic database as a dataset, or both. That separation between cartographic product and reusable geographic content became increasingly important in the years that followed.
Land-resource data opens
The same pattern was appearing in environmental and scientific information. On 24 August 2010 Landcare Research launched the Land Resource Information System Portal. The LRIS Portal provided online access to nationally significant environmental and land-resource datasets used for mapping, analysis and modelling. Its holdings included material derived from the New Zealand Land Resource Inventory, Land Environments of New Zealand, soil databases and digital elevation information.
Many of these datasets had long histories inside specialist science and government work, so their value predated 2010. The distribution model changed. A user could search for a dataset, inspect metadata and download GIS data through an online service rather than treating access as an exchange between specialists who already knew which group held the information.
Manaaki Whenua’s later institutional account notes that the first LRIS Portal release was developed with and brought to market rapidly in 2010. Making data accessible depended on software platforms capable of cataloguing, previewing, converting and delivering geospatial data in forms people could use.
The portal model had another advantage. Data could be accompanied by metadata, documentation and licence information in the same environment. Open access without context can simply move confusion from one desk to another. A soil layer or land-resource dataset still needed explanations of its scale, attributes, methods and limitations. Making the file easier to obtain did not remove the need to understand what it represented.
That combination of file and explanation was especially important for older scientific datasets. A layer created from field surveys carried assumptions about scale, classification and date that were not obvious from its geometry alone. Moving such data into an online portal made it easier to find, but also made it easier for users outside the original research community to apply it in ways its creators had not anticipated. Metadata therefore became part of responsible openness and practical reuse.
A licence for reuse
By 2010 New Zealand also had a more general framework for answering the legal question. The State Services Commission issued the first version of the New Zealand Government Open Access and Licensing framework, NZGOAL, in August 2010. It encouraged State Services agencies to make copyright works available for reuse on the most open licensing terms available where restrictions did not apply.
For copyright works, NZGOAL recommended the Creative Commons Attribution licence as the default. Users could apply a common set of licence terms across participating agencies and datasets. A standard licence could state that material could be copied, adapted and redistributed, including commercially, provided attribution requirements were met.
NZGOAL provided guidance for releasing suitable government information. The framework explicitly recognised restrictions. Agencies still needed to consider copyright ownership, privacy, confidentiality, security, statutory obligations, third-party rights and other limits. Open licensing worked only where an agency had the authority to license the information and where release itself was appropriate.
Licensing and distribution were separate changes. A portal solved practical questions about discovery and delivery. A licence solved a legal question about reuse. A useful open-data environment needed both. A perfectly designed download service could still leave users hesitant if the rights were vague, while an excellent licence was of limited practical value if the data could only be obtained through a slow manual process.
The LINZ Data Service
The LINZ Data Service combined these changes in a public data service. An official open-government case study dates its launch to 1 July 2011. The service delivered property and ownership, topographic, hydrographic, geodetic and electoral address data in commonly used formats, with much of the information licensed under Creative Commons Attribution 3.0 New Zealand.
joined LINZ as Deputy Chief Executive Location Information in 2009. In 2016 she and Chief Geodesist presented LINZ’s open-data journey, covering the LINZ Data Service and the release of aerial imagery. The presentation named the institutional leaders and described the service’s licensing, data and technical work.
The service changed the ordinary workflow around national spatial information. Instead of treating every request as a special extraction or physical delivery, users could search, inspect and download datasets online. Web services also allowed some data to be connected directly to other systems. Later APIs extended this access through direct application connections. Authoritative spatial information was becoming something other systems could build on.
By September 2012 that idea had moved beyond one-off downloading. A GISList discussion recorded that LDS now supported incremental change-sets, allowing users to retrieve only records that had changed rather than repeatedly downloading an entire dataset such as New Zealand Primary Parcels. LINZ was also preparing an open-source replication tool for MSSQL 2008, ArcSDE and FileGDB, with reprojection and attribute or spatial filtering built into the workflow. at the Ministry for the Environment immediately asked whether the same approach could eventually be applied to raster tiles. Open access was beginning to include synchronised operational copies as well as downloadable files.
This was a substantial administrative change as well as a technical one. Under older distribution models, staff might receive a request, determine the appropriate extract, prepare files and arrange delivery. The customer then had to wait for the transaction to be completed. A self-service portal moved much of that work to the user while also allowing the custodian to publish one maintained version for many different audiences. The same infrastructure could serve a surveyor, a council analyst, a software developer and a community project without creating a separate supply process for each.
Choice of formats also mattered. Open access was far more useful when a user could obtain information in forms that fitted normal GIS and database workflows rather than a single proprietary delivery format. Standards and interoperability had their own history, but the open-data services benefited directly from that work. Access became practical because legal permission, discoverability and technical usability were beginning to line up.
A 2013 Government announcement reported early uptake. By 19 June that year the service was offering more than forty datasets for direct integration with business systems and had more than 5,600 registered users. Those users included people in geospatial, surveying, utilities, information technology, contracting and engineering as well as community organisations and central and local government.
Thousands of registered users and dozens of fundamental datasets indicate that the model had moved beyond an experiment. Users were downloading, combining and maintaining the data in working systems.
