The debate on whether (geo)spatial is special has been going on for some time. In 2012 James Fee reiterated “Spatial has never been special”. Justin Holman responded stating “Spatial is indeed special…but GIS software skills will soon be obsolete”. Let’s resume the debate and find out what’s special, and common about geospatial.
Many will agree that geospatial requires specialized knowledge and skills in areas that are specific to the domain. Here are 5 that stand out for me:
- Geodata – Geospatial data is geo-referenced to a location on, below or above the earth’s surface. Coordinate measurements are ambiguous and difficult to work with, since there is a multitude of coordinate systems and geodetic datums.
- Spatial autocorrelation – Waldo Tobler’s first law of geography states: “Everything is related to everything else, but near things are more related than distant things”. Hence, spatial observations are dependent within a range that needs to be inferred from the data.
- Spatial analysis – The locational component of geodata enables new types of analysis based on the spatial distribution of single or multiple data layers. It facilitates the discovery of spatial trends, patterns and relationships and solves problems through spatial modeling.
- Spatial statistics – A powerful collection of tools used to optimize data sampling, interpolate point data (e.g. Kriging), measure distributions (e.g. hot-spot analysis), analyze patterns (e.g. clustering), and model spatial relationships (e.g. geographically weighted regression).
- Cartography – Anyone can make a map, but a map that communicates to the target audience needs to adhere to cartographic design principles (e.g. legibility). Cartographic map production also relies on the use of specialized cartographic tools (e.g. generalization).
What is special may be abracadabra to you, but geospatial has a lot in common with other IT solutions. To tie the scores, let me mention 5 of them:
- Data dimensions – Whether the emphasis is on quality or quantity, good data needs to be fit for purpose. Aspects such as accuracy, validity, completeness, consistency, timeliness and integrity are important for all data types.
- Value chain – Value will be added incrementally when data is analyzed, visualized and interpreted. Any good IT solution should provide actionable information and/or recommend a set of business actions.
- Visualizations – A picture says more than a 1000 words, and we live in a world that is increasingly attracted to images. A map can be just as powerful as a graphic, chart, or infographic and complements the visuals that you are already using.
- Technology trends –The general IT industry and the geospatial industry are facing the same trends of Cloud computing, web and mobile Apps, Agile development, and Big Data. These trends will cause the emergence of new and the disappearance of old IT niche disciplines.
- Scientifically sound – Any BI (Business Intelligence) solution regardless whether it uses geospatial, needs to be data- and process-driven. It should allow for cross-sectional and longitudinal data studies and support inductive and deductive research methods.
Geospatial does indeed require specialized knowledge and skills, but has plenty in common with other IT solutions. IT and geospatial already support and complement one another and technology trends will likely result in further integration. Going forward, the geospatial industry should retain and develop advanced competencies, while IT and other disciplines absorb the basic competencies. Attempts to turn geospatial into a holy grail or members-only club ought to be rejected if the industry wants to remain relevant. Join the debate and let us know what you think.