When speaking or writing about GIS all too often the concept in peoples' minds for Geographic Information Systems can be rephrased as "the software for map data". While software is integral to GIS, thinking of spatial software as the GIS is a narrow view, unnecessarily limiting our conversation and understanding.

I have some ideas about what the "system" in geographic information systems really means, which I'll throw into the punch bowl below, but let us hear yours first.

Secondly, is it too late to bring into public consciousness an expanded truer definition which will arise in answer to the GIS moniker, or do we need to find a new term to encompass it? if the latter, what do you suggest?

  • +1 The System is such an important part, it's often in there twice "GIS System". Still, defining boundaries to any information system runs the risk of cultivating an information silo. en.wikipedia.org/wiki/Information_silo Oct 26 '10 at 17:08

To use old "Introduction to ArcGIS 1" suggestions. GIS is built on 5 separate parts which work together in a seamless and cohesive way to allow the flow of data into information which can be used for decision making. These 5 parts are

  • People
  • Data
  • Hardware
  • Procedures
  • Software

The idea I have is that this term does not need to be rolled out to everyone other than to explain how most of the Web Map Business Models we are seeing like Google and Bing are GES. Geographic Exploration Systems (a phrase I cannot remember the origin of but I think is apt). A GES differs in removing the People, Procedures and in some cases Hardware from the equation and giving a "one size fits all" solution to everyone.

These are just my ideas, and sorry for being so wordy


  • Interesting that that Sofware is at the bottom of the list. ;-) I'd order it: Data, People, Procedures, Software, Hardware. Actually, this is begging for a diagram because it isn't a linear or hierarchal relationshop. (gotta love an age when 3 paragraphs is considered "wordy"!) Sep 30 '10 at 16:46
  • I ordered it that way as the software can only ever work as effectively as the hardware it is on, so which piece of software you use is dependent on the Skills of the people, the Data you have available including purchasing resources, Hardware you have or can afford, and the task you wish to perform within the work procedures and requirements you have. Just my opinion
    – CDBrown
    Oct 1 '10 at 2:17
  • 1
    I think what you term 'GES' could be better called Geographic Visualization Systems (GVS). Currently virtual globes and web maps have real value in visualizing data and have little/no functionality in analyzing data. I don't see GVS as a challenge to GIS (as you seem to?) because they are good at doing different jobs, GIS remains the way most spatial analysis needs to be done.
    – Trevesy
    Oct 2 '10 at 8:50

I usually stick to just four parts, or essential elements, to a GIS:

•Hardware •Software •Data •People

The •Procedures in my mind, are just the ways that people interact with the system.

I've also seen it whittled down to just three (minus the people part) but I think that's going too far.

As for the GIS moniker, I call myself a geospatial analyst, since GIS as a profession is dead. GIS as a tool, however, will be with us for a long long time. :-)

  • To my mind procedures cover the interactions between all the parts data/software, hardware/software, etc. Leave that out and we don't have a system. Sep 30 '10 at 20:32

The People/Data/Hardware/Procedures/Software headers for what makes up the system is a good answer. A 6th header that needs splitting out from the 'People' heading IMHO is 'Users'. Producing outputs that are usable and/or cartographically sound is often the neglected part of a GIS project because solving all the other challenges is hard enough. The client often doesn't appreciate the importance of usability, so user testing doesn't take place and mistakes happen. Case in point, the EA flood map which is linked from this page


It's there to allow people to see if they live in a flood prone zone. In the key you will see light blue = 'Extreme Flood Extent'. User interviews have shown that users interpret 'extreme' to mean 'extremely big flood' whereas the EA actually mean those flood events are 'extremely rare' (1 in 1000 years probability).


I said I'd throw in my own contribution: In short, what is often referred to as "the GIS" is actually just the software, while the true value comes from the people, data and workflows (analysis). Software is significant, the system doesn't work without it, but in essence is just fluff. To bring out this last point: what are you still using today that you were 10 years ago? Data? check. People? check. Arcview v1? nope (released 1995). ArcView v3? a little (released 1997). ArcInfo Workstation? Mapinfo? Manifold? some. Arcmap? yes, ...for now.

The software list is ESRI centric, reflecting my experience, but the same trend can be observed elsewhere, in all software, across all domains. The open source GIS ecosystem didn't even exist 10 years ago (though it should be noted GRASS has been around since 1982 and is still under active development). In short, focussing on the software du jour is to miss the point, very nearly entirely.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.