The service also exposed the continuing limits of openness. Not every layer could be treated identically. Property ownership information, for example, carried legal and privacy considerations that required different treatment from an ordinary topographic layer. The general move towards reuse did not remove the obligation to protect people or obey statutory controls.
Commercial value moves
Open government data changed where commercial suppliers added value. A raw cadastral parcel layer is not the same thing as a cleaned property-intelligence service. An address point is not the same thing as a maintained customer database. A census table is not the same thing as a demographic market model.
Companies could combine open government data with proprietary information, clean errors, standardise fields, create APIs, host services, build applications and provide analysis. Those activities could become more valuable when the underlying national framework was easier to obtain because less effort was wasted recreating basic geography.
This also changed the position of government agencies. If a fundamental dataset was already being collected as part of a statutory or public function, charging a high price for downstream digital reuse could suppress activity without creating better data. Open distribution did not remove the cost of producing or maintaining the authoritative source. It changed how that cost was recovered and how broadly the resulting information could circulate.
The boundary moved. Some information that had once been sold as a specialist input became public infrastructure, while commercial value shifted towards convenience, interpretation, linking, currency, specialised support and applications. Government and commercial spatial information continued to coexist, but they occupied different parts of the value chain.
Maintaining open data
A free download could still be expensive to use. Large datasets required storage and bandwidth. Different coordinate systems had to be understood. Field names and schemas needed interpretation. Updates had to be managed. An organisation might need to transform data before it fitted an operational database, and someone still had to decide whether the information was accurate enough for the job.
Open data therefore rewarded capability. A skilled GIS team could take a free national dataset and make it useful quickly. An organisation without spatial expertise could download the same file and discover that the hard part had only moved. The cost had shifted from purchasing access towards understanding, maintaining and integrating the information.
Open-source software could reduce some software costs and expand technical choices, while open government data could reduce some information costs and simplify reuse. Neither removed the need for professional judgement, infrastructure or staff time. The barriers became lower, but they did not vanish.
Controlled information
The open-data period also made limits more visible. Some information should not be released simply because publication is technically easy. Personal information, protected locations, security-sensitive infrastructure, commercially confidential material and data affected by statutory restrictions require different treatment.
Spatial information can be particularly revealing because location allows datasets to be combined. A table that appears harmless on its own can become sensitive when joined to property, demographic or infrastructure information. The same GIS capability that makes open data useful can also make re-identification or unintended disclosure easier.
Māori data governance raises a further issue that cannot be reduced to privacy. Data relating to whenua, wāhi tapu, taonga, whakapapa or other culturally significant knowledge may involve questions of authority, tikanga and collective interests that are not resolved by Crown copyright or a Creative Commons licence. Technical possession of coordinates does not settle who should decide how knowledge is represented, shared or reused.
Open data required continued investment in maintenance, documentation and support. Custodians needed to establish access conditions, licences and stewardship obligations for open data. The early open-data programmes were strongest where access, licensing, metadata and stewardship were treated together rather than as separate afterthoughts.
This also placed more responsibility on users. A Creative Commons licence might grant broad legal permission, but it did not certify that a dataset was complete, current or appropriate for a particular decision. Open data made reuse possible; it did not transfer professional judgement from the user back to the custodian. Every layer still had a lineage, a scale, an update cycle and a purpose, even when the download button made acquisition feel effortless.
formed in 2015, with and among the founding contributors identified in the research. Its charter brought Māori authority, governance and benefit into discussion of data access. These questions applied to geographic information about land, environments and communities as well as other forms of data.
Teaching people to use the open-data environment
Open data was useful only when people could find, understand and reuse it. LINZ’s capability programmes addressed that practical layer directly. GeoBites sessions covered authoritative national datasets, web services, property and address data, marine information, LiDAR and data-management practices. Ngā Poutama Matawhenua used the same open-data environment for Māori land, freshwater, climate resilience, historical imagery, coastal applications and other kaupapa Māori mapping. This material should be cross-linked to the detailed LINZ Data Service and open-data history rather than repeating that platform history here: the capability programme shows how open spatial data was taught and applied.
Sources · 1
Chapter source notes
1. Statistics New Zealand product records around 2001–02 establish the distinction between free browser mapping and paid reusable statistical/geographic products before the major access reforms. is the principal this baseline.
2. The State Services Commission's New Zealand Government Open Access and Licensing framework, NZGOAL version 1, August 2010, is the authoritative policy the open-licensing framework. It recommends Creative Commons Attribution as the default open licence for reusable State Services copyright works where restrictions do not apply.
3. LINZ Data Service chronology is supported by the 19 June 2013 Beehive announcement recorded as P9C-S15. It states that LDS launched in 2011 and by 2013 offered more than forty free/open fundamental datasets with more than 5,600 registered users. Those are period government figures and remain attributed as such.
4. The December 2014 data.govt.nz LDS showcase provides a dated near-period interface and open-data example. Its screenshot must not be used as though it showed the exact 2011 launch interface.
5. LRIS and related research-data services support the broader shift towards reusable environmental and land-resource information. Chapter 33 discusses the access/licensing history; Chapter 35 discusses the LRIS application design, and Chapter 41 discusses later service/API architecture.
6. The chapter keeps four ideas separate: data free of charge, data legally reusable, data downloadable in usable formats and data consumable through maintained services. A dataset can satisfy one of these conditions without satisfying all four.