High tech network data security center background.jpg
Blog: Making space for spatial data on the web

Frans Knibbe’s blog Making space for spatial data on the web

In an ideal world, spatial data is easily shared everywhere on the worldwide web. Unfortunately, this is still a long way off. There is currently no agreement on a single method for dealing with spatial data on the web, but that’s exactly what we need. In this blog, Geodan Research’s Frans Knibbe explores the potential of a new universal web structure for spatial data.

Recently, I had the privilege of participating in discussions on how the web could become a better place for data at the annual W3C TPAC conference. A noble goal, because sharing all kinds of data can give mankind the knowledge and wisdom that it so desperately needs. Good methods have already been found for sharing raw data, making it much easier to find, understand, process, and combine data. Recent examples of this include the Data on the Web Best Practices and a draft recommendation from the W3C. In addition, Google has provided a manual on how scientific datasets can be provided with a suitable context with terms from schema.org to increase findability. For spatial data in particular, the Spatial Data on the Web Working Group (SDWWG), in which Geodan also participates, is expanding the general Data on the Web Best Practices guidelines to Best Practices for Spatial Data on the Web.

Spatial data

Something special is happening with spatial data on the web. Those looking to select the best methodology have an immense number of practices to evaluate first. Just look at the many ways in which geographical data (spatial data with a relation to the earth's surface) can be shared on the web. And there’s more, because spatial data is not the exclusive domain of geographers. Aspects of spatiality can be found in virtually all areas of human activity. As such, many domain-specific ways have been developed to express spatial data. All in all, there are a huge number of standards and practices.Spatial data is everywhere. It would be a shame if all that useful data remained in its own dark corners of the web, without the possibility of combining it. Space is, just like time, a universal phenomenon and all these data points therefore have the potential to complement each other perfectly. If it were possible to align all this diverse data, it would result in a ‘dataverse’, which would excite even the most ardent of space explorers. Unfortunately, however, space is not yet a dimension in the data web that can be navigated well.

A structure for spatiality

What seems necessary is a universal model of spatiality. Something that is simple and powerful, and ready for the web. Something that everyone can quickly deploy without requiring too many adjustments to existing practices. One of the goals specified in the charter is developing a general spatial structure, which is why the SDWWG has started developing a standard method for working with spatial data on the web. The idea is that it will be a new version of GeoSPARQL's current ontology. The new structure will have to contain more elaborate definitions of the key concepts for spatial data, which could then serve as the foundation for, for example:

  •  data exchange formats
  • data types for storage media
  • definitions of spatial functions and spatial filters
  • general APIs (i.e., not application-specific or domain-specific)

We have to ensure that new developments don’t result in even more new, additional standards, and must come up with a structure that’s compatible with existing standards, instead. These must to be found in the OGC (or ISO/TC211) corpus and in the web domain.

The new universal standard is currently being developed. For a better look at the development process, you can take a look at this WebProtégé project. One of its most fascinating features is that only two core concepts (or classes) seem to be necessary for a multifunctional model for spatiality: a spatial entity and a geometry.

Spatial entity

A spatial entity can be defined as something that has a certain presence or size in space. That space can have one, two, or three dimensions and can be real or virtual. Examples of spatial entities are the planet Saturn, the spacecraft Voyager 2, the Eiffel Tower, a drawing on a sheet of paper, the earth's magnetic field, an amoeba, the lost island of Atlantis, and you. Ultimately, it’s the person who wants the data who decides whether it makes sense to see something as a spatial entity.

An important characteristic of spatial entities is that they can enter into mutual relationships. Different types of spatial relations can be distinguished, such as topographical relations, for example. In this type of relationship, spatial entity A can enclose spatial entity B or be located next to spatial entity C. Distance relations are yet another type of relation: a spatial entity can be located at 524 meters from another spatial entity, or, somewhat more vague, it may be far away from the next spatial entity. Finally, there are directional relations: spatial entity A can be to the north of spatial entity B, or a story above spatial entity C. One can easily imagine that the possibility of expressing spatial relations in a general way would be very beneficial for connecting data on the web.

A spatial entity can be identified by a code, such as a toponym or an address, but it can also be described by one or more geometries, which is the second core concept.



A geometry can be described as an ordered collection of n-dimensional points. These can be used to describe the shape or location of a spatial entity. A geometry can have the following known or unknown properties:

  •  a collection of coordinates
  •  a reference to a coordinate system
  • dimensionality (a geometry can be 1, 2, or 3-dimensional)
  • a type (such as a point, polygon, multipoint, multicurve, ...)
  • a detail level or spatial resolution (in case the geometry models the shape or location of a spatial entity)

Coordinates consist of numbers, which means that computers are very good at processing them. Coordinates can be used in many types of calculations and geometry is easy for a computer to display, for example on a map, or via Virtual or Augmented reality. Just like spatial entities, geometries also have spatial relationships. However, in geometries, these relationships can be calculated. With geometric data, the distance between geometry A and geometry B can be calculated, or one can calculate whether geometry A is or is not in geometry B.

Simple cooperation

It’s easy to see spatial data as a complex subject, and it’s true that there are many complex models for spatiality. However, there are ways of making spatial data on the web simple, and simplicity need not come at the price of restricting the possibility of cooperation. The spatial structure that is currently being developed will probably contain much more than just definitions of the spatial entity and geometry. As is the case with all web ontologies, though, you are free to only use the parts that you need. Stating that an element of a dataset is a spatial entity is a powerful statement in and of itself, especially in a world where automated data analyses are flourishing. Likewise, a universal online definition of the concept of geometry can provide numerous possibilities for effectively combining spatial data that is available in so many different formats.

Humanity is about to step outside the boundaries of its home world and engage in personal exploration of other planets. Would it be possible for us to allow spatial data to really flourish in our information space before that time?


Geodan has two branches in the Netherlands

One location in Amsterdam and one in 's-Hertogenbosch. The general mailing and visiting address is President Kennedylaan 1, 1079 MB Amsterdam.


GPS Lat/Lon:Geodan ‘s-Hertogenbosch
51.69174 5.299683
GPS Lat/Lon:Geodan Amsterdam
52.342346 4.91305

Geodan ‘s-Hertogenbosch
Buitenhaven 27-A
5211 TP ‘s-Hertogenbosch
 +31 (0)73 – 6925 151


Geodan Amsterdam
President Kennedylaan 1
1079 MB Amsterdam
 +31 (0)20 – 5711 311