30 years of geospatial standards

ISO/TC 211 has been developing and maintaining a set of standards for the geospatial community for 30 years. Some standards are well known, others are used without people realizing it, and – because each standard changes over time – some do not yet fulfil their potential. This article reflects on the changes over the past three decades, including the benefits of harmonization, the maturing of the standards development process, and the evolution of the outreach activities. 

ISO/TC 211 is the International Organization for Standardization’s technical committee responsible for geographic information. The vision is to support a sustainably prosperous future by developing and maintaining, in cooperation with others, a set of standards that enable better management of geographic information. ISO/TC 211 doesn’t do this alone. Key collaboration partners include the Open Geospatial Consortium (OGC) which is also 30 this year, UN agencies such as the Food and Agriculture Organization and Universal Postal Union, and professional bodies such as the International Association of Geodesy (IAG), the Committee on Earth Observation Satellites (CEOS) and Group on Earth Observations (GEO).

For example, the OGC adopts several foundational TC 211 standards as Abstract Specifications providing the framework in which they then develop implementation standards and domain-specific standards. Some of these are then fast-tracked to be ISO standards, meaning a technically equivalent specification is available as a formal ISO standard where needed. This is the basis of the ISO/OGC Co-operative Agreement in operation since 1998. Well-known implementation standards include OGC Web Feature Service (ISO 19142) and the newer OGC API Features (ISO 19168).

Over the past few years, the United Nations Committee of Experts on Global Geospatial Information Management (UN-GGIM) has become an increasingly important collaboration partner, helping ISO/TC 211 understand the use of geographic information across a wider range of countries and focusing thinking on the UN’s Sustainable Development Goals (SDGs). All these inputs help to ensure that ISO/TC 211 standardizes demonstrable best practices rather than committee-driven (‘ivory tower’) thinking.

Harmonization

ISO/TC 211 standards are often model-driven, in that they are created around a Unified Modeling Language (UML) model of the data, using a subset of the Object Management Group’s UML standard. The committee’s approach to using UML is documented in two of its foundational standards: ISO 19103 Conceptual schema language, and ISO 19109 Rules for application schema.

Within ISO/TC 211, several groups are focused on ensuring that the set of standards work together, by having a harmonized UML model, consistent eXtensible Markup Language (XML) schemas (where relevant) and other supporting technical tools. XML is a mature way to structure self-documenting data; some would say ‘over mature’, but it remains core to a lot of data and information processing. Of course, each standard changes over time, so this is a continual round and at any time some standards are out of step.

The standards development process has matured over the past three decades, as the communities have learned to work even better together. For example, ISO 19115 initially ran for some four years, including five rounds of voting to resolve some 1,500 comments. There were then a further four years between ISO 19115:2003 and the first XML implementation standard for it, ISO 19139:2007. Nowadays, a standard is expected to be published within two years, with implementation in the next year. In the case of OGC fast-tracked standards, there are usually implementations ready as the standard goes to publication. ISO/TC 211 has played a leading role in trialling ISO’s Online Standards Development tool, which seems likely to further reduce the publication cycle by approximately six months.

Using the standards makes it easier for members of the geospatial community to ensure their data harmonizes with other data within their domain and across domains. For example, the European Commission’s INSPIRE project used the ISO/TC 211 UML standards to produce data specifications across a range of themes. This resulted in substantial consistency and eased integration of data from different publishers and about different kinds of geography such as road networks, settlements and land cover.

Outreach

Standardization is only useful if people know about it, so outreach is an important part of the committee’s work. In fact, the committee’s first advisory group, established in 2001, was on outreach. A key part of the outreach programme is the biennial Standards in Action workshop, which now has an archive of presentations spreading back across most of the past 30 years. Outreach gives the committee the opportunity not just to explain what standards exist and how they can be used, but also to listen to experiences of using them – both good and bad. For example, ISO standards are considered open because it is not necessary to pay patent licence fees to implement them or to use software which does.

Some of the success stories are when the standards themselves become invisible. For example, ISO 19125-2 Simple feature access – SQL option was published in 2004 and withdrawn in 2018 because it had become a core part of the SQL standard: ISO/IEC 13249-3:2016 Information technology - Database languages - SQL multimedia and application packages - Part 3: Spatial. ‘Spatial’ is now a normal part of relational database technology. Other invisible standards are essential to core parts of geographic information but remain invisible to almost all users. These include ISO 19161-1 International terrestrial reference system (ITRS). Previously standardized by the IERS in France, this standard (and ISO 19111) ensures that coordinates and coordinate reference systems can be used to consistently locate things.

Outreach tends to focus on the standards that communities can benefit from using, such as those which support communities in agreeing how to model and share their data. This includes foundational standards like ISO 19109 that can be used to model any object-based data, standards for coverage data, and domain-oriented standards like the ISO 19144 series for creating and managing classification systems for land cover and land use.

The open standards can help across most of the Findable, Accessible, Interoperable and Reusable (FAIR) journey. They make it easier to find data (metadata, catalogues), access the data (web services, APIs), use the data with other data (thanks to the interoperability supported by consistent reference systems, models and encodings) and reuse the data (clearer licence information in the metadata, and documented data-quality measures to indicate what the data is good for). Together, this means that ISO/TC 211’s most widely used standard is ISO 19115 Metadata. This has been adopted across Europe, the Americas and Australasia, by defence, meteorology, geology and the EC’s INSPIRE initiative. There are likely millions of records across thousands of servers using at least ten different software implementations. ISO 19157 Data quality follows closely, with the results of data quality assessment increasingly being available alongside the metadata. Very few of those who create or use those records will have read the standard; they rely on software implementations which have been demonstrated to comply with the standard.

ISO is on a journey from publishing paper and human-readable PDF files to machine-readable standards. ISO/TC 211 has been a trailblazing committee in the publication of machine-readable supporting artefacts. These include data models and tool configurations that can jumpstart the data modelling process, XML schema files that can help ensure that data instances conform to the standard, and a web concept dictionary of relevant terms (on Geolexica). Building on this, the committee plays a key part in ISO/IEC’s SMART project, which is working to show ISO how to provide standards that are machine readable, interpretable and testable.

Conclusion

The growing realization of the value of good – well-structured and quality-assured – data is driving greater use of open standards for data. Increasingly, data is needed that not just people can interpret, but also that machines can be trusted to interpret. This creates the need for standards for the structure and meaning of that data which in themselves can be understood by machines. Like people, artificial intelligence can only make a useful contribution if it has good access to reliable data and information. Therefore, the ISO/TC 211 team is looking forward to many more years of standards development in support of a sustainably prosperous future.

Join the community!

You can save design effort by basing implementations on open standards. Your experience of working with geographic information will be important to the future development of those standards – that’s a big part of what is meant by ‘open’ in this context.

Read more about ISO/TC 211 at https://committee.iso.org/home/tc211, and join the community – either through your professional body, your national standards body, or one of many other liaison communities listed on TC 211’s website.

Acknowledgements

This article contains public-sector information from Ordnance Survey 2024, released under an Open Government Licence